A RIGHT OF SELF-TERMINATION?*
J. David Velleman
Department of Philosophy
University of Michigan
Getting cancer changed my feelings about people
I remember hearing a fellow philosopher expound, with a wave of his cigarette, on his right to choose whether to live and die smoking, or to quit and merely survive. I was just beginning a year of chemotherapy, and mere survival sounded pretty good to me. But I was the visiting speaker, and my hosts were unaware of my diagnosis. Several of them lit up after dinner as we listened to their colleague's disquisition -- they with amused familiarity, I with an outrage that surprised even me and would have baffled them, if I had dared to express it. That I didn't dare is a cause for regret even now, ten years after the fact.
One objection was already clear to me at the time.
A few months with cancer had taught me that a tumor rarely invades a region smaller
than an extended family.
Physically, the cancer was confined to my body, but even in that respect it was difficult to regard as mine. The tumor cells were growing in my bone marrow, which didn't live up to its poetic billing as the core of my being. The marrow in my bones, I discovered, was as foreign to me as the far side of the moon: it was, in a sense, my far side -- unseen, insensate -- its depth inside me being a measure of remoteness rather than intimacy. Of course, this fertile gunk in my pelvis and skull was also my sole source of blood cells, and my life depended on it. But so did the life of my sons' father, my wife's husband, my parents' son, my brothers' brother, and I was never sure who among us would suffer the greater harm if that life ran out of gunk.
Listening to my host laugh at his future cancer, I wondered whether he realized how many others would share it. What I would have said on their behalf, however, wouldn't have expressed my strongest feelings, which were felt on my own behalf, in a sense that I couldn't articulate. I was somehow offended, insulted. Watching smoke curl from the lips of people unmindful of my mortality, I felt as I probably would feel listening to anti-Semitic remarks directed at another person by a speaker unaware that I, too, was a Jew. I was witnessing an insult to a group of which I was also a member.
This symposium isn't about the right to smoke, of course; it's about
the right to die. Not surprisingly, however, these rights tend to be articulated
in the same terms. A person claiming either right might describe it, for instance,
as a right "to live and die in the light of ... his own convictions about why
his life is valuable and where its value lies."
I can't recall whether the speaker in my story used these exact words, but I seemed to hear his voice again when I read them in The New York Review of Books, under the title "The Philosophers' Brief." This brief had been submitted to the United States Supreme Court in support of a challenge to statutes outlawing physician-assisted suicide. Reading it, I once again felt a collective slight, and this time I couldn't miss which group was being slighted.
So I think that I can now explain why I was once offended by one philosopher's defense of smoking, and the explanation leads me to reject The Philosophers' defense of assisted suicide as well. As for assisted suicide itself, however, I don't know what to think. The complexities of the issue have thus far defeated my attempts to arrive at a settled position. On the policy question posed by this symposium, then, I am neither Pro nor Con. I'm, like, Not So Fast.
The principle quoted above, which would settle the
issue quickly, can be derived from two broader principles. The first principle
is that a person has the right to make his own life shorter in order to make it
better -- to make it shorter, that is, if doing so is a necessary means or consequence
of making it a better life on the whole for him. The second principle is that
there is a presumption in favor of deferring to a person's judgment on the subject
of his own good. Together, these principles imply that a person has the right
to live and die, in particular, by his own convictions about which life would
be better for him.
For the smoker in my story, of course, shortening his life was not a means of making it better but rather a likely consequence of an activity that made it better, in his opinion, despite making it shorter, too. But in most of the cases for which assisted suicide is advocated, shortening a patient's life is intended as a means of making it better, because the continuation of the patient's life would detract from its overall value for him. When the first principle is confined to this latter context, it can be rephrased as the assertion of a patient's right to end his life on the grounds that it is no longer worth living.
I think that this principle is mistaken. Before I criticize it, however, I should speak briefly to the second principle stated above, which I can accept. I think that a person's considered judgment about his good is a judgment to which we generally ought to defer.
More specifically, then, I think that we generally ought to defer to a person on the question whether his life is worth living, since the living-worthiness of a life measures the extent to which the continuation of that life would be good for the person living it. The person living a life is the best judge of the value that its continuation would afford him -- not an infallible judge, of course, but usually more reliable than anyone else is likely to be. Indeed, his judgment of this value is to some extent self-fulfilling, since his merely liking or disliking aspects of his life can to some extent make them good or bad for him.
The reasons for deferring to a person's judgment about his good go beyond his reliability as judge. Respect for a person's autonomy may require that we defer to his considered judgment about his good even when we have reason to regard that judgment as mistaken. Letting him live his own life may sometimes entail letting him make his own mistakes about what's good for him -- including, perhaps, mistakes about whether it would be good for him to go on living. Forbidding a person to make such mistakes can be objectionably paternalistic, because it would usurp his role as the primary agent of his own affairs.
Thus, if a person had the right to end his life on the grounds that it wasn't worth living (in accordance with the first principle, above), then he would have the right to be guided by his own judgment on that score (in accordance with the second principle). But I reject the principle that a person has the right to end his life solely on the grounds of the benefits he will thereby obtain or the harms he will avoid.
One reason for rejecting this principle is that
a life confers benefits and harms on people other than the person living it. Does
a person have the right to deprive his children of a parent simply because life
isn't worth enough to him?
I want to set aside this question, however, because it tacitly concedes the assumption that the values at stake in life-or-death decisions are relative to personal interests; it merely invites us to consider a wider circle of potential beneficiaries. The values that we need to consider, in my view, aren't relative to personal interests and consequently have no beneficiaries.
One might insist that values must have beneficiaries,
because they wouldn't exist if there weren't someone who could appreciate them:
nothing would be good or bad in a universe devoid of sentient beings.
But the fact that values wouldn't exist without potential valuers does not entail
that they must accrue to someone.
Values are relative to potential valuers because they are normative, in the first instance, for valuation. That is, for something to be valuable just is for it to be such as ought to be valued in some way -- respected, loved, admired, wanted, treasured, or the like. The very concept of value therefore contains the concept of a valuer, actual or potential.
