www.richmond-philosophy.net - © COPYRIGHT 2002-10 RUTC - Designed and maintained by Mat Carmody.

A Defence of Internal Reasons

Lisa Grover

Abstract

The argumentative aim of this paper is to challenge John McDowell's argument that deliberation is not the only method of gaining new reasons for action and, hence, that there are external reasons. This is important because the debate as to whether there are external reasons could undermine Bernard Williams' central argument that all reasons are internal reasons.


My methodology is to question McDowell's notion of conversion. I argue that if conversion is allowed to be a legitimate route from your existing subjective motivational set to a reason for action, then it seems that anything could be a reason for action and that any attempt to limit the scope of conversion collapses into a version of internalism about reasons. However, I do argue that McDowell has identified a question as to whether the legitimate acquisition of new motivations is exhausted by the process of rational deliberation. I continue to analyse the work of Elijah Millgram, who develops McDowell's argument and claims that there are certain reasons we can only gain through experience. I reach a different conclusion to that of Millgram and argue that the idea of gaining reasons through experience can be compatible with an internalist approach.


Introduction

The debate as to whether reasons are internal or external focuses upon a disagreement about the truth conditions of statements of reasons: if we state that an agent has a reason to help someone pick up some papers that she has dropped, what makes this statement true? The internal reasons theorist argues that such a statement is true, that the agent has a reason to act, when he has, broadly speaking, some motive, desire or pro-attitude that would be furthered by that action. For example, the agent wants to be helpful and helping the lady pick up her papers will satisfy this desire. The external reasons theorist denies that the truth of such a statement depends upon the existence of a relevant desire. Under this view, the statement that an agent has a reason to help someone pick up her papers can be true even if the agent has no desire whatsoever to help. The truth of the reason statement is independent of the agent's desires.


This debate is important because it links to questions within contemporary moral theory as to how reasons motivate an agent to act. The internal reasons theorist argues that there is a close connection between having a reason and a motivation to act upon that reason because the truth of a reason statement depends upon the agent's desires. For example, if we state that an agent has a reason to help someone pick up some papers that she has dropped, the internal reasons theorist will argue that this is true when he has some desire that would be furthered by that action. Under such an interpretation, the internal reasons theorist need tell no further story as to how the agent becomes motivated to act upon the reason, as the agent can only be said to have the reason if he has access to the relevant motivation. The external reasons theorist faces the challenge of explaining how the agent comes to be motivated by a reason because he denies the dependency between reasons and desires.


What Williams says

The distinction that Bernard Williams makes between internal and external reasons is central to this debate (1981, 101-113). He formulates an internal reasons statement as:


'A has a reason to φ only if there is a sound deliberative route from A's subjective motivational set to A's φ-ing' (2001, 91).

For example, A may have a desire for a gin and tonic within his subjective motivational set. He can deliberate between various methods of satisfying this desire and conclude that the best way to satisfy this desire is to drink the gin and tonic that is in the glass on the table. In this case, A will have a reason to drink the gin and tonic that is in the glass on the table. Williams thinks that a subjective motivational set is broad, containing not only desires, but 'such things as dispositions of evaluation, patterns of emotional reaction, personal loyalties, and various projects, as they may be abstractly called, embodying commitments of the agent' (1981, 105). He thinks that deliberation is important because it can give rise to new internal reasons previously hidden from the agent, it can remove elements from a subjective motivational set (for example, a false belief) and it can lead to the realisation that in fact there is no reason to act in a certain way. Williams derives from his thoughts about deliberation that a subjective motivational set is not static, as it can be changed by the process of deliberation (1981, 105).


Williams thinks that there are only internal reasons for action and he argues that the concept of someone having a reason for action depends upon them being able to reach the conclusion that they have a reason to do a certain action by a 'sound deliberative route' from their 'actual motivational set' (1995a, 35). Williams argues that under the internalist view this formulation is a necessary condition of the truth that an individual has a reason for action. The externalist does not think that this is a necessary condition and that it is possible for someone to have a reason for action for which there is no desire (either direct or through sound deliberation) within their subjective motivational set, or S.


