2004-01-19 - 1:25 a.m.

This is chapter 2. Following my general characterization of contract theories in chapter 1, I will specialize in Thomas Scanlon's version of contractualism. Part I is a discussion about the basic ideas in the contractualist formula of 'what we owe to each other', including 'reasonable rejection' and 'principles'. The examination report swiftly points out that my discussion is focussing too much on individual principle rather than Scanlon's own idea of 'a set of principles'. However, I would not try to further amend the final draft since it will certainly affect the flow of my main arguments (and because I am lazy). Part II deals with a familiar criticism towards contractualism-- the charge of circularity. I will argue how Scanlon's contractualism could avoid these charges.

Personally, I think this is the worst chapter in my thesis. One of the reasons is the task of explaining the basic idea of a theory to set up a stage for later discussions is itself very boring. Nevertheless, this chapter is important and do worth reading if readers are interested in the normative content of contractualism. Comments welcomed!

Chapter 2

Principles and Reasonable Rejection

In this chapter, I will examine the important notions in Scanlon��s contractualist formula. It is unfortunate that many invalid criticisms come from misunderstandings of these important but complex notions. It is also necessary for me to clarify these notions before I can go further in examining contractualism as a non-consequentialist normative theory and its relationship with consequentialism in later chapters. My examination will be focused on ��principles�� and ��reasonable rejection��. It is hoped that in the end of this chapter we can have a more precise picture of ��not-reasonably-rejected-principles���X the central idea of Scanlon��s contractualism. At the end of this chapter, I will also evaluate a common criticism of contractualism: the charge of circularity. I will argue that although contractualism employs substantive judgments in deciding reasonable rejectability, it does not commit the charge of circularity as claimed by many critics.

Scanlon holds that an act is wrong ��if its performance under the circumstances would be disallowed by any set of principles for the general regulation of behavior that no one could reasonably reject as a basis for informed, unforced general agreement.�� (153) In other passages, Scanlon seems to agree that ��if it would be disallowed by any principle that no one can reasonably reject�� is equivalent to ��if it would be permitted by any principle that someone can reasonably reject.�� (4) If an act is wrong when it is disallowed or permitted by some principles, then it is natural to ask two questions: (i) what is meant by ��disallowed or permitted by a principle��? (ii) By what principle is the act permitted or disallowed? The first question concerns how an act is to be evaluated by ��principles��. More importantly, it examines the role of principles in evaluating our actions. The second question, on the other hand, concerns how and by whom these principles are decided. They are the principles that no one can reasonably reject as a basis for informed, unforced general agreement. The two questions correspond to my two points of focus: principles and reasonable rejection respectively. I will proceed to the details of these two questions separately.

2.1 Principles

In this section, I will first have a brief sketch of Scanlon��s idea of ��moral principles�� and elaborate how principles are to be interacted with one another. Afterwards, I will examine Scanlon��s two main theses concerning ��moral principles��: (i) every moral judgment is backed up by a moral principle, and (ii) every moral judgment that claims that an act, X is wrong is to claim that any principle permitting X can reasonably be rejected (or any principle requiring not to do X cannot be reasonably rejected). I will then consider some criticisms concerning Scanlon��s account of moral principles to conclude this section.

Scanlon holds that an act is wrong iff it is disallowed by any not-reasonably-rejected principle, or is permitted by any reasonably-rejected principle. At first glance, Scanlon��s contractualist account can be described as a kind of rule contractualism. Moral wrongness is determined by moral rules founded on a contractualist basis. However, such characterization is too simple to grasp Scanlon��s idea entirely. Principles, according to Scanlon, cannot be understood merely in terms of its verbal formulation. Principles are not just stable rules for us to apply to settle a wide range of moral problems, forbidding or permitting certain actions; moreover, they represent ��general conclusions about the status of various kinds of reasons for action��. (199) They state the conclusive reasons for certain actions in normal circumstances, and may rule out certain actions by ruling out the reasons on which they would be based. Take the principle of fidelity as an example. In normal circumstances, we should take the fact that we have promised to do something as a conclusive reason for doing it. Here, the principle of fidelity serves not only as a rule in telling us what to do and what not to do; it also states that the fact that we have promised to do something is normally a sufficient reason in prescribing us to do the promised thing. Other reasons, such as breaking a promise would favor the one who promises, or doing so will have a better consequence for everyone, are normally not sufficient to outweigh the reason that he has promised to do so. Therefore, the clearest part of the principle of fidelity is a claim about the importance of certain reasons for action, namely, the fact that I have promised is important (normally sufficient) for carrying out the promise. It is not a simple rule requiring us to fulfill the promise in whatever circumstances. There is no simple verbal formulation for such a rule without a more complex understanding about the status of reasons behind.

We may have a better understanding of moral principles in Scanlon��s account by considering how principles can survive exceptions and how they interact with other principles in case of conflict. Consider the principle of fidelity again. This principle states the general conclusion that having promised is normally a sufficient reason for us to fulfill the promise. ��Normally�� here leaves room for judgment and exceptions. Scanlon thinks that if the principle of fidelity requires us to keep promises no matter what, it itself is a principle that we can reasonably reject. For example, if keeping a promise would lead to a great harm which is not foreseeable before the promise was made, then not keeping the promise may survive as an exception to the principle of fidelity. The principle of fidelity does not rule out all actions of promise breaking. In making particular judgment of right and wrong, we are drawing on this complex understanding of principles.

