Cognitive Acts and the Unity of the Proposition

ABSTRACT

In this paper I do four things. (1) I explain one clear thing that ‘the problem of the unity of the proposition’ might mean. (2) I lay out a few different versions of the theory of propositions as cognitive acts, and explain why this problem arises for the version of that theory which has been defended in different forms by Peter Hanks and Scott Soames. (3) I argue that the natural ways in which the act theorist might try to solve the problem fail to solve it; (4) I propose a way to fix the problem, and then I explain how the problem re-emerges in the act theorist’s treatment of propositional attitude relations.

KEYWORDS:

• propositions
• unity of the proposition
• act theories

Disclosure Statement

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

Notes

1 See, for example, Gaskin [2008].

2 See, for example, Plantinga [1974] and Merricks [2015].

3 This is closest to the ‘unity question 1’ discussed by King [2009, 2013].

4 This is perhaps clearest in Frege [1923/1963: 1], where he proposes that ‘we look upon thoughts as composed of simple parts, and take these, in turn, to correspond to the simple parts of sentences.’ On the other hand, this passage is immediately preceded by the remark that ‘we really talk figuratively when we transfer the relation of whole and part to thoughts.’

5 See Hanks [2011, 2015] and Soames [2012, 2013, 2014, 2015].

6 This is one of the central points of disagreement between the theories of Soames and Hanks; since (as far as I can see) it does not matter much for the discussion in the next few sections, I simply adopt Soames’s view for ease of exposition. This difference between their respective views will be important to the discussion in section 4 of judgment, and I return to it there. For Hanks’s criticism of Soames’s view of predication, see Hanks [2015: sec. 1.3].

7 This is Soames’s preferred view (personal communication). It is less clear to me whether Hanks’s view is best understood as a version of the quasi-Fregean act theory or instead as a version of the complex-act theory. On the one hand, his way of representing propositions suggests the quasi-Fregean interpretation; on the other hand, he thinks that we predicate properties like being on fire of individuals rather than of some other mysterious sort of entity, which suggests that he would not endorse that interpretation.

8 Hanks [2013] gives an account of what distinguishes first-person reference to an object from other ways of cognizing objects. Soames leaves the notion unanalysed.

9 Or, perhaps, to perform each of these acts within a certain interval.

10 One might try to get around these problems by giving a dispositional version of the intentional theory. On one version of this idea, one’s $\phi$ing is part of one’s plan for $\psi$ing iff one is disposed to intend to $\psi$ because of an intention to $\phi$ and one is disposed to believe that $\phi$ing is a means to $\psi$ing. This would solve the over-intellectualisation and regress problems. But it would run into basically the same problem as faced the conjunctive-act theory. I could predicate being-aflame of an object (myself) seen in a mirror while also thinking of myself in a first-person way and while possessing the dispositions just described without entertaining (1P). (Thanks to two anonymous referees for very helpful comments about the arguments in this section.)

11 In what follows, I am indebted to the useful overview of that literature by Mayr [2011: ch. 5]. It is standard to distinguish between the problem of causal deviance as it applies to basic actions (as in the case of the climber) and as it applies to non-basic actions. The latter is often handled by reference to the agent’s plan for carrying out the relevant non-basic action. (See, for example, Bishop [1990].) As appeal to plans would involve appeal to propositional attitudes of the subject, I ignore this kind of treatment here.

12 In the theory of action, a version of this idea is defended by Brand [1984].

13 As King [2013: 89] in effect notes (although he is discussing a slightly different version of Soames’s theory).

14 Although I’m a bit sceptical about whether the proposed sort of explanation is much of a theoretical virtue. For some discussion of the point, see Speaks [2014].

15 I should add that the reasons why Hanks adopts a view of this sort have nothing to do with the kind of issue under discussion here; rather, they come from scepticism about a certain kind of content-force distinction.

16 See, for example, King [2009].

17 Thanks for very helpful comments on previous versions of this material to Peter Hanks, Lorraine Keller, Gideon Rosen, Scott Soames, audiences at the University of Leeds and USC, the participants in Boris Kment’s and Shamik Dasgupta’s graduate seminar at Princeton, and two anonymous referees.