Reasons for Action and Psychological Capacities
- First Online:
- Cite this article as:
- Lowry, R. Ethic Theory Moral Prac (2012) 15: 521. doi:10.1007/s10677-011-9307-6
- 108 Views
Most moral philosophers agree that if a moral agent is incapable of performing some act ф because of a physical incapacity, then they do not have a reason to ф. Most also claim that if an agent is incapable of ф-ing due to a psychological incapacity, brought about by, for example, an obsession or phobia, then this does not preclude them from having a reason to ф. This is because the ‘ought implies can’ principle is usually interpreted as a claim about physical, rather than psychological, capacities. In this paper I argue for an opposing view: if we don’t have reasons to do things that we are physically incapable of doing, then neither do we have reasons to do things we are psychologically incapable of doing. I also argue that extending the ‘ought implies can’ principle to psychological capacities makes the principle more attractive.
KeywordsReasons for actionCapacitiesOught implies canKorsgaardPettitSmith
Most moral philosophers agree that if a moral agent is incapable of performing some act ф because of a physical incapacity, they do not have a reason to ф. ‘Reasons’ in this paper refer to those considerations which can justify action, rather than considerations that merely explain action. Most moral philosophers believe that our reasons are constrained by our physical capacities because they believe that ‘ought implies can’ (OIC). For instance, most would agree that quadriplegics do not have reasons to run after burglars, precisely because quadriplegics are physically incapable of running after burglars. It is not however only those with physical incapacities whose condition prevents them from performing actions they would otherwise be capable of performing. People with phobias and obsessions are sometimes rendered psychologically incapable of performing some action, by these phobias or obsessions. I will take an agent to be psychologically incapable of ф-ing if the agent is incapable of getting herself to be motivated to ф. Interestingly, despite the general agreement among moral philosophers that moral agents physically incapable of ф-ing thereby lack reasons to ф, not many agree that moral agents psychologically incapable of ф-ing thereby lack reasons to ф. This is because OIC is usually interpreted as a claim about physical, rather than psychological, capacities. In Gert’s words, “When ‘ought to ф’ means, roughly ‘is rationally required to ф’, it simply cannot plausibly be taken to imply ‘is psychologically able to ф’.” (p.60).
It certainly seems possible that Å could desire that A ф in C and yet, due to A’s own incapacities—remember, we are not supposing that A herself has and exercises all of the capacities that ensure that her desires conform to the principles of reason, only that Å has and exercises these capacities—A might actually be. . . incapable of coming to believe that this is so. (2006, p.150)
In this paper I argue that if we accept the OIC doctrine, then we should reject this view represented by Korsgaard, Pettit and Smith. As the OIC doctrine has been argued for extensively elsewhere,2 I will not rehash these arguments here. I will confine myself to arguing that if one does accept the OIC doctrine then one should conclude that it applies to physical and psychological incapacities alike. If we accept the OIC doctrine, we should thus also believe that if, despite the appeals of others, I cannot get myself to cease obsessively washing my hands, I do not in fact have a reason to cease the hand washing. I will argue for this view by first exploring four arguments for distinguishing between physical incapacities and psychological incapacities when making reason attributions. I argue that none of these arguments can justify the claim that OIC applies only to physical capacities. The evaluation of these arguments is contained in the next section. Sections 3 and 4 argue that extending OIC to psychological capacities makes the OIC doctrine more attractive, and section 5 is devoted to rejecting one objection to this extension.
2 Does OIC Apply to Psychological Capacities?
In order to properly understand the claim that we should not distinguish between physical and psychological incapacities when making reason attributions, and ensure that there is no equivocation in the meaning of ‘capacity’ in this paper, more must be said about what it means for someone to be capable of doing something. Arguably, the best worked out account of capacity is offered by Michael Smith. Smith claims that a capacity consists in intrinsic qualities which will cause the agent to exercise this capacity in a whole raft of circumstances very similar to the agent’s actual circumstances. Consequently, when determining whether an agent has a capacity to, for example, answer a certain scientific question in circumstances C, we must investigate whether she answers in a raft of circumstances very similar to C.3 The similar circumstances will be ones in which the agent’s intrinsic qualities remain the same but she is unaffected by external interferences, such as stress or forgetfulness. If an examination of these counterfactual situations reveals that the agent systematically answers in these situations, this demonstrates that she has the capacity to answer, regardless of whether this capacity is utilised in the agent’s actual circumstances.4 According to this view of capacity, we are justified in claiming that a quadriplegic is incapable of running after the burglar, because in a whole raft of circumstances very similar to their actual circumstances, the quadriplegic does not run after the burglar.
One justification for distinguishing between psychological and physical incapacities when making reason attributions might reveal itself when we investigate the behaviour of those with phobias or obsessions. More specifically, we may be justified in making this distinction if, in circumstances very similar to the actual circumstances of the phobic and obsessive people, we find that such people are not crippled by their phobias and obsessions. Given Smith’s account of capacities outlined above, this would demonstrate that a phobic or obsessive person has the capacity to perform the action that they fail to perform in their actual circumstances. In order for it to be true that a phobic person has the capacity to perform the action that the phobia prevents her from performing in her actual circumstances, it must be true that in close possible situations (where things are slightly different, such as the agent has calmed herself, or taken medication), the agent will be doing what her phobia prevents her from doing in her actual circumstances. There may of course be many cases where a person merely has to practise her relaxation techniques, or take some medication, in order to perform the actions that she fails to perform when she does not take these measures. If this were always the case with phobic people, then we should indeed think that phobic people are not incapable of performing the actions with which their phobias are associated. However, it seems unlikely that this will typically be true. We know from experience that some phobias and obsessions are more persistent; they cannot be overcome by merely popping a pill, or practising relaxation techniques. In these cases, an agent afflicted with a phobia or obsession will still be afflicted in similar situations where she has taken medication, practised relaxation techniques etc. This first argument is consequently unconvincing; we need to look elsewhere if we are going to claim that we should distinguish between physical incapacities and psychological incapacities when making reason attributions
Suppose that we accept that some agents with phobias and obsessions are rendered incapable of performing actions according to the account of capacity outlined above. We may still argue that unlike agents with physical incapacities, agents with phobias and obsessions are not rendered incapable in any fundamental sense, of performing actions. And this fundamental sense, it may be argued, is not captured by Smith’s picture of capacities. Rather, it ignores an important difference between an incapacity caused by our physical state and an incapacity caused by our psychological state.
What could this fundamental sense of incapacity be? How could we justify the claim that an incapacity caused by our physical state is a kind of fundamental incapacity, while denying that an incapacity caused by our psychological state is a fundamental incapacity? The following observation may offer some support. Consider an acrophobe who is terrified of standing higher than 2 m above ground. It seems that even if an acrophobe does not perform actions that expose her to heights in a whole raft of similar circumstances to those that she is in, there may be some rare and quite unusual circumstances where she will perform these actions. For example, suppose that under normal circumstances she would never be able to get herself to walk onto the balcony of a 10 story building. Tonight however, she is sleep-walking in a hotel and in an attempt to find the bathroom, she quite unwittingly takes the lift to the 10th floor and walks onto the balcony of this 10 story hotel. While this scenario is doubtless an unusual one, it may happen in certain rare circumstances. In contrast, it is difficult to imagine any realistic circumstances in which a quadriplegic would run after the burglar. So it seems there are some rare, perhaps “fluky”, circumstances where acrophobia will not impede the acrophobe from performing an action that her acrophobia normally prevents her from performing. On the other hand, it seems far more unlikely that a quadriplegic will perform an action that his quadriplegia normally prevents him from performing. Another way of putting this point is that the notion of capacity outlined above does not take the possibility of ‘flukes’ to be relevant to capacities, and we might argue that the possibility of a fluke points to an important difference between incapacities caused by our physical state and incapacities caused by our psychological state.
In order for this discussion of flukes to be relevant to the claim that we should distinguish between physical and psychological incapacities when making reason attributions, we must also make the further claim that this difference between incapacities caused by our physical state and incapacities caused by our psychological state is relevant to reason ascriptions. More specifically we will need to argue that the fundamental difference between incapacities caused by our physical state and incapacities caused by our psychological state translates to a fundamental difference in the way these incapacities affect reason ascriptions.
The necessary form of ‘ought-implies-can’ is not available in the domain of rationality. When ‘ought to ф’ means, roughly ‘is rationally required to ф’, it simply cannot plausibly be taken to imply ‘is psychologically able to ф.’ Some people suffer from mental illnesses (addictions, phobias, etc.) that essentially compel them to do things that they rationally ought not do, or prohibit them from doing things that they rationally ought to do: their irrational actions are precisely the result of the fact that they are psychologically unable to act otherwise. Perhaps such mental illnesses erase the moral ‘ought,’ but they leave the rational ‘ought’ intact. (2004, p.60)
But why should we think that Gert is right about this? Why should it matter to the reasons we have whether people are prohibited from performing actions as a result of physical incapacities or as a result of addictions, phobias etc.? Would the possibility of flukes offer a justification for thinking that incapacities caused by addictions, phobias etc. do not affect reason ascriptions? It is hard to see why the possibility of a fluke would imply that an agent has a reason, if in the absence of this possibility she would not have the reason. Accepting this view would mean that we would have to interpret the OIC principle as the restriction that one ‘ought to ϕ if there is at least a possibility that one may ϕ by fluke’. This interpretation fails to capture the intuition behind the OIC principle. For instance, suppose we accept the OIC principle, and our friend is in a coma. We know that even the prognosis for partial recovery is poor. There is however a very small chance that at any moment he might emerge from his coma without any permanent damage. It seems that according to the interpretation of the OIC principle under consideration here, the comatose patient has a reason to do things, such as read a book or phone his family, precisely because there is the ‘fluky’ possibility that he may awake at any moment and then do such things. But isn’t this exactly the sort of case where we want the OIC principle to tell us that the comatose patient doesn’t have a reason to phone his family, or read a book? We want the OIC principle to imply that the comatose patient has no such reasons because on any normal understanding of what he can and can’t do, it seems that he cannot make a phone call or read a book. The fact that there is a very small possibility that he may wake and then make a phone call or read a book, does not seem to affect this judgment.
A further problem with understanding the OIC principle as ‘ought to ϕ if there is at least a possibility that one may ϕ by fluke’, is that it also seems to conflict with what we may call the ‘motivational requirement’ on reasons. This is the requirement that we must be able to act on our reasons, because they are reasons. Bernard Williams describes this requirement as follows: “If there are reasons for action, it must be that people sometimes act for those reasons, and if they do, their reasons must figure in some correct explanation of their action.”5 Another way of expressing this requirement would be to say that if an agent is to have a reason, the reason must be able to be that agent’s reason for acting on a particular occasion. To see what this requirement implies, suppose that the acrophobe has a reason to venture onto the 10 story balcony in order to help a friend in need. We know that she can only do so in the case of a fluke like the sleep-walking incident described above. More particularly, the only way the acrophobe can walk onto the balcony is when she does so unwittingly. Consequently, the motivational requirement on reasons will not be satisfied in this case. It will not be satisfied because this requirement demands that the acrophobe be able to venture onto the balcony because her friend needs her help—and this seems to require the acrophobe entering the balcony wittingly. In summary then, it is unlikely that flukes can provide a good justification for thinking that unlike physical incapacities, psychological incapacities do not affect reason ascriptions. Such a justification would fail to capture the intuition behind the OIC principle, and be incompatible with the motivational requirement.
Another way to argue that physical incapacities and psychological incapacities affect reason ascriptions differently would be to straightforwardly deny that reasons relate to psychological capacities. But this way of arguing doesn’t seem promising either; psychological capacities are clearly important to the reasons one has. If ‘ought implies can’ merely implied that an agent must be physically capable of ф-ing in order to have a reason to ф, and not that an agent must be psychologically capable of ф-ing, a bird physically capable of tapping on the window with its beak in order to warn me of danger, may have a reason to do so. But this is clearly implausible. This point will be returned to in Section 4.
A has reason to ф in certain circumstances C, we might say, only if Å desires A to ф in C, where Å has and exercises all of the capacities that ensure that her desires conform to principles of reason. Otherwise put, A has a reason to ф in certain circumstances C only if it would be good-from-the-standpoint-of-reason for A to do so. (Pettit and Smith 2006, 167.)
Suppose that A is so depressed that she not only has no desire whatsoever to improve her life, but that her belief that Å desires that she (A) improves her life leaves her completely cold. The depression incapacitates her. It removes her capacity for self control: stops her being able to respond, appropriately, to the incoherence involved in her having the belief but lacking the desire. Does A have a reason to improve her life in these circumstances none the less? (2006, 156)
The external reasons theorist might insist that, in a sense, she does, simply in virtue of the fact that Å, who suffers no such incapacity, does want A to improve her life. (2006, 156)
So here Pettit and Smith explicitly state that an agent who cannot respond to her putative reasons because of her lack of capacity for self control may still have these reasons, simply because her ideal advisor might desire her to act on such ‘reasons’. But this response just pushes the question one step back: instead of asking why we should think that the difference between physical incapacities and psychological incapacities is a difference that is relevant to what we have reason to do, we can ask: why should we think that an ideally rational version of oneself would want their actual self to ф even if they were psychologically incapable of ф-ing, whereas they would not want their actual self to ф if they were physically incapable of ф-ing? Without a good argument for thinking that only physical incapacities affect reason ascriptions, it is hard to see why an ideally rational version of oneself would have these desires.
So why should we think that OIC applies only to physical incapacities? If we accept OIC, why should it matter for reason ascriptions whether it is a physical incapacity, or a psychological incapacity that prevents one from performing an action in rafts of circumstances similar to those they are in? I believe that there are no good answers to these questions. We should reject the claim that physical incapacities, but not psychological incapacities, affect reason ascriptions. The next section argues that understanding OIC as applying to psychological capacities makes OIC more attractive than if it is thought to only apply to physical capacities.
3 The Purpose of Reasons
Whatever responsibility is, considered as a metaphysical state, unless we can tie it to some recognisable social desideratum, it will have no rational claim on our esteem. Why would anyone care whether or not he had the property of responsibility (for some particular deed, or in general)? (1984, p.163)
Similarly, it seems that reasons that cannot influence or guide, will not be tied to any recognisable social desideratum. So we might also ask here: why would anyone care whether or not an agent has such a reason?6
If a person couldn’t have a reason to do what she cannot do, and if the impossibility of ф-ing would eliminate the reason for doing so, the derivative reasons would be eliminated as well. But in many cases it seems that they are not. But if there are derivative reasons relating to the reason to ф, even if one cannot ф, there must be a non-derivative reason for ф-ing as well, and that is, there is, in those cases, a reason to do what one cannot do.(2010, p. 236)
However, it seems that we can explain what Heuer refers to as ‘derivative reasons’ without claiming that these reasons must be derived from other reasons. A reason which relates to ф-ing needn’t be derived from a reason to ф. It may instead be derived from the value of ф-ing. For instance, if there is nothing I can do to cease obsessively washing my hands, I may still have reason to do something that I can get myself to do; something related to the value of ceasing this activity, such as scheduling an appointment with a psychiatrist. This reason may be derived from the value involved in ceasing to obsessively wash my hands, even though, due to my incapacities, I have no reason to cease this activity. Non-derivative reasons to which we are impervious are not needed to explain derivative reasons.
The second response might be given by someone objecting that we are overlooking a different practical defence available for those reasons to which we are impervious. In particular, someone might argue that our psychological capacities can be influenced and that discussing reasons may be an effective means of doing this. For instance, a mother could continue to try to persuade her son to recognise certain reasons she believes to exist. In so doing, her son may develop the capacity to see these reasons. It seems however that this consideration does not provide a role for reasons beyond that which can be achieved through talk of values and good circumstances. It would be just as useful for a parent to talk of the value of certain pursuits or ways of life; nothing is gained by conducting this appeal in terms of reasons.
