Wednesday, 7 June 2017

Autonomy and Beneficence


In this posting I want to investigate what we mean by autonomy and the relationship between autonomy and beneficence. I will firstly examine two different accounts of autonomy. In order to do so I will briefly outline the differences between a content neutral account of autonomy and a substantive one. I will then raise some difficulties with accepting a substantive account of autonomy. Next I will examine the relationship between a content neutral account of autonomy and acting beneficently. I will conclude that preference should be given to respecting autonomy over acting beneficently when these two values clash. I will then consider what specifically makes rape and slavery so wrong to support my conclusion. Lastly I will examine the implications of accepting this conclusion for the doctrine of informed consent, the age at which someone should be able to vote and the right of the disabled to make their own decisions.

John Stuart Mill defined the only way power can be rightfully exercised over another.

“The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant.” (1)

Mill’s definition can be used to provide a definition of a content neutral account of autonomy. An autonomous person should always be free to exercise his will freely, provided this exercise doesn’t harm others. Such a definition of autonomy might be classed as a primitive or basic account. In what follows a basic account of autonomy will refer to a Millian account It might be suggested that advances in technology and medicine mean that such a basic account is an outdated one. It might be further suggested that these advances mean a more substantial account of autonomy is required. A substantive account. A substantive account of autonomy places some constraints on what an autonomous person can autonomously choose even when his choices don’t harm others. For someone’s decision to be autonomous it must accord with certain accepted norms in addition to the norm of not harming others.

I now want to present two arguments against adopting a substantive account of autonomy. My first argument will suggest that adopting a substantive account of autonomy would mean that autonomy becomes a superfluous concept. Let us consider a substantive account of autonomy in which someone’s decisions are only accepted as autonomous ones provided they would be regarded as reasonable by most reasonable people. It might be objected it would be hard to define which people are reasonable and what such reasonable people might find to be reasonable. For the sake of argument let us ignore this objection. Why does autonomy matter? It matters because it differentiates between those decisions we should always respect and others. Let us accept that an autonomous choice should be a reasonable one which would be accepted by most reasonable people. In this situation it seems that to talk of respect for autonomy becomes mere rhetoric. In this situation the concept of autonomy is doing nothing useful. Someone wanting to know which decisions he must respect, by deciding if they are autonomous decisions must first know that which decisions would be accepted by all reasonable people. I would argue that this is all he needs to know in order to respect someone’s autonomy. In this situation we simply don’t need the concept of autonomy. I would suggest the same argument can be applied to any other norms which might be applied in any other substantive account of autonomy such as, someone’s best interests, acceptability to society or giving no offence to religious standards. It appears to follow that those who advocate the need for a more sophisticated account of autonomy, a substantive one, make the concept of autonomy a redundant one.

I have argued above that respecting someone’s autonomy cannot simply mean acting to further someone’s best interests because if we do so the idea of respecting autonomy seems to do no useful work. I might act in my dog’s best interests but this doesn’t mean I respect him or believe he is an autonomous dog. Caring about someone or someone doesn’t seem to be the same as respecting autonomy. What then is meant by respect? Respecting someone doesn’t simply mean saying nice things about him for if this was so there would seem to be no difference between respect and flattery. I would suggest respecting someone involves admiration and that admiration is linked to valuing. For instance, if I respect someone I might do so because I admire his honesty, determination or his ability to make good decisions. I can admire someone because he is a certain kind of person but what qualities do I admire and value if I admire someone simply as person? What is a person? Christine Korsgaard argues that a person is not identical to the human being the person supervenes on. She suggests that,

“When you deliberately decide what sort of effects you will bring about in the world, you are deliberately deciding what sort of cause you will be. And that means you are deciding who you are.” (2)

It would be difficult to call anyone who was unable to make any decisions a person. However, whilst the ability to make decisions is a necessary condition for personhood it isn’t a sufficient one. Some who makes all his decisions randomly or based on mere whims might be regarded as a wanton according to Frankfurt (3). To be a person someone must have the capacity to make decisions based on what he cares about or values. What does the above tell us about respecting someone simply as a person? It would seem that if respect involves admiration then respect for a person involves admiration for a creature which can make his own decisions based on what he cares about. I have suggested that respect, admiration and valuing are linked. I would now suggest that respecting a person requires accepting his decisions. If we don’t do so, our supposed admiration and valuing of him as the kind of creature who can make his own decisions based on what he values, becomes mere rhetoric.

