Flow of Ideas


Is it right to define an action as an event caused by a reason?

Alexander Rikowski

An essay written as an undergraduate in the Department of Philosophy, King's College London

London, June 2010

An action is an event which is something that a creature does, and such an event is to be distinguished from an event which happens to a creature (Wilson, 2007, p. 2). For example, an event such as a pigeon dropping faeces on my head would be an event that happens to me. However, if I was to shoot that pigeon with a gun, then the pulling of the gun’s trigger may be said to be an action of mine, as it would be something which I did. I shall be taking an ‘event’ to simply mean a happening. Many Philosophers, including Davidson, think that an action is caused by a reason. Davidson (1963, pp. 685-686) says that a person has a reason for doing something by having a “pro attitude” aimed towards a particular kind of action, along with a belief that “…his action is of that kind” (1). We are told that a pro-attitude includes desires, urges and values. Davidson says that a “primary reason” would involve a combination of a pro-attitude plus a belief, with this combination being the cause of an action. He says: “The primary reason for an action is its cause” (2). However, I shall be arguing alongside Hornsby, Anscombe and Melden that, it is not right to define an action as an event caused by a reason. I argue that although some reasons explain why creatures do what they do, such reasons are not causes.

An action is something that a creature does, and a creature could be something like a spider or a human person. Wilson points out that, in some sense, even spiders may be recognised as doing things for reasons. As he illustrates: “When a spider walks across the table, the spider directly controls the movements of his legs, and they are directed at taking him from one location to another” (3). The spider moves his legs to reach some goal. Also, I myself may do an action I am not even aware of. For example, I may blink my eyes without realising that I am doing so. Perhaps I did blink my eyes for some reason. However, when Davidson speaks of “primary reasons”, he is speaking about the intentions of agents. As he says: “To know a primary reason why someone acted as he did is to know an intention with which the action is done” (4). An agent has a primary reason by having a pro-attitude aimed towards some action whilst believing that that action is an action of a certain kind. Davidson gives the example of a person wanting to flip a light switch whilst believing that the act of flipping the switch is a certain kind of action—it is a kind of action that turns a light on. Davidson says that the intentional action of flipping a light switch is an action caused by a primary reason. Believing that actions are caused by intentions is the root of believing that every action is caused by some reason. It will be my aim to show that no action is caused by an intention.

Melden (1961) provides us with an example to focus on. He gives the example of a person raising his arm in order to indicate a turn whilst he is driving. The man signalling would be an event, as it would be a happening (Melden, 1961, p. 18). He signalled by raising his arm. Melden also points out that: “certain muscle movements took place—this is how the arm got raised” (5). However, Davidson would say that the raising of the arm was caused by a primary reason—the primary reason made it happen. But the problem is that such a reason cannot be identified with an action we do. As Melden illustrates: “We cannot identify what one does with what one makes happen” (6). If we say that an action is an event caused by a reason, we are still unable to identify the reason as an event, as it would not be a happening. Imagine someone being asked what is happening and they responded by talking about some reason as though the reason itself is a happening. Such a response would surely make no sense. A reason is not something which happens at all. But it also makes no sense to say that something which does not happen is the cause of something that does happen. Thus, it is nonsense to say that an action is an event caused by a reason.

Some would object to what has been said by claiming that, the primary reason is in fact an event. They say that the primary reason is an act of volition. “So I move my muscles by performing an act of volition which in turn produces a muscle movement” (7), says Melden. An act of volition is supposed to be an act of willing to do something, and it may be said to involve a belief and a pro-attitude. If willing is an act, then it must be a different act to that of muscle movement, explains Melden. Yet, he also points out that there would have to be a difference between an act of volition and the cause of that act. We would be leading to an infinite regress in claiming that the cause of the willing is itself caused by some other willing, as we would need to say what that last willing is caused by (Melden, 1961, p. 45). It may be said though, that, willing itself has no cause. Yet even if this is true, it could be maintained that this would still mean that the man raising his arm would not actually be a case of him doing some action. The man only willed to raise his arm, and his arm being raised was only an effect of an act of volition (Melden, 1961, p. 46). We are lead to the strange claim that the raising of an arm cannot be an action, as the closest it can be to an action is being the effect of one. However, even if we assume that this claim is true, an action would still not be an event caused by a reason. An action would instead be a reason that causes bodily movements.

Yet, it could be said that I have two types of actions. It could be held that some of my actions are the acts of volitions which cause my bodily movements, whilst others are my actual bodily movements. It would seem that at least some actions may be defined as events caused by a reason then. But what sense is there in saying that I have acts of volitions? After all, an act of volition cannot be described in the way that bodily movements can. It seems that all we can say about acts of volitions is that they cause bodily movements. But this suggests that nobody really understands what an act of volition is actually meant to be (Melden, 1961, p. 47).

Melden states: “Let us grant that there is some peculiar mental activity of willing” (8). He explains that we would surely have to say that each specific act of volition causes a particular bodily movement. We may be said to learn from experience which particular bodily movement an act of volition causes. But, as Melden says, if this is so, then: “I should have no reason to suppose, when I first performed that act of volition v1, that m1 [bodily movement 1] rather than m2 would follow” (9). We generally think that an act of volition involves some desire/urge and a belief directed towards some action, but if I have never performed the act of volition v1 before, then I could not have had the intention of causing m1. This all indicates that it is not at all obvious what acts of volitions are supposed to be. Yet, a bigger problem is that we can only describe an act of volition by associating it with a certain bodily movement (Melden, 1961, p. 52). If an act of volition truly is a distinct event from the bodily movement it is supposed to have caused, then it must be possible to describe such an event without associating it with a bodily movement. However, since such a description is impossible, we are lead to the thought that there is not actually such an event as an act of volition.

Anscombe acknowledges that there are intentional actions. Yet, she also explains that: “The mistake is to think that the relation of being done in execution of a certain intention, or being done intentionally, is a causal relation between act and intention” (10). She explains that we can realise this mistake from recognising that an intention need not be a distinct mental state that is prior to some action. She gives the example of a person who pushes a telephone button extra hard because the button is jammed. However, only afterwards did the dialler realise why he pushed the button extra hard. Someone could have afterwards asked the telephone dialler: ‘why did you push the buttons so hard?’ And the dialler could have responded by saying: “…to unjam it” (11), says Anscombe. The fact that such an explanation is given indicates that the action was intentional. He pushed the button extra hard for a reason—the reason was to unjam it. But this reason was only realised by the dialler after the act was done. Since the reason for the action was only realised afterwards, the telephone dialler could not have used the reason to cause the action.

When the car driver raised his arm, he also did so for a reason. But although the reason for an action is an explanation for why the action was done, this reason is not itself an event that causes some action. As Anscombe says: “there is such a thing as intentional action and when there is, the intention or intentions involved in it belong in an account of such action” (12). What this means is that the reason for why someone did what they did is part of an account or explanation of why the action was done, and such explanations form part of our understanding. A person may indicate to us that an action he did was intentional by giving us an explanation for why he did it. But explanations are not to be confused with actions themselves. Anscombe shows us that actions are one thing, but the explanations we have for why they occurred are something else.

To further recognise that intentions are involved within accounts and explanations only, Hornsby (2004, p. 3) raises the point that we can have an intention not to do something. She gives the examples of a person deciding not to have a chocolate, and of a person deciding not to answer a telephone call. She also gives the example of a person choosing to ignore someone else. However, although our actions are events that we do, these examples show us that there is no necessary link between intentions and actions. As Hornsby writes: “when we ask why someone did something, expecting to learn about what they thought or wanted, we don’t always need to consider whether or not there was a positive performance on their part” (13). I may ask someone why they did not answer their telephone, even though them not answering the phone is itself not an action they actually did. But it would make no sense for me to say that some intentions are causes of things that do not happen, as a cause is a cause if and only if it is the cause of an effect. The only plausible explanation is to say that, it is the role of intentions to explain why agents either did or did not do certain actions. Intentions are explanations rather than actual events.

