Why Killing Can Be Morally Comparable to Letting Die
This is one of the core issues in the assisted death debate
One of the fundamental debates regarding assisted death is the distinction between killing and letting die. Proponents don’t think there’s a moral difference between the two such that whenever letting die is permissible, so is assisted death. In contrast, opponents take the position that killing an innocent human being is always wrong even though letting someone die is sometimes permissible.
To be clear, I use the term “assisted death” to refer to the practice of assisted suicide and euthanasia. Opponents of assisted death tend to believe that assisted suicide is also wrong even though the doctor is not killing the patient (like she is in the case of euthanasia). The reason is that they are opposed to people killing themselves and thus are against others helping them kill themselves. Hence, even in the case of assisted suicide, the opposition is based on a moral position against killing.
To show the moral comparability between killing and letting die, I will first argue that both can be equally wrong and then show why both can be equally justified.
The Bathtub Case
Consider the famous scenario proposed by James Rachels. There are two people who stand to gain a large inheritance if their cousin were to die. The first person, Smith, drowns their cousin in the bathtub while the second person, Jones, sits back and watches their cousin accidentally drown in the bathtub but is ready to actively drown the cousin if needed. The former is a case of killing whereas the latter is a case of letting die.
Rachels’ position is that there is no moral difference between the two. I find this intuitive. What makes the two scenarios equally morally wrong is the actors’ evil intentions, their vicious characters, and the negative consequences. If Rachels is correct, then it shows that killing doesn’t have to be worse than letting die in at least some cases.
In response, Scott Hill has pushed back against this thought experiment because he doesn’t think the two scenarios are analogous. According to Hill, “the subject who lets die has both the ability to kill and the ability to let die while the subject who kills lacks the ability to let die” (p. 767). So he proposes the following scenarios instead.
In the first scenario, instead of simply drowning the cousin, “Jones sneaks into the bathroom planning to drown his cousin, and, as Jones enters, the child hits his head and falls face down in the water. But Jones refuses to let his cousin die. Instead, Jones insists on killing him” (p. 768).
In the second scenario, Jones lets his cousin die after the cousin accidentally hits his own head while taking a bath.
Hill believes that the killing in the first scenario is worse than the letting die in the second. He states,
There is something extraordinarily vile and disturbing about intending to kill a child for money, discovering that without one’s intervention the child will very likely die anyway, but then absolutely insisting that the child must die by one’s own hand. This is worse than simply standing by, hoping that the child will die on his own, so that one does not have to kill him to get what one wants. If killing and letting die were equivalent, then one would expect that it would not matter whether Jones insists on letting the child die on his own, if possible, or insists on pulling the child out of the bathtub to make sure that his cousin will not die on his own and then choking or slitting the throat of his unconscious cousin. The acts in that case would be the same. So, in that case it would not matter morally. But it does matter morally. (p. 768-769)
I would agree with Hill here. There is something disturbing about wanting to kill a child for money that makes it worse than the letting die case. However, I don’t think these two scenarios are analogous for a couple of reasons. First, there’s an element of violence in the killing case that’s absent in the other. Hill specifically mentions choking or slitting the cousin’s throat. And second, in the killing case, Jones insists on killing the cousin whereas that intention is lacking in the letting die case. This suggests that Hill’s thought experiment does not control for the person’s preferences either.
To control for the violence component, consider the following scenarios presented by Zak Kopeikin.
Suppose that Smith and Jones are committed to their plans of killing their cousins for money, and that both want their cousins to die with as little violence as possible. Each of their cousins slips and begins to drown when entering the bathrooms. Smith believes that lightly holding down his cousin will be less violent, and so he gently holds down his cousin with one finger; Jones believes that letting his cousin drown is less violent, and so he lets his cousin drown (while being prepared to lightly hold down his cousin if necessary). Their beliefs are both wrong—lightly holding down the cousin involves neither more nor less violence than not doing so because the cousin does not thrash or resist—and so the only differences are Smith’s and Jones’s different beliefs and the actions resulting from those different beliefs—namely, a killing vs a letting-die. And although the contents of their beliefs differ, their desires are relevantly similar: both desire to accomplish their evil goals with as little violence as possible. (p. 197)
Kopeikin believes that this hypothetical shows that when the violence factor is controlled for, our intuitions suggest that killing is not worse than letting die. Additionally, we arrive at the same conclusion when we control for the preference component where neither person cares whether they kill or let die. Consider the following scenarios.
Heads-Indifference. Smith sneaks in and sees his cousin lying unconscious in the water drowning. Smith stops. He is indifferent to whether or not his cousin dies by his own hand. Smith removes a coin from his pocket and decides to flip it. If ‘heads’ is the result, he’ll drown the cousin; if ‘tails’, he’ll let the cousin drown. The coin lands as ‘heads’. Smith shrugs and proceeds with the outcome of the coin-flip. He gently holds down the unconscious child until the child dies.
Tails-Indifference. Up to the coin-flip, everything proceeds in the same way for Jones. His coin lands ‘tails’. He shrugs and proceeds with the outcome of the coin-flip: he stands by, ready to gently hold down the unconscious child, but doesn’t need to do so. The child drowns as Jones watches. (p. 198)
In these indifference cases, Kopeikin believes that our intuitions tell us that they are morally equivalent despite one being killing and the other letting die.
A skeptic may interject here and state the following: The reason why we have to control an agent’s preferences to get this result is because we assume that the preference for killing is worse than the preference for letting die. And the simple explanation for that is this: the action of killing is morally worse than the inaction of letting die.
In response, Kopeikin is skeptical of our intuitions here due to how our perceptions are influenced by the media. He states, “Consider intentional killings first: we’re regularly inundated with stories of gruesome, horrific, intentional killings involving racism, homophobia, pure greed, and other terrible motivations” (p. 200). But these are absent in cases of intentional lettings-die. The latter are portrayed as compassionate or honoring patient wishes. As a result, “we plausibly develop a heuristic that intentional killings tend to be morally worse than intentional lettings-die” (p. 200).
Overall, I tend to agree with Kopeikin. Once we control for violence and the agent’s preferences, I don’t see a moral difference between killing and letting die. And I also find it plausible that our perceptions of killing and letting die are influenced by the negative reports of the former and the more positive reports of the latter.
Another person who has pushed back on Rachels’ argument is Francis Kamm, who argues that the bathtub case does not show that there is no moral difference between killing and letting die. Kamm states, “A test of its truth is to see if it would be permissible to impose the same losses on Jones and Smith, if these losses were to bring their victims back to life. I do not think it would. Although it might be permissible to kill Smith, it would not be permissible to kill Jones” (p. 17).
In response, I don’t share Kamm’s intuitions here if we control for violence and preference. Consider the coinflip cases again where the agents don’t have a preference for killing or letting die. It’s not obvious to me at all that Smith deserves more punishment than Jones does.
The Road Case
Kamm proposes another pair of cases.
(1) We know that if we drive down one road, we will kill someone who cannot move out of the way. The only alternative is to go down a side road, where we risk hurting ourselves. (2) We know that to save someone from drowning, we must go down a side road, where we risk hurting ourselves. I think that an agent is obligated to face a larger personal risk to avoid killing than to avoid letting die. If this is right, then there is a fundamental moral difference between killing and letting die. (p. 18)
The idea behind these cases is this: if killing and letting die were morally the same, then we would not feel as if we needed to take more of a personal risk in one case than another. But since we do, it follows that the moral balance is not the same in both cases. Specifically, the weight of not killing weighs heavier than not letting die.
As presented, I tend to agree that we have more reason not to kill than to let die; however, the cases are too abstract to be sufficient guides. How would I save the person from drowning if I were to drive down the side road? How do the mechanics of that situation work? I can’t imagine it. In regard to avoiding killing the person stuck on the road, that I understand. Presumably, if I drive down the other road, I won’t hit him. But how would I let someone drown by taking the main road? Without more details, it’s hard for me to get a clear idea of how I feel.
Another potential issue is a proximity asymmetry. In the first case, if we take that one road, we hit the person with our car. In the second case, if we take that one road, a person will drown, but Kamm doesn’t clarify where this person is. When I read the case, I imagined that someone out of sensory range would drown. So let’s change the specifics such that the person who drowns is right in front of you. Now, again, I don’t know how this would work mechanically; I have trouble imagining the details, but let’s assume this was the case. Once we do, it’s not obvious to me that we have more of an obligation to take the side road to avoid killing than to let die.
To be clear, I’m not committed to the position that distance does or doesn’t morally matter. I’m merely claiming that a proper comparison between killing and letting die must control for all other factors. So if the case of killing is as up close and personal as killing someone with your car, it must also be in the case of letting die.
What Justifies Letting Die?
The discussion so far has attempted to show how killing and letting die can be equally wrong. I will now show why killing is justified when letting die is justified. So let’s start by showing why the latter can be justified and then see if it can equally apply to the former.
There are two reasons that justify letting someone die: autonomy and well-being.
One of the principles of bioethics is respect for autonomy. Autonomy means “self-rule,” so respecting autonomy means respecting people’s self-rule. The basic idea is that we acknowledge that people are sovereigns over their own bodies such that we ought to respect people’s medical decisions. This includes honoring refusals. So if a patient is refusing life-saving or life-sustaining care, we ought to respect that, assuming they’re making an autonomous choice.
Two other principles of bioethics are beneficence and nonmaleficence. The former refers to the duty to help and the latter refers to the duty not to harm. I see them as two sides of the same coin of well-being. Needless to say, the value of well-being also supports letting some people die because, for some, continuing to live entails extreme suffering. By withholding or withdrawing treatment, we prevent further unnecessary suffering.
The reader should now realize that the two arguments that justify letting some people die can equally apply to the case of killing. First, people can consent to being killed, so the value of respecting autonomy applies. And second, killing someone can also prevent further unnecessary suffering.
Deontology
Those who resist this argument generally rely on some deontological commitment against the act of intentional killing of innocent human beings. They claim that it doesn’t matter if the patient formally consents or if death would prevent further unnecessary suffering. What matters is that the act itself is wrong.
My response is to ask what justifies this deontological constraint against killing. In other words, why is it necessarily wrong?
To be clear, I’m perfectly willing to grant that killing innocent people is generally wrong but that’s because it generally violates people’s autonomy and is harmful. But that doesn’t have to be true. So what’s the justification? Unless one claims that God forbids it, I don’t see how the position is defensible. I have yet to encounter a convincing argument.
Consider the case of euthanizing animals. Almost everyone is in favor of that in some cases, including opponents of assisted death. I think we all acknowledge that it may get to a point where it’s in the best interest of an animal to die. If that’s true, then why wouldn’t it apply to a human being? And if it does apply to humans, then why is it morally impermissible to euthanize a human being with their consent?
Philosophers traditionally construe human beings as the rational animal, meaning that what makes us who we are is our rational nature. If this is true, then it must be something about our rational nature that justifies the deontological constraint against killing. But I don’t see how it does that. Why does the fact that we’re the rational animal mean that it’s always wrong to kill an innocent human being intentionally?
The deontological commitment against killing innocents seems to be paired with a belief in the noninstrumental value of life. Many believe that life is good, and therefore being alive is good. I deny this is necessarily true, but for the sake of the discussion, let’s assume it is. The belief in the noninstrumental or intrinsic value of life can’t justify the prohibition of killing.
First, just because life is noninstrumentally good doesn’t mean that it’s always wrong to destroy it, especially when we do it to avoid the evil of unnecessary suffering. There would need to be some additional argument here.
And second, if we accept the claim that life is noninstrumentally good, then that would require us to accept that life is noninstrumentally good for animals too. If so, that would mean it’d be wrong to euthanize animals, which seems false. This suggests that what has to be doing the argumentative work is the fact that we’re rational. So this goes back to my earlier question. Why does the fact that we’re rational mean that it’s always wrong to kill an innocent human being intentionally? I see no reason to accept this.
Well-Being
I suspect the opposition to assisted death is fundamentally rooted in the belief that death necessarily harms the person who dies. It’s based on the belief that death is prudentially bad. If you didn’t think that, if you thought that death could be in someone’s best interest, then you wouldn’t be committed to a deontological constraint against it.
Think about it this way: Is there a deontological constraint against doing good? Is there some moral rule that says that it’s always wrong to improve people’s health? Of course, not. A prohibition of x only makes sense if we assume that x is bad. This is why I think the core of the debate needs to be around deprivationism.
Deprivationism says that death is bad for you because, and to the extent that, it deprives you of future well-being that you would’ve had had you not died. It also says that death is good for you because, and to the extent that, it deprives you of future ill-being that you would’ve had had you not died. Therefore, according to deprivationism, death is not necessarily bad for you.
I adopt deprivationism because I think it makes sense, but I suspect my opponents don’t adopt it. If that’s true, then they need to come up with an argument against it. Otherwise, we have good reason to think that death can be prudentially good for the person who dies, in which case there wouldn’t be a deontological constraint against killing.
But what about harm to the person assisting the suicide, and the harm to the larger society due to the devaluation of human life? My response is this. Given the moral comparability between killing and letting die, any harm to the person or society from assisted death equally applies to the case of letting people die, and yet we still think it’s justified to honor patient refusals. Once you grant that killing can be morally comparable to letting die, much of the opposition to assisted death falls.
Conclusion
Killing is not necessarily morally worse than letting die. In fact, both can be equally wrong. Furthermore, what justifies letting die equally applies to killing. Both of these claims combined give strong reason to think that killing and letting die are not morally different (even though they’re conceptually different) when all other factors are controlled for. It follows that whenever letting die is morally justified, so is killing, all else equal.
Despite the reasons given, opponents will maintain that there is a deontological constraint against intentionally killing innocent human beings, but I have yet to encounter a convincing argument in favor of it. In fact, there’s good reason to reject this prohibition—namely, on the grounds of consistency. We think it’s okay to euthanize animals sometimes, and there doesn’t seem to be a good reason to think this couldn’t equally apply to humans. The fact that we’re rational doesn’t seem to suffice. And remember, we’re talking about the cases where the patient consents to being killed.
Ultimately, I think the fundamental disagreement is over well-being. Opponents of assisted death believe that death is necessarily harmful to the person who dies. However, deprivationism gives us good reason to reject this claim. And once we do, the deontological constraint becomes irrelevant. Actually, it renders the killing vs. letting die distinction irrelevant as well, but I’ll elaborate on this claim in a future article.