2004-01-19 - 1:38 a.m.

This is the best chapter in my thesis-- a chapter praised by Prof. Hansen as having doctorate level. Section 3.2 and 3.3 has been extracted to be a sample paper for me to apply graduate schools in the States. It is a chapter about consequentialism and contractualism. I compare the two dominating moral theories in terms of their different normative implications and foundations. I also discuss a proposal of accomodating contractualism by consequentialism which is suggested by Philip Pettit. This is a rather long chapter and I will paste 3.1 here first and 3.2 and 3.3 in a latter entry.

I have written an undergraduate thesis in 2001 on utilitarianism, a mainstream consequentialist theory. It helped me to win the Best Philosophical Writing Prize at that year and guarantee me a place in HKU's graduate school (amid poor grades in other subjects). It provides me with a soild background to write this chapter. I was asked during the oral examination that either contractualism or consequentialism convinces me as a more plausible moral theory. My answer is contractualism; but I do think the debate is not over yet, and it would still be an interesting battlefield for moral philosophers to argue in future.

Again, due to technical problem, I cannot include footnotes here. Please email me for a full version of the paper if you are interested.

Chapter 3

Contractualism and Consequentialism

One of Scanlon��s main aims in presenting his contractualist account of ��what we owe to each other�� is to undermine the massive influence of consequentialism in moral philosophy. Consequentialism in its purest and simplest form holds that the morally justified option in any choice is the one that can produce the best overall outcome, as judged from an impersonal standpoint giving equal weight to the interests of everyone. In other words, it holds that acts, moral rules, social institutions, practices, law or any other evaluative focal points are morally justified if and only if they can maximize neutral values. According to Kagan��s framework in normative ethics, consequentialism at the factoral level holds that the only morally relevant factor in determining the moral status of an evaluative focal point, say act, is the goodness of its consequences. At the foundational level, consequentialism makes the further claim that the option is morally justified because it can produce best overall consequences, i.e. it is justified by the goodness of its consequences. In this chapter, I will briefly discuss how Scanlon��s contractualism diverges from consequentialism in both the factoral level and the foundational level. Section 3.1 will discuss contractualism as a non-aggregative theory and section 3.2 will discuss it as a non-teleological theory. After locating the main divergence between contractualism and consequentialism, I will focus on a current debate between Scanlon and Philip Pettit for the possibility of a consequentialist accommodation of contractualism. Taking Scanlon��s stand, I will argue against such a possibility and claim that the contractualist account of ��what we owe to each other�� is an autonomous account which cannot be explained within a consequentialist framework.

3.1 Contractualism as a Non-aggregate Theory

The first divergence between Scanlon��s contractualism and consequentialism is their different implications for first-order moral questions. Scanlon holds that an act is wrong (in the sense of ��what we owe to each other��) if and only if it is disallowed by any non-reasonably rejected principles. An act is right (not wrong) if and only if it is distributively, not just aggregatively justifiable to any party affected by the principle. The right act in this sense is unlikely to be the consequentially optimific one because an act may be aggregatively for the best and yet involves unreasonable sacrifice on one of the involved parties and thus is distributively unjustifiable. Alternatively, an act may be distributively justifiable to everyone involved and yet it is not for the overall good aggregatively speaking. For example in the Organ Transplant Case, an aggregatively good option to choose may be the one that kill the innocent and save the five dying patients using his own organs; a distributively justifiable option, on the other hand, is the one not to kill the innocent because he can reasonably reject any principle allowing his being murdered. While goodness of outcome is the only morally relevant factor for consequentialism at the factoral level, contractualism admits of constraints that cannot be broken even if breaking the constraints can bring about better consequences.

Scanlon��s contractualist account of ��what we owe to each other�� is non-aggregative in nature because it holds that the rejectability of principles depends on reasons from various individual standpoints. A principle can be rejected if any individual��s reason of rejection is stronger than other individuals�� reason of rejection for other alternative principles. What are the advantages of this non-aggregative account over those aggregative accounts like consequentialism? Since reasonable rejection depends on a series of pair-wise comparisons among each individual��s reason for or against a certain principle, it can guarantee separate and equal consideration of each individual��s interests. We can illustrate this by using Scanlon��s own famous example. Let me call this ��the case of individual sacrifice for aggregate interests��.