Williams thinks that we can write the requirement of correct information and reasoning into the notion of a sound deliberative route because all rational deliberative agents have an interest in this and are constrained by the requirements of reasoning (1995a, 37). He says 'the aim of getting things right in such ways (taking correct means to his ends) is part of any agent's interest as a rational deliberator, and the aim can be assumed to figure in any rational agent's S' (2001, 92). He argues that we cannot build moral considerations into the notion of a sound deliberative route because there would have to be a separate argument as to why morality has to be part of someone's motivational set. He thinks that if moral requirements were considered part of sound deliberation, then 'there would be no significant difference between the internalist and the externalist accounts' (1995a, 36). He continues to argue that although it may be possible to construct an argument that the requirements of morality are part of what it is to be rational, it is only truth and reasoning that are essentially constitutive of rational agents. He does not exclude altruistic and ethical considerations from an agent's S; he just argues that we cannot assume that these are necessarily part of what it is to be rational.


McDowell on establishing the externalist position

Williams uses the example of Owen Wingrave to illustrate the problem he perceives with external reasons (1981, 106). In this example, Owen's father thinks that Owen has a reason to join the army, based upon family pride, but Owen has no motivation to join the army, nor would he conclude this from any deliberation. His father thinks that the reason for Owen to join the army is family pride. He thinks that Owen has this reason in an external sense, because he thinks that his son's lack of motivation does not make the reason statement false. Williams does not think an agent can come to believe he has a motivation to act without an internal reason statement becoming true of the individual, because the individual will now have the relevant motivation in his subjective motivational set. Williams does not think that if an individual deliberated rationally, whatever motivations he originally had, he would be motivated to act because there is no original motivation for him to deliberate from.


John McDowell considers the move from not being motivated by an external reason to being motivated by it that Williams thinks an external reasons theorist cannot explain (1995, 72). Williams pictures the external reasons theorist as thinking that an individual gains a new motivation by coming to believe the external reason statement through a process of correct deliberation; Williams says that 'the external reasons theorist essentially wants that the agent should acquire the motivation because he comes to believe the reason statement, and that he should do the latter, moreover, because, in some way, he is considering the matter aright' (1981, 108-9). Williams discounts this possibility because there is no existing motivation for the agent to deliberate from. McDowell questions why this move to gain the new motivation has to be due to correct deliberation. He also questions why Williams excludes scenarios such as being persuaded and 'swayed by reasons' (1995, 72). McDowell thinks that it is quite open as to how the individual becomes motivated; all that is important to the external reasons theorist 'is that in coming to believe the reason statement, the agent is coming to consider the matter aright' (1995, 73).


He continues to consider an individual who has not had a correct upbringing and thinks that in his case we do not have to think that there is some piece of reasoning we can use to start making him see things aright (1995, 73-4). He agrees here with Williams that in this case the individual will not gain this new motivation unless the reason appeals to something already in his motivational set. McDowell thinks that to get him to see matters aright would involve some sort of conversion, a possibility that he thinks is excluded by Williams' argument. He characterises such a conversion as a change in motivation not caused by reasoning practically from existing motivations to find an internal reason that was previously unknown. He thinks that in such a conversion, where someone ends up seeing matters correctly, an individual is being made aware of an external reason. McDowell thinks Williams rules out the possibility of a conversion because he assumes that the external reasons theorist demands that a change in motivation must derive from reasoning (1995, 75).


Problems with McDowell's argument

I think that there is something a little odd about the notion of conversion; there must be a very fine line between a legitimate conversion and brain-washing, hypnotism or drugging someone. McDowell characterises conversion as 'an intelligible shift in motivational orientation', citing a religious experience as an example of a comprehensible factor that would explain such a shift (1995, 74). It appears that brainwashing someone, giving him drugs to alter his personality or hypnotising him are equally comprehensible processes that could cause a change in his motivations. However, it is not clear that if as a result of one of these processes he, for example, became motivated to be helpful we would consider that he is becoming aware of an external reason, as it is unclear how this reason can be said to have been a reason for the agent before the motivational shift took place. I think that we would consider such processes manipulative and question the legitimacy of using such methods to induce someone to see matters aright. It is unclear how conversion could be defined in such a way that it is distinct from these other non-deliberative processes.