A defender of a substantive account of autonomy might object to the above and argue that we can respect someone’s autonomy by respecting him as the sort of creature that can make some of his own decisions. He might proceed to argue we need only accept those of his decisions which don’t harm his best interests. He might suggest that by doing so we then are still respecting his autonomy, we are just according it less importance. I would reject such a suggestion and. will now present two arguments to support my rejection.

Firstly, let us assume that we can respect someone’s autonomy by only accepting those of his decisions which are in his best interests. It might then be argued that provided we do so we are still respecting him as a person. But if we do so are we respecting someone simply as a person or a particular kind of person? For instance, I can respect someone simply as a person whilst at the same time failing to respect him as a particular kind of person.  I believe for instance that he is a bad parent. It seems if we respect someone’s autonomy, by only accepting those of his decisions which are in his best interests, that we are respecting a particular kind of person. We are only respecting those persons who make good decisions. Alternatively, we might only respect someone when he makes good decisions. Does this matter? Let us assume we only respect the autonomy of those people who make good decisions and that we should adopt a beneficent attitude to those who don’t. I have argued above that what defines someone simply as a person is his ability to make his own decisions, to shape his life. It appears if we only respect the autonomy of people who make good decisions that we fail to recognise some people simply as persons. Next let us assume that we only respect someone’s autonomy when he makes what we regard as good decisions. If we do so I can employ the argument used above and question whether respect for autonomy really does any useful work. It follows if we respect peoples’ autonomy by only accepting those of their decisions which we think are in their best interests that either we won’t be respecting some people simply as persons or we are only respecting people as a part time persons.

I now want to argue that if we respect someone’s autonomy by only accepting those of his decisions which we believe to be in his best interests that we aren’t acting in a fully caring way. It might be objected that we are only acting in this way because we really do care about people. In response I would suggest that in this situation because we decide what is in someone’s best interests we might be accused of epistemic arrogance. However, let us lay this suggestion aside and assume that respecting someone’s autonomy in this way doesn’t mean we are exhibiting epistemic arrogance. I would still suggest that this form caring is a deficient form. I accept if we act in such a way we are acting sympathetically but I would argue we aren’t acting empathically. True empathic care means we must care about what someone cares about rather than simply what we believe to be in his best interests. Someone might suggest sympathetic caring is as good as empathic caring. I would reject such a suggestion. I can care about someone or something sympathetically simply because I want him to flourish. This is the way someone might care for a dog he loves. Empathic caring isn’t so simple. If I care about someone empathically I must care about what he cares about in addition to what I believe are in his best interests. Empathic caring is a more complicated way of caring than caring based on sympathy. However, because something is more complicated doesn’t automatically mean it is better. People don’t want to be cared for in the same way as dogs. But why, surely it’s good to be loved, cherished and beneficently cared for? People don’t want to be cared for in the same way as pets because they value being recognised as persons which requires recognising them as the kind of creatures who can decide their own future. It follows if we care about people as persons we must care about what they care about and this requires caring about them in an empathic way.

