We can think of Davidson’s theory of action as being an instance of both psychologism - the thesis that intentional actions are partly constituted by the having of certain mental states -and causalism - the thesis that intentional actions are constituted by their causes.

Wilson’s aim is to re-open the dispute between causalists and non-causalists which many had regarded as conclusively settled by “Actions, Reasons, and Causes.”

1 The contrast between causalism and non-causalism

A first step in doing this is say more about what this dispute comes to. Wilson thinks that we should not attribute to non-causalists the view that reasons are not causes in any sense of intentional actions. ‘Cause’ has “different uses” (30), and sometimes is used to mean something like ‘rational ground.’ So what we need is some restrictive understanding of causation which the causalist, but not the non-causalist, claims the relation between reasons and actions satisfies.

Wilson offers us the following:

What kind of causation is this? Presumably some kind of efficient causation, thought of as a relation between events. But Wilson dos not tell us more than that at this point.

2 Wilson’s non-causalist answer to Davidson’s challenge

2.1 Intentional action and action in order to

As Wilson rightly says, the main argument of “Actions, Reasons, and Causes” can be thought of as a challenge: if the explanatory relation between reasons and action is not an event-causal relation, what can this explanatory relation be? Davidson implies that there is no real alternative. Wilson’s aim, beginning on p. 31, is to sketch an alternative. This is what he says:

At this stage, it looks like Wilson is suggesting that we replace Davidson’s causal analysis with a non-causalist analysis like the following:

As Wilson suggests, Davidson would likely regard this less as a replacement of the causal theory of action than as an early stage in the development of that theory. For Davidson would go on to analyze the right-hand side of this biconditional in terms of causation (in the right way) by beliefs and desires.

Wilson often suggests that the real issue between he and Davidson is about the reducibility of these teleological claims to event-causal claims. If this were the whole of the issue, then I think that it would be hard to be satisfied with Wilson’s claim that there is a non-causalist answer to the challenge of “Actions, Reasons, and Causes.”

2.2 Intention in action

But I think that Wilson’s position has more to be said for it than this. The key passages, I think, are the discussion of intentions in action on pp. 32 ff. There he suggests that there is a link between ing in order to do something and having a certain intention in ing. He endorses, with some qualifications, the following series of equivalences:

The idea is then that, given the equivalence between intentionally ing and there being some x such that one s in order to x, we can give an account of intentional action in terms of any member of this set of equivalent claims. (This is why, I take it, Wilson spends so much time discussing the nature of the relevant intentions in action - because we can give an account of the nature of intentional action in terms of the relevant intentions in action.)

The account Wilson seems to have in mind is something like this:

 (See also Wilson, The Intentionality of Human Action, p. 143.) He makes two main points about how the right-hand side of this analysis is to be understood: (1) that the intention is de re with respect to the particular action token being undertaken by the agent, and (2) that this is an intention rather than some other de re pro-attitude involves the agent consciously directing the movement at the objective in question.

I am not sure whether this gloss captures Wilson’s view or not. But if it does, one way to get what’s going on is to place it in the context of Anscombe’s three senses of ‘intention’, by contrast with the other views we have discussed. Remember that these were (i) intentional action, (ii) intentions to act, and (iii) intentions in acting. This is a slight oversimplification, but we can say that whereas Anscombe took (i) as primary and most causal theorists take (ii) as primary (sometimes identifying these intentions with complexes of beliefs and desires), the kind of view we have been developing in our discussion of Wilson takes (iii) as primary.

There is one important sense in which Wilson’s view, if correct, has answered Davidson’s challenge. Davidson said that, even given that having something as a reason is a matter of being in certain mental states, having a certain reason be the reason for an action requires there to be some link (which Davidson claimed to be causal) between those mental states and the action. After all, it is not enough to simply be in the mental states and perform the action; I may believe that getting up in the morning will get me to class on time, desire to get to class on time, and get up early, but not having getting to class on time be my reason for getting up early. This was part of the explanatory gap that Davidson tried to close by requiring that the belief/desire pair cause the action.

Wilson seems to have the makings of a solution to this problem in the remarks under the heading of (1) above; but his solution is not in finding some relation between the mental states and the action (like Davidson’s causal relation) but rather in the special character of the content of the relevant mental state. The key point is that the intention in acting is de re and directly refers to a specific action token, unlike Davidsonian means-end beliefs, which are usually thought of as general beliefs about action-types. This de re character of the intentions is supposed to, so to speak, connect up the intentions with the specific action being undertaken by the agent, and so to erase the need for a linking causal relation.

A possible counterexample to this claim (from Wilson’s The Intentionality of Human Action): cases of secondary deviance.

A problem with part (2) of Wilson’s view. We can hardly build the notion of conscious direction of action into the content of the relevant intention in action, so the attitude of intention in action will not be reducible to any other attitude. (This is contrary to causalist views like the early Davidson and Setiya, which reduce intentions to beliefs, desires, combinations of the two, or states with some of the characteristics of each.

3 Objections to Wilson’s view

3.1 Actions done for their own sake

For Wilson’s view of action to even get off the ground, we need the truth of the biconditional with which we started:

But it is just not obvious that this is true. Consider, e.g., arational actions. Do I perform these actions in order to do something else? A similar point can be made about actions done for their own sake. If this phenomenon is to be taken at face value, it’s hard to see how it fits into this biconditional.

A possible solution to this problem in the fact that the relevant intentions in action are de re.

3.2 The analysis of intention in action

Why intentions in action do not seem a comfortable resting place for analysis; disadvantages of Wilson’s non-causal form of psychologism.

Psychologistic and non-psychologistic analyses of ‘in order to’ and forms of the teleological theory of action. The problem with accommodating Wittgenstein’s view of intention in action in §647 of the Investigations (quoted by Wilson) in Wilson’s version of the teleological theory. That passage instead points to a non-psychologistic account of ‘in order to’ as a way of explicating intentions in acting.