LINGUIST List 13.891

Mon Apr 1 2002

Diss: Semantics: Garrett "Evidentiality and ..."

Editor for this issue: Karolina Owczarzak <>


  1. eg3p, Semantics: Garrett "Evidentiality and Assertion in Tibetan"

Message 1: Semantics: Garrett "Evidentiality and Assertion in Tibetan"

Date: Sun, 31 Mar 2002 14:21:35 +0000
From: eg3p <>
Subject: Semantics: Garrett "Evidentiality and Assertion in Tibetan"

New Dissertation Abstract

Institution: University of California at Los Angeles
Program: Department of Linguistics
Dissertation Status: Completed
Degree Date: 2001

Author: Edward John Garrett 

Dissertation Title: 
Evidentiality and Assertion in Tibetan

Dissertation URL:

Linguistic Field: 
Semantics, Pragmatics, Philosophy of Language

Subject Language: Tibetan

Dissertation Director 1: Carson Schutze
Dissertation Director 2: Dominique Sportiche

Dissertation Abstract: 

This dissertation examines the three major evidential categories in
Standard Tibetan: ego, direct, and indirect. Indirect is argued to be
a performative epistemic modal; its performativity accounts for its
highly restricted distribution. Direct is argued to be complex,
consisting of a demonstrative component, which requires that the
marked situation be stage-level, and a pragmatic component, which
requires that the marked situation have been observed. Evidence from
conditionals and from the Amdo variety of Tibetan show that these two
components can be disentangled from each other. Ego is argued to be a
morphologically zero, default, "elsewhere" case, which indicates
either immediate or groundless knowledge.

Aside from its contribution to Tibeto-Burman linguistics, this
dissertation touches on various theoretically important issues in
pragmatics and the philosophy of language. First, in discussing ego,
it is argued that a property-based semantic view of attitudes de se
has no advantage over a proposition-based pragmatic theory. Special
uses of ego with names de se and performatives highlight this

Second, it is suggested that questions be analyzed as in the
traditional answer set approach to the semantics of questions; but
rather than taking a question to denote a set of propositions, it is
argued that a question should denote a set of assertions

Third, a new division among conditionals is proposed. Based primarily
on the behavior of Tibetan evidential constructions in conditional
protases, but also on the behavior of English "will" in the same
position, a category of "interactional" conditionals is introduced.
Interactional conditionals differ from "hypothetical" conditionals in
that the speaker does not simply represent the protasis as unknown,
but as something which can and should be immediately verified or
countered by another discourse participant.

Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue