An oft-posed moral dilemma runs something like this:

A runaway train is careering at full speed towards one of two sets of people on the track. Ahead of the train’s current position the track divides into two, the direction of the train being controlled by a set of “points” (sometimes called a “switcher”). After the divide one person is trapped on the first track and five people are trapped on the second track. Person A has no ability to stop the train, but he can control the switcher so that he can divert the train from one track to the other. Doing so will necessarily cause the train to strike and kill the people on that track but will save the people on the other track. What should A do?

Nearly every discussion of this problem launches head first into tackling the issue of whether five lives should be saved at the expense of one or vice versa. Seldom considered is whether A should do anything at all and whether any of the six people in peril have a right to command his assistance. We will therefore postpone discussion of the “many versus few” problem until this more fundamental issue has been resolved.

Ignoring, then, the fact that A’s assistance to the people on one track will necessarily cause the deaths of the people on the other, if A can be said to have an obligation to help any of the six people then this obligation must vest in these people a corresponding right to be helped. This right is manifest in the six people being able to demand the use of A’s resources, diverting them from what A would prefer to do with them towards what the six wish to do with them.

But where does this obligation and corresponding right come from? In the first place, if A did nothing and the train crashed into some or all of the people, A would not have caused this. The train was already running away, the people were already trapped on the track. No positive act of A has done anything to cause this situation. Regardless of whatever else A was doing the train would still have crashed and killed them.

It could still be argued that A’s failure to intervene was an active cause of the crash, i.e. what he didn’t do as opposed to what he did do was a necessary factor. The problem with this, though, is that there are thousands, if not millions, of other things that didn’t happen to intervene and prevent the crash. The train didn’t run out fuel; the wagons were not heavy enough to slow the train down the train to a safe speed; a meteor didn’t fall from heaven and crush the train in its path. Why is the lack of A’s intervention any more of an important absence than any of these examples? It might be retorted that A was in the position to act, that he had the ability to change the outcome whereas everything else depended upon the intervention of nature. But another cause of the crash was the fact that everyone else in the world didn’t help either. Why are they not equally obliged as A is to intervene? May be they didn’t have the ability that A did to prevent the outcome successfully or may be they were too far away, but how far away and how “impossible” does it have to be for them to intervene before the positive obligation for A to do so alone becomes active? How can this be measured non-arbitrarily? At what point do we single out A as being morally culpable?

Indeed, in the absence of any positive act of A being the cause of the crash, the only potential basis of the right of the six people to his intervention is that they have a “need” and A has the ability or “means” to fulfil it. Ignoring then the fact that other people in the world may also have the power to intervene is this a solid basis for this alleged right? The first problem is, again, one of measurement – how do we judge A’s “means” to intervene? Does he have to possess resources that will fully ameliorate the problem or is he still required to act if he can partially intervene? What about his knowledge? What if he doesn’t know what his actions will do or whether they will have any effect? Is he still obliged to do something? The second, more pressing problem, is judging the “need” of those who wish to be helped. What should the “need” consist of? Must it be urgent and life threatening? If so, how urgent and life threatening? Where precisely and non-arbitrarily is the cut off point, one side of which A has no obligation or culpability at all and on the other side he does? What if the “need” consists only of something that affects the quality of a person’s life rather than the life itself? Should this be included or excluded? And again at which level of “quality” and under whose judgment of it should A be deemed as being obliged to intervene?

Lest anyone considers all of the preceding paragraph to be too abstract this type of analysis relates directly to the redistributive efforts of Governments through taxation and the aggrandizement of the welfare-warfare state. The poor have a “need” that can be fulfilled by taking from the rich. People who are oppressed abroad “need” to be saved from the horrors of an evil dictator or an invading foreign power so we are taxed in order to pay for military help to be deployed.

Aside from the question of whether this is just, the economic effect of determining rights and obligations by “need” and “means” can only be wealth destroying. If X has a “right” to have his “needs” met by Y’s “means” then X producing the means to meet these needs himself will become, to him, relatively more costly and less beneficial. He will therefore produce fewer means. Similarly, if Y knows that a part of his productivity will be confiscated to meet X’s needs, then production of these means will become, to him, relatively more costly and less beneficial and he too will produce less. Further, because being in the position of having “needs” is now rewarded with other people’s resources, having “needs” instead of “means” becomes a more attractive option. There will therefore be more people with “needs” and fewer people with the ability to meet them. Overall, therefore, societal wealth will decline. Applying this to the runaway train scenario, if the six victims know that they have a right to be helped in the event of peril, relatively they will have less of an incentive to prevent the perilous situation from arising in the first place. Similarly A, knowing that he will be obliged to divert some of his resources to helping the six will, relatively, avoid putting himself in the position of being able to help by producing less. The result will be more needs and fewer resources to meet them and, hence, decreased societal wealth.