The experience of valuing something can be beneficial, as in the case of appreciating the aesthetic value in a work of art. But the concept of value, in positing a potential valuer, doesn't necessarily require that he would benefit from the experience. Things can be venerable, for example, whether or not there is any benefit in venerating them; and they can be awesome whether or not one would gain by holding them in awe. So the fact that value must be capable of registering with someone, who would thus appreciate it, does not mean that it must be capable of accruing to someone, who would thus gain by it. Value requires a potential valuer but not a potential beneficiary.
In fact, our appreciation of values that are relative to the interest of a beneficiary may depend on a prior appreciation of a value that is not relational in this sense. This dependence emerges when we try to explicate the concept of interest-relative value, or what is good for a person.
The concept of what is good for a person turns out to be fairly resistant
to explication. We might initially think to equate what's good for a person with
whatever would be rational for him to care about. But this equation would end
up implying that all rational concerns are self-interested, by definition. In
order to allow for the possibility of rational selflessness, we have to acknowledge
that not everything that would be rational for someone to care about is necessarily
in his interest.
Various philosophers have therefore attempted to define what's good for a person as a proper subset of the things that would be rational for him to care about, such as the subset including only those things which require his existence. It may or may not be a drawback in these definitions that they would exclude from a person's good such things as posthumous fame. In any case, these definitions are still too inclusive, since the things involving a person's existence that are rational for him to care about include, for example, particular sacrifices that he can make for other people.
The only convincing analysis of a person's good,
to my knowledge, is one recently proposed by Stephen Darwall, who argues that
what's good for a person is what's rational to want for his sake.
"For the sake of" is a phrase that marks the subordination of one concern
to another: to care about one thing for the sake of something else is to care
about the former out of concern for the latter. To want something for the sake
of a person is thus to want it out of concern for the person himself. Darwall's
analysis says that a person's good is what would be rational to want out of concern
for that person.
Darwall argues -- convincingly, to my mind -- that a person's good is a rational object of desire for anyone who cares about that person. By the same token, he argues that even the person himself is rationally obliged to care about his good only insofar as he cares about the person whose good it is -- that is, himself.
Think here of the familiar connection between how you feel about yourself and how you feel about your good. Sometimes when you realize that you have done something mean-spirited or shameful, you come to feel worthless as a person; you may even hate yourself; and one symptom of self-hatred is a loss of concern for your own welfare. It no longer seems to matter whether life treats you well or badly, because you yourself seem to be no good. Your desire for your good thus depends on your concern for yourself -- and rationally so, according to Darwall's analysis.
Note that self-loathing isn't the feeling that you are worthless to yourself. Indeed, the value that things afford to you is precisely what no longer seems to matter, and so your having no value for yourself wouldn't seem to matter, either. The reason why value accruing to you no longer seems to matter, however, is just that you don't seem to matter, period. You have lost your appreciation for the value that things have in relation to your interest because you have lost a sense of embodying value in yourself.
Now, things could still be good for you, according
to Darwall's analysis, even if you didn't embody any value; since they could still
be such as would be rational for someone to want if he cared about
you, however baseless the latter concern might be. But things that were good for
you would not actually merit concern unless you merited concern; and if you didn't,
then despite their being good for you, they wouldn't ultimately be worth wanting,
after all. As I put it a moment ago, what's good for you wouldn't matter if you
This account of a person's good therefore implies -- rightly, again, in my opinion -- that what's good for a person is not a categorical value, any more than what's good for a purpose. What's good for a purpose is worth caring about only out of concern for the purpose, and hence only insofar as the purpose is worth caring about. Similarly, what's good for a person is worth caring about only out of concern for the person, and hence only insofar as he is worth caring about. A person's good has only hypothetical or conditional value, which depends on the value of the person himself.
Of course, we assume that a person's good does matter. But we make
this assumption only because we assume that people matter -- that everyone has
a value that makes him worth caring about. Darwall's analysis of a person's good
reveals how our appreciation of value that accrues to someone depends on a prior
appreciation of a value inhering in him.
The latter value cannot be relative to personal interests, on pain of setting off a problematic regress. If this value were relative to someone's interest, then it would matter only to the same extent as that beneficiary. This regress of values would continue until it reached a value that was not relative to anyone's interest and that consequently mattered for its own sake. In fact, however, the regress never gets started, because we assume that every person already matters for his own sake, because of embodying an interest-independent value.
A value of this kind, which a person has in
himself but not for anyone, is the basis of Kantian moral theory. Kant's
term for this value is "dignity," and he attributes dignity to all persons in
virtue of their rational nature. What morality requires of us, according to Kant,
is that we respect the dignity of persons.
The dignity of a person is a value that differs in kind from his interest. Unlike his interest, for example, his dignity is a value on which his opinion carries no more weight than anyone else's. Because this value does not accrue to him, he is in no better position to judge it than others.
Similarly, respect for a person's autonomy does not require deference to him on questions of his dignity, as it does on questions of his good. On the contrary, respect for a person's autonomy just is an appreciation of a value in him that amounts to a dignity, in Kant's sense of the term, precisely because it commands respect. If a person denies embodying such a value, he can hardly claim that we should defer to him out of an appreciation for a value such as he denies. He cannot claim, in other words, that out of respect for his autonomy we should defer to his judgment that he possesses nothing worthy of our respect.
Nor is it paternalistic to challenge a person's judgment about his dignity, as it is in the case of his good. Challenging a person's judgment about his good is objectionable because it undermines his role as the agent of his own affairs; but his value as a person is not just his affair. Although his good is a value that accrues to him alone, in the first instance, his value as a person inheres in him among other persons. It's a value that he possesses by virtue of being one of us, and the value of being one of us is not his alone to assess or defend. The value of being a person is therefore something larger than any particular person who embodies it.
That's what I miss in so many discussions of euthanasia and assisted
suicide: a sense of something in each of us that is larger than any of us, something
that makes human life more than just an exchange of costs for benefits, more than
just a job or a trip to the mall. I miss the sense of a value in us that makes
a claim on us -- a value that we must live up to.
I don't deny that there are circumstances under which it would be better for one's life to end and permissible to hasten its ending. What I deny is that one may end one's life simply because one isn't getting enough out of it. One has to consider whether one is doing justice to it.
If a person possesses no value that he must live up to, or do justice to, then
his life becomes a mere instrument, to be used or discarded according to whether
it serves his interest. His moral claim to his own life then looks something like
[A] patient's right to life includes a right not to be killed. But
that right gives [him] a protected option whether to live or die, an option
with which others cannot legitimately interfere; it does not give [him] a duty
to live. If a patient decides to die, he is waiving his right to live. By waiving
his right, he releases others (perhaps a specific other person) from a duty
not to kill him.
This can't be right. It portrays morality as protecting a person's options without
protecting the person himself, except insofar as his own existence is one of his
options. Surely, however, options are worth protecting, not for their own sake,
but for the sake of the person whose options they are. So how can morality treat
the person as worth protecting only for the sake of protecting one of his options?
If he doesn't already merit protection, how can they?
The quotation above is drawn from a recent essay
by Frances Kamm, who goes on to answer Kantian objections as follows:
Suppose life involves such unbearable pain that one's whole life is
focused on that pain. In such circumstances, one could, I believe, decline the
honor of being a person. ... We might acknowledge the great (and normally overriding)
value of being a person ... [and yet] allow that some bad conditions may overshadow
its very great value.
Here Kamm is claiming that someone can view life as a mere option even while accepting
the Kantian view of his value as a person. The problem with this passage is that
it misstates the Kantian view.
When Kamm says that the value of a person normally "overrides" the value of other goods, but can be "overshadowed" by conditions that are exceptionally bad, she implies that it can be balanced against the person's interest. And when she goes on to speak of this value as an "honor" that the person can decline, she implies that it is actually part of a person's interest, since an honor accrues to a particular person, whose role as its beneficiary entitles him to accept or decline it.
But the dignity of a person isn't something that he can accept nor decline, since it isn't a value for him; it's a value in him, which he can only violate or respect. Nor can it be weighed against what is good or bad for the person. As I have argued, value for a person stands to value in the person in something like the relation that holds between the value of means and that of the end: in each case, the former merits concern only on the basis of concern for the latter. And conditional values cannot be weighed against the unconditional values on which they depend. The value of means to an end cannot overshadow or be overshadowed by the value of the end, because it already is only a shadow of that value, in the sense of being dependent upon it. Similarly, the value of what's good for a person is only a shadow of the value inhering in the person, and cannot overshadow or be overshadowed by it.
These are abstract considerations, but they are
concretely illustrated by the story with which I began. When my host claimed that
he benefited more from the pleasures of smoking than he would be harmed by an
early death, my first thought was that he had failed to consider harms and benefits
to people other than himself. On second thought, however, I resented his assumption
that harms and benefits were the only values at stake.
My host's remarks implied that an early death, of the sort that he was risking and I was hoping to forestall, would be a loss that could be offset by sufficient gains. But what would it matter how much I lost or gained if I myself would be no loss? My gains or losses would merit concern only on the basis of concern for me -- which, being the basis of concern for them, could not then be offset by that concern. Hence my gains or losses wouldn't matter unless I had a value that could not be offset by theirs.
My host was implicitly denying the existence of such a value. For he claimed that death was worth worrying about only in respects for which he could be compensated by the pleasures of smoking. He was thus implicitly denying the interest-independent value of a person, without which it couldn't really matter whether I lived or died.
Of course, he was denying the existence of this value in his own case, not in mine; but our cases were indistinguishable on this score. By implicitly denying his own interest-independent value, my host was somehow trivializing or denigrating himself as a person. Sometimes people's self-denigrating remarks just embarrass us, but in other instances they can be sufficiently principled to give offense. Recall my earlier reference to anti-Semitism. Anti-Semitism can manifest itself in self-denigrating remarks, if it is the anti-Semitism of a self-hating Jew. My host's disregard for his own value as a person offended me as another person, just as someone's denigrating himself as a Jew would offend me as another Jew.
I think Kant was right to say that trading one's
person in exchange for benefits, or relief from harms, denigrates the value of
personhood, respect for which is a criterion of morality (Kant would say, the
criterion). That's why I think that smoking is a vice -- at least, when practiced
for the reasons offered by my host. It's also why I think that suicide is immoral
when committed on the grounds that life isn't worth living.
Mind you, I don't go around snatching cigarettes out of people's mouths. And I'm not sure that I would forcibly try to stop someone from committing suicide solely because it would be immorally self-destructive. The impermissibility of someone else's conduct doesn't necessarily give me permission to interfere with it. By the same token, however, I think that encouraging or assisting others in impermissible conduct is itself impermissible. That's why I think that the tobacco industry is engaged in an immoral enterprise. And it's why I think the same of Dr. Kevorkian, who has done more than anyone to help people die by their own convictions.
Note that these moral judgments distinguish between self-destruction and mere self-harm. As I have said, I believe that people are sometimes entitled to act on mistaken judgments about their own interest; and to this extent, at least, they are entitled to harm themselves. But the behaviors that I have criticized don't merely damage the agents' interests; indeed, they may not damage the agents' interests at all, if the agents are right about the costs and benefits involved. These behaviors are to be criticized, in my view, because they are premised on a disregard for the value of the agents themselves.
The same criticism would apply, for example, to agents who put up their own freedom as collateral in order to obtain loans. People have no right to sell themselves into slavery, no matter what their convictions, but the reason is not that they would thereby be harming themselves; the reason is that they would be violating their own personhood.
These moral judgments depend, of course, on my belief
that a person has an interest-independent value; and they may consequently seem
to impose my Kantian values even on people who don't believe in them. Don't people
have the right to live and die by their own convictions as to the value of their
If the question is whether people are morally permitted to end their existence solely because they find it unrewarding, then I have already answered in the negative, on the grounds that they would then be violating their own interest-independent value as persons. But of course the present question is meant to be taken differently, as suggesting that we defer to people's judgments about whether they have an interest-independent value, in the first place. Under this interpretation, the question is not whether people are permitted to violate their own dignity but whether they are entitled to be believed when they insist that they have none. I have answered in the negative to this latter version of the question as well. The reasons for deferring to people about values relative to their interests do not apply in the case of interest-independent value.
This answer may seem to beg the question, since it presupposes the existence of the very interest-independent value that is at issue. What I have now argued, however, is that we cannot avoid presupposing the existence of this value anyway, since it's needed to account for the importance of interest-relative values. We cannot justify someone's death on the grounds that it's good for him, while also denying the existence of another value, embodied in him. For if he were himself a cipher, evaluatively speaking, then what's good for him would be, in the same manner of speaking, good for nothing.
I admit that talk of someone's value as a person
sounds like religion rather than philosophy. Such talk is a secular version of
religious talk about the sanctity of human life.
Historically speaking, however, most moral discourse has religious sources. The question for secular ethics is whether we can rationally accept the values bequeathed to us by religion while being skeptical of their theological basis. A question that's equally pressing, though less widely acknowledged, is whether we can selectively accept some of these values while discarding others. My view is that our values will be incoherent so long as they lack a counterpart to the sanctity of human life.
This view will immediately seem to entail reactionary consequences, such as a rejection of euthanasia and abortion in any form. But a secular value that corresponds to the sanctity of human life needn't be exactly the same value or yield exactly the same consequences. In particular, it need not attach to biological life or biological humanity per se; and so it needn't rule out abortion, for example, simply because the fetus is both alive and human. What secular morality must regard as sacrosanct, I have suggested, is not the human organism but the person, and a fetus may embody one but not the other.
Recognizing the interest-independent value of a person wouldn't necessarily rule out euthanasia or suicide, either. On the contrary, recognizing such a value is essential to one familiar argument in favor of these practices -- namely, the argument for dying with dignity.
The idea that dignity can justify a person's death
may seem incompatible with the Kantian conception of dignity as a value inhering
in the person. Wouldn't a person's value always militate in favor of saving his
This apparent conflict is due, however, to a confusion about the normative implications of dignity. Dignity is what Kant called a "self-existent" value -- a value to which we are obliged to respond only when it already exists, and then only by paying it reverence or respect. The value of persons does not oblige us to maximize the number of people in existence; it obliges us only to respect the people that do exist. And respecting these people is not necessarily a matter of keeping them in existence; it is rather a matter of treating them in the way that is required by their personhood -- whatever way that is.
The Kantian objection to suicide, then, is not that it destroys something of value. The objection is not even to suicide per se, but to suicide committed for a particular kind of reason -- that is, in order to obtain benefits or escape harms. And the objection to suicide committed for this reason is that it denigrates the person's dignity, by trading his person for interest-relative goods, as if it were one of them. This interpretation of the objection to suicide leaves open the possibility that a person's dignity may justify suicide in other contexts, if suicide would constitute an appropriate expression of respect for one's person. Kantianism would then be able to endorse the notion of dying with dignity.
Actually, the phrase "dying with dignity" is potentially
misleading. We don't think that a person's death is morally acceptable so long
as he can carry it off with dignity. Rather, we think that a person's death is
acceptable if he can no longer live with dignity. The operative concept is undignified
life, not dignified death.
When a person cannot sustain both life and dignity, his death may indeed be morally justified. One is sometimes permitted, even obligated, to destroy objects of dignity if they would otherwise deteriorate in ways that would offend against that value. The moral obligation to bury or burn a corpse, for example, is an obligation not to let it become an affront to what it once was. Librarians have similar practices for destroying tattered books -- and honor guards, for destroying tattered flags -- out of respect for the dignity inherent in these objects.
Of course, the value inhering in mere things, such as books or flags, must be different from that inhering in persons by virtue of their rational nature. But all of these values belong together as a class, the class of dignity-values, whose defining characteristic is that they call for reverence or respect.
These examples suggest that dignity can require, not only the preservation of what possesses it, but also the destruction of what is losing it, if the loss would be irretrievable. Dignity, unlike well-being, does not come in degrees that we are obliged to maximize; as we have seen, it is not a value whose existence we are obliged to promote at all. To treat a dignity-value as capable of degrees, all of them worth preserving, would be to treat it like an ordinary good -- which would in fact be disrespectful. Respect for an object of dignity can sometimes require its destruction.
The question, then, is what constitutes the loss
of dignity for a person. The dignity in question has nothing to do with being
dignified, with keeping up appearances, or with sustaining any particular social
status. It has nothing to do with what people ought to admire or esteem in one
another, or with what they actually respect. It is rather what they ought to respect,
in the way that they can manifest only by treating one another morally. According
to Kant, what people ought to respect in this way is one another's rational nature.
Ironically, Kant's view is borne out by Kamm's example, in which "life involves such unbearable pain that one's whole life is focused on that pain." Kamm assumes that this case invites us to weigh the disvalue of pain against the value of being a rational agent. In fact, however, Kamm has described a case in which pain is more than painful, since it not only hurts the patient but also becomes the sole focus of his life. Pain that tyrannizes the patient in this fashion undermines his rational agency, by preventing him from choosing any ends for himself other than relief. It reduces the patient to the psychological hedonist's image of a person -- a pleasure-seeking, pain-fleeing animal -- which is undignified indeed. And Kamm is clearly envisioning that this severely reduced condition of the patient can be ended only by his death.
I suspect, then, that if euthanasia seems justified in Kamm's example, the reason is not that relieving the patient's pain is more important than his dignity as a person; the reason is rather that pain has already undermined the patient's dignity, and irretrievably so. The example thus supports dying for the sake dignity, not for the sake of self-interest.
I often wonder whether proponents of assisted suicide
don't overstate the moral significance of pain. Pain is a bad thing, of course,
but I doubt whether it can justify anything close to euthanasia or suicide unless
it is (as Kamm calls it) unbearable. And then what justifies death is the unbearableness
of the pain rather than the painfulness.
What do we mean in calling pain unbearable? What is it not to bear pain? It certainly isn't a matter of refusing to feel the pain, of shutting one's eyes to it, as one might to an unbearable sight, or of walking away from it, as one might from an unbearable situation. Not to bear pain is somehow to fall apart in the face of it, to disintegrate as a person. To find pain unbearable is to find it thus destructive not just of one's well-being but of oneself.