In summary, if we interpret OIC as applying to psychological capacities as well as physical capacities, then we only have reasons to which we are not impervious. The reasons implied by this view therefore serve a practical purpose, for they are reasons that can influence, guide, deter and encourage the agent for whom they are reasons. Such reasons are therefore vindicated through the role that they play in society and our lives.
4 Explanatory Power
Another advantage of understanding OIC as applying to psychological capacities, as well as physical capacities, is that this gives us the appropriate explanation for why inappropriate candidates for (justificatory) reasons, such as birds and cats, don’t have reasons. We don’t claim that birds and cats have reasons to do things such as warn us of burglars, because while they may be physically capable of doing so, they lack the relevant psychological capacities needed for acting on such ‘reasons’.
Normative concepts exist because human beings have normative problems. And we have normative problems because we are self-conscious rational animals, capable of reflection about what we ought to believe and to do. . . Normative concepts like right, good, obligation, and reason are our names for the solutions to normative problems, for what it is we are looking for when we face them. (1996, p.46–47)
By drawing attention to why normative concepts exist, this discussion explains why we do not ascribe reasons to agents such as birds and cats. Note however that the notion of agents’ psychological capacities is crucial to this explanation. Korsgaard claims in this passage that the types of being that are eligible for having reasons are agents who are ‘capable of reflection about what we ought to believe and to do’. According to this view too, the exclusion of birds and cats from the set of suitable candidates for reasons is explained via an appeal to agents’ psychological capacities.
There is another way to explain why we do not ascribe reasons to agents such as birds and cats. This explanation appeals to the notion of biological normality. It might be claimed that agents have reasons when they are the type of being, which, when normal, has certain attributes such as rationality, autonomy and self-consciousness. According to this explanation, birds and babies do not have reasons because, when normal, they do not have such attributes. The explanatory power of such a response is limited however. To see this, consider Streumer’s example. Fred, who is completely and irreversibly paralysed, hears someone screaming for help outside his house. According to the ‘biological normality’ explanation, Fred may have a reason to go outside and help this person, even though he cannot help the person. As Streumer points out however, the idea that Fred has such a reason is implausible.7
In summary, if we understand OIC as applying to psychological and physical capacities, rather than merely physical capacities, then reasons serve a practical purpose. Understanding OIC in this way also offers a good explanation for why inappropriate candidates for reasons, don’t have reasons. In the next section I deal with a problem that we might think understanding OIC in this way introduces: the concern that this interpretation of OIC debars us from explaining the defective behaviour of agents who cannot ‘see reason’.
5 An Objection
Suppose we are convinced that we have to accept OIC because of the implausibility of claiming that quadriplegics have reasons to run after burglars. We might worry however that accepting a version of OIC that applies to psychological capacities prevents us from offering the appropriate explanation of certain defective behaviour. For instance, I suggested at the start of this paper that if we accept OIC and I am incapable of motivating myself to cease to obsessively wash my hands, then we should conclude that I do not have a reason to cease my obsessive hand washing. Alternatively, if we accept OIC but deny that it applies to psychological capacities, we may explain my behaviour as defective on the basis that I am irresponsive to my reason. If we accept a version of OIC that applies to psychological capacities we cannot claim that I am irresponsive to my reason because this version of OIC implies that I lack this reason.8 We may think that this interpretation of OIC thus prevents us from offering the appropriate explanation of my behaviour.
While rationality requires us ‘to do our very best’ this is a matter of ‘our best’. It is not ‘the best that can be done’ as distinct from ‘the best that can be done by us, given the capacities and resources at our disposal’. (1988, p.8)
The idea is that rationality places demands on people relative to their circumstances. An agent’s circumstances are construed broadly enough to include not just the information available but also the agent’s capacity to digest this information in a meaningful way. According to this view, rationality is relative to our capacities. Rationality “does not make demands beyond the limits of what is genuinely possible for us—it does not require accomplishments beyond the limits of the possible.”10 Our capacities will determine what is possible for us. If we lack certain capacities, we will not be subject to the rational demands associated with those capacities, nor will we qualify as a fully rational being.
Perhaps the most plausible prerequisite for full rationality, or for being rationality-assessable, is an agent’s capacity to recognise and respond to reasons.11 The absence of this capacity seems to be a large part of that which sets the functioning of fully rational agents apart from the functioning of less rational agents, such as birds and babies. If we are right to categorise birds and babies as arational on the basis that they do not have the capacity to recognise and respond to ‘reasons’, we should also view humans to be arational when they lack these same requisite capacities for full rationality (at least to the extent that they lack these capacities and at least with regard to those considerations which would count as reasons if they had the relevant capacities). With regard to the hand washing example, this discussion implies that we should view my irresponsiveness to rational argument as evidence that I am an arational agent. Moreover, the appropriate explanation of my behaviour will be one that describes my dysfunctional behaviour as a symptom of an absent prerequisite for full rationality.
In summary, interpreting OIC as applying to psychological capacities, does not debar us from describing the defective aspects of my obsessive hand washing behaviour. Rather, it coheres with the appropriate explanation: I am arational with regard to my hand washing obsession and thus fail to be rationality-assessable in this matter. We can thus dismiss this objection; the inability to claim that I am irresponsive to my reason for ceasing hand washing does not force us towards any counterintuitive conclusions. Instead, my arationality with regard to my hand washing obsession explains why I lack a reason to cease the hand washing.
In conclusion, I have argued that if one accepts the OIC doctrine then one should conclude that it applies to physical and psychological incapacities alike. I argued for this by exploring four arguments for distinguishing between physical incapacities and psychological incapacities when making reason attributions. I argued that none of these arguments can justify the claim that OIC applies only to physical capacities. I also gave two positive arguments for the claim that extending OIC to psychological capacities makes the OIC doctrine more attractive. The first argument addressed the practical merits afforded by this interpretation of OIC. The second argument demonstrated that interpreting OIC as applying to psychological capacities offers an appropriate explanation for why inappropriate candidates for reasons, such as birds and babies, don’t have reasons. Additionally, I discussed an objection that might be made in response to interpreting OIC as applying to psychological and physical capacities. This was the worry that such an account removes the tools required for the appropriate explanation of certain defective behaviour. The objection was not problematic. Contrary to the view commonly held then, we should conclude that if ‘ought’ implies ‘can’, then ‘ought’ implies ‘psychologically can’.
The similar circumstances or possible worlds which we need to investigate are ones which share a similarity of a certain kind. Possible worlds that are similar to the actual world except that an agent has different intrinsic qualities which cause her to answer the question, are not worlds that establish the agent’s capacity to answer. Instead, we need to examine those worlds where we hold constant the qualities that are both intrinsic to the agent and relevant to her answering the scientific question. These worlds will establish whether the agent has the capacity to answer the question. For further discussion see: Smith 2004, and Lowry 2011.
What would count in favour of a practice of ascribing reasons to agents who are quite impervious to such reasons? It may be thought that the universality of reasons pushes in that direction. Surely anyone has reason to thank a person who has helped them. True—but does “anyone” include Tom? If this seems evasive remember that “anyone” plainly does not include the wounded bird, or the cat. Whether it includes Tom is precisely the question. (2007, p.90)
We might also characterise rational agents in the same way as Nozick (1993) and Audi (2004): as agents who are responsive to reasons. If rational agents are characterised in this way, a theory which denies that reasons are sensitive to psychological capacities can accommodate the claim that my hand-washing behaviour is irrational. However, a theory which accepts that reasons are sensitive to psychological capacities cannot take my behaviour to be irrational on the grounds that I am irresponsive to my reason, because the theory implies that I have no such reason.
‘Rationality-assessable’ is not the term used by Rescher.
While Rescher suggests that the relevant capacities may include a capacity of choice and a capacity to implement one’s choices in action, I am making a weaker suggestion here. I am suggesting that the more primitive capacity to recognise and respond to reasons, may be a prerequisite for being rationality-assessable. For Rescher’s discussion see Rescher (1988, p. 11).