This understanding of principles allows principles to be immune to counter-examples. If we conceive of the principle ��you should not kill�� as a simple rule forbidding all acts that will lead to a person��s death, then this rule is opened to counter-examples. In cases of self-defense, suicide or euthanasia, killing is not necessarily morally forbidden if not morally permitted. These cases could be explained when principles are conceived as claims of reasons. Killing for my personal advantage is not a sufficient reason for aiming at it, but in case of defending myself against attack from an insane person is sufficient for doing so. In deciding what counts as a legitimate exception to a certain principle in particular cases, Scanlon thinks that we have to appeal to ��a shared sense of what the point�� of the principle is. ��Anyone who understands the point of promising�Xwhat it is suppose to ensure and what it is to protect us against�Xwill see that certain reasons for going back on a promise could not be allowed without rendering promises pointless, while other exceptions must be allowed if the practice is not to be unbearably costly.�� (200)

This complex structure of a principle can be extended to cases of conflict of principle. Scanlon thinks that his account is to be contrasted with a different model, under which, ��moral thinking employs more fully specifiable principles which always apply but which must be balanced against one another in cases of conflict, the outcome depending on their relative moral weight.�� For Scanlon, moral principles cannot conflict. Apparent conflicts between principles are caused by inadequate understanding and by incomplete specification of principles. The matter can be resolved by a more complete specification of either principle. For example, if we face a situation in which the only way to save one person is to kill another, innocent person. In this case, the principle requiring help and the principle forbidding killing are in fact not conflicting if we have a more complete understanding of both principles. A proper understanding of the principle of help will include an exception to that duty in cases where helping would involve killing another. For the principle forbidding killing, we can conclude that to help another person to survive is not a sufficient reason to break the principle. This does not mean the principle forbidding killing is always more ��weighty�� than the principle requiring help. It is only in this case that the consideration of helping another is not sufficient to grant an exception to the principle forbidding killing. In every particular case, we make a judgment about the sufficiency of the reasons for an action. This judgment, as claimed by Scanlon, is guided by, and expresses, our understanding of a moral principle.

After sketching what Scanlon means by ��moral principles��, we can go deeper to examine his claims about the relationship between moral principles, reasons and particular moral judgments. Scanlon thinks that every moral judgment is backed up by moral principle. (197, 199) An act is wrong, according to contractualism, if we cannot justify it to others. It may be asked, why the justification of our actions should proceed through principles but not individual acts instead? Why one has to justify a principle licensing general performance of his act at the same time when he is justifying his own individual act? For Scanlon, such questions can be answered satisfactorily only if principles are understood in the wider context in terms of reasons. To justify an action is to offer reasons supporting it and to claim that they are sufficient to defeat any objections that others may have. Scanlon thinks that to do so is also to defend principles supporting such an action since these principles represents the claim that reasons of permission are sufficient to defeat reasons of prohibition for such an action in the prevailing circumstances. (197) To defend an action in terms of reasons is to defend the principles backing up the act.

On the other hand, Scanlon thinks that reasons are universalizable, which holds that if someone has a reason to do something X given a certain circumstance C, then anyone else whose situation is similar to C in relevant respects will also have the same reason to do X. (73-74) Therefore, we are committed to defending the general practice of an individual action even when we are justifying this individual action alone in terms of reasons. It is because when we are defending the reasons that license a certain act, we are at the same time defending the reasons licensing actions of similar sorts. When conceived as general claims about reasons, principles help to explain why justification of an act has to proceed via principles but not through individual acts at all.

Following from the claim that every moral judgment is backed up by a moral principle, Scanlon continues to claim that every moral judgment that claims that an act, X is wrong is to claim that any principle permitting X can reasonably be rejected (or any principle requiring not to do X cannot be reasonably rejected). In a particular case, to determine whether X is wrong in the contractualist account of ��what we owe to each other��, we have to check whether X is permitted or disallowed by a set of principles that is non-reasonably rejected. To justify X, we have to offer a principle that backs up X; however, such a principle is not necessarily the one that no one could reasonably reject. For instance, to justify telling a white lie to a dying patient, we can offer a principle that backs up this act, namely the principle that permits telling a lie for the sake of the happiness of a person who will never know the truth. This principle is a consideration to justify the act, but it is not necessary a non-reasonably-rejected principle in this case. For example, the patient may reasonably reject such a principle because he wants to know the truth more than to be happy without knowing the truth. He values the dignity of not being cheated more than that of being happy. In this circumstance, our principle is ��defeated��: the claim that ��the reasons permitting us to tell a lie is sufficient to outweigh the reasons of prohibition�� is defeated. The non-reasonably-rejected principle is the one raised by the patient, namely the principle that his dignity should be respected and he should not be cheated. Telling a white lie here is wrong because it is disallowed by a principle that no one can reasonably reject. It is disallowed by the whole set of principles (broadly conceived) which is not reasonably rejected.

To decide whether a particular act, say X, is wrong is to check whether it is permitted or disallowed by a set of non-reasonably rejected principles in that case. Does it correspond to Scanlon��s claim that X is wrong iff any principle permitting X can reasonably be rejected (or any principle requiring not to do X cannot be reasonably rejected)? We have to resolve an ambiguity here to argue for the connection. There are two interpretations for ��X is wrong iff any principle permitting X can reasonably be rejected��. The first interpretation reads ��any principle�� simply as ��any one principle��. Under this interpretation, X is wrong if there is one principle permitting X that can be reasonably rejected. For any one of the principles under which X is allowed, if it can reasonably be rejected, then X is wrong. Similarly, X is wrong if any one of the principles requiring not to do X cannot be reasonably rejected.

For the second interpretation, ��any principle�� is read as ��every principle��. X is wrong in this circumstance, if it is the case that, for every principle, whenever it permits X, it can be reasonably rejected. It simply means that X is wrong if and only if all principles that permit X can reasonably be rejected. Similarly, ��any principle requiring not to do X cannot be reasonably rejected�� is read as ��for every principle that requires not to do X, they cannot be reasonably rejected.

