Conversational Contrast and Conventional Parallel: Topic Implicatures and Additive Presuppositions

Kjell Johan Sæbø
Department of Germanic Studies
University of Oslo

In various forms, topic foci have been ascribed a contrastive
implicature (Bolinger 1961, Rooth 1992, Büring 1998, Hetland
2002). When a topic focus does generate a contrastive implicature, it
can be called a contrastive topic. Often, such an implicature is
(partially) confirmed in the - preceding or succeeding - context. This
is a locus for contrastive particles (Saeboe 2003, Umbach
2003). Sometimes, however, it is contradicted (cancelled) in the -
preceding or succeeding - context. This is a locus for additive
particles, like "too".

In my talk, I will focus on how additive particles are necessary in
situations where a contrastive topic implicature is or would otherwise
be contradicted in the context, an issue discussed by Krifka
(1999). According to Krifka, the verifying context for an additive
particle generates a contrastive implicature which is contradicted by
its succeeding context, the additive particle sentence, but the
additive particle alleviates this violation of "Distinctiveness" by
adding an explicit affirmative element.

I argue that this account is inadequate in two respects: First, if the
"too" sentence contradicts a contrastive implicature, this is a
cancellation, and a cancellation is not supposed to cause a serious
problem. If an affirmative element is called for, it can be one of
several, including "actually", "indeed", "in fact". So the strength of
the necessity of the additive particle is not accounted for. Second
and more seriously: When a particle is necessary, the verifying
context far from always generates a contrastive implicature, because
often, the presupposed alternative is not a topic focus at all.

My alternative proposal is that it is the additive particle sentence
which generates a contrastive implicature if the additive particle is
absent, an implicature which is contradicted in the preceding context,
the verifying context for the additive particle. This contradiction is
not a cancellation. This explains why the additive particle is
necessary. The answer to the question why the additive particle is
sufficient to render the discourse coherent is this: The additive
particle does not just add a presupposition; it also adds the
presupposed alternative to the topic of the assertion. In this way,
although this topic accumulation is informationally redundant, it
causes the contrastive implicature generated by the assertion to not
concern the contextually given alternative. Since the unification of
the presupposed alternative with a contextually given alternative is
effected in the verification of the presupposition, this account
argues for an analysis of additive particles in terms of anaphoric
presuppositions, and against Zeevat (2003), who takes the necessity of
additive particles to indicate that they are not presupposition
triggers but something else, "context markers".


If you would like to meet with the speaker, please contact:

Elena Karagjosova

This seminar series is jointly organized by the Department of
Computational Linguistics and Phonetics and the European Post-Graduate
College in Language Technology and Cognitive Systems.

A current version of the program for this term can be found at:

http://www.coli.uni-sb.de/colloquium/