Jeff McMahan

This book made me think, as can be inferred from my approximately 180 notes and notably more highlights, 20 hours to read and at least as much time thinking about the ideas, and many short essays, at least half of which were arguing with the author.

Despite my differences with McMahan, I give this book five stars because it made me a better thinker, about both the ethics of killing on the margins of life and about ethics in general. One thing you should know if you're going to read this book is that it is dense. I found myself reading it incredibly slowly. Often I would have to a sentence, think about it for a minute, and read it again because the ideas were packed in so tightly. It's worth the struggle. Given the topic of this book, it can also sometimes be hard to read if the topics have personal relevance. While I read this book, I lost one pregnancy and had a somewhat challenging pregnancy with another. I had to take multi-month breaks because I just could not emotionally handle reading about life and death when worrying about the life inside of me.

Here is a brief summary of the book, and it's only the shallowest of skims into the ideas discussed. Also, I almost certainly get some of the ideas wrong. They're complicated :-)

The first section is on identity. What is the essence of what we really are? McMahan calls this being a person, although he acknowledges that this particular term can bring up associations that muddle our thinking (e.g., the assumption that a person is human). There is a long discussion of the soul which dismisses it as the essence of what we are. Although the idea of a soul has relevance for those who believe in it, there is no really coherent definition of it that does not ultimately depend on faith. Somewhat similar in its implications is the idea that we are essentially a human organism -- that being genetically human is the essence of what we really are. This account fails because it grants personhood to any living thing with human genetics and fails to grant it to sufficiently sentient non-human organisms.

The discussion of these two conceptions of our essence dismisses the idea that personhood begins at conception, although one can accept a mysterious soul as a matter of faith. Although McMahan's focus is ethics, the idea that there is no view of personhood from conception that is generally defensible is relevant for the legal status of a human organism. Essentially, saying human organisms are persons from conception is imposing a faith based belief on others.

The third account of what we really are relies on psychological continuity. This account has many elements that will go into McMahan's view of what we are. In this view, identity is based on a sequence of strongly connected psychological states. Where McMahan diverges is that psychological continuity is generally taken to be all or nothing. This and other concerns about psychological continuity lead McMahan to move away from a focus on identity to a focus on egoistic concern (the sort of concern that we have for ourselves as selves and not for others). The idea here is that what connects a person to some past or future state is a combination of psychological continuity and psychological concern. McMahan further notes that egoistic concern is not binary but can wax and wane over time. The initial development of a person from a embryo illustrates this on one end of life and mentally degenerative diseases illustrate this on the other end of life. Note that while egoistic concern provides the basis for identity, it can diverge from identity, although not necessarily in real cases. But science fiction inspired examples such as a transporter accident which duplicates someone help illustrate the idea. If a person knew they were to be duplicated, it would be rational for them to be egoistically concerned about both but hard to argue that they would share an identity with both duplicates.

Leaving aside the discussion of the complexities of this, McMahan eventually develops the embodied mind account of identity. He argues that we are essentially defined by the physical, functional, and psychological continuity of an embodied mind. This account explicitly dismisses transfer of psychological continuity as a basis of what we essentially are. I do not agree that physical continuity should be considered strictly necessary, but in all practical cases, they're the same, so I'm willing to grant that it doesn't make a difference for the rest of the discussion. The more important implication of this view is that it accounts for the observation that a binary treatment of psychological continuity seems insufficient for our sense of what makes a person and that is better captured by the idea that there is an embodied mind which has varying degrees of continuity based on it's physical existence, it's functional capacity, and its content. In particular, this allows that the essence of what we are does not require the sort of advanced psychological continuity that does not appear until months or even years after birth.

Section two talks about death. Why and when is a death bad? He explores several models such as the common conception that death is bad because it thwarts the total good of a life or the good that remains in store for a person. Accounts that try to look at an overall life suffer from a couple problems. From an external perspective, we tend to perceive a death as less bad after a life well lived, which is consistent with the thwarted goodness view. However, we also tend to perceive the death of an infant as bad but less bad than the death of a young adult. Related to this, an infant arguably suffers less from death because they have no anticipation of their future.

Through a long series of discussions, McMahan captures these intuitions in the time-relative interest account. Death is bad because it thwarts the future interests of a person in realizing future good. Note that it is their interest that matters. Even if their future life only contains suffering, if they think that living itself provides sufficient good, then it is in their interest to live even if someone else might make a different decision (this becomes quite relevant in the later discussion of assisted suicide). These interests, however, are not constant over time. The amount of interest that a person has in their future varies with their egoistic concern, i.e., it varies based on the amount of physical, functional, and psychological continuity that exists between their current and future self. Thus, the interest based not only on the future good lost but only on the strength of the connection. This accounts for why the death of someone who has lived a full life is less bad (less good thwarted, even when the basis for egoistic concern remains strong) and why the life of an infant is very bad but less bad than that of a young adult (lots of good is thwarted in both cases, but the young adult has a much higher time-relative interest in the good that is lost). While this account still leaves some tricky problems to solve (many of which are discussed later in the book), it does a better job of accounting for our intuitions about the badness of death than a discussion based solely on goodness lost.