It is only in the case that when the non-parenthesis condition is interpreted as ��iff every principle permitting X can reasonably be rejected�� and when the parenthesis condition is interpreted as ��iff any one principle requiring not to do X cannot be reasonably rejected�� that Scanlon��s claim is correct. I have argued that an act X is wrong iff it is disallowed by the non-reasonably-rejected principle in that case. To say that X is disallowed by the non-reasonably-rejected principle is equal to saying that all principles in which X is not disallowed cannot be the non-reasonably-rejected principle. It means that X is wrong if all the possible principles supporting it are reasonably rejected. It echoes with our previous argument that to justify an act is to defend the principle supporting such an act. On the other hand, to say that ��any one principle that disallow X cannot be reasonably rejected in the circumstance C �� is to claim that there is a sufficient reason for disallowing X in C. There is a non-reasonably-rejected principle in this case in disallowing X. Obviously; it is equivalent to saying that X is disallowed by a non-reasonably-rejected principle in C. I conclude that, my interpretation that ��an act is wrong iff it is disallowed by any non-reasonably-rejected principle, not wrong iff it is permitted by any non-reasonably-rejected principle�� is logically consistent with Scanlon��s claim that ��an act is wrong iff every principle permitting it can reasonably be rejected (or any one principle requiring not to do it cannot be reasonably rejected)��.

It is important to notice that Scanlon��s claim is ��an act is wrong iff for every principle permitting it, it is a principle that can be reasonably rejected.�� Thomas Pogge, who has presented an outstanding criticism of Scanlon��s contractualist formula in his ��What we can reasonably reject�� , commits an error in charging Scanlon for this claim. Pogge claims that only if the notion of ��permitted�� and ��disallowed�� is revised to ��authorized�� and ��fail to be authorized�� that the whole formula can make sense. He argues that if an act X is wrong iff a principle permitted X can reasonably be rejected, or iff X is disallowed by any non-reasonably-rejected principle, then there will be no wrong act at all. This is because for any act, ��there will be at least one not-reasonably-rejectable principle that does not disallow this act. For example, most acts of telling malicious lies are not disallowed by the principle that forbids murder�Xand most acts of committing murder are not disallowed by the principle that forbids malicious lies.�� For an act to be wrong, he thinks, the principle to be reasonably rejected must be one that ��actively authorizing�� the act.

Literally speaking, it seems that ��an act is wrong iff for every principle permitting it, it is a principle that can be reasonably rejected�� will unavoidably lead to counter-examples of the sort suggested by Pogge. A principle forbidding killing only forbids classes of actions that are classified as ��killing�� and it will not disallow lying and most other acts that are considered as wrong, but it seems it is hard to say that this principle is the one that we can reasonably reject. For the sake of clarity, I think the revision is not bad. However, I do not really think Scanlon��s account cannot avoid these counter-examples. Strictly speaking, the principle forbidding killing is a principle that can reasonably be rejected in a case where telling lie is the only action of evaluation. In a case where A tells a malicious lie to B, without any consideration that may involve the principle forbidding killing, A cannot justify his telling lie to B by saying that ��no one, including B can reasonably reject the principle forbidding killing, and my action of telling lies is permitted by such a principle.�� He cannot do so because, the reasonably-rejectability of a principle, as I will point out in the next section, depends on the reasonable rejection of alternative principles. B may raise a principle forbidding telling malicious lies here, and the reason for rejecting this principle is clearly weaker than the reason for rejecting the principle forbidding killing as a authoritative guide to the general regulation of our behavior in this case, even telling lie is permitted (or even authorized, if it is the case) by the principle forbidding killing. In a case where not telling a lie would kill another person, the principle forbidding killing may, to a degree, be less rejectable than the principle forbidding telling lies. However, in this case this principle only permits telling a lie in order not to kill another. It does not permit telling a lie in whatever circumstances.

And for the claim that ��an act is wrong iff it is disallowed by any non-reasonably-rejected principle�� should be revised as ��an act is wrong iff it would fail to be authorized by any non-reasonably-rejected principle��, I think Pogge does not provide us with a sufficient argument. Even if a wrong act is not disallowed by a non-reasonably-rejected principle, like telling a malicious lie is not disallowed by the principle forbidding killing, it is not a counter-example to Scanlon��s claim that ��an act is wrong iff it is disallowed by any non-reasonably-rejected principle��. Even if telling a lie is not disallowed by many non-reasonably-rejectable principles, it does not follow that there is no non-reasonably-rejectable principle that disallows it. Even if it is disallowed by only one principle that cannot be reasonably rejected, it is still wrong.

Pogge may claim that here is where the real problem lies. If we have to prove an act to be wrong, we have to go through all possible principles that permit it and claim that all of them can be rejected. To prove an act to be not wrong, we also have to go through all possible principles disallowing it and claim that none of them can be not reasonably rejected. This is not realistic at all. My suggestion is to shift the direction of proof: to prove that an act is wrong, we go through principles that disallow this act and find whether there is one that cannot be reasonably rejected. If there is any, it is sufficient to claim that the act is wrong. To prove that an act is not wrong, we go through principles that permit it and find whether there is one that cannot be reasonably rejected. Also, one principle of this kind is sufficient to claim that the act is not wrong. In discussing whether an act is wrong, I think we can appeal to this kind of bilateral reasoning in order to achieve equilibrium.

There are more detailed discussions of the role played by moral principles in contractualism in Scanlon and Jonathan Dancy��s article, ��Intention and Permissibility�� (Aristotlian Society 2000; Supp, 24). I will not explore them because they are not relevant to my main argument here. Nevertheless, they provide important and useful materials for further research in Scanlon��s idea of ��moral principles��.