The case of individual sacrifice for aggregate interests:

��Suppose that Jones has suffered an accident in the transmitter room of a television station. Electrical equipment has fallen on his arm, and we cannot rescue him without turning off the transmitter for fifteen minutes. A World Cup match is in progress, watched by many people, and it will not be over for an hour. Jones��s injury will not get any worse if we wait, but his hand has been mashed and he is receiving extremely painful electrical shocks. Should we rescue him now or wait until the match is over?�� (235)

I think no one will disagree that we should save Jones immediately, even though we may have disturbed millions of people who are enjoying the World Cup match. However, it seems ordinary consequentialist reasoning may involve a difficulty in accounting for this judgment. The interests of Jones seem to be out-weighed by the aggregative interests of the millions of people. Suppose the suffering of extremely painful electrical shocks for Jones counts for a degree of harm 1000, and the degree of loss for each individual who are watching the match is 0.01. Assuming there are 10 millions of people watching the match, the degree of harm in not saving Jones is 1000, but the degree of harm in saving him is 100000 (0.01x 10millions). Consequentialist may be forced to say that in this case, individual��s interests should be sacrificed for the aggregate interests of the others.

I am not sure whether there is any non-aggregative consequentialism that can resolve the above difficulty in a satisfactory way. Even if there were, contractualism could still provide a more straightforward and precise explanation of our intuitions in the case of individual sacrifice for aggregate interests. Jones can reasonably reject any principle permitting us not to save him just for the interests of the people watching the match. It is because when compared separately, his reason for not being harmed is much greater than the reason for enjoying the match by any other single individual. According to contractualism, it is wrong not to save Jones in this case.

Contractualism as a non-aggregative normative theory has the advantage of preventing unlimited sacrifice by individuals for aggregate interests. The inter-personal comparison of reasons guarantees separate and equal consideration of each individual��s claims on a one-on-one basis. This is particularly important in the domain of ��what we owe to each other��. In deciding what duty we owe to every other individual, we are considered as equal agents with our claims compared with those of each other separately. No one��s claim can have a greater moral urgency than another��s claim simply because there are more people sharing the same claim. The contractualist explanation of why we should have Jones being saved immediately even at the expense of the enjoyment of millions of people can be extended to cases like the Organ Transplant Case where we should not kill an innocent to save five or even more dying patients. The justification lies not on the aggregate value of outcomes but on our duty to each other individuals.

Another advantage of contractualism as a non-aggregative account is that it can prevent inter-personal trade-off of interests. When an individual raises an objection to a principle, he may make an appeal to its effect on his interests when such principle is adopted. Since contractualism does not allow inter-personal but only intra-personal aggregation�Xthe aggregation of all harms and benefits ��within�� the individual only, no inter-personal trade-off of harms or benefits among individuals is allowed. For consequentialists, the failure to prevent inter-personal trade-off of interests is sometimes a fatal objection because it neglects the differences in individuals�� identity. For example, if taking an innocent A��s life is the only method to save B��s life, a consequentialist may be committed to say that there is no moral difference between the two resulting state of affairs of either killing A or not. If we take A��s life, the result will be the loss of one life. If we do not take A��s life, the result will also be the loss of one life. The only difference is that it is the loss of A��s life in the former case but it is the loss of B��s in the latter. Such difference may not constitute a moral difference for a simple consequentialist. A��s loss of life is said to be compensated by B��s life.

According to contractualism, however, A can reasonably reject any principle permitting the taking of his life to save B��s life. Even though A and B both have a claim to life, A��s reason for his claim is stronger than B��s because he is an innocent and is not responsible for B��s loss of life. He can reasonably reject that principle because the gain of B��s life cannot compensate for his loss of life.

In the area governing our inter-personal relationships with each other individual, contractualism possesses the advantage of respecting the moral status of each individual. Individuals are not separated from their own interests and the interests of an individual cannot be replaced by the interests of another individual. Since the rejection of principles depends on an agent-relative perspective, any principle that neglects the difference between different people��s interests and allows trading-off of interests among different agents can reasonably be rejected. The distinct status of individuals is recognized when we decide what duty we owe to them.

