In this paper, I will explore if there is a morally relevant difference between doing and allowing harm. To do this I will examine what ethical theories give this difference weight and which do not. I will apply the theories of deontology and consequentialism to two cases: one case where there appears to be a morally relevant difference between doing and allowing harm, and one where there does not seem to be one. I will end the paper with my personal perspective claiming that there is indeed a morally relevant difference.
In order to explain what the distinction between doing and allowing harm amounts to, we must first discuss consequentialism and deontology. I will begin with a discussion of consequentialism. The only factor with intrinsic moral significance for consequentialists is the goodness of outcomes. That is to say that an agent is morally required to choose whatever option will bring about the best outcomes. It should be noted that consequentialists don’t entirely rule out common sense morality. They recognize that there may be some value in acknowledging a rule of thumb about not doing harm for example. But a rule about not doing harm only has instrumental value. That is it can be used to help determine the goodness of an outcome. In and of themselves, these sorts of things don’t have any intrinsic normative significance (70). Thus, the distinction between doing and allowing harm for consequentialists isn’t one of intrinsic moral signficance. Doing harm may very well be relevant in determining whether an act is good or bad, but the same goes for allowing harm. Doing harm is not intrinsically worse than allowing harm. All else equal, if doing and allowing harm leads to the same outcome, the difference between the two is morally irrelevant.
On page 71 Kagan brings up the “chop up Chuck” case. Essentially, healthy Chuck goes into the hospital for a routine visit. In that same hospital are five patients dying of organ failure. Chuck’s organs are compatible with the people dying from organ failure. Is it okay to chop up Chuck and use his organs to save the five? Would a consequentialist say it’s okay to kill Chuck because it’s best if five live instead? After all, it doesn’t seem to matter if we’re doing harm to Chuck rather than allowing five to die. But this still seems morally wrong.
Considering it seems morally terrible to kill poor innocent Chuck to save five, there must be some other factor at play here. Perhaps the moral status of an act also depends on whether or not we are doing harm to somebody. Commonsense morality suggests that it is more important to avoid doing harm in this case than it is to save the five. “Commonsense morality insists that the possibility of saving five lives — even though this is the best result overall — is not the only factor of intrinsic significance. It is also relevant that the only way to bring about the best results involves doing harm” (72). If it were between failing to produce the best result and having to do harm to produce the best result, according to commonsense morality, what’s more, important is avoiding doing harm. Moreover, commonsense morality does not limit itself to the idea that doing harm has intrinsic moral significance, this factor also possesses weight. In other words, the significance of harm doing is so relevant “that all other things being equal, if an act involves doing harm, it is simply forbidden” (72). Now all this really means is that there is a “constraint” against doing harm. Moreover, because commonsense morality is deontological, any theory that incorporates constraints must be deontological (72-73). So for the deontologist, the distinction between doing and allowing harm comes down to the constraint against doing harm. Harm doing possess intrinsic moral significance. In other words, doing harm (all else equal) may well be worse than allowing harm. Moreover, if there is to be a constraint against doing harm, doing harm must have intrinsic normative significance. “There can be a constraint against doing harm only because the very fact that an act involves the doing of harm weighs heavily against performing it” (84).
Now that we’ve gone over both consequentialism and deontology, it’s clear that the two ethical theories view doing and allowing harm in very different lights. In order to explore if there is a morally relevant distinction between doing and allowing, we will revisit the chop up Chuck case. I will analyze the case using both the consequentialist and deontological perspectives. I will begin with the deontological one. First, I must note that in Chapter 3.4 Kagan intends to use the chop up Chuck case as a way to demonstrate that there is a morally relevant difference between doing and allowing.
Originally, Kagan’s chop up Chuck case involves killing one to save five. However, to make things more equal I will adjust the case so we are analyzing if it would be okay to kill Chuck to save one person dying from heart failure. Chuck cannot possibly live without his heart so it’s his life versus the other guy’s. It seems like a deontologist would be against us killing chuck because of the constraint against doing harm. While it’s still important and bad that one person dies, it’s more important not to do harm to Chuck. In other words, we must allow the other man to die of organ failure. At least for moderate deontologists, they would claim that the threshold against the constraint against doing harm has not been met. The amount of harm to Chuck does not outweigh the good produced by saving the other guy. Considering they’re both innocent, their lives are equal. Moreover, there is a great deal of risk involved. What if Chuck dies and the other guy dies because he rejects his transplant organ. Then, two people are dead. But what if it is five people on the line instead of one again? Is it okay to kill Chuck in this case? The answer here really depends on how you analyze the situation. Absolute deontologists believe that no matter how bad the consequences may be, we must not harm innocent people (79). Of course, even with absolutist deontology, there are some loopholes (which I won’t get into), but in this case, it doesn’t even matter whether you apply the local or global approach to doing harm. Chuck unwillingly dies and isn’t better off even on balance. But what if the case changes so that Chuck’s death saves one million people? A moderate deontologist might say that if enough good is at stake, it’s okay to infringe on the constraint and harm Chuck. This may make it seem like there must not be a morally relevant difference between doing and allowing harm if we can do harm to Chuck in order to save one million people. However, is important to remember that it is in and of itself morally relevant that we had to kill Chuck. Deontologists believe in the intrinsic value of goodness of outcomes and harm (80). In the case of the one million people, the good results simply outweigh having to harm Chuck.
Now that we have analyzed the Chop up Chuck case using a deontological lens, I will move on to the consequentialist analysis. If the case was just about killing Chuck to save one person, it’s probable a consequentialist would say it doesn’t matter if you kill Chuck or allow the other guy to die. Remember, for a consequentialist, it doesn’t matter in and of itself whether you’re doing or allowing anything. What matters is the goodness of outcomes. However, that being said, would it really lead to the best outcome if we chop chuck up and steal his organs? Chances are it wouldn’t. Nobody would ever want to go to the hospital again which would in turn lead everyone to have bad health, some might even die. If that is so, killing Chuck isn’t wrong because there’s a difference between doing and allowing, it’s because it leads to a horrible outcome. Perhaps if we increase the number of people Chuck can save or make it so that nobody finds out about what happens to him then it’s morally permissible to kill Chuck. According to the consequentialist, there need not be any infringing on constraints or anything of the sort. It would be okay to kill Chuck because of the good results it would produce.
Next, I’ll present a case in which Kagan intended to show that there may not be a moral difference between doing and allowing. On pages 98 and 99 Kagan brings up the case of the drowning cousin. You know that if your cousin dies you’ll inherit the family fortune. In the first scenario, you drown your cousin in the bath. In the second scenario, your cousin is drowning and you stand there ready to push him back under if needed, but luckily for you, he drowns all on his own. Instinctually, it seems wrong to say one of these is morally better than the other. In the first scenario, you are doing harm and in the second scenario, you are allowing harm. How can it be that the difference between doing and allowing harm is morally relevant here?
Using a consequentialist lens, what’s morally relevant isn’t whether you did the harm. The only thing that matters is the goodness of outcomes. In this case, the outcome no matter what you do is that your cousin drowns. If you don’t get caught, there isn’t really a relevant moral difference between you drowning your cousin and just watching him drown. Your cousin would’ve drowned either way. Perhaps we shouldn’t drown our cousin and should instead save him because if we do drown him it sets a pretty evil precedent that would likely lead to bad consequences for our future. But according to a consequentialist, the reason for helping your cousin isn’t because it matters whether or not your doing and allowing anything. It’s because maybe your cousin’s death would have worse outcomes overall.
A deontologist on the other hand would analyze this case very differently. Assuming that the outcome is the same either way (your cousin dies and you inherit the family fortune with no guilt or other side effects), then it is still morally less bad to allow him to drown. It is clearly still awful and morally impermissible to let your cousin die, but it is more important that we avoid doing harm. The reason it’s less bad to allow drowning is simply because of the intrinsic normative significance against doing harm. There is a constraint against doing harm and if we must pick between doing and allowing harm in a situation with equal outcomes, we must always allow harm over do harm.
I think the distinction between doing and allowing harm is morally significant. I most closely identify with the stance of the moderate deontologist. I recognize a moral constraint against both doing and allowing harm. I believe, however, that in application, the constraints are different. Actively killing one innocent patient to save another life does not overcome the constraint. On the other hand, I do think it would be morally acceptable to kill one innocent person to save a million innocent lives.
However, there are still problems with taking a moderate deontological stance. While it seems relatively obvious that we can only infringe on the constraint against doing harm if it is outweighed by another moral factor, at what point is the threshold reached? At what point does it become permissible to kill Chuck? How many people need to be at stake? How low does the risk need to be? It seems like knowing these sorts of things is important. I acknowledge that I have no rational calculus to predetermine this balancing equation. In making the decision, roles and intent would be strong factors that I would evaluate. For example, if I were a doctor, I would take my Hippocratic Oath seriously and it would be a higher threshold to overcome the constraint. Similarly, if I stood to gain from this decision I would scrutinize my intent, trying to ensure my decision for action was not consciously or subconsciously biased.
In conclusion, I think that common sense morality is onto something with the moral distinction between doing and allowing harm. Although I can’t proffer rigid criteria for application, I do think there are tools to analyze the decision on a case-by-case basis. This includes considering local and global consequences and closely scrutinizing the moral constraint.