Having covered all that, why is killing bad? The time-relative interest account implies that the degree to which a death is bad varies by a person. Yet our intuition says that killing someone who has lived a long and fulfilling life is just as bad as killing a young adult or infant. The badness of killing seems to be distinct from the the badness of death. To respond to this, McMahan introduces the idea there is a threshold of respect, a point at which egoistic concern is strong enough that, regardless of the badness of death, we must respect the moral authority of others. This seems to be based on psychological continuity being being strong enough that, to over simplify, one can object to one's killing rather than merely to the pain that such a killing would cause. Below the threshold of respect, the badness of killing seems to vary with the strength of the time relative interests -- thus, this is why we see the killing of so-called higher animals as worse than that of so-called lower animals, regardless of how bad we personally see those killings overall. McMahan argues that for beings below the threshold of respect, the badness in killing them can be traded off against other moral considerations. E.g., that it would be ok to kill an animal to save a person but it would not be ok to kill a person to save another person.

So where are we at this point: People are physically, functionally, and psychologically continuous beings (regardless of whether physical continuity is necessary or, as in my view, it's a multi-variable function of these three factors). Death is bad because it thwarts the future interests of a person, but that interest varies with the strength of egoistic concern, which is built on the mentioned continuities. If a being exceeds the threshold of respect -- if, essentially, they are capable of caring that they are dying -- then killing is always equally bad. Below that threshold, the badness of killing varies with the strength of the time relative interests and can be traded off against other concerns. That brings us to killing at the margins of life: abortion and euthanasia.

There is a lot discussed the abortion section. McMahan argues that until there is something for a mind to be embodied in -- a brain with at least minimal functionality -- then there is no one to kill and so an abortion does not kill a person. What is killed is, so to say, an unoccupied human organism. It seems that, depending on what is considered core functionality of the brain, this does not occur until 20 - 28 weeks. 20 weeks being when synapses start to form, 28 weeks being when cortical activity resembling rudimentary consciousness appear.

Late abortion is more complicated. This makes sense since even those who advocate for abortion rights, including late abortion rights, generally see the issue as more complicated, especially once the fetus has passed viability. Given the view of personhood described so far, an abortion after 20 weeks involves killing, as McMahan puts it, someone instead of something. Yet the connection is fairly weak. Is this killing at this point governed by the time relative interest account -- where the badness of the killing is variable and can morally be traded off against other interests? Or is it like killing someone above the threshold of respect? The gist of the discussion is that the late fetus is still below the threshold of respect because their psychological connection to their future self is essentially non-existence. This brings up the tricky question of infanticide -- birth doesn't make that strong of a difference in psychological connectivity. What McMahan concludes here is that the reasons against killing infants are not, in fact, as strong as the reasons against killing older persons. However, because infants are not potentially unwelcome guests in the body of another, the moral trade-offs are not strong enough to justify killing infants. However, he observes that in times and places where infants were sometimes left out to die in times of severe shortage, a different standard does seem to apply to infants than older children (even though the older children presumably use more resources).

One argument against this view is that while fetuses have a weaker time relative interest than older people, they have the potential to develop a stronger interest, and thwarting that potential makes it a wrongful killing. That potential, this objection goes, makes the fetus more valuable than an animal even when it's psychological capabilities are still notably less. (Note that this objection applies, although less strongly, even before minimal brain capability is present.) The gist of this discussion is that in so far as potential is morally relevant, it is relevant in a way that is similar to the interests that a late fetus has in its future. It provides a reason to consider the fetus's interests -- to consider it a someone rather than a something -- but not one strong enough to put it above the threshold of respect. In particular, we have stronger moral obligation to avoid thwarting the interests of others than to avoid thwarting their potential interests.

McMahan then goes on to compare his account with alternate accounts of the ethics of abortion. I'll only discuss the response to Thomson's violinist. McMahan notes that this is potentially compatible with his account and that the two together could make a strong case for abortion rights. While the time-relative interest account merely provides a case that the fetus's interests can be traded off against other concerns, it does not argue how those trade-offs should be considered (and does, in fact, note that it is a very difficult judgment call). Thomson argues that the right to bodily autonomy gives broad authority over deciding when another can use a body for life support. McMahan notes a number of common objections to Thomson's analogy and notes that they are fairly problematic if, as Thomson does for the sake of argument, the fetus is granted the status of a person. However, if the fetus is governed by the time-relative interest account, then these objections become weaker.

The four objections are: The woman bears some responsibility for the pregnancy. The woman bears some responsibility for the fetus needing aid. The woman has a special relation to the fetus as its biological parent. An abortion kills the fetus rather than merely allowing it to die.