But then we make a mistake if we describe the patient in unbearable pain as if he were his rational old self, weighing the harm of pain against the benefits of existence. If his pain is truly unbearable, then he isn't his rational self any longer: he is falling apart in pain. Even if he enjoys some moments of relief and clarity, he is still falling apart diachronically, a temporally scattered person at best.
I don't think that we serve the patient well, in these circumstances, by claiming broad rights of self-determination in his name. He may indeed be entitled to help in dying, and he will certainly have to participate in the relevant decisions. But let us keep in mind that these decisions would be premature if the patient were not already in the twilight of his autonomy, where self-determination is more of a shadowy presumption than a clear fact.
I do not know how to frame a public policy or law
that would distinguish between the cases in which I think that euthanasia or suicide
is morally permissible and the cases in which I think it is not. Of course, the
law would not have to follow the moral vicissitudes of the practice so closely
if they were covered by a right of self-determination. If there were a broad class
of cases in which the patient had the right to decide for himself whether death
was justified, then we could legalize euthanasia or assisted suicide in those
cases, even though it might not be justified in all of them. If a patient then
opted for death when it wasn't justified, he would still be acting within his
rights, which the law would have been justified in protecting.
But I do not believe that a person has the right, in general, to choose between life and death; nor do I believe that a person's rights suddenly expand when he becomes terminally ill. So I don't see how a case for legalization can be founded on rights of self-determination, and I am once again faced with the difficulty of legalizing death for the sake of dignity without also legalizing it for the sake of self-interest.
I certainly don't think that the law should forbid
activities simply because they have the potential of being self-destructive in
some circumstances. I don't think that mountain-climbing should be outlawed --
or smoking, for that matter. The problem is that killing, unlike mountain-climbing
or smoking, impinges on the dignity of persons essentially and not just in some
unfortunate circumstances or cases. The result is that the law on killing, like
the law on slavery, unavoidably expresses our collective valuation of personhood
Supporters of euthanasia and assisted suicide sometimes liken them to the other intrinsically injurious treatments to which a patient may consent for his greater good -- the cutting and stabbing and drugging and poisoning that are the physician's stock in trade. Then they ask: What's so special about killing? Isn't killing just another medical intervention to which a patient should be allowed to submit when it serves his interest?
My inclination is to answer this question with another: What's so special about slavery? Isn't enslavement just another cost that a person should be allowed to risk in pursuit of his interests?
Surely, there is something special about slavery. Though we may indeed have a right to live and die in light of our own convictions, it doesn't extend to convictions about the price for which our freedom would be worth selling. Nor does it extend, in my view, to convictions about the price for which our lives would be worth ending. And self-interested reasons for ending our lives are, in the Kantian sense, reasons of price rather than dignity.
When it comes to the design of public policy on
assisted suicide, I am inclined to think that considerations about the morality
of the practice itself may be swamped by considerations about the collateral effects
of legalization. As I have argued elsewhere, simply allowing people to opt for
death may eliminate the conditions that make some people's lives worth living,
thereby creating new candidates for killing.
Legalization would therefore do harm to people who are currently no more than
bystanders to the debate.
These collateral harms might have to be tolerated if there were a fundamental right to choose between life and death. We can't deprive all people of a choice to which they're morally entitled just because some people would be better off without it.
What I have argued here, however, is that there isn't a fundamental right to choose between life and death. There may still be a moral justification for death in some cases, but it doesn't rest on a right of self-determination. And without such a right, the case for legalization must proceed more slowly -- far more slowly than The Philosophers would like us to believe.
REPLY TO FRANCES KAMM
Frances Kamm's critique of my essay is mainly devoted
to refuting arguments that I do not recognize as mine. The passages from which
Kamm draws these arguments are passages in which I criticize evaluative reasoning
that neglects or implicitly denies the value of a person. Eventually I conclude
that committing suicide on the basis of such reasoning is immoral, as is assisting
a suicide committed on that basis. I also find fault with philosophers, including
Kamm, who use such reasoning to justify the practice of assisted suicide. In the
first instance, however, I merely explore the inadequacies of the reasoning itself
in various contexts, without drawing any conclusions about assisted suicide. Unfortunately,
Kamm finds arguments against assisted suicide everywhere.
Thus, for example, she finds an argument against assisted suicide in my response to her own statement, in an earlier publication, that the right to life is a morally protected option not to be killed. Here I say that a person's option not to be killed wouldn't be morally protected, in the first place, if he didn't possess a value that would afford him direct protection from certain kinds of treatment, which I assume would include certain kinds of killing. I thus suggest that insofar as morality protects a person's life, it must go beyond protecting his option of living; but I say nothing in this context about how much further it might go, or which kinds of killing it might forbid. Kamm assumes that I must be jumping to some conclusion against assisted suicide, and so she sets about reconstructing the trajectory of my jump. But there is no argument against assisted suicide here, only an argument against reducing a particular right to a protected option.
Kamm constructs another argument against assisted suicide out of two brief allusions on my part to the immorality of slavery. The point of these allusions is to illustrate my claim that moral thinking becomes impoverished when it calculates what is good or bad for a person without considering the value of the person himself. If we consider only the harm of enslavement, I argue, we fail to see what is wrong with it, and what makes it morally different from other kinds of harmful treatment. Here Kamm finds an argument by analogy against assisted suicide. I give no such argument.
The main argument that Kamm discovers in my paper
is the one that she calls the first Reductio Argument. Here again, Kamm pulls
an argument against assisted suicide out of a passage in which I am objecting
to someone else's moral reasoning. But this rabbit does seem to come out of my
hat, and so it merits more attention.
Kamm draws the first Reductio Argument from my remarks about a smoker who defends his habit by calculating its impact on his interests. What I say about this person is that his interests wouldn't matter unless he mattered -- that is, unless he had a value that was prior to, and not commensurable with, the value of his interests. When this person justifies his death solely in terms of his interests, I say, he weighs himself in the balance against those interests, thereby denigrating his value as a person -- and, by implication, the value of all persons.
