MOST people believe there is an important ethical distinction between actively harming someone (i.e. taking an action that harms them) and passively harming them (i.e. not taking an action to help them), and in particular that active harm is inherently worse (morally) than passive harm. It can be difficult, psychology and practically, to believe that no such distinction exists. If allowing someone to starve to death is as bad as causing them to starve to death, then one begins to wonder whether someone who spends a lot of money on a fancy car (instead of saving children from starvation by donating that money to a charity) is ethically better than someone who, for example, refuses to feed a child that they have adopted, and the child dies because of it.
Despite the disturbing consequences, let’s consider whether a case can be made against the idea of active harm and passive harm being inherently different from a moral standpoint. Consider the following:
As in the Poe classic, a man is tied to a board, and above him a pendulum, with a sharp blade attached to it, swings back and forth. We will compare two scenarios. In both, imagine that you are standing near the man, and by some means are being monitored by a machine that operates the pendulum. In both cases, assume that you have complete information about the situation, and that there is no good reason that this man should die. Furthermore, assume that he has no hope of immediate escape (beyond what is stipulated in the scenarios) and that his death will involve enormous suffering.
Scenario 1. (Passive Murder)
As the pendulum swings, it is slowly being lowered by the machine that operates it. If the pendulum gets low enough, the man will die. The machine monitors your motion. It has a rule programmed in it such that if your first movement is towards the man, the pendulum will be shut off. If, on the other hand, your first movement is away from the man, the machine will not react to your movement in any way (i.e. the pendulum will continue to be lowered and the man will die).
Scenario 2. (Active Murder)
As the pendulum swings, it stays at a fixed height (i.e. it is not being lowered toward the man). The machine monitors your motion such that if your first movement is away from the man, it will react by beginning to lower the pendulum (which will eventually kill the man). If, on the other hand, your first movement is towards the man, the machine will not react in any way (i.e. the pendulum will continue to remain at a fixed height, so the man will live).
Let’s compare these two scenarios. In Scenario 1, the man is already going to die unless you act by moving towards him. Hence, if you move away from him, you will be failing to help him, but you will not be actively killing him since the machine is already in the process of killing him and will not change its operation in any way if you move away. On the other hand, in Scenario 2, the man is not going to die unless you move away from him. Hence, if you move away from him you are actively killing him, since you are knowingly activating the machine, causing it to begin lowering the pendulum. Therefore, if failing to help a person is ethically superior to actively harming them, then moving away in Scenario 1 (where it has no effect) is ethically superior to moving away in Scenario 2 (where it causes the machine to initiate the death process).
At the same time we observe that despite this active versus passive distinction, the two scenarios are almost identical. In both cases, if you move towards the man he will not be killed, and if you move away from him he will be killed. In both cases it is trivially easy to perform either action. In both cases you have complete knowledge of the consequences of your actions. In both cases you are deciding whether he lives or dies. The only important difference between the scenarios is that in one, the machine happens to be lowering the pendulum already, whereas in the other it happens to be doing nothing. However, after you act (by moving towards the man or away from him) the resulting situation will be identical whether you are in Scenario 1 or Scenario 2.
If you believe that active harm is inherently better than passive harm, then you must therefore conclude that walking away in Scenario 1 is superior ethically to walking away in Scenario 2. Is this something you are truly willing to accept? In both cases walking away will mean the man will die, and preventing his death is equally easy in both cases. Can we really conclude that the initial motion of the pendulum is sufficient to change the ethical characteristics of the problem?