If conversion is allowed to be a legitimate route from your existing S to a reason for action, then it seems that anything could be a reason for action. This in itself would not be a problem for the externalist who wished to maintain that this condition was only necessary and not sufficient, but McDowell seems to want to claim, using his example, both that there is not a reason for all agents to listen to twelve-tone music and that the particular agent could only obtain this reason through conversion. This raises the question of why this particular agent is a candidate for conversion, but not all other agents. If all agents are possible candidates for conversion then any reason you can dream up, including a reason to listen to twelve-tone music, must be external reasons for all agents. If only certain agents are candidates for conversion, we need an explanation of why this is and I am not sure that this could be achieved without reference to the existing S of the agent, thus collapsing into an internal reasons claim.


Williams wonders whether McDowell means to bring in further qualifications that draw a link between taste in music and being well brought up (1995b, 191). This is suggested by McDowell's comments that 'considering the matter aright' has to occur within an ethical outlook and is 'not an argument that would somehow win over someone unmoved by what one wants to represent as external reasons' (1995, 80). The idea seems to be that an agent is within a certain outlook and has an associated S. Williams would argue that the agent's (internal) reasons for action are derived by deliberation from his S. McDowell's claim is that within this outlook there are (external) reasons that the agent cannot access by deliberation, but only by 'conversion'. It is difficult to see how an agent could have such an incomplete grasp of the outlook that conversion would be necessary for him to come to see matters aright. If the agent had a certain outlook, surely he would have access to existing relevant motivations to deliberate from?[1]


Take the example of being well brought up and taste in music and assume, as Williams suggests, that being well brought up entails having certain musical tastes. McDowell's claim seems to be that within the outlook of the well brought up person, it would be possible for an agent to miss the force of the argument that he has reason to listen to twelve-tone music; the only way for him to see this would be through conversion, not through deliberation. My question is: how could the agent be described as having the outlook of a well brought up person if he did not have the relevant reasons to do with taste in music that are part of such an outlook? It seems that either he has enough relevant factors within his S to be described as having the outlook, hence either has those reasons or has the motivation within his S to deliberate and recognise those (internal) reasons, so does not need conversion. Or he does not have enough relevant factors within his S to be described as having the outlook. In this case, it makes sense to think that he may need some form of conversion, but McDowell excludes individuals outside of the outlook from being subject to its reasons. I find this puzzling, as it is these people that are unmoved by such reasons that the external reasons theorist seems to want to target. However, if conversion was made a legitimate route to gain new reasons, it seems that anyone could be converted to believe any reason.


I think that McDowell has here identified a potential flaw in Williams' argument. It is not clear that the only way of acquiring, changing or losing desires, evaluative dispositions, emotions and loyalties is through deliberation from my existing set of desires, evaluative dispositions, emotions and loyalties. McDowell suggests that it is not important how the individual comes to see matters aright; this could occur through becoming habituated through a correct upbringing, conversion or being swayed through conversations with others. It seems true that these are all non-deliberative processes that could affect our desires, evaluative dispositions, emotions and loyalties. However, there are problems with McDowell's notion of limiting conversion to those with a certain moral outlook that seems to result in a type of internalism. However, this has raised the question as to whether the legitimate acquisition of new motivations is exhausted by the process of rational deliberation.