The first objection is dismissed as being obvious that the woman did not consent to having the fetus in her body merely by having sex. The second objection is similar yet critically different. It is not an argument about consent but rather about responsibility. If I cause someone to be in a life threatening situation, then I bear some responsibility for that, even if it is completely invalid to say that I consented to taking on that responsibility. McMahan argues that while creating a life in a person who does not want to support it does indeed put the new life in a threatening situation, but because that life would not have existed otherwise, the responsibility is not exactly the same as if, for example, someone accidentally pushed another into deep water. Also, one could argue, the pregnant woman's action is responsible for creating the fetus, but it's the fetus's growth, not the woman, that is creating it's ongoing need for aid. This is similar to saying that even though the parents created an infant, once the child is grown, no one would say that the parents created the infant that they were the ones causing it to still need significant aid. Still, if you grant the fetus personhood, then this objection is fairly strong and weakens Thomson's argument.

The third objection is that special relations can strengthen our obligations to a person. McMahan observes that while being a parent is something that is understood to be consensual, we do generally seem to think being a biological parent has some responsibility. That, for example, in a situation where a biological parent cannot transfer their parental obligations to someone who consents to take them on, the biological parent retains those obligations even if they do not want them. Thus, while it may be that a person has no moral obligation to stay attached to a violinist, they may have an obligation to stay attached to their biological child, even if they do not consent to being a parent. Again, this objection is not a definitive argument against abortion rights, but does weaken the Thomson argument when full personhood is assumed.

The final argument, that killing is worse than letting die, McMahan responds to by noting that if the person who is removing the aid also provided the aid, then they are letting the other person die, regardless of whether the death itself is more active or passive. Rather, what seems to make a difference is whether or not the provider/remover is responsible for the person needing aid. If they were, then removing the aid feels more like an act of killing (because they are removing aid that they created the need for). If they were not responsible for the other person needing aid, the act feels more like letting die, even if we would still intuitively prefer for the death to be passive.

The last section discusses euthanasia. McMahan spends awhile discussing what it means to die (as opposed to earlier, where the discussion focused on why dying is bad). He contrasts biological death, brain death, and permanent loss of cognitive capacity. Although brain death is the currently agreed upon standard, he argues that permanent loss of cognitive capacity should be considered the relevant criteria for the death of a person. It is hard to define what is minimally necessary for this capacity to be lost. However, being in a permanent vegetative state where the brain can, at least somewhat, regulate the body but there is no cognitive activity seems sufficient.

From this, he concludes that advance directives asking for life support to be removed should be respected. He also argues that advance directives against removing support should be respected even though someone in this state does not retain their full moral status as a living person. Even though the person died, we should generally respect people's wishes for their remains.

What about cases where a person wants to die? McMahan notes that he is only considering cases where the person's death would be a net benefit to them. Where the prospects for future life are definitely bad, such as an incurable disease that causes great suffering. In addition, McMahan differentiates between voluntary (consent given), nonvoluntary (consent is impossible to either give or deny), or involuntary (consent refused). The last is clearly bad because it violates respect for the person. McMahan concludes that if it would be rational and morally acceptable for someone to seek death, it would be morally acceptable for a third party to aid them. It is even arguably better to let them be assisted (by willing third parties) than to say that their desire for death is justified but must be executed by themselves via means that are likely to be more painful to themselves and others. In short, if suicide is ever justified, then we should consider euthanasia justified in such cases.

However, McMahan argues that there may be a compelling argument against euthanasia and suicide, even when life contains a prospect of nothing but suffering and a person has no desire to live. Causing death may violate the worth of a person. Note that, for the sake of discussion, McMahan defines worth as being based on having the capacities that make-up personhood and so differs from arguments about the sanctity of life which, in this presentation, are attributed to the human organism regardless of capacity. Killing, in this view, uses a person instrumentally, destroying a person merely for the sake of their good. McMahan's response to this is more a question than an answer: let's take that it is important to respect a person's worth. Why should we suppose that killing violates that worth when it does not violate the will of the person killed? He notes that respecting someone's worth requires treating them in a way that their personhood requires, even if it goes against their wishes. But what that entails is hard to define. You can argue that killing does not respect the person, even if it respects his good and autonomous will. You can also argue that if you are respecting both his good and autonomous will, you are respecting his worth. Key, is that neither account in inherently more plausible.… (meer)

Explores 'foundational' claims about personal identity, death, and killing (not merely letting die), as well as ethical issues around 'organ harvesting' and abortion, PAS, advance directives, life planning around the dementias, and rational social policy. How many of these Oxford people head to the state universities in the USA; this one went to Illinois.

"Who are we, and why it matters" - rel="nofollow" target="_top">http://www.stonybrook.edu/commcms/philosophy/people/faculty_pages/docs/Kittay_Ma...… (meer)