Grammar constrains acts of predication

ABSTRACT

Båve has argued that act-type theories of propositions entail unwanted ambiguity of sentences such as ‘Donald loves Joan’. King has argued that act-type theories of propositions entail an unwanted abundance of propositions. I reply that a version of the act-type theory can avoid these objections. The key idea is that grammar constrains the acts that can be performed by the utterance of a sentence. I present enough of the details of this version of the act-type theory to show how it can be used to respond to Båve's and King's objections. I conclude that this is a promising way to develop the act-type theory of propositions.

KEYWORDS:

• act types
• ambiguity
• grammar
• propositions
• meaning

Acknowledgements

I thank Arvid Båve, Ainhoa Fernández, Manuel García-Carpintero, Peter Hanks, Lorraine Keller, and Indrek Reiland, and an audience at the Act-type Theories of Propositions Workshop at the University of the Basque Country.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 See Hodgson (2021) for a survey of act-type theories.

2 It might be said that this conception of ambiguity does not fit with the standard notion, because according to the standard notion ambiguous sentences do not typically express both meanings simultaneously. This would not affect Båve's substantive point, or mine.

3 Heim & Kratzer use bold instead of quotation marks to mention expressions; I will reserve use of bold for meanings.

4 I use ‘property’ to cover entities of all arities. I would be happy to use ‘relation’ instead. Or to call those with an arity of greater than one ‘relations’, and restrict ‘property’ to those with an arity of one.

5 I have replaced ‘relation’ with ‘property’.

6 Collins (2011, chap. 5), in the context of a discussion of the metaphysics of propositions, defends a theory of syntax such that all phrase structures are binary branching. If so, my assumption is justified.

7 Hanks (2015, 86) uses these ideas in his treatment of constructions such as (1). Hanks appeals to work on thematic roles, citing Dowty (1989). Collins (2018) also presents Hanks’ position in terms of thematic roles, citing Chomsky (1981) as the source of the idea. Soames attributes his version of the idea to Montague (1973). There is therefore, a developed position that can be appealed to, and it is one to which some act-type theorists are already sympathetic. The discussion in Heim and Kratzer (1997, chap. 3) explains their thinking about thematic roles and related issues. Heim & Kratzer's theory imposes constraints on the roles played by arguments via the denotations they assign. They explicitly reject the possibility of variation of this hierarchy (Heim and Kratzer 1997, 55). The idea that I propose for the act-type theorist is essentially the same as the one that they discuss, and attribute to Grimshaw (1990). Verbs encode a hierarchy of arguments. Furthermore, the phrase structure fixes which object is assigned to each argument: slots are filled by immediate syntactic neighbours in an order fixed by the encoded hierarchy.

8 A perfect number is the sum of its divisors: $1+2+3=6$.