http://orcid.org/0000-0001-8778-524X
![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
A widespread view in the philosophy of mind and action holds that intentions are propositional attitudes. Call this view ‘Propositionalism about Intention’. The key alternative holds that intentions have acts, or do-ables, as their contents. Propositionalism is typically accepted by default, rather than argued for in any detail. By appealing to a key metaphysical constraint on any account of intention, I argue that on the contrary, it is the Do-ables View which deserves the status of the default position, and Propositionalism which bears the burden of proof. I go on to show that this burden has not been met in the literature.
Notes
1. The ideas in this paper have benefitted from discussion with a number of people. In addition to those mentioned in footnotes below, thanks go especially to Alex Grzankowski, and in addition to Tim Crane, Alexander Greenberg, Jane Heal, Jen Hornsby, Dave Jenkins, John Maier, Daniel Morgan, Silvan Wittwer, and to the audiences at the London Mind Group (Autumn 2017) and Questioning Propositionalism in Philosophy and Linguistics, a workshop at Birkbeck, University of London (Spring 2018).
2. See also Shah and Silverstein (Citation2013); Coliva (Citation2016), 1; Fodor (Citation1978), 506; Williamson (Citation2017); McDowell (Citation2010); Vermazen (Citation1993); Velleman (Citation1989); Searle (Citation1983, ch. 3).
3. Sometimes we do use ‘that’ clauses, but these tend to be in the subjunctive rather than the indicative – I might intend that the audience take me seriously. That-clauses in the subjunctive do not obviously express propositional contents – for discussion (in relation to desire), see Alvarez (Citation2010, 67–68).
4. The identities of the (supposed) propositional contents of intentions are sometimes delivered in a different way, by appealing to what one would say in expressing one’s intention (see e.g. McDowell Citation2010, 417). I can express my intention to wash the dishes with ‘I will wash the dishes’ or ‘I’m washing the dishes’, and might express my intention for Mynn to wash the dishes with ‘Mynn will wash the dishes!’ (Anscombe Citation1957, 3; Velleman Citation1989, 61; Grice Citation1971, 271). Both ways of identifying the (supposed) propositional contents of intentions have been appealed to as part of an argument for the standard view. I will consider and reject both arguments in section 4.
5. This is the basic shape of the standard account. Some add extra frills, for example holding that an intention to φ is an intention that one will φ as a result of this very intention. I consider this idea in 2.2.
6. What I am here calling ‘Propositionalism’ might be rejected by someone who nonetheless accepts the idea that intentions are propositional attitudes. One person who does so is Luca Ferrero (Citation2013; thanks to an anonymous reviewer for this journal for alerting me to this work). Ferrero objects to the idea that in general, an intention to φ is an intention that one will φ. Rather, he thinks, the content of this intention is the intender’s goal, understood as the state of affairs thereby aimed at, and brought about if the intention is successfully executed. The goal of my intention to wash the dishes, for example, would be the state of affairs of the dishes’ being clean, and the propositional content of my intention would be that the dishes are (will be?) clean. My target in this paper is the orthodox version of the idea that intentions are propositional attitudes, so I won’t assess Ferrero’s account in the main text. Occasionally I will comment on his position in notes.
7. ‘Do-ables’ is my label; intention-contents so-understood are also referred to in the literature cited as ‘acts’ or ‘actions’. I prefer the label ‘do-ables’ because it nicely captures the inherent potentiality of what is intended.
8. My sense is that this final version of the Do-ables View is, in the end, preferable (see also Alvarez Citation2010, 67; Hornsby Citation2016; Thompson Citation2008, 121). I reject the pure do-ables view because it collapses the distinction between own-action and non-own-action intentions, viewing all intentions as own-action intentions (for more on why this is problematic, see section 2.2 below).
9. See (Baier Citation1970). What I here call ‘Baier’s Distinction’ is not explicitly discussed by Baier, but she does discuss its very close cousins, such as the distinction between intending to φ and intending to get oneself to φ (e.g. Citation1970, 653). More generally, my reference to Baier here is aimed to highlight how much common-ground our discussions share. But we do not completely overlap for a few reasons, one of which being that whilst I am objecting to Propositionalism, Baier is objecting to Chisholm’s conception of intending as (always) intending to bring it about that p, which is not strictly a propositionalist account – although it shares some pitfalls with the latter. For another difference between Baier’s and my discussion, see note 16.
10. Ferrero’s account (see again note 6) can accommodate Baier’s Distinction in a similar way, holding that ordinarily an intention to wash the dishes has as its content the state of affairs of the dishes’ being clean, but that one’s intention can also concern the state of affairs of one’s cleaning the dishes. I compare Propositionalism to the Do-ables View rather than to Ferrero’s propositionalist alternative because the Do-ables View is both the most commonly-held alternative – and so the main rival – to Propositionalism, and because it is prima facie better-motivated than Ferrero’s propositionalist account. Whilst he argues against standard Propositionalism and in favour of his account of how to identify the propositional contents of intentions, Ferrero simply assumes without argument that intentions’ contents are propositional: he argues that the content of an intention is the goal the intender has in intending, but it is an assumption in his discussion that goals are states of affairs, and so propositionally structured (Citation2013, 73). So Ferrero’s position is unmotivated unless he can provide an argument for the claim that intentions are propositional attitudes of some variety. (I undermine several candidate arguments for this conclusion in sections 4.4 – 4.6.) The Do-ables View, by contrast, is motivated initially by the fact that it fits nicely with our natural-language talk about intention. I have conceded that this alone is not proof of the Do-ables View, but it does give it the edge over Ferrero’s view.