From these remarks Kamm fashions an argument that turns on the following premise:
(3) If it is permissible to dispose of x independent of concern
for any other worthwhile thing (while x retains the characteristics that
supposedly make it important that he have what is good for him for no other
reason than that it would be good for him), x does not matter. [p. 13]
Somehow, my statement that a person's interests wouldn't matter unless he mattered
has suggested to Kamm the statement that he wouldn't matter if his destruction
were permissible. And my objection to justifying his death solely on the grounds
of his interests has suggested a premise about justifying it "independent of concern
for any other worthwhile thing." The insertion of this phrase is especially puzzling.
What I object to is destroying a person on the grounds of concern for particular
worthwhile things -- that is, solely out of concern for his interests.
The objection to destroying a person solely for the sake of his interests is that
it treats the person as commensurable in value with those interests. I find no
trace of this objection in Kamm's Reductio Argument. Insofar as I understand the
latter argument, I agree with her that it confuses the value of a person with
the value of his existence. I attempt to forestall this very confusion in my essay,
The value of persons does not oblige us to maximize the number of
people in existence; it obliges us only to respect the people that do exist.
And respecting these people is not necessarily a matter of keeping them in existence;
it is rather a matter of treating them in the way that is required by their
personhood -- whatever way that is. The Kantian objection to suicide, then,
is not that it destroys something of value. [p. 14]
This passage is sufficient to show that the confusion inherent in the Reductio
Argument, like the argument itself, is not mine.
Note that the argument implicit in the passage from
which Kamm draws her Reductio Argument is more like the one that she calls the
Exchange Argument. Among all of the arguments against suicide that Kamm attributes
to me, the Exchange Argument is the only one that I actually give. This argument
says that to kill oneself solely for the sake of obtaining benefits is, in effect,
to trade or exchange oneself for those benefits, and hence to treat one's value
as commensurable with theirs. Kamm offers two objections to this argument.
Kamm's first objection is that the intrinsic value of an artwork does not forbid
us from selling the artwork for money. This case is not relevantly similar to
that of self-interested suicide. What one gives up in selling an artwork is possession
of the artwork, in return for possession of something else. The artwork still
exists, to be appreciated by its new owner. Passing the ownership of an artwork
to another appreciative owner is perfectly compatible with regarding its beauty
as priceless in the Kantian sense (and, further, with regarding oneself as no
more than the steward of its beauty). What would be incompatible with respecting
the beauty of an artwork, in this sense, is destroying it in order to produce
ordinary goods -- say, by burning a painting in the fireplace because one has
run out of kindling. But then, burning an artwork for kindling would ordinarily
be objectionable. Not coincidentally, it's also what would be analogous to self-interested
Kamm's second objection to the Exchange Argument is that killing oneself in order to obtain benefits does not involve commodification of the sort committed by someone who sells a kidney for cash. But the objection to self-interested suicide is not that it involves treating oneself literally as a commodity or as having a price in the literal, monetary sense; the objection is that it involves treating oneself as commensurable in value with ordinary harms and benefits and hence as having a price in the Kantian sense. The fact that self-interested suicide doesn't involve literal commodification is thus irrelevant to the Exchange Argument.
Kamm rightly points out that, although exchanging a kidney for money is objectionable, exchanging a kidney for pain-relief is not. I wonder how Kamm expects this example to support her case. She sees that a kidney shouldn't be treated like a used car; does she think that a person should be treated like a kidney? A kidney is indeed something to be discarded when it does more harm than good. But that's the difference between a kidney and a person. And that -- the fact that a person may not just be ditched like bum organ -- that is the basis of the Exchange Argument.
The Exchange Argument is a version of the more general
Kantian argument against treating a person merely as a means. Kamm tries out various
interpretations for the Kantian notion of treating a person as a means, but none
of her interpretations completely succeeds. Maybe I can explain why.
At the heart of Kantian ethics is the insight that there are different ways of valuing things, and that these different modes of valuation have distinctive expressions in thought and behavior. This view sets itself in opposition to the consequentialism that currently prevails in Anglo-American ethics, according to which there is only one way of valuing things -- namely, by ranking them as possible objects of preference or choice. This latter view implies that valuation always sets the terms for a possible exchange, of lower-ranked alternatives for higher, with the possible exception of alternatives that are not eligible for exchange because they are strictly preferable to all the others.
Kamm continually wonders how much we must value a person in order to treat him as an end, as if the Kantian notion were that of according the person, or his existence, a privileged position in our preference ranking. But the Kantian notion is that of valuing a person in a special way, not to a special degree -- indeed, in a way that doesn't admit of degrees. It's the notion of a privileged mode of valuation, which is naturally expressed in some actions and naturally incompatible with others.
When the Kantian objects to treating people in ways that are incompatible with respect for them, his objection rests on an interpretation of what these objectionable forms of treatment mean, what sort of valuation they express. (His objection depends, that is, on an interpretation of the agent's maxim.) The Kantian says: To destroy something just because it no longer does one more good than harm is to treat it as an instrument of one's interests. This statement about what an action means, in the context of its reasons, cannot be reduced to a statement about how it ranks things or what preferences it expresses.
I suspect that Kamm is puzzled about the notion of treating persons as means because she expects a kind of understanding that isn't to be had. She would like to see treating persons as means reduced to some pattern of preferences or choices -- to something that can be tested simply, without evaluatively loaded interpretations of the agent's behavior. No such reduction is possible.
Kamm sometimes considers how a person's value entitles
him to be treated, but she thinks that such considerations militate in favor of
assisted suicide rather than against. For she thinks that the way to respect the
rational nature of a person is by deferring to his choices, as exercises of that
nature. Thus, assisted suicide "may protect the person, even if it eliminates
him, for it protects the fulfillment of his reasoned choices" (16).
This statement begs the question. The question is whether the self-interested choice of suicide can really be a "reasoned choice" -- this question being foundational, in Kantian ethics, to the question whether such a choice is morally permitted. The answer is that the self-interested choice of suicide cannot be an exercise of rationality, because it entails treating oneself as an instrument of one's interests, which is incoherent. That's why this choice is not morally protected. One's value as a rational being cannot require that others defer to one's irrational disregard for that same value.
Maybe Kamm thinks that self-interested suicide doesn't entail disregard for one's own value, because deference to one's choices is all that one's value demands of anyone, including oneself. Maybe she thinks, in other words, that respect for oneself as a rational creature requires only that one permit oneself to make choices, so long as they are rational in the sense of serving one's interests.
