Argument: Withdrawing life-support should not be to end life, but for other purposes
Patrick Lee. "Personhood, Dignity, Suicide, and Euthanasia". The National Catholic Bioethics Quarterly." Autumn 2001, Vol.1 No.3 - "It has sometimes been objected that there is no morally significant difference between actively killing a dying person and letting him die.7 Therefore (the argument continues), since everyone admits that it is sometimes morally right to let someone die (that is, choose not to use a lifesaving treatment), then it also is morally right in some cases to choose to kill a dying person. However, this argument is surely mistaken. If killing were the same as withholding or withdrawing lifesaving treatment, then everything we did other than lifesaving attempts would be cases of killing. For in every choice we make to pursue some good other than the saving of someone's life we are doing something which has the side effect of not saving someone's life. If this objection were correct, then, as John Finnis points out, the choice to take one's children for a walk, thus passing up the opportunity to take a plane to Calcutta to save street children, would be as murderous as deliberately blanketing those same children with machine gun bullets.8 We cannot pursue all of the basic goods all of the time, but we are morally required to maintain an openness or appreciation for all basic human goods, including human lives. And that is incompatible with a choice to destroy one instance of a basic good for the sake of others.
It is important to see that the difference between intentionally killing and causing death as a side effect is not primarily a difference in physical behavior. The same physical behavior -- for example, injecting a patient with morphine -- might in one case be carrying out a choice to relieve pain with the side effect of hastening death, and in another case carry out a choice to kill in order to relieve pain. Although the external results are the same, there is a tremendous moral difference between the two cases.
So not doing something -- an omission -- can in some cases be a way of intentionally killing someone. Clearly, if someone withholds needed insulin from his wife in order to end her life, he intends her death just as much as if he had deliberately dropped arsenic in her orange juice. Similarly, if a relatively nonburdensome lifesaving treatment is withheld or withdrawn then the reason probably is that one wants the death -- in this case the omission is the means chosen for the sake of the death.9
Of course, really causing death as a side effect is not always morally right either. How does one decide when it is? Frances Kamm has argued that everyone already admits that we may let someone die, if doing so would be for the patient's overall benefit. In other words, we sometimes make the judgment (according to Kamm) that, "in this particular case, the greater good for the patient is relief of pain, and the lesser evil is loss of life ... "10 But if that is so, she argues, it should also be morally right to kill the patient in order to bring about this greater good. I believe Kamm is right that if that were the basis for not adopting those means, then it would be inconsistent to say that the letting die can be permissible but not the killing. However, contrary to what Kamm assumes, people often make such judgments (that they should withhold lifesaving treatment) on the basis of something other than the denial of the patient's intrinsic dignity (which is what the judgment that a patient would be better off dead amounts to).
Think of a concrete case. The patient who forgoes chemotherapy does not usually say to himself, "The total consequences of living two months without chemotherapy will be objectively better, overall, than all of the consequences of living six months with chemotherapy." No, there usually is no futile attempt to calculate what all of the consequences will be in the two different scenarios. Rather, people usually make such a judgment on the basis of the belief that their responsibilities to family and others could best be carried out in one way rather than the other. The criterion for whether one should do something that causes bad side effects is not whether doing so will produce the greatest net good, since that is a judgment that cannot be objectively assessed, but whether doing so is just and consistent with all of one's responsibilities."