Therefore, in the absence of any positive cause by A of the runaway train (or any contractual agreement that stipulates A will intervene) none of the six person’s has any right to A’s intervention at all and correspondingly A has no obligation to provide such intervention.

Let us now, however, turn to another aspect of any intervention by A – the fact that saving one set of persons will necessarily mean that the other set is killed. Ignoring for the time being the fact that one of the tracks contains more people than the other, if any of the people have a right to intervention then what this entails is that the act of saving them will cause the deaths of one or more others1. What right does of the any of the six people have to be saved at the expense of another’s death? A person might be said to have a “right to life” but what of the “right to life” of the other person? Why is this not important? Furthermore, if A intervenes his intervention will now be a positive cause of deaths of other people. How can this be a moral obligation?

However, the live issue, and the one that always forms the core of the debate, is the fact that one of the tracks contains more people than the other and hence diverting the train onto this first track will kill more people. Is A, if he intervenes, justified in saving more lives at the expense of killing fewer?

There are numerous problems with arguing in the affirmative. First, how we do know that five lives are worth saving more than one? If something is worth something then by definition it is, in some way, valued. But valuations are made by individual human beings and so precisely whose valuation of these lives are we talking about? We could start by only considering the value that the potentially doomed people have of their own lives. But the problem is how do we measure this value and compare it between the individuals? In other words just how we do know that the five people together value their lives more than the single person does? What is the measurement of this value and how do we add it together? What if the five people were all suicidal or indifferent to living whereas the other individual possessed a great zest for life and the last thing he would ever want to do is extinguish it? Is it not possible, in that case, that killing the five would be a better outcome than killing the one? In any case it is never possible to determine as valuations are ordinal not cardinal and can never be added, subtracted, multiplied or whatever for not only a lone individual but also across many individuals.

Further, if we argue that saving a greater number of people is the correct moral choice must we not also consider how many people would be negatively (or even positively) affected by the deaths? What if the single person had a large and close family who would be devastated at his death whereas the other five were single and there is no one to grieve for them? Or what about their past or potential contribution to “society”? If the five people are lazy and idle vagrants who have never wanted to do a day’s work in their lives and have always eaten from the hand of the taxpayer whereas the single person is hard working and productive, wouldn’t more people benefit from the deaths of the five rather than the one? Or, rather, wouldn’t the deaths of the five be less of a tragic loss rather than the death of the one? What if the five people were Hitler, Stalin, Pol Pot, Mao, and Kim Jong-Il whereas the single person was Mother Teresa? Surely we would be justified in letting all of them die if we can save the saintly nun?

Moreover, where the identities of the people are unknown to A how is he supposed to judge all of this in the heat of the moment (assuming he could at all)? Indeed how is he supposed to act morally at all in making this decision?

However, to reiterate, whose valuations are considered is irrelevant as the main problem with any type of “many versus few” analysis is the impossibility of making inter-personal utility comparisons. Even if we say that we are morally justified in killing the people whose lives have the least value to both themselves and others there is no measurement of this value and if there is no measurement then these values cannot be added to together and compared. We cannot, therefore, say that killing any number of people leads to a “better” or more moral outcome if it kills others, how ever many this may be. Any argument to the contrary must necessarily be based on these subjective, immeasurable valuations.

A possible reductio of these points might be something like this: what if the choice was between killing all the people in the world against killing only a handful? Surely it would be better to let the few die in order to save the entire world? But we are simply confronted with the same problems – better to whom and how do we measure it? What of the people (extreme environmentalists, for example) who consider the human race to be a filthy, disgusting, depraved parasite upon the Earth, engaging in endless destruction of the planet and of each other? What of those who believe that a drastic reduction in population is necessary if the world is to continue following a sustainable path? How do we know that the strength of these people’s desire to see the population largely killed off is less than the desire of everyone else to continue living? What if the few survivors pledged their dedication to building a new human race free of the flaws (real or imagined) of the present human race, resulting in a world without war, conflict, poverty and the like? Wouldn’t that be a “better” outcome than saving the present world?

In conclusion, therefore, none of the people on the tracks has any right to any intervention by A; none of them, further, has any right to be saved if such an act kills others; and, finally, A is not justified in saving people at the expense of killing fewer people.

——

View the video version of this post.

——

1 We should point out that if A did not intervene at all and the train careered onto its pre-ordained track then some people would be saved and others would be killed. But in this case the saving of some is not the cause of the death of the others – it was merely a case that some people were in the wrong place and others were in the right place, in just the same way as a person in the middle of the street is likely to get hit by a car while those on the pavement will not. What we are discussing here is the right of any of the people on the tracks to be saved through an intervention that necessarily will kill other people.

Advertisements