One problem with this thought is that it would eliminate any application for the concept of self-respect. It would imply that there is nothing that self-respect forbids one from doing to oneself provided that one gains by doing it; whereas the concept of self-respect just is that of a constraint on what one may do for gain. If any self-serving choice were ipso facto self-respecting -- if it were impossible to debase oneself in the pursuit of one's interests -- then there would be no such thing as self-respect. (And if there were no such thing as self-respect, Kant might ask, how could there be such a thing as morality?)
More importantly, to permit oneself the choice of suicide for the sake of one's autonomy is to treat a single exercise of that autonomy as worth the sacrifice of one's autonomous self. It's to treat oneself like a stick of dynamite, which realizes its nature by blowing itself up. That's fine for dynamite, which is a means of producing explosions, but an autonomous agent is not a means of producing choices, any more than he is a means of promoting interests.
Kamm believes that the premises from which I reason
about assisted suicide would lead to unacceptable consequences, because they would
rule out other practices that are clearly permissible. These practices include
relieving pain with lethal doses of morphine (MPR, as Kamm calls it) and "euthanizing
a cat to stop its pain." In fact, my reasoning would permit these practices in
many cases, just as it would sometimes permit directly helping a person to die.
Kamm underestimates the range of cases to which these permissions would apply. Because I describe the patient in an example of hers as "a pain-fleeing animal," she concludes that I would countenance suicide "only when it cannot be the choice of a rational, responsible person" and not when "rational nature is still being exercised but has a very restricted scope" [25 - 26]. But my description of her example is just that; it is not a general criterion for the permissibility of the practices in question.
My view is that rational autonomy is a complex state that is subject to infinitely subtle grades of deterioration, which can justify a person's death, on the one hand, but can also undermine his capacity to choose it, on the other. When exactly the justification becomes sufficient, and when exactly the patient's capacity becomes insufficient, are questions that I do not address. They arise in what I have called the twilight of the patient's autonomy, where he both is and is not a responsible agent.
These complications do lead me to doubt whether assistance in dying, when morally permissible, is best conceived as assistance in committing suicide. The general enthusiasm for this conception seems to spring from the hope of enlisting the patient to bear responsibility for his own death, leaving the survivors with clean hands. But I think that anyone who is ready for assistance in dying is usually past the point where he can bear full responsibility for the decision, no matter how he may participate in it. He wouldn't ordinarily be a candidate for assistance in dying unless he was somehow diminished as a person, and thus diminished in responsibility.
Still, to say that any candidate for assistance in dying must be diminished as a person is not to say that he cannot be a person at all. The view stated in my essay is that assistance in dying is morally justified to spare the patient from degradation. This view could hardly justify withholding such assistance until there was nothing left to degrade.
In Kamm's opinion, however, my view is still too
restrictive, because it wouldn't accept pain as sufficient justification for MPR
or assistance in dying. My essay already expresses disagreement with Kamm on the
moral significance of pain, and I must say that her new remarks on the subject
continue to trouble me.
For example, she says: "I think there is more urgency in helping a person in unbearable
pain to die than in helping a demented person not in pain to die" [p. 25]. To
be sure, pain is a more urgent condition than dementia if the former is unbearable
and the latter is not. But does Kamm assume that dementia is always bearable?
Is pain the only thing that we cannot bear?
I suspect that when Kamm speaks of pain, what she really means is suffering, or
perhaps pain-and-suffering. What she means to say, I think, is that a person who
is suffering with pain requires relief more urgently than a demented person who
isn't suffering. I agree. But I imagine that dementia can occasion profound suffering,
and I see no reason to think that the suffering occasioned by dementia is a less
urgent matter than that occasioned by pain.
Moreover, I think that suffering is precisely that which, as it increases, tends to make one's condition unbearable. I am therefore inclined to understand suffering in the same terms that I applied, in my essay, to the unbearable -- that is, as a distressing perception of actual or threatened disintegration in the self. To suffer with pain is to feel oneself falling apart under it, overwhelmed by it, coming undone. So understood, suffering doesn't necessarily accompany pain, and it doesn't accompany pain exclusively. But it does necessarily touch one's dignity -- the value that one has by virtue of being a person and that is jeopardized when one falls apart.
With these observations in mind, let me turn to
the practices that Kamm thinks my view would restrict or forbid unduly. I'll begin
with euthanasia for cats.
Kamm speaks of "euthanizing a cat to stop its pain." Now, I have owned several cats, some of which have been in pain on occasion, and two of which were eventually euthanized. Never have I considered, nor do I think that most veterinarians would consider, euthanizing a cat for reasons that could be summed up by phrase "to stop its pain" -- as if euthanasia were just another form of analgesia. What is usually required as justification for euthanizing an animal is that it is suffering. And anyone who has seen an animal suffer is aware of having seen it reduced to a state unworthy of it, a state that is not just unfortunate but somehow unfair, an insult supervening on its injury.
In euthanizing an animal, then, one has the sense of doing it a kindness, of course, but also of doing it justice -- which is all that I would ask on behalf of persons. Of course, doing justice to a person is different from doing justice to a cat, since they are very different creatures. But the basic principle is the same: kindness must be tempered with respect.
As for MPR, my position is the same as it is toward
other lethal interventions. When MPR
is clearly permissible, the reason is that the patient's pain is part of an inexorable
process of deterioration, which we may cut short, or risk cutting short, without
disrespect to the person's dignity.
Kamm says "It will be a big problem for any objection [to assisted suicide] if it also requires us to give up the permissibility of giving morphine for pain relief when this foreseeably causes death" . But Kamm herself would deny the permissibility of this practice in some cases (lethal doses of morphine for a broken leg); and I accept its permissibility in some cases (lethal doses of morphine to relieve terminal suffering). The question is not whether MPR is permissible but when it's permissible, which depends on why it is to be permitted.
Kamm thinks that MPR should be permitted solely on the grounds of the patient's interests; I think that the patient's dignity must be taken into account. I see nothing in Kamm's critique, or in her earlier article, to show that my reasoning rules out MPR when it should be permitted.
*Work on this paper was supported by a fellowship from
the National Endowment for the Humanities, and by a sabbatical leave from the
College of Literature, Science, and the Arts, University of Michigan. An earlier
and very different version was presented to the Philosophy Department and the
Center for Ethics and Humanities in the Life Sciences at Michigan State University.
I received helpful comments on that version from Elizabeth Anderson and Stephen
Darwall, both of whom have also contributed significantly to my thinking on
this subject through their published work. I also received comments from Bette
Crigger and an anonymous referee for The Hastings Center Report. For
comments on the present version, I am grateful to Sally Haslanger, Connie Rosati,
Tamar Schapiro, and Brian Slattery. Back
1. New York Review of Books, March 27, 1997, pp. 41
- 47. The brief was submitted in the case of WASHINGTON et al. v. GLUCKSBERG et
al. Links to briefs and opinions in this case can be found on the World Wide Web
at http:// ethics.acusd.edu/ euthanasia.html. Back
2. I discuss evaluations of this kind in "Well-Being and
Time," Pacific Philosophical Quarterly 72 (1991): 48 - 77. Back
3. See Peter Railton, "Facts and Values," Philosophical
Topics 14 (1986): 5 - 31. Back
4. See Elizabeth Anderson, Value in Ethics and Economics
(Harvard: Harvard University Press, 1993). Back
5. "Self-Interest and Self-Concern," Social Philosophy
and Policy 14 (1997): 158-178; reprinted in Self-Interest, ed. Ellen
Paul (Cambridge: Cambridge University Press, 1997). This paper is also the source
for my statement of the problem in the preceding section. Back
6. The points made here and in the following paragraph
appear in Anderson, Value in Ethics and Economics, p. 26.
7. This point, too, is made by Anderson, ibid.
7. Here I am making a leap that requires more justification
than I can provide in the present context. I am equating the value that we appreciate
in caring about a person with the value that we appreciate, somewhat differently,
in respecting that person in the Kantian sense. I defend this equation in "Love
as a Moral Emotion," Ethics109 (1999): 000 - 000. Back
9. "A Right to Choose Death," Boston Review, Summer
1997, 20 - 23, pp. 22 - 23. Back
10. The interpretation of Kant expressed in this paragraph
is not uncontroversial. I defend it at length in "Love as a Moral Emotion."
11. For a Kantian argument along these lines, see Thomas
E. Hill, Jr., "Self-Regarding Suicide: a Modified Kantian View," in Autonomy
and Self-Respect (Cambridge: Cambridge University Press, 1991), 85 - 103.
12. Actually, I am inclined to think that the dignity of
books or flags is borrowed from the dignity of personhood; but this question
is beyond the scope of the present paper. Back
13. I believe that this feature of dignity-values explains
why the permissibility of euthanasia and assisted suicide is limited mainly
to cases of terminal illness. Felicia Ackerman has recently claimed that such
a restricted permission is unstable ("Assisted Suicide, Terminal Illness, Severe
Disability, and the Double Standard," in Physician-Assisted Suicide: Expanding
the Debate, ed. M. Pabst Battin et al. [New York: Routledge, 1998],
149 - 61). She argues that assistance in dying must be permissible either for
all competent adults or for none. I agree with Ackerman that the arguments usually
offered in favor of assistance in dying cannot be restricted to cases of terminal
illness, although their proponents often adopt that restriction anyway, without
justification. As Ackerman shows, for example, the arguments of "The Philosophers'
Brief" support assisted suicide for everyone if they support it for anyone.
But I think that the Kantian view can justify the restriction, and that its
ability to do so counts in its favor. Back
14. Kamm asks this question in section IV of "A Right to
Choose Death." Back
15. "Against the Right to Die," Journal of Medicine
and Philosophy 17 (1992): 665-81. Back
16. This appendix responds to criticisms contained in Kamm's
contribution to the present symposium ("Physician-Assisted Suicide and the Ground
of Value," ___). I am grateful to Kamm for providing this opportunity to clarify
my essay, which was not originally written for a philosophical audience. For
comments on this reply, provided on very short notice, I am grateful to Nomy
Arpaly, Steve Darwall, and Connie Rosati. Back
17. This brings me to one misunderstanding for which I
am probably responsible. Kamm seems to think that I am opposed to considering
a person's interests at all in justifying his death. The question I raise, however,
is "whether people are morally permitted to end their existence solely because
they find it unrewarding" (13). My answer to the question is that "I reject
the principle that a person has the right to end his life solely on the grounds
of the benefits he will thereby obtain or the harms he will avoid" (4). The
objection, then, is not to considering a person's interests but to considering
them solely, or exclusively; specifically, the objection is that destroying
a person solely on the grounds of his interests would be an affront to his dignity.
I then go on to argue, however, that when a person's dignity is compromised,
it may militate in favor of his destruction. I therefore support "dying for
the sake dignity, not for the sake of self-interest" (16). Unfortunately, this
last phrase seems to imply that considerations of self-interest are to be excluded
entirely. So interpreted, the phrase is too strong. For when a person's dignity
serves as grounds for his death, then his interests may also be taken
into account, without any disrespect to his dignity. Back
18. Here, as in my essay, I draw on Elizabeth Anderson's
Value in Ethics and Economics. Back
19. Here I must distinguish my view about the morality
of these practices from my view about their legalization. See note 22, below.
20. For this account of suffering, see Eric J. Cassell,
"Recognizing Suffering," The Hastings Center Report, May-June 1991, 24
- 31. Back
21. See note 17, above. Kamm presumes (17) that cats do
not have the sort of intrinsic value that would entitle them to respect. To
be sure, cats do not have that value which belongs to persons by virtue of their
rational nature. But a cat's interests would have no claim on our concern unless
the cat itself had a prior claim. The question is what sort of concern is appropriate
to have for a cat -- a question that lies beyond the scope of this paper.
22. This statement applies to the moral position outlined
in the present essay. It does not apply to the position that I stated in "Against
the Right to Die," which is about the legal status of euthanasia. I do not think
that the considerations adduced in that paper apply equally to MPR. Hence I
do not favor similar legal treatment for these practices. Back