2.2 Reasonable Rejection

We have discussed what is meant by ��permitted (or disallowed) by a principle��. The next question is, ��when and how can we reasonably reject a principle?�� Briefly speaking, Scanlon holds that a principle can be reasonably rejected when any individual��s reasons in rejecting this principle is stronger than any other individual��s reasons in rejecting alternative principles. There are several essential features for what counts as a valid reason in reasonable rejection, I will spell them out separately in terms of the form and content of these valid reasons.

First, a valid reason in reasonable rejection must be one in the form of a personal reason, arises from the standpoint of an individual in certain positions. The reasons for rejecting a principle must come from an individual basis but not from an agent-neutral perspective. Individuals can only reject a principle on their own behalf but not on behalf of a group. They cannot appeal to the interests of others in citing objections to a principle. The limitation of valid reasons to various individual reasons is an important feature in Scanlon��s contractualist account of ��what we owe to each other��. ��What we owe to each other�� is a specific, narrower domain within the broader concept of ��morality��, and it denotes what duties we owe to each other individual. It is particularly important that in determining principles that govern our inter-personal relationship, each individual��s claims should be considered separately and equally.

The emphasis on individual reasons is one of the key features that distinct contractualism from consequentialism. First, not only consequentialist elements can constitute reasonable rejection. A principle may be rejected if it prevents someone from pursuing his legitimate, vital interests even if such prevention would lead to a better consequence for many people overall. Also, the restriction to individual reasons in reasonable rejection disallows any inter-personal aggregation of reasons. An individual��s reason in rejecting a principle would not automatically become stronger just because there are more people sharing similar concern. I will come back to these points in later chapters.

Apart from individual reasons, a valid reason in reasonable rejection is also restricted to the form of a generic reason. Generic reasons are reasons that ��we can see that people have in virtue of their situation, characterized in general terms, and such things as their aims and capabilities and the conditions in which they are placed.�� (204) When comparing various individual reasons in assessing the rejectability of principles, we do not appeal to the particular aims, preferences and characteristics of specific individuals; rather, we have to appeal to what these individuals in their positions have reason to want in general. Scanlon opts for generic reasons for two reasons. First, in assessing principles for the general regulation of behavior, a principle licensing a certain act will also license the general practice of that act, and the effects of its adoption is much larger than the effects of the individual practice of such act. We do not know in advance which particular individuals will be affected in various ways when a principle is adopted. Therefore, in assessing the rejectability of a principle, we have to appeal to reasons that we think those individuals who actually end up in the positions of those being affected, whoever the particular individuals turn out to be, will appeal to.

Another aim for the adoption of generic reasons is to rule out appeals to morally irrelevant and even undesirable preferences of particular individuals in assessing rejection of principles. Scanlon does not think that individuals�� preferences are not important in deciding reasonable principles. It is only those contingent preferences that cannot be generalized in the sense that everyone in the position of the individual will have that are excluded as generic reasons. For example, in assessing the principle requiring the rich to aid the poor financially, we have to look at the reasons of any individual who actually occupy the positions of the rich and the poor, and to appeal to what they may have reasons to want in these positions. We cannot appeal to the preferences or aims of particular individuals in assessing the principle because different people will have different preferences towards the same principle. Rather, we have to appeal to the reasons that an individual may have in the positions of the rich and the poor, or in the positions of offering aid and receiving aid.

The appeal to generic reasons may be criticized as not liberal enough. So far, what can be counted as a generic reason depends on our understanding of what reasons individuals may have in different positions. However, in our actual deliberation of whether a principle is reasonably rejected, our understanding of the reasons of people occupying different positions is always not enough. We always fail to acknowledge all the reasons that people may have reason to have, and sometimes underestimate the force of some of their generic reasons. Scanlon admits that one important role of moral theory is to correct this bias. Yet he does not think it would pose a theoretical difficulty to contractualism. The way to resolve it is to expand our experience of others�� point of view, ��including the experience of hearing and discussing the claims that people actually �Kadvance.�� Theoretically speaking, we still have to appeal to generic reasons in assessing principles.

The form of legitimate reasons in assessing principle must be expressed in terms of individual and generic reasons. But what considerations are counted as legitimate grounds that constitute reasonable rejection? This moves us to the discussion of the content of these individual generic reasons. It is well-known that contractualism does not reduce all grounds for reasonable rejection to a single idea of well-being. Although individuals�� well-being is a salient consideration in some cases, there are other grounds that an individual can cite as reasons in rejecting principles. From the discussions of various principles in Scanlon��s book and his later papers, I can identify certain considerations that Scanlon considers as relevant grounds for rejecting a principle. An individual can reasonably reject a principle not only by appealing to the effects on his well-being, but also to how the burdens are imposed on and how the benefits are distributed. An individual can reject a principle if he will be treated unfairly by the principle, under which others are favored arbitrarily even if such principle will have no effects on his actual well-being. An individual may also reject a principle if he is treated only as a means but not an end in himself. In addition, notions about rights or entitlement can be and will be taken into consideration in deciding whether a principle can reasonably be rejected. There may be other reasonable grounds as well, which I cannot spell out all here. The diversity of grounds for reasonable rejection raises difficulties in comparing the relative strength of reasons arising from different grounds, it also invites the charge of circularity. I will come back to these difficulties in section 2.3.