Contractualism provides a better explanation than simple forms of consequentialism in accounting our moral intuitions in the above cases. However, the non-aggregate nature of contractualism may become a burden in some other cases where aggregative reasoning seems to be appropriate. I will explore some of them here:

(1) The case of Equal Harm :

A number of people face harms roughly equal in magnitude (e.g. the danger of losing their lives). They are divided into two groups; one of them has a much larger number of people. Suppose we have no special obligation or relation with them, and we can either save the larger or the smaller group but not both, having the same little cost on us. Which group shall we save?

(2) The case of Small Inequalities in Harm

This case is similar to the case of equal harms. Its only difference is that the harms faced by each member of the larger group are only slightly smaller than the harms faced by each member of the smaller group. For example, people in the smaller group are having the danger of losing their lives while people in the larger group are having the danger of permanent paralysed.

Can Scanlon��s contractualism account for our moral intuitions in the case of equal harm and the case of small inequalities in harm? Our intuitions seem to support that we should save the larger group because other things being equal, we should avoid causing a greater loss of lives for others. However, it seems Scanlon��s contractualist formula cannot provide such an answer because the harm faced by each member of either the larger or the smaller group is the same in the case of equal harms, and it seems no one in the larger group has a reason that can out-weigh the reasons of anyone in the smaller group. In the case of small inequalities of harms, it seems the reasons of a member of the smaller group can even out-weigh the reason of every single member in the larger group. Even if the harm faced by each member from the larger group is faced by more people, the harm for each member from the larger group is still slighter than the harm faced by each member of the smaller group. The objection to being harmed by a member from the smaller group seems to be stronger than the objection raised by any member from the larger group if compared on a one-on-one basis. If it is really the case, then according to ��what we owe to each other��, it is not wrong not to save the larger group in the case of equal harm, and it is even wrong to save the larger group in the case of small inequalities in harm.

Scanlon can respond by admitting that it is indeed not a requirement for ��what we owe to each other�� to save the larger group in the above cases. So far, he can claim that we do not violate our moral duty to each other in not saving the members of the larger group because each of their claims is of no greater moral urgency than the claim of any single member in the smaller group. It is morally supererogatory to save the larger group but this supererogation is not implicit in our moral duty to each member of both groups. Yet, Scanlon does not want to accept this implication because he thinks his contractualism can in fact account for our intuitions in these cases. He wants to argue that ��what we owe to each other�� in fact requires us to save the larger group in these cases. Before examining Scanlon��s own argument, I will first discuss some other possible proposals in arguing that it is in fact a requirement of ��what we owe to each other�� to save the members in the larger group .

One proposal is to compare the reasons of the two groups directly. This means to compare the reason of the larger group and the reason of the smaller group in rejecting principles directly. Such a move allows not only personal reasons, but also reasons from groups of people as legitimate reasons for rejection. We consider the case not only from the standpoint of each individual, but from a group of individuals. This proposal has the advantage of accounting for our intuition that it is not wrong to save the larger group; however, it gives up a central feature of contractualism, namely, the equal and separate considerations of each individual��s reason supporting or objecting to such action. If we accept reasons from a group of people as legitimate reasons for rejection, we will open the gate of unlimited aggregation. In Jones��s case, we may say that we should not save Jones if the number of people enjoying the World Cup match is numerous. Insisting on personal reasons as legitimate ground for reasonable rejection, Scanlon himself rejects such proposal. (230-231)

Another proposal is to appeal to the combined effects in adopting a principle. Consider Sophia Reibetanz��s argument:

���Ksomeone could reasonably appeal, in rejecting a principle, to the combined effects on herself and others of adherence to that principle. Suppose that such appeals can constitute reasonable grounds for rejection of a principle, at least when the effects in question are more harmful than the combined effects of some alternative principle. Consider now any principle requiring or permitting us to save the smaller number in cases of equal harm or cases of small inequalities in harm. Any such principle would lead to a greater total amount of harm than would a requirement to save the greater number in these cases. So it would be true�Kthat any requirement or permission to save the smaller number�Kcould reasonably be rejected by someone.�� (My emphasis)

