At one with our actions, but at two with our bodies

Abstract

Jennifer Hornsby's account of human action frees us from the temptation to think of the person who acts as ‘doing’ the events that are her actions, and thereby removes much of the allure of ‘agent causation’. But her account is spoiled by the claim that physical actions are ‘tryings’ that cause bodily movements. It would be better to think of physical actions and bodily movements as identical; but Hornsby refuses to do this, seemingly because she thinks that to do so would be to endorse the so–called ‘standard causal story’. But Hornsby misses a possibility here, for we can insist on this identity claim without endorsing the standard story if we embrace an account which parallels the disjunctive account in the philosophy of perception. This will leave us with a picture of physical action that saves the insights of Hornsby's account without succumbing to its distortions.

Acknowledgements

I would like to thank Jane Calvert, Nadine Cipa, John Dupré, James Mark, an anonymous referee and (above all) Jennifer Hornsby, for their helpful comments.

Notes

1. Danto puts things in this way: ‘an action [is] a movement of the body plus x … and the problem … is to solve in some philosophically interesting way for x’ (1981, 5). Danto was writing at a time when it was (falsely) assumed that whenever there is an action, there is a movement of the body. The case I am considering may have ditched that assumption, but it has retained the form of Danto's ‘problem’.

2. Hornsby is in line with an orthodoxy here, for many contemporary ‘analytical’ philosophers also disavow the assumption. To take an example at random, Noë claims to be ‘no fan of the project of philosophical analysis’, and then asserts: ‘it's doubtful that there has ever been an analysis (that is, a breakdown into necessary and sufficient conditions) of any philosophically interesting concept’ (2003, 94).

3. One finds the (supposed) idea of this (supposed) relation in the work of Richard Taylor (1966), Thomas Reid (1969), Roderick Chisholm (1976) and Timothy O'Connor (2000). These philosophers differ in what they take the relata in this (supposed) relation to be. Taylor holds that the agent causes her action, and not merely her bodily movement; Chisholm holds that she causes her bodily movement; and O'Connor contends that she causes neither of these, but rather ‘the event of a coming-to-be of a state of intention to [act]’ (2000, 72, n. 11). The important point, however, is not so much the ontological status of the effect as the ontological status of the cause—that it is a continuant, and not an event.

4. I hope that these comments will go some way towards undermining O'Connor's (2000) assertion that he is simply describing our ‘common sense picture of the world’. I prefer to think of him rather as gesturing towards a merely apparent idea.

5. O'Connor's exact words are ‘a movement triggering volition to move my arm in a specific sort of way’ (2000, 26). To appreciate why I am entitled to replace these words with ‘an event of my moving my arm’ we must remember that the account similar to Hornsby's that O'Connor is trying to rebut is Carl Ginet's (1990). Ginet argues that when there is a physical action there is a certain kind of ‘volitional event’ that either is or ‘forms the core of’ a physical action if and only if it appropriately ‘triggers’ an appropriate series of further events that include the appropriate bodily movements. Ginet describes this ‘volitional event’ as an event of ‘the person's trying to do something’. And he insists that, in the case of the physical action of someone's moving her arm, the event of her moving her arm is the event of her trying to move it if and only if this ‘volitional event’ appropriately causes the appropriate series of events (one of which must be her arm's moving). As the ‘volitional event’ of which O'Connor speaks ex hypothesi causes these events, so this event is an event of my moving my arm. These features of Ginet's account anticipate those of Hornsby's that I discuss in Section III. (And, if you are not familiar with either account, reading that section probably will be a condition for understanding this footnote.) The differences between the accounts turn mainly on fine points of action-individuation—Hornsby thinks that all physical actions are identical to tryings, whereas Ginet insists that only those that belong to a certain restricted class are. Whilst not without interest, this difference is not relevant to the matter in hand, as Ginet is clear that an action of someone's raising her arm does fall into this class, and this is the only sort of action that concerns us here.