Different individuals may raise different generic reasons based on various grounds in rejecting principles in a certain situation. It is important to note that reasonable rejection depends on the comparative strength of individuals�� reasons in rejecting different principles. Whether a principle is rejected depends not only on the reason of any individual in rejecting this principle, but also depends on whether this reason is stronger than any other reasons raised by other individuals in rejecting alternative principles. A non-reasonably-rejected principle is therefore the ��least rejectable principle�� in terms of individual reasons in every single case. This leads to the important idea that there does not exist a threshold level of cost in reasonable rejection ��such that it is reasonable to reject any principle that would lead to one��s suffering a cost that great.�� (196) Even if a principle forbids an individual to do something that is necessary to save his life, it is not always the strongest reasons in particular situations. It depends on whether other individuals have a stronger reason in rejecting this principle than his rejection to some alternative principles.

It is also important to note that such comparison is not one that compares the relative strength of reasons for or against a certain principle. It is a comparison among the reasons of rejection between different principles. Suppose in a situation there are only two individuals A and B, who propose two principles X and Y respectively in settling a particular moral conflict. The reason of A in supporting X has a degree of strength of 100S and the reason of B in supporting Y has a degree of 50S. The reason of A in rejecting Y is 10S and the reason of B in rejecting X is 50S. In this case, the ��net reasons�� for or against principle X is 50S (reasons for X- reasons against X: 100S-50S) and the ��net reasons�� for or against principle Y is 40S (reasons for Y- reasons against Y: 50S-10S). It seems that the reason for adopting X is stronger than the reason for adopting Y if we compare them in terms of the net difference between the strength of support and the strength of rejection for the principle alone. Yet according to Scanlon, what we should opt for in this case should be principle Y. It is so because the reason in rejecting Y (10S by A) is less strong than the reason in rejecting X (50S by B). Even A may have a strong reason in supporting X, he may be less reluctant to accept Y as the guiding principle in that case than B in accepting X. For example, A may be benefited greatly under principle X while having a small cost under Y. However, under X, B may have a large sacrifice that constitutes a very strong reason for him to reject X while under Y, he may have a relatively smaller cost. Scanlon��s emphasis on ��principles that can be justified to every one�� naturally leads us to accept Y that is relatively less rejectable from the standpoint of an individual. We cannot justify to B if we force him to accept X when clearly there is an alternative principle that can avoid his sacrifice and others do not have a stronger objection to this principle than his objection to X. This raises an interesting question about what Scanlon meant by ��alternative principles��. I will explore on them later.

Last but not least, reasonable rejection is also characterized by the aim of finding principles for the general regulation of behavior that no one similarly motivated can reasonably reject. This aim of finding mutually acceptable principles in regulating behavior provides a meta-reason for us to take into consideration of the reasons of other individuals in determining the least rejectable principles. ��A claim about what it is reasonable for a person to do presupposes a certain range of reasons which are taken to be relevant, and goes on to make a claim about what these reasons, properly understood, in fact support.�� This aim is presupposed in the contractualist framework of moral reasoning to ��bring other reasons in its train��. (192) It is why it is reasonable for all of us not to act according to a principle that can be rejected with stronger reasons by other individuals, even if we ourselves have a strong reason in supporting such a principle.

I will now turn to some interesting discussions about reasonable rejection. First, as I have argued above, the rejectability of a principle does not depend on a comparison between the reasons supporting it and the reasons objecting it. It depends on the comparative strength of rejection of this principle against alternative principles. A non-reasonably-rejectable principle in a particular situation is therefore the ��relatively least rejectable principle�� among the alternatives, but actually what is meant by alternative principles?

Scanlon holds that the alternatives to a principle permitting a certain act are not limited to the principle prohibiting that act. Therefore, even if the reasons in supporting a principle and the reasons in objecting to it are at a tie, when there exists another principle whose reasons of rejection is weaker than this principle, it can be reasonably rejected. For example, in his imaginary shipwreck case (195-196), two swimmers arrive at a life jacket, which is necessary for saving either one but not both of their lives. One of them is much stronger than the other. Can the principle permitting the use of force to seize the jacket reasonably be rejected? It seems that the reasons in rejecting this principle by the weaker person is only equally strong as the reasons in rejecting a principle prohibiting the use of force by the stronger person since their lives are both at stake and their considerations are balanced, holding other factors constant. If alternative principles for the principle permitting the use of force are limited to the principle prohibiting doing so, then we are forced to admit that the reasons in rejecting both principles are tied. However, if there exist some principles that can settle the matter better, for example, if it is possible to require them to draw lots or take turns, then this alternative principle will be the least-rejectable principle in that situation. If it is the case, then the principle permitting the use of force can reasonably be rejected as it is not the least rejectable principle in this case. The stronger person is not permitted to seize the jacket without drawing lots with the weaker person or try his very best to take turn with him. In this case, the least rejectable principle will be the one that can settle the conflicting claims of the two swimmers in the ��best�� way. A principle permitting the use of force may have the merit of ��recognizing the symmetry of their claims and the need for some decisive solution��, (196) therefore if it is the only available principle, it is less rejectable than a principle that prohibits the use of force and suggests nothing to do, resulting in the death of both persons. However, if there exist some principles that can resolve the conflict in a more reasonable way, the principle permitting them to struggle with force can reasonably be rejected.

Scanlon seems to think that there are always some ways in coming up with further less rejectable principles to break the tie. But is it possible to have principles that both are not reasonably rejectable? Let me explore a possible case here.

In the case of shipwreck, apart from the principle permitting the use of force, a large number of principles can be proposed to settle the conflicting claims. Different principles will propose different settlements by appealing to various factors, such as the previous moral merits of the two people, the number of family members and their relationship with them, their future achievements, their chance of survival without the jacket, and so on. Is it always the case that we can only have one non-reasonably-rejected principle? Suppose the weaker person say, ��man, you are much stronger than me, and your chance of finding another jacket nearby is much greater than me, therefore you should let the jacket to me and find another.�� Let me name this as the principle of survival chance. In fact, I think this principle is actually less rejectable than the principle permitting the use of force, however, for the sake of argument, let me assume that they are just equally rejectable, and are both less rejectable among the alternatives.