Reibetanz��s argument is open to two interpretations; and I think both of them are unsatisfactory. The first interpretation is that what ��someone�� in Reibetanz��s argument means is anyone, including people neither of which come from the larger nor the smaller group. It means that any ��outsider�� can reasonably reject the principle permitting us to save the smaller group, just because the general adoption of such principle would bring greater harm in general. The first difficulty with this interpretation is that it is not clear whether this reason for rejection of the principle permitting us to save the smaller group from some ��outsiders�� is stronger than the objection raised by a member from the smaller group in rejecting the principle permitting us not to save the smaller group. The strongest reason an outsider can raise to reject principles permitting us to save the smaller group is that if such principle was adopted, there would be greater harm not for himself but for everyone in general. Anyone may have a greater chance of being harmed because people may have a greater chance to be a member of the larger group. People��s chance of being harmed will enhance if such principle was adopted. However, if compared with the objection raised by a member of the smaller group, whose life is at stake and who will receive actual harm, I think the objection of a member of the smaller group is much stronger than the objection from any outsider. Therefore, even if we appeal to the combined effects on an outsider and the others in general, I think it cannot constitute a reasonable rejection on any principle permitting us to save the smaller group. The second difficulty of this interpretation is that if we allow the effects on others to be appealed to as our own personal reason for rejection, then there may be no distinction between my personal reason and the reasons from other people. All people can have the same appeal to the general effects on all of us. Moreover, if we allow the concern of others to be counted as reasons for my own, then it will unavoidably open contractualism to the danger of unlimited aggregation. I can appeal to the effects on someone else as my reason for rejecting certain principles. This is clearly contrary to the spirit of Scanlon��s contractualism, which guarantees separate consideration of each individual��s claims.

The second way of interpreting ��someone�� in the argument is to read it as someone from the larger group. If it was the case, then it may plausibly avoid the first difficulty mentioned above since the combined effects on the member from the larger group and also on others may possibly out-weigh the reason raised by any member from the smaller group. However, it still cannot avoid the second difficulty and more importantly, it seems it makes the same mistake as the proposal suggesting comparison between the two groups directly. This is because if he appeals to the claims of the other people from the larger group together with his own, he is trying to raise a reason for the group collectively. Again, it is contrary to the individualistic spirit of Scanlon��s contractualism.

I mention the above two proposals in order to highlight Scanlon��s own argument. He cannot appeal to the reasons of the larger group as a whole nor to the reasons of the combined effects of the principles, yet he still thinks the reason to be saved from the members of the larger group is stronger than the reason to be saved by the members of the smaller group, and thus any principle permitting us not to save the larger group can reasonably be rejected. For the case of Equal Harm, he employs the Tie-breaker Argument :

���Kany principle dealing with cases of this kind would be reasonably rejected if it did not require agents to treat the claims of each person who could be saved as having the same moral force. Since there is�Ka positive duty to save in cases in which only one person is present, this means that any non-rejectable principle must direct an agent to recognize a positive reason for saving each person. Since a second reason of this kind can balance the first�Xturning a situation in which one must save one into one in which it is permissible to save either of two people�Xthe reason presented by the needs of a second person in one of these two groups must at least have the power to break this tie.�� (232)

Scanlon claims that a member from the larger group could reasonably reject any principle allowing us not to save the larger group since such a principle simply neglects his presence in the larger group, taking no account of the value in saving his life. If the two groups having equal harms are of the same size (i.e. having the same number of people), then the reason for saving either one of them but not the other is the same. The additional member of the larger group becomes a tie-breaker to the situation. His presence provides an additional reason for saving the members in the larger group.