Even if the above is accepted an objector might argue that if we care about someone empathically that whilst we must always care about what he cares about, in some situations we should give priority to acting beneficently. This argument supposes a particular concept of beneficence. This concept holds that to act beneficently is to act in someone’s best interests. It also holds that to act beneficently towards someone doesn’t always means acting in what he perceives to be in his best interests. This means we must act in accordance with some accepted standard, perhaps a standard that most reasonable people would accept. But if we act beneficently in this way who are we acting beneficently towards? We are certainly acting beneficently to human beings but towards persons? We are acting as if someone can be a part time person. It might be objected that there can’t be such a thing as part time person. I find this objection unconvincing. Children can make some decisions for themselves whilst their parents make others in their best interests. Children might be regarded as part time persons. Nonetheless I would suggest that most adults don’t want to be part time persons being a person is central to them. Perhaps this is one reason why children want to grow up. Being a person is central to most people’s interests. Can we be said to be acting truly beneficently towards someone if we are prepared to ignore what he perceives to be central to his interests? I would suggest we can’t. If we accept the above, it follows that acting truly beneficently requires acting in accordance with someone’s perceived best interests and not what we perceive to be in his best interests. It further follows if we act in a way that serves someone’s best interests, as we see them, that we are acting in a caring way, however our caring even if well intentioned is an incomplete form of caring.

At this point a further objection might be raised. It might be suggested that I’m presenting a misleading view of substantive autonomy. A substantive account of autonomy might be better defined as an account that places some constraints on what an autonomous person can choose, even when his choices don’t harm others, in some limited circumstances but in all other circumstances we should respect his choices. My objector might agree that in the past a basic account of autonomy was sufficient to protect our freedoms. He might now suggest that technological progress and modern medicine mean we have a need for a more sophisticated account of autonomy and that a substantive account satisfies this need. I have questioned above whether any substantive account of autonomy is actually an account of autonomy. I would suggest that any such proposed account is in reality an account of how to balance respect and caring about someone. How to balance respecting autonomy and acting beneficently. My objector might suggest that it is perfectly legitimate to balance these two. In response I would argue that whilst someone might well have a legitimate aim of respecting autonomy and acting beneficently when these two values don’t clash that this clash. A clash of these values depends on a particular account of beneficence. To act beneficently according to this account is to act in someone’s best interests and this doesn’t always means acting in what he perceives to be in his best interests. I have argued above that acting in this way is a deficient form of beneficence and is an incomplete form.

I have argued that we should reject a substantive account of autonomy for two main reasons. Firstly, if we adopt a substantive account of autonomy this account makes itself redundant. Secondly, if we adopt such an account we are not acting in a truly beneficent or caring way. Accepting the above means we must always accept someone’s basic autonomous decisions. It also means we cannot ignore such decisions or coerce someone into changing such a decision. Accepting the above also means we must sometimes accept bad decisions. Autonomous decisions needn’t be good decisions. In such cases we should attempt to persuade the decision maker to change his decision when it is unwise, however if our persuasion fails we must be prepared to accept the decision.

I now want to consider what’s wrong with slavery. It might be argued that the wrongness of slavery is self-evident. Slaves are abused and cruelly treated. However, R M Hare (4) used a thought experiment to show this need not always apply. He imagined an island called Juba which was ruled by a benevolent elite for the good of all with no abuse or cruel punishments. He also imagined an island called Camaica on which everyone was free but all lived in abject poverty. He speculated that some free citizens of Camaica might prefer to be slaves on Juba. If we accept such a situation is possible, even if unlikely, and we believe slavery is wrong what reasons can we advance for this wrongness. What is wrong is that the slaves on Juba are not regarded as the kind of creatures who can determine their own future and this harms them because as I have argued above for any person the ability to determine his own future is central to his interests. However, I will now argue that the concept of autonomy violated is our basic concept of autonomy. Is it conceivable that the substantive autonomy of the slaves on Juba could be respected? A substantive account of autonomy might allow a slave’s decisions to be accepted as autonomous ones and respected provided they are in his best interests and any decision a slave makes which aren’t in his best interests aren’t regarded as autonomous ones. If we accept a substantive account of autonomy, then the autonomy of the slaves on Juba would be respected. The slaves on Juba would be treated as children or part time persons. Wasn’t colonialism a bit like this? Beneficent colonialism was a bit like Hare’s imaginary Juba and treated the people colonised as children or only part time persons.