Developments of McDowell's argument

Elijah Millgram suggests an alternative to McDowell's story about people being converted to the correct way of seeing things as a method of seeing matters aright that does not involve deliberation (1996, 207). He thinks that he can provide an answer to Williams' rhetorical question, which says: 'What is it that one comes to believe when he comes to believe that there is a reason for him to φ, if it is not the proposition, or something that entails the proposition, that if he deliberated rationally, he would be motivated to act appropriately' (1981, 109). Millgram thinks that we can answer that we can gain a motivation to act in a certain way through experience. He thinks that this avoids the 'bolt-from-the-blue' feeling of McDowell's examples. He gives the example of Archie to illustrate how experience could give him a new motivation to act (1996, 204). Archie is insensitive and cannot appreciate the rewards of being sensitive, such as 'the flourishing of trust in his friendships, or the ability to respond sympathetically and effectively to others' concerns' (1996, 204). He has no desire for such rewards in his S and cannot understand why others value these rewards. However, Millgram thinks that if Archie were to experience these rewards, then he would see that being sensitive would improve his life, thus giving him a new motivation to act in such ways. We think that Archie has a reason to change his ways and act more sensitively, even though this reason is not grounded in his S.


Millgram considers whether you can object to his position on the grounds that the individual is ignorant of an element in his S. He gives the example of finding someone's habits irritating after moving in with them; an internal reasons theorist would argue that this disposition to find the habits irritating was always in his S, but he was ignorant of it. Millgram thinks this response is unsuccessful because of the internalist's appeal to the role of reasons in explanation. He gives the example of 'before time t my heart was set on X, and after experiencing X at t I wish I had never laid eyes on X' (1996, 208). He thinks that before time t only my desire for X can explain my action. I think that Williams would agree that the desire does explain his action X, but he was ignorant about an element in his S, so did not have a reason to do X. However, Millgram's point is that before time t there was no way of deliberating from his existing S to get to a reason to not X; this could only be achieved by experiencing X at t.


I think that Millgram is correct to be suspicious of conversion. It would appear to have some kind of 'magical force' and it must be possible that anyone could be converted to believe any reason. Experience is a weaker notion, but appears less problematic. However, it is not clear that appeal to experience can establish the existence of external reasons. There is a sense in which experience is external to the agent, in that what is experienced, or the object of experience, is external to the agent. However, without actually being experienced by the agent, the object of the experience is inert. In experiencing something, the agent brings to the situation their own perspective and perceives the situation in their own particular way, so the experience itself is internal to the agent who experiences it. What an agent experiences and how he can experience is, therefore, to some extent limited by his existing S, similar to the limits it places on his deliberation and imagination. For this reason, different individuals can learn different things from the same situation; the external and internal nature of experience appears inseparable. An interest in correct information and reasoning still holds, but there is more than one way of ensuring that you obtain this, through experience as well as deliberation.


What might Williams say about the idea that we learn and change our S through experience? He believes that an agent's subjective motivational set can be changed, but only through rational deliberation. What if it can also be changed by experience? What implications does this have for his argument? Would it mean that before time t you had no reason for action, but after time t, when your S has changed, you gain this reason for action, but through experience, not rational deliberation?


Millgram thinks that the internalist will say that we learn things from experience only because we have not exercised our imagination enough (1996, 214). He thinks Williams suggests that if the facts were clear and the bearing of the facts on S were made clear, then it would be evident what we wanted. Identification of what we want is only difficult because of the complexity of the facts, the complexity of each individual's S and the difficulty of linking the two. Millgram thinks that imagining what something is like is limited by our cognitive resources. He thinks that there are things you can only find out through experience, not imagination; we cannot fully imagine what some things would be like without experiencing them, so it will not always be clear what we want by imagining from our existing S. I agree that Williams is being over-optimistic about the extent of our imaginative capacities.


Conclusion

If experience and conversion are included as legitimate methods of gaining new motivations, then the possibility of external reasons is not excluded. We end up in a situation where anyone who could be converted, so potentially everyone, would have a reason to do particular things, no matter what their S. McDowell attempts to limit the scope of this externalist position to those who share a moral outlook, but I think that this argument is problematic. However, if we limit the argument to include only experience as a legitimate method of gaining new motivations, as Millgram suggests, I argue that this is no longer an externalist position. I think that a version of internalism can be maintained, because what can be experienced is limited by the existing S of the agent. Here I am adapting Williams's argument to include experience as well as deliberative reasoning as methods that can uncover internal reason statements. Thus the internalist argument would read: 'A has a reason to φ only if there is a sound deliberative or experiential route from A's subjective motivational set to A's φ-ing'.


I think that the experiential route would parallel the sound deliberative route introduced by Williams. Williams does not spell out exactly what he means by a 'sound deliberative route', but I think that it is clear that he does not mean to provide a full account of deliberation and means it to be a fairly flexible notion (2001, 91-2). Similarly, I do not think we need a full account of experience and that the notion can be flexible to a certain degree. The concept of someone having a reason for action would depend upon them being able to reach the conclusion that they have a reason to do a certain action by having a certain experience from the position of their actual subjective motivational set. Taking the example of Archie, he currently has no desire for the rewards of sensitive actions in his S. If Archie were to experience these rewards, then he would see that being sensitive would improve his life. There is no argument that can be made to make Archie deliberate and see this fact. However, the experience only enables him to recognise this fact because the experience combines with existing elements in his S, such as a desire to improve his life. Although the experience makes Archie aware of a reason to act sensitively, this reason is grounded in elements of his current S.


Both the sound deliberative route and the experiential route give a statement about an individual having a reason for action normative force. Williams argues that this is important because we need to be able to talk about what people should be disposed to do, not just what they are currently disposed to do (1995a, 36). Under the proposed view, what people should be disposed to do will not solely depend upon how they should deliberate but also on what they should experience. One aspect of a 'sound deliberative route' that Williams insists upon is that it is important to exclude the possibility of an agent having an internal reason based upon a false belief (1995a, 36). He thinks that we can write the requirement of correct information into the notion of a sound deliberative route because all rational deliberators require correct information to ensure that they deliberate successfully about their ends (1995a, 37). I think that we can also write the requirement of correct information into the notion of an experiential route because all rational agents have an interest in correct information and some information can only be gained through experience. This parallels Williams's claim that a desire for correct information can be assumed to be part of the S of any rational deliberator, because a rational deliberator has to be interested in correct information to ensure that he gets things right.


Lisa Grover
University of Kent
lh249@kent.ac.uk

Further reading

McDowell, J., 1995, 'Might There Be External Reasons?' in World, Mind and Ethics, eds. J.E.J. Altham and R. Harrison (Cambridge: Cambridge University Press): 68-85


Millgram, E., 1996, 'Williams' Argument Against External Reasons', Noûs, Vol. 30, No. 2: 197-220


Williams, B., 1981, 'Internal and External Reasons', Moral Luck (Cambridge: Cambridge University Press): 101-113


Williams, B., 1995a, 'Internal Reasons and the Obscurity of Blame', Making Sense of Humanity and Other Philosophical Papers (Cambridge: Cambridge University Press): 35-45


Williams. B., 1995b, 'Replies' in World, Mind and Ethics, eds. J.E.J. Altham and R. Harrison (Cambridge: Cambridge University Press): 186-194


Williams, B., 2001, 'Postscript: Some Further Notes on Internal and External Reasons' in Varieties of Practical Reasoning, ed. E. Millgram (Cambridge, Massachusetts: MIT Press): 91-97



Notes

[1]  It could be argued that if an agent had a certain outlook, he may not be able to access existing relevant motivations to deliberate from. For example, the agent may have an unconscious desire that he is unable to access that could nonetheless motivate his action. Williams considers that an agent may be ignorant of an element (D) in his subjective motivational set (1981, 103). However, he thinks that for the agent to have a reason to act there must be a rational, deliberative link between the action and D. He thinks that if D is unknown because it is an unconscious desire, there will not be a deliberative link, even though it may still provide the explanation as to why the agent did the action. In such a case, the agent had no reason to act, although his action can be explained.  [back]