For the case of Small Inequalities in Harm, a further Rough Equality Proposal can be introduced. Two harms are to be treated as different in size, for moral purposes, only if the one is substantially greater than the other. For example, when one is facing the loss of life and one is facing permanent paralyzed, their harm, though slightly different, are treated as roughly equal in moral sense:

��If one harm, though not as serious as another, is nonetheless serious enough to be morally ��relevant�� to it, then it is appropriate, in deciding whether to prevent more serious harms at the cost of not being able to prevent a greater number of less serious ones, to take into account the number of harms involved on each side.��(239)

With the Rough Equality Proposal, we can also account for the case of small inequalities in harm. It is because according to such a proposal, small inequalities in harm do not constitute harms with different magnitude. Together with the Tie-breaker Argument, this yields the same result as in the case of Equal Harms.

Is Scanlon��s response adequate? Consider the Rough Equality Proposal first. In fact it is not quite clear what the vague terms ��substantially greater (or smaller)�� or ��serious enough to be relevant�� mean. Scanlon mentions that the harm of losing three fingers is not substantially greater than that of losing only two fingers, but how about losing four or five? It seems that whether two harms are treated as having the same magnitude or as roughly equal just depends on what we need in arguments. Nevertheless, it is still reasonable to accept the claim that two harms are morally relevant or roughly equal in size if there is only a slight difference in the seriousness of harm. The intuition behind this is that it would be an open question if a person would choose to die or to be permanently paralyzed, and he could rationally choose to die in such a situation. This can also well explain why the inconvenience in enjoying TV is not morally relevant to Jones��s pain, yet on the other hand permanent paralysed can be treated as roughly equal to the loss of life. Accepting the Rough Equality Proposal, the case of Small Inequalities in Harm is reduced to the case of Equal Harm.

For the Tie-breaker argument, there are more controversies. Suppose the smaller group is made up of one member A only; and the larger group has two members B and C. If we say that the reason presented by C in the larger group have the power to break the tie between A and B, it seems that it is just a disguised argument appealing to the reasons of the larger group collectively. Michael Otsuka argues that Scanlon has illegitimately appeal to the combined reasons of B and C together in outweighing the reasons of A:

��The Kamm-Scanlon argument for saving the greater number considers C's claim in combination with B's claim so that they together tip the balance in favour of saving B and C...the argument seems to appeal to nothing more than C's claim to be saved, which appears by itself to justify saving the greater number. But this appearance, as I have shown above, is an illusion: C tips the balance in favour of saving B and C only when C's claim is combined with B's claim...C has a claim to be saved by virtue of an appeal to the difference that B and C make when considered together or in combination rather than one by one. But this is just to appeal to the claim of a group of individuals.��

Otsuka thinks that the reason of C itself cannot justify the saving of the larger group without being combined with B��s reason. The crucial question is: whether the reason of C in itself is stronger than A��s reason of being saved. A��s, B��s, and C��s reasons to be saved are the same when they are considered separately: when we compare the reasons of A and C directly, they are balanced, so do we compare the reasons of A and B. All three have the equal claim to be saved. This will not be affected by their presence with one another. Suppose C does not present in the group with B but moves to A, forming a larger group of A and C, does it mean A��s claim to be saved will automatically become stronger? I think it should not be the case. The claims of A, B and C are still equally strong. The arbitrary presence with someone else is not a reason in itself to claim to be saved. This is obvious if we are aware that we are dealing with the moral duty we owe to each of them. Otsuka is right in claiming that the reason held by C in itself is not stronger than A��s reason.

Nevertheless, Scanlon may admit that we should save the larger group not because the reason held by C in itself is stronger than the reason held by A. In fact, he does not argue in this way. He thinks we have an equal moral duty to save A, B, and C separately, holding other things constant. C��s reason to be saved should be acknowledged even without the presence of A and B. When A and B��s claim to be saved are balanced and it is not wrong to save either one of them, we still have a reason to save C. This is because C��s claim to be saved would not be affected by the tie between A and B. His claim would be ��undefeated�� in the scenario and we have a moral duty to save C; as a result of contingency, we also have to save B because they are in the same group.