Let us now explore the wrongness of rape using a thought experiment similar to that of Hare. Let us consider a gentle rapist and a compliant victim. The physical harms caused by such a gentle rape are minimal nonetheless the crime doesn’t seem to be a minor one to us. What reasons can be advanced for the seriousness of a psychically gentle rape? It might be pointed out that the harm lies not the violence inflicted but the threat of violence, the violation of bodily integrity or both of these harms. I accept these points. Let us consider the violation of bodily integrity first. The simple fact that the victim’s body was penetrated is irrelevant, this could occur during consensual intercourse. What matters was that her body was penetrated against her will and this involves failing to respect her autonomy. Let us now consider the threat of physical harm causing psychological. Let us assume the victim is aware that she will not be harmed provided she complies. She complies and is raped. She isn’t psychically harmed and because she complied moreover she had no reason to fear psychical harm so any psychological harm is not due to fear of being psychically harmed. In spite of this I would argue psychological harm occurs. It occurs because she isn’t seen as the kind of creature who has a right to decide what to do with her own body, she isn’t considered as a person, her basic autonomy isn’t respected.

If we accept a non-substantive or basic account of autonomy as the only meaningful account of autonomy what implications does this have for the doctrine of informed consent? Is the doctrine of informed consent based on respect for autonomy? If the doctrine of informed consent is based on a substantive account of autonomy, then I would suggest the doctrine isn’t actually based on respecting autonomy for the reasons given above. In this situation the doctrine of informed consent is concerned with balancing acting beneficently and respecting autonomy. The concern is to stop people making bad decisions rather than respecting autonomous ones. This balancing act assumes beneficent care means acting in someone’s best interests as seen from a particular vantage point, perhaps what most reasonable people would consider to be in someone’s best interests. I have argued above such a concept of acting beneficently is an incomplete concept and that true beneficence requires always accepting basic autonomous decisions. Autonomous decisions don’t have to be good decisions. However autonomous decisions are not made randomly or based on mere whims. Autonomous decisions are based on what we care about, based on what matters to them.

Accepting the above has important implications and I will now briefly examine three of these. We might divorce the doctrine of informed consent from respecting autonomy and simply say that the doctrine is concerned with furthering patients’ best. This would be an honest approach. However, if we do so when we ask patients for their consent are we really asking for consent or acquiescence? Alternatively, we might accept that the doctrine of informed consent is based on respect for basic autonomy. If we do so it seems to me that a patient can make an autonomous decision simply to trust his doctor’s advice, after all we trust lawyers, accountants and other professionals all the time. It also seems that the information needed to make a basic autonomous decision is less than that currently supplied when taking informed consent. This might have more to do with a fear of litigation rather than a misguided concept of autonomy, see montgomery and the information needed for informed consent . The information required for informed consent should be patient driven and be determined by how much information he needs and wants to make an autonomous decision. Secondly democracy depends on voters’ ability to make an autonomous decision. If we accept a basic concept of autonomy, then perhaps the voting aged should be lowered. Perhaps it should be lowered to the age needed to give sexual consent. Lastly the United Nations convention on the rights of persons with disabilities want more people with cognitive and psychosocial disabilities to make their own decisions, see United Nations . Let us accept that an autonomous person has the right to make his own decisions. It follows that how many people with cognitive and psychosocial disabilities should be able to make these decisions depends on the concept of autonomy employed. If as I have suggested a basic concept is employed then more disabled people should be able to make their own decisions as an autonomous decision is not the same as a good decision. The emphasis should be on helping such people make good decisions rather than making good decisions on their behalf.



  1. John Stuart Mill, 1974, On Liberty, Penguin, page 69
  2. Christine Korsgaard, 2009, Self-Constitution, Oxford University Press, page 1
  3. Harry Frankfurt, 1999, Necessity, Volition, and Love. Cambridge, page 114
  4. R. M. Hare, 1978, What is Wrong with Slavery, Philosophy and Public Affairs 8.
  5. Matthew Burch, 2017, Autonomy, Respect and the Rights of Persons with Disabilities in Crisis, Journal of Applied Philosophy, Vol 34(3) 

No comments:

Engaging with Robots

  In an interesting paper Sven Nyholm considers some of the implications of controlling robots. I use the idea of control to ask a different...