I said that an action is an event which a creature does. Davidson says that an intentional action is an action caused by a primary reason, which is itself a combination of a pro-attitude and a belief. However, the mistake Davidson made was in failing to properly distinguish the difference between actions and the understandings / explanations we have of them. We say that creatures do things for reasons. Yet, it is a mistake to say that such reasons are the actual causes of actions. A driver raised his arm because he wanted to indicate a turn—this is the reason that both we and the driver may have for him raising his arm. The action was intentional because the driver had an explanation for why he did it. Perhaps raising his arm actually caused the indicator light on his car to flash. However, actual causal relationships are not themselves explanations—they are instead what we are trying to explain. We cannot say that some mind/brain states are primary reasons involved in causal relationships, as it is the role of reasons to explain such relationships (Anscombe, 1983, p. 182).

As soon as we ask why the driver raised his arm, the reason given would be an explanation as to why that event occurred. Reasons for actions are either accounts or are involved within accounts, but it makes no sense to say that some accounts are causes of actions (Anscombe, 1983, p. 182). Accounts can only be explanations, and they do not interfere with actual causal relationships. Even if I have a reason to do a particular action prior to me actually doing that action, the reason itself would only be an account of why that action was done. Since the reasons for actions are either accounts or are involved within accounts, there is thus no sense in saying that an action is an event caused by a reason. For a definition to be a right definition, it needs to be a sensible thing to say. But it is nonsense to say that an action is an event caused by a reason. Therefore, it is not right to define an action in this way. Many philosophers have mistakenly defined an action as an event caused by a reason. But as Hornsby says: “I can’t help thinking that, these days, it takes a really great number of philosophers to contrive in the persistence of a really great mistake” (14).


Anscombe, E. “The Causation of Action”, published in Knowledge and Mind: Philosophical Essays, edited by Ginet. C and Shoemaker. S, Oxford: Oxford University Press, (1983).

Davidson, D. “Actions, Reasons, and Causes”, published in The Journal of Philosophy, Volume LX, No. 23, (November 7, 1963).

Hornsby, J. “Agency and Actions”, published in: Agency and Action, edited by H. Steward and J. Hyman, Cambridge: Cambridge University Press, downloaded from website: http://eprints.bbk.ac.uk/95/, (2004).

Melden, A. Free Action, London: Routledge & Kegan Paul Ltd, (1961).

Wilson, G. “Action”, published in Stanford Encyclopedia of Philosophy, website: http://plato.stanford.edu/archives/fall2009/entries/action/ (2007).


(1)Davidson, D. (1963), Actions, Reasons, and Causes, p. 685.

(2)Davidson, D. ibid, p. 686.

(3)Wilson, G. (2007), “Action” published in Stanford Encyclopedia of Philosophy, p. 2.

(4)Davidson, D. ibid, p. 689.

(5)Melden, A. (1961), Free Action, p. 18.

(6)Melden, A. ibid, p. 43.

(7)Melden, A. ibid, p. 45.

(8)Melden, A. ibid, p. 48.

(9)Melden, A. ibid, p. 51.

(10) Anscombe, E. “The Causation of Action”, published in Knowledge and Mind, (1983), p. 179.

(11) Anscombe, E. ibid, p. 180.

(12) Anscombe, E. ibid, p. 182.

(13) Hornsby, J. (2004), Agency and Actions, downloaded from website: http://eprints.bbk.ac.uk/95/, p. 5.

(14)Hornsby. J, ibid, p. 16.

This paper was originally a Pre-Submission for the 2009/10 BA Philosophy of Mind Paper, King’s College, London for the BA in Philosophy

Copyright © Alexander Rikowski, November 2010

© Copyright, Flow of Ideas, Ruth Rikowski and Glenn Rikowski. Website by [whiteLayer]
Call for Authors
Printer Friendly
Order by DateAlphabetical[Close menu]