Suppose finally the stronger person win the jacket by force. Could the weaker person complain that it is wrong? Could the weaker person reasonably reject the principle permitting the use of force because the stronger person��s rejection of the principle of survival chance is not stronger than his rejection of the principle permitting the use of force? Here, Pogge discovers an ambiguity about the term ��reject��. On one hand, one might say that both of them can reasonably reject the principle of each other. Both may claim that, ��I cannot reasonably reject your principle only if you could reject my alternative principle on stronger grounds. But your rejection is not stronger than mine in rejecting your principle; therefore my rejection of your principle is alright.�� In this case, both can reject the principle proposed by the other by proposing an alternative principle against which no stronger grounds can be adduced. The result is both of them act wrongly according to their own principle, because the two principles are both reasonably rejectable from another��s viewpoint.

On the other hand, one might claim that they both do not act wrongly according to their own principle. Both may claim, ��you can reasonably reject my principle only if you could present an alternative principle that my rejection would be weaker than your rejection to my principle. Since the grounds you reject my principle is not stronger than my rejection to your principle, you cannot reasonably reject my principle.�� As a result, both of them do not act wrongly because both principles cannot be reasonably rejected.

In the former case, the two principles are both not the least reasonably rejectable principle, with ��least rejectable�� having a more stringent requirement: a principle is least rejectable when there is no alternative principle whose reason of rejection is as weak as its reason of rejection. In the later case, the two principles are both least rejectable principles, with ��least rejectable�� having a less stringent requirement: a principle is least rejectable when its reason of rejection is not stronger than the reason of rejection of its least rejectable alternative. Scanlon, in his discussion about relativism, claims that ��there is a need for some principles to govern a particular kind of activity, but there are a number of principles that would do this in a way that no one could reasonably reject.�� (339) This favors reading ��reject�� in the latter sense. He also confirmed this interpretation in correspondence with Pogge.

This brings unnecessary complexities in Scanlon��s account. In cases where there are two or more least rejectable principles, we can choose either one of them to follow. If one of these principles prescribes us to do x, then x is not wrong; however, if x is disallowed by another least rejectable principle P, we have to admit that x is not wrong but at the same time P is non-reasonably-rejectable principle. This sounds odd and may involve further difficulties in more complicated cases. Why not combine the two or more least rejectable principles into one composite principle? In the above case, if both the principle permitting the use of force and the principle of survival chance are least rejectable, then we may form a composite principle: a principle permitting the use of force or appealing to the survival chance of both parties. In this sense, we may maintain the thesis that a certain action is wrong if it is disallowed by a non-reasonably-rejected principle, without engaging in further difficulties. This further principle may not be helpful in resolving the conflicts of different parties, but it can make Scanlon��s account more concrete.

Another interesting discussion concerning reasonable rejection centers on the phrase ��principles for the general regulation of behavior��. In deciding the reasonable rejectability of principles, shall we take into consideration that they are for the ��general regulation of behavior��? If a principle is least-rejectable from individuals�� standpoints, but it is not the one that everyone will follow in actual circumstances, does it constitute a reasonable rejection to this principle as a ��principle for the general regulation of behavior��?

There are at least two possibilities where compliance to a principle may be relevant in deciding its reasonably rejectability. Firstly, principles in particular situations may be very fine-grained, and they may be too fine-grained so that people cannot understand and apply them easily, hindering their practicality as principles for the general regulation of behavior. Scanlon has once claimed that, ��finer-grained principles will create more uncertainty and require those in other positions to gather more information in order to know what a principle gives to and requires of them. For example, the principle of fidelity to promises protects us against being bound in ways that we do not like by specifying that only voluntary undertakings are binding. But individuals differ in their ability to foresee possible difficulties and to resist subtle pressures to enter an agreement. The protection offered by the requirement of voluntariness is therefore of different values to different people.�� (205) From this, can we say finer-grained principles can reasonably be rejected because they are too fine-grained?

Suppose in a particular conflicting situation, a person is told that his proposed principle should be rejected because it is too fine-grained and cannot reliably guide people in general. He can defend it by claiming that, ��yes, this principle may be too fine-grained to be applied for general regulation of behavior, but it is clear that it is applicable in this situation, so no one can reject it because it is too fine-grained.�� Scanlon will agree that this fine-grained principle would not be rejected in this case just because it is too difficult to be applied in general cases. But how can he settle the claim that it is not a good principle for the general regulation of behavior?

We have to resort to Scanlon��s more complex understandings of principles. Principles have no exact formulations in form, and they signify the status of various reasons for actions. For a very fine-grained principle, it may be difficult to apply and understand and therefore not suitable as a principle for general regulation of behavior. Yet we should note that this particular fine-grained principle is only one of the forms of a simpler, more general principle. If we understand the complex structure behind, say the principle of fidelity, we can also understand a finer-grained version of it, such as a principle requiring us to keep our promise in a situation that keeping a promise will lead to very undesirable consequence, but this consequences is clearly foreseeable before the promise is made. This makes a fine�Vgrained principle a part of a more general principle, which is suitable for general regulation of behavior. We should not take the fact that a principle is easily and usually misapplied as a reason in rejecting a principle�Xit is a result of people��s lack of understanding of general principles. Rather, we have to improve people��s understanding about the complex structure of general principles so that finer-grained principles will not be misapplied easily and become acceptable as ��principles for the general regulation of behavior��.

