Topic situations and domain restriction
University of Massachusetts Amherst
Topic Situations are discussed in several guises in the linguistics literature. Austinian Topic Situations (Austin, 1950) are familiar in situation semantics, where sentences are true of partial worlds, not entire worlds, and people hold attitudes toward partial worlds (Barwise and Perry, 1983, Kratzer, 1989, 2017). McKenzie (2015) discusses non-canonical switch reference in Kiowa (citing Watkins, 1993) using the example in (1), where the ‘same subject’ marker gàu is used to indicate that the conjoined clauses refer to the same writing situation.
(1) Kathryn gà=gύt gàu Esther=àl gà=gύt
“Kathryn wrote a letter and Esther wrote one too.”
The same sentence with the ‘different subject’ morpheme nàu indicates that the conjoined clauses refer to different situations.
Frazier and Clifton (2018) proposed the topic situation hypothesis in (2) (see also Bestgen and Piérard, 2014, Schwarz, 2019).
(2) Topic situation hypothesis: Initial temporal and locative PPs in English introduce topic situations. By default, following material is included in the topic situation until a new topic situation, or incompatible information, is encountered. Other things equal, the topic situation supplies the preferred context for implicit domain restriction.
The results of several interpretation and naturalness judgment experiments support the predictions of the topic situation hypothesis.
We will then look more closely at domain restriction in an attempt to understand the linguistic and processing mechanisms involved in supplying unstated context. Studies of time dependent nouns (In 1960, the senator …) confirm that initial temporal phrases, which by hypothesis introduce a topic situation, are more likely to restrict the interpretation of the noun than do sentence-final temporal phrases, which by hypothesis do not introduce a topic situation. However, noun class does not seem to be particularly important: clearly relational nouns are not particularly likely to be restricted by the topic situation, as might have been expected if domain restriction were a bottom-up process triggered by the nominal. The results perhaps fit best with Schwarz’s account where definite DPs have a situation variable which may be interpreted using the topic situation, may be bound by a higher operator, or may remain unbound. We will conclude by looking at whether indefinites, and generic sentences, pattern with definites.
Barwise, J., & Perry, J. (1983). Situations and Attitudes. Cambridge, MA: MIT Press
Bestgen, Y. & Piérard, S. (2014). Sentence-initial adverbials and text comprehension. In L. Sarda, S. Carter Thomas, B. Fagard & M. Charolles (eds.) Adverbials in Use: From Predicative to Discourse Functions (pp. 151-168). Louvain-la-Neuve: Presses universitaires de Louvain.
Frazier, L., & Clifton Jr., C. (2018) Topic situations: Coherence by inclusion. Journal of Memory and Language, 103, 176-190.
Kratzer, A. (2017). Situations in natural language semantics. In E. N. Zalta (Ed.), Stanford Encyclopedia of Philosophy (Winter 2017 Edition). Stanford, California.
McKenzie, A. (2015). A survey of switch-reference in North America. International Journal of American Linguistics, 81, 409–448. DOI 10.1086/681580
Schwarz, F. (2019). Definites, domain restriction,and discourse structure in online processing. In K. Carlson, C. Clifton Jr., and J. Fodor (Eds), Heidleberg and New York: Springer.