Scanlon��s argument can be reformulated as below:

Given A is alone and B and C are in the same group,

(P1) A has a reason to be saved, call this reason R(a)

(P2) B also has a reason to be saved, call this reason R(b)

(P3) C also has a reason to be saved, call this reason R(c)

(P4) Everyone��s reason to be saved should be recognized

(P5) R(a) and R(b) are balanced

(P6) We still have an undefeated reason to save C

(C) Therefore, we have a reason to save the larger group because we have a reason to save C

I do not think Scanlon��s argument is sound. Even if (P5) is true, it does not follow that (P6) is also true. Separate and equal consideration of each party involved is guaranteed in contractualism. When we compare R(a) and R(b) and find that they are balanced, we should continue to compare R(a) and R(c). R(c) will also be balanced by R(a) if they are compared independently. According to ��what we owe to each other��, C cannot claim that he instead of A should be saved. The claim of C against A will not be affected when the claim of A is balanced by the claim of B. R(c) is not undefeated (in the sense that it is not counter-balanced by another reason) and therefore (P6) is false. On the other hand, C cannot complain that his claim is not being acknowledged. R(c) is acknowledged when it is compared with R(a) on an equal basis. We may have more reason to save the larger group because collectively speaking, we have two individual reasons (R(b) and R(c)) to save the larger group and only one reason (R(a)) to save the smaller group. However, the reason why we should save the larger group does not come from ��what we owe to each other���Xit is not because we owe B or C a special duty to save either one of them. It should be noted that if A and B do not present and there is only C, we do have a moral duty to save C because of R(c); nevertheless, in this particular scenario, if we consider R(c) and neglect R(a), we are not treating ��the claims of each person who could be saved as having the same moral force.��

A fair principle (according to ��what we owe to each other��) to resolve this moral dilemma should be the one that gives A, B, and C equal chances in being saved, perhaps by flipping a coin or by weighted lottery. In this sense, we owe A, B, and C an equal chance of being saved, no matter which group they come from. As human beings who are in relevantly similar situation, their claim to be saved should be respected equally. They can claim for justification for our actions to them qua being a person with moral standing; therefore their claim should not be discriminated just because they are in a minority position. A weighted lottery giving a one-third chance to each individual can be carried out to decide who is going to be saved. As a result, there is a two-third chance of being saved for the larger group and a one-third chance for the smaller group. Any additional member to either group would change the chances of being saved for other members. This proposal has the advantage of giving everyone��s life a positive weight since the presence of each of them will have an influence on the final result. It also has the advantage of recognizing the need of a procedure to make decisions rather than doing nothing.

I have to emphasize that I am not going to rule out the moral significance of number in all circumstances. I just want to stress that in the above ��dead or alive�� case, if we reflect deeply on the considerations of each individual, we may have to admit that there is no additional reason to save a member from the larger group rather than a member from the smaller group because ��being in the larger group�� is not itself a reason that counts in favor of saving a member in the larger group. However, my position is not to deny the possibilities that ��being a member of a larger group�� may be morally relevant in other situations. In certain circumstances, ��being a member of a larger group�� may be a morally relevant consideration. My disagreement with Scanlon is only a practical disagreement over whether it is reasonable to reject a principle that permits us to save a member from the smaller group�XScanlon thinks yes but I do not think so; I am not disagreeing with the method of deciding ��what we owe to each other�� by appealing to non-reasonably rejected principles. I do not mean to provide a counter-example to ��what we owe to each other��. The whole system of reasonable rejection is still consistent and appealing in other cases.

In conclusion, the non-aggregate feature of contractualism signifies its main divergence with consequentialism in normative questions. ��What we owe to each other�� represents a genuine domain in accounting our moral duty to each other individuals, serving as a constraint to consequentialist optimific options. Although Scanlon��s ambition in accounting the moral significance of number in the case of Equal Harm and the case of Small Inequality in Harm fails, it does not mean that his account is internally inconsistent. After all, as I have mentioned above, Scanlon can welcome that it is not our moral duty to save the larger group in these cases, even though we should save them in an all-things-considered sense. In fact, Scanlon does agree that in some cases, what we have most reason to do is not to act according to ��what we owe to each other��. I think Scanlon should not be too ambitious in certain situations; otherwise, the moral requirement of ��what we owe to each other�� may become too demanding generally speaking.

?lt;/P>

previous - next

?lt;/P>