6. It is in the light of the (supposed) problem of deviant causal chains that Hornsby says that the action must ‘appropriately cause’ the movement. Certain philosophers—Mele (2000), for instance—have tried to give a non-circular account of what the ‘appropriate’ causal connection might be. Unsurprisingly, Hornsby makes no such attempt, preferring instead to tell the following ‘circular’ story: ‘the right kind of causal connection here … is the kind there is when someone's arm going up is an exercise of her capacity to raise her arm at will’ (2004, 132). This makes it seem as if she thinks that the event of Jane's arm's going up is her action—for, assuming that her exercising of this capacity is an event, could it really be one of any other kind? Such a view is not officially Hornsby's own; but (in my opinion) it should be.

7. Hornsby (1980) is full of this ‘inside the body’ talk. But Hornsby (1997c) disowns it, seemingly because, in the light of a commitment to the mind/world dualism that McDowell (1998b) attacks (see Section IV), it can suggest that actions are located on the ‘mental’ side of this dualistic divide. But, even if we retain this commitment, this talk only carries this suggestion if we suppose that the boundaries of a person's mind are the boundaries of her skin—and there is nothing in the talk per se that must lead us to suppose this. Hornsby says that she would ‘like to suppose that [this talk] can be retracted while leaving almost everything else [of what she says in the 1980 book] intact’ (1997c, 232, n. 1). But, if I am right, even though she can reject the suggestion, she cannot retract the talk.

8. Hornsby has suggested that the bodily movements that actions cause are of a different kind to the bodily movements that there might be ‘in a world bereft of beings who do things for reasons’ (1999, 393). But it would be a mistake to think that this ‘disjunctive’ conception of bodily movements could help Hornsby avoid the alienating consequence—as if the objects whose powers are wholly distinct from our powers of agency could be made any less alien by the fact that they could not exist in a world where there are no powers of agency for their powers to be wholly distinct from.

9. Watson remarks that, ‘what seems wrong with Hornsby's account is something it shares with dualism: it implies that I stand to my body as a captain to his ship … the implied gap between me and my body is what seems unacceptable’ (1982, 467). Similar points, which seem to me to fit Hornsby's account well, can be found in McDowell (1994, 74–75).

10. The personal/sub-personal distinction is originally due to Dennett (1969). Recent defences (and discussions) can be found in McDowell (1998c) and Hornsby (2000).

11. I think we should be suspicious of the story's suggestion that we can form a conception of a person—the embodied entity whose movements occur in response to its beliefs and desires—without yet conceiving of something capable of engaging in physical action. Without this suggestion, the story's attempt to explain reductively the very idea of physical action in the way it does cannot succeed. There is no space to argue for this suspicion here; but let me point to Wiggins (1987) and chapters 4 and 7 in Wiggins (2001), which provide much that such an argument would need.

12. Various different versions of this account can be found in work by Hinton (1973), Snowdon (1980–81), McDowell (1998a, 1998b) and Martin (2004). The version I offer here is (what I take to be) McDowell's.

13. Auditory, tactile and olfactory alternatives are easy to envisage.

14. It is important that these facts concern an agent (Jane). It would be simply wrong to think we can take the idea of bodily movement disjunctively if the body that moves is not the body of an agent. Yet, I am urging that if the body that moves is the body of an agent, then our understanding of what it is for it to move should be understood along the lines of the disjunctive account. Our understanding of what it is for a body to move should not be neutral between cases where the moving body is the body of an agent and cases where it is not.

15. Here I am adopting for my own purposes a suggestion that Hornsby makes regarding the disjunctive account in the philosophy of perception, when she writes of the disjuncts as ‘giving a determinate kind of which its seeming to [Jane] as if … might be thought of as a determinable’ and that the ‘disjunctivist considers it an error to take the determinable as fundamental’ (1997a, 104).

16. She notes that it goes back as far as Wilkins (1668).

17. Watson (1982) offers the only other attempt to rebut this argument of which I am aware. He tries to use what he sees as the general inadequacy of Hornsby's account to motivate the thought that the proper conclusion to draw from the linguistic truth is that we might need to embrace ‘agent-causation’ after all. Clearly I want no truck with that; so let the suggestion I am making here be a response to Watson as well as Hornsby.

18. It is also worth noting that in this case there can be no event of Jane's body's merely moving_I. Of course, if there were such an event, it too would also be an event of Jane's body's moving—on the understanding that to characterise the event in this latter way is to characterise it as a member of a determinable, and not a determinate, kind.