In general, we have to avoid unwanted arbitrariness in ethics. Arbitrariness is closely related to the notion of asymmetry. Symmetry means that an object looks the same after performing a transformation. Consider a perfect sphere with a uniform color. This object has rotational symmetry: if you close your eyes, I rotate the sphere, and you open your eyes again, you are not able to tell whether I rotated the sphere. No point on the sphere is special. Now you can randomly select a point on this sphere and draw a small line there. Suddenly, all rotational symmetry is gone. You can see when the sphere is rotated. In drawing the line that breaks the symmetry, there is arbitrariness, because you did not follow a rule when selecting the points and the direction of the line. The points on the line are special points of the sphere.
If we see an asymmetry in ethical principles, we have to be cautious, because it might involve unwanted arbitrariness. Therefore, most ethical principles should not have an asymmetry. There are however two fundamental principles that do have an asymmetry, one principle in population ethics, and one in rights ethics. For those two principles, we have to check whether they involve unwanted arbitrariness. In particular, as there is arbitrariness, we have to check whether there is something unwanted with those principles.
The first asymmetry is present in population ethics. Consider the phrase: “it is always bad to bring a person into the world who has a negative welfare, but it is not always good to bring a person into the world who has a positive welfare.” This phrase contains an asymmetry, because switching the words “bad” and “negative” with the words “good” and “positive” does not give the same meaning. According to this ‘procreation asymmetry’ in population ethics, adding an extra person with a negative utility (negative preference) always lowers the aggregate utility (total welfare), but adding an extra person with a positive utility does not always increase aggregate utility.
This asymmetry is derived from one of the most basic ethical principles: the maximum self-determined relative preferences principle or variable critical level utilitarianism (and all its variations and special cases). In such theories, the aggregate utility or total welfare consists of the sum of everyone’s relative utilities, where a person’s relative utility is his or her utility (e.g. individual welfare) minus a (self-chosen) critical level. The asymmetry kicks in when we condition that the critical levels may not be negative: they should always be zero or positive. The justification for a non-negative critical level, is that if a person chooses a critical level, that person should be willing to accept a life with a utility equal to that chosen critical level. No-one is willing to accept a life with negative utility. Everyone prefers a higher utility (higher welfare, higher preference) above a lower utility, so there is an asymmetry: increasing welfare is wanted, decreasing welfare is not wanted. This asymmetry is based on what is wanted, so as a result the arbitrariness is not unwanted.
The second asymmetry is present in rights ethics. Consider the phrase: “Harming you always violates your rights, but not helping you not always violates your rights.” This phrase contains an asymmetry, because switching the words “harming” with “not helping” does not give the same meaning. According to this asymmetry in rights ethics, negative rights (not to be harmed by others, or to not help others) trump positive rights (to be helped by others or to harm others). When there is a conflict, the right not to be treated by others in some way is more important than the right to treat others in some way.
Similarly, there is an asymmetry between active and passive rights. Consider the phrase: “You have the right not to be harmed by others, but not the right to harm others.” Switching “not to be harmed by others” with “to harm others” changes the meaning. According to this asymmetry in rights ethics, passive rights (to be treated by others in some way) trump active rights (to treat others in some way).
These two asymmetries are derived from another one of the most basic ethical principles: the mere means principle or basic right not to be used against one’s will as a means to someone else’s ends. Consider four rights: the positive, active right to kill others against their will, the positive, active right to use others as means against their will, and their negative, passive counterparts: the right not to be killed by others against one’s will and the right not to be used as means by others against one’s will. I will argue that only the latter (the negative, passive right not to be used) is special. This exclusiveness breaks the symmetry with its positive, active counterpart. (We do not need to consider the more trivial positive, passive right to be killed by others, nor the negative, active right to not kill others.)
This negative, passive right not to be used against one’s will, is related to the mere means principle. It can be explained using the trolley problem. In the first trolley dilemma, called ‘the switch’, a runaway trolley is about to kill five people on the main track. The only way to save themselves, is if one of the five people pulls a switch that directs the trolley to the side track, killing one person. In the second trolley dilemma, ‘the bridge’, there is no side track, but the five people on the main track can save themselves by pulling a heavy person from a bridge (suppose they can pull a lever that topples the bridge). The heavy person falls in front of the trolley, the trolley is blocked and the person is killed.
