intonation and syntax

[gil]

Dear all,

Does anybody know of any languages in which grammatical relations such
as subject and object (or thematic roles such as agent and patient) are
systematically distinguished by intonation?

More specifically, I would be interested in examples of either of the
following two cases:

(a) [the strong case] languages in which the *only* formal device
distingushing grammatical relations (or thematic roles) is intonation.

For example, in such a language, a string of words such as

(1) JOHN HIT BILL

might be able to take two different intonation contours, C1 and C2, such
that (1) with C1 means "John hit Bill" (but not "Bill hit John"), while
(1) with C2 means "Bill hit John" (but not "John hit Bill").

(b) [the weak case] languages in which grammatical relations (or
thematic roles) are distinguished by intonation as well as by some other
formal device (such as word order or case marking).

For example, in such a language, strings of words such as

(2) JOHN-nom HIT BILL-acc
(3) JOHN-acc HIT BILL-nom

might be systematically associated with distinct intonation contours C1
and C2, such that (2) could only occur with C1 (meaning "John hit
Bill"), while (3) could only occur with C2 (meaning "Bill hit John").

[The reason for this query is as follows:  Elsewhere I have argued that
in at least one language, Riau Indonesian, there are no formal devices
distinguishing grammatical relations (or thematic roles).  In this
language, then, a string such as (1) could mean either "John hit Bill"
or "Bill hit John".  However, linguists frequently react to the above
claim with the comment "What about intonation?".  Whereas some linguists
simply pose the question whether the different meanings may be
distinguished by intonation, others go further and insist that there
must be such a difference, ie. that Riau Indonesian should be an example
of (a) above  -- though nobody has yet suggested exactly what the
different intonation contours might be.  In fact, so far I have not been
able to discover any systematic intonational differences between the
different meanings, in Riau Indonesian.  However, I am puzzled by the
frequency in which the issue of intonation is raised, all the more so in
view of the fact that I am not familiar with a language in which
intonation works this way, which suggests that such languages are at
best relatively uncommon.  Thus, the purpose of this query is to get a
feel for whether there is any cross-linguistic typological motivation
for the suggestion that intonation plays a role in distinguishing
grammatical relations (or thematic roles), or whether such suggestions
are -- as I suspect -- typologically unfounded.]

Thanks,

David

--
David Gil

Department of Linguistics
Max Planck Institute for Evolutionary Anthropology
Inselstrasse 22, D-04103 Leipzig, Germany

Telephone: 49-341-9952321
Fax: 49-341-9952119
Email: gil at eva.mpg.de
Webpage:  http://monolith.eva.mpg.de/~gil/