11. This suggestion was first made to me by Hugh Mellor, and has also been pressed by various anonymous reviewers commenting on earlier versions of this paper.
12. Thanks to an anonymous reviewer for this suggestion.
13. For a particularly extreme version of this tendency, recall the quotation from Davis from the beginning of this paper, which characterises the definiendum of an account of intending as ‘S intends that p, where p stands for any [indicative] sentence’ (Citation1984, 43; my underlining; see also Williamson Citation2017, 168).
14. This problem also leads me to reject the idea that all intentions have do-ables as their contents (section 1).
15. I also don’t want to deny the important fact that we sometimes think of intentions as including one another. We can describe me as intending to bribe Mynn in order to get her to wash the dishes. But we can also (truly) describe me simply as intending Mynn to wash the dishes – even though my intention won’t be executed unless I do something do get Mynn to wash them. Again, compare the belief case: we can describe me as believing that [I need to buy cartridges because I’m running out of ink]. But we can also describe me simply as believing that I need to buy cartridges – even where I believe this because I also believe that I’m running out of ink. For discussion see Anscombe (Citation1957, 46–47).
16. Compare Baier here: ‘In our language […], the agent’s special privilege is, not to insert himself into all his intentions, but to leave himself out of the basic ones.’ (Baier Citation1970, 658) Baier here conflates the class of basic actions with the class of actions one can perform without acting on oneself – the mistake I complained about in previous section. This is part of the reason why although my discussion owes a lot to Baier’s in spirit, I don’t follow its letter too closely (see fn. 9).
17. John Maier, Alison Fernandes, and Daniel Morgan have all independently suggested something like this option to me.
18. David Lewis, in a different context, argues that while we should care about qualitative parsimony, there is no motivation for quantitative parsimony. That is, we should be wary of introducing new types of entity if avoidable, but once we have committed ourselves to a given type of entity, it is no more problematic to think that there are a great number of them than it is to think that there are only a few (Lewis Citation1973, 87). Presumably the same point can be applied to theoretical tools as to entities, so it might be argued that the current argument – call it the explosion argument – fails because the lack of parsimony it observes is merely the unproblematic quantitative kind. Two points in response: first, I don’t want to put too much weight on the explosion argument. Perhaps its function is to emphasise the problem with the ‘IA’ idea itself rather than to add further problems. But second, it’s not clear to me whether the distinction between qualitative and quantitative parsimony can be drawn in a principled way – at least not in this case. Whether introducing e.g. ‘CampbellA’ in addition to ‘IA’ is introducing a new type of theoretical tool depends on how we describe ‘IA’. Is it a form of reference? If so, then introducing ‘CampbellA’ is introducing more of the same, since ‘CampbellA’ is also a form of reference. Is ‘IA’ a form of self-reference? Then introducing ‘CampbellA’ is introducing a new type, since ‘CampbellA’ is not a form of self-reference.
19. Thus the failure of arguments 4–6 is a problem also for Ferrero’s propositionalist account (Ferrero Citation2013); see again notes 6 and 10.
20. Williamson (Citation2017, 167) does so obliquely; Stanley is much more explicit in his appeal to the standard syntax, although in the context of knowing how to, rather than intending to, φ (Stanley Citation2011, ch. 3). It is fair to assume that Stanley would view the argument as carrying over to intending to φ; see Hornsby (Citation2016, Citation2017) for criticism.
21. Castañeda is not a Propositionalist per se. He holds that the contents of intention are neither do-ables, nor propositions, but ‘proposition-like’, and self-referential (Castañeda Citation1972, 41). He argues against the Do-ables View (as it appears in Baier Citation1970) on this basis.
22. Price (Citation2016) argues that we ought to understand Aristotle as (implicitly) committed to the Do-ables View (only implicitly because Aristotle does not discuss things in these terms).
Additional information
Notes on contributors
Lucy Campbell
Lucy Campbell works primarily in epistemology and in the philosophies of mind and action. Particular research foci include the metaphysics of agency, Elizabeth Anscombe's epistemology and action-theory, psychological and practical self-knowledge, and the metaphysics of knowledge. Her PhD, "Action, Intention, and Knowledge", is a monograph on practical knowledge, and was completed in Cambridge in 2015. At the time of writing, she is a Stipendiary Lecturer in Philosophy at Exeter College Oxford, a College Lecturer in Philosophy at St. Peter's College Oxford, and an Associate Member of the Faculty of Philosophy, University of Oxford.