The second possibility that compliance may be relevant in reasonable rejection is the case that actual compliance of a principle may yield very different effects than the general compliance of it. Elizabeth Ashford raises a case concerning the principle of aid in this case to argue for the demandingness of contractualism as a normative theory. She argues that if we take into considerations of the actual compliance of a principle concerning the duty of the rich in aiding the poor, the least rejectable principle will be much demanding than the least rejectable principle assuming general or universal compliance. Given full compliance, if each rich people contribute a fair share of his income to aid the poor, than it is sufficient to help the poor to maintain a minimum level of subsistence. However, this principle is not generally observed and therefore it is reasonable to reject this fair-share principle and demand a more demanding principle from the viewpoint of each poor person.

She explains it with a hypothetical case of rescue. Suppose there are ten children drowning in a pond, and five agents are available for help. The water is very cold and therefore the burden for rescuing the children for each agent increases as the number he saves increases. An ideal principle of rescue assuming full compliance will require each of the five agents to save two children. However, four of the agents are ��slackers�� and left the scene after they have saved one. Given actual compliance of principles, Ashford thinks that the principle requiring each of the agents to save two can reasonably be rejected from the standpoint of the children. A reasonable principle may be the one that requires the agents to save at least three or four children.

The first problem of this argument is that it assumes a theoretically more demanding principle would lead to more compliance in actuality. A principle that demands saving three by each can actually motivate each of them to save at least two, and a principle that requires the rich people to contribute more than a fair share will actually lead them to contribute at least a fair share. This may not be the actual case. Pogge has counter-argued that ��it does not follow that partial compliance stipulation favors an more demanding principle of mutual aid�K(because) a less demanding principle engendering greater compliance may be better (produce a higher yield) than a more demanding principle that would be widely disregarded.��

The second problem is, it seems that both the principle requiring the saving of two children in the case of rescue and the principle requiring a fair share to be contributed from the rich are still the least-reasonably-rejectable principles in the respective cases. In the case of rescue, if there is only one agent who is available to help, he may not be able to reject any principle requiring him to save at least most of the children. However, even all other potential agents refuse to help, it simply does not mean that the person who stays (call him the conscientious agent) is the only person in the position of help. Other potential agents who have left are still in the position of help even if they actually refuse to help. Nothing happened has prevented them from being able to help. So why can��t this conscientious agent raise an alternative principle, requiring (or even compelling) those other potential agents to help?

Those children would not reject this principle because they will not care how the burden of help is distributed among those agents of help unless they would be saved. The only objection can be raised is from those potential agents who refuse to help. But this rejection is not strong enough when we compare it with the rejection of the conscientious agent for any alternative principle, since he has already contributed a fair share while they did not. On the other hand, these 'slackers' cannot reasonably reject the reasonable claim of the children since their contribution is insignificant to themselves but significant to those in need. Therefore, the principle requiring these slackers to help is less rejectable than the principle requiring the conscientious agent to take the full share, even in the actual world where full compliance is not guaranteed. These two principles would be indifferent to the children, but the conscientious agent can reasonably reject any principle permitting the slackers in avoiding help because it shifts the necessary burden that should be shared by all to the conscientious agent only.

The crucial point lies in the fact that those slackers are really in the position of help. But if in fact these irresponsible slackers really left, and the conscientious agent turns out to be the only agent available to help, does it mean that the conscientious agent is responsible only for his fair share�X to save two children only? Not necessarily. If he turns out to be the only person available to help after those slackers left, Scanlon would also think that he should save as many as he can. In this case, the reasonableness in rejecting or supporting a principle will also depend on the magnitude of the burden it would impose on a benefactor. The conscientious agent should save as many of the children as he can until the burden imposed on him exceeds a reasonable limit.

I think the general mistake in the arguments that principles may be rejected because they are too fine-grained and principles would be much demanding when we take into considerations of actual compliance is they mistaken how a principle is justified. A principle is justified, under contractualism, because it is the one that no one can reasonably reject but not because it can generate greatest acceptance-utility. A more general principle, or a more demanding principle taking into considerations that people will not universally comply to a less demanding principle, may yield a higher acceptance-utility as a principle for general regulation of behavior. Yet they are not the contractually least reasonably rejectable principles in each particular case. For Scanlon, how to make more people accept the least rejected principles in each particular case is another problem, but it simply does not mean that principles that may yield a higher acceptance-utility constitutes a reason in rejecting principles for the general regulation of behavior.

2.3 Circularity

Scanlon holds that an act is wrong if and only if it is permitted by any principle that can reasonably be rejected. Many philosophers criticize that it is a circular account of moral wrongness. I will examine these criticisms in this section.

There are two possible charges that Scanlon��s contractualism is circular. The first claims that if we appeal to any prior notion of rightness or any notion with moral content to tell us what can be reasonably rejected, then the contractualist framework is unnecessary, since the framework just tries to explain what has already been assumed. Let me describe this kind of circularity as trivially circular. The second sort of circularity claims that the contractualist framework involves substantive elements and judgments that are not beyond question, even though wrongness is not directly determined by these elements and judgments. Contractualism is substantively circular in this sense. I will argue that Scanlon��s account is not trivially circular, even if it involves many substantive elements in reasonable rejection.

It may be useful to compare two models of reasonable rejection before I start examining whether the two charges of circularity are valid. In deciding the rejectability of principles, we have to appeal to the generic reasons of various individuals when the principles are generally accepted for regulation of behavior. Here, two important questions may arise: (i) what are the possible grounds for reasonable rejection? And (ii) how to determine the relative strength of these grounds in rejecting principles? The two models provide different answers to these two questions.