Consider everyone having the negative, passive right not to be killed by others, even when being killed by others minimizes the number of people dying. In that case, pulling the switch and pulling the heavy person from the bridge are not allowed, because the persons on the side track and the bridge have the right not to be killed by others. (If the trolley kills the five people, they are not killed by the persons on the side track or on the bridge, because if those persons were absent, the five people would still die.) Now consider the positive, active version of this right: everyone has the right to kill others. In this case, pulling the switch and pulling the heavy person are allowed. Which of those two rights is most important?
The issue with these rights to kill and not be killed, is that there is always at least someone who prefers the absence of other people. If the one person on the side track has the negative right not to be killed, the five people on the main track prefer the absence of the one person, because then they are allowed to pull the switch. Similarly, if the five people have the positive right to kill the one person, the one person prefers the absence of the five people, because in their presence the five people could choose to kill the one person. There is a symmetry between the positive and negative rights: either A wants B to be absent, or B wants A to be absent. In other words: both the rights to kill and not be killed have costs for others: adding people who have this right might limit the freedom of other people. When the side track is empty, the five people can save themselves by pulling the switch. But adding a person with the negative right not to be killed on this side track harms the five people on the main track: they can no longer save themselves.
Every decision in the switch case is always costly or harmful to at least someone. The only option we have, is to minimize costs. So we can either choose the positive right to kill or the negative right not to be killed, depending on the situation. When harms are minimized by the positive right to kill, as in the switch dilemma, we can choose the positive right. In the switch dilemma, with the positive right only one person prefers the absence of other people, whereas with the negative right not to be killed, five people prefer the absence of only one person. So in the switch case, the positive right prevails. But due to the symmetry, if there were five people on the side track and only one person on the main track, the negative right not to be killed prevails and the one person on the main track is not allowed to pull the switch.
Now consider the positive, active right to use other people against their will as means to one’s own ends. In the switch case, this right is never violated, because even when the person on the side track is killed, that person was not used as a means. In the absence of the person, the five people could still pull the switch and be saved, so the presence of the person on the side track is not necessary. The right to use others does not determine the permissibility of pulling the switch.
The right to use others becomes non-trivial in the bridge dilemma. The presence of the heavy person on the bridge is required to save the five people. Without the person, the trolley cannot be blocked, so the heavy person is used as a means (as a trolley blocker). When the five people have the right to use others, the heavy person prefers the absence of the five people on the main track, because their presence endangers the heavy person on the bridge: they can exercise their right to use the heavy person. The five people on the other hand, prefer the presence of the heavy person: without that person, they cannot pull anyone in front of the trolley so they cannot save themselves. So again, this positive, active right to use others is harmful: if there are people who have this right, the freedom of other people shrinks.
So we are left with one right: the negative, passive right not to be used by others as means against one’s will. This right is very special, because it harms no-one. If the heavy person on the bridge has this right, the five people on the main track are not made worse off compared to the situation where the heavy person was absent. In the switch dilemma, the right not to be used by others does not determine the permissibility of pulling the switch. Even if the one person on the side track has this right, the five people on the main track do not violate this right when they pull the switch, because they do not use the person on the side track as a means.
In summary: there is a symmetry between the positive, active right to kill others and the negative, passive right not to be killed. But there is an asymmetry between the positive, active right to use others and the negative, passive right not to be used. The latter asymmetry is the result of preferences of individuals: what individuals want. We can choose the negative and passive instead of the positive and active version of the right about usage, but this choice is arbitrary. However, due to the asymmetry in preferences, this arbitrariness is not unwanted: it fits with what people can want about absence or presence of other people. In this sense, the choice for the negative right is not arbitrary: we followed to rule to choose the right that best fits people’s preferences.
Hence we select the basic right not to be used by others against one’s will as a means to someone else’s ends. This basic right underlies the asymmetries that we see in rights ethics. First, we have the asymmetry that negative rights trump positive rights. Consider the positive right to be helped by others. This right means that others have a duty to help you. However, if you force others to help you, those persons are used against their will: their presence is necessary in order to help you, and they do not want to help you (for example because they want to help someone else instead). So their basic right is violated. The fact that you cannot force someone to help you, but you can force someone not to use you, means that your positive right to be helped by others is weak. Second, we have the asymmetry that some passive rights can trump active rights. The right not to be harmed by others is stronger than the right to harm others, when ‘harming others’ is interpreted as ‘using others as means against their will’.