The Complaint Model claims that the only possible ground for reasonable rejection is our well-being and the strength of rejecting a principle depends on the burden on an individual��s well-being imposed by that principle and on the individual��s absolute level of well-being. Scanlon originally adopts this model in ��Contractualism and Utilitarianism�� Under this model, an individual can reasonably reject a principle if his burden (the amount he would be worse-off in well-being) under which the principle is adopted combined with his absolute level of well-being constitutes a complaint greater than the complaint had by any other individual about some alternative principles. The strength of the reasons in rejecting a principle is therefore represented by a function of the effects on, and the absolute level of, the individual��s well-being.

Scanlon himself has departed from this model and suggests the new Generic Reasons Model in What we owe to each other. As I have suggested in the above section, the possible grounds for rejecting principles from an individual��s standpoint do not only arise from considerations about well-being but also depend on what people have reasons to want and concern. Various factors like our concern about fairness, concern about entitlements and rights, and concern about responsibility will be taken into account in addition to the level of well-being of individuals involved. In calculating the relative strength of different factors, there is no specified method but we have to appeal to our intuitive judgments in each particular case. (217-218)

The appeal to various ��generic reasons�� invites the charge that contractualism is trivially circular. An act is said to be wrong because there are some generic reasons against it; but what are the grounds for these generic reasons? It is claimed by the critics that they must be moral grounds. Not all reasons can be counted as generic reasons, and if the generic reasons in rejecting principles are non-moral reasons, they cannot really give an account of moral wrongness. If an act is wrong because it is unjustifiable to others, it must be unjustifiable in terms of moral reasons. And if generic reasons turn out to be moral reasons, then what is wrong in ��what we owe to each other�� is pre-determined by these prior moral reasons. For example, Colin McGinn claims,

��The first and most pressing problem with Scanlon��s theory is that it is circular. It presupposes what it is meant to explain, in the most obvious way. An act is said to be wrong if it may be criticized by others. But on what grounds is it criticizable? On moral grounds, surely. But then aren��t we back where we begin?��

If the reasons in rejecting principles are moral reasons, then what is wrong will depend on a pre-determined delineation of what are moral reasons and what are not. And if what is wrong is determined by these moral reasons, then the hypothetical agreement in the contractualist framework would be redundant. Here is Simon Blackburn��s criticism,

���Kthe role of contract diminishes further if we turn our attention to those reasons for rejection. Why not suppose that they themselves provide the very reasons an action is wrong, short-circuiting any residual appeal to contracts with others? Suppose it is reasonable to reject my principles because, for instance, they lead to vast inequalities of wealth. Why then isn��t that very feature that makes my principles wrong? Why go through the detour of dragging in the hypothetical agreement with others?��

McGinn and Blackburn may be right if the reasons for reasonable rejection do not come from an agent-relative perspective. A reason is agent-relative if ��it makes an ineliminable and pronominal back-reference to the person to whom the reasons applies��. If Scanlon holds that an act is reasonably rejected because there are some agent-neutral reasons against it, for example, because it will lead to vast inequalities of wealth, then what is wrong is determined by this reason but not by the contractualist framework. However, as pointed out by Michael Ridge, the reasons for rejecting principles are individual, agent-relative reasons but not agent-neutral reasons; therefore Scanlon��s account is immune of this criticism. If the reasons in rejecting principles are agent-neutral, then there will be no difference between the reasons for different individuals. It is unnecessary for us to calculate the strength of reasons from different individuals�� standpoints and what we have to do is to weigh the strength of different agent-neutral reasons directly. In this sense, what is wrong is presupposed by these moral reasons. Yet Scanlon has explicitly claims that the reasons in rejecting principles must be individual reasons and such move rules out any appeal to impersonal reasons.

Let me illustrate this with an example. In a distribution of welfare among three people A, B and C, who are all in equal need with equal claims to the resource. A fair principle will guarantee a 3-3-3 distribution among them. Under a principle with distribution 5-4-0 among them respectively, only C has a reason to claim that such principle is unfair. B cannot reject this principle by claiming that it is unfair. An individual cannot reasonably reject a principle by claiming that such principle is unfair, unless he can claim that such principle is unfair to him agent-relatively.

It is also important to note that individuals cannot rely on a single agent-relative reason solely to reject principles. There is no absolute threshold for reasonable rejection and the reason that it is unfair for C in the above example is not sufficient to reject the principle without considering possible generic reasons in supporting the principle by A or B. For example, if the assumption is revised and there is a desperate need for A and B in saving their lives by the amount they receive, then A or B��s generic reasons may have to compare with C��s claim that it is unfair (in the sense that they all have equal claims or rights to the wealth) in order to determine whose reason is stronger.

The appeal to individual reasons and the comparative nature of reasonable rejection makes Scanlon��s account a far more complicated mechanism than just a disguised container of moral reasons. On one hand, agent-relative reasons arise from what agents have reasons to want; and what agents have reasons to want are not limited to what they have moral reasons to want. For example, Scanlon has once claimed that a principle that will make it impossible for an agent to give special attention to her own projects can reasonably be rejected. (204) This corresponds to what Thomas Nagel calls a non-moral ��reason of autonomy��. On the other hand, a reason itself, even if it is a moral one, is not sufficient to determine whether an act is wrong without comparing with the generic reasons of other individuals. Even if someone may raise a ��moral�� reason against a principle from his own individual standpoint, it cannot solely determine the wrongness of the act licensed by that principle. Obviously, ��what we owe to each other�� does not depend on comparison of agent-neutral reasons but on agent-relative reasons among individuals ; therefore it is not trivially circular in the sense that the output is already employed in the input.

Nevertheless, critics may still claim that ��what we owe to each other�� is substantively circular since Scanlon holds that what we have reasons to want for generic reasons and how to weigh the relative strength of different generic reasons both depend on substantive judgments. He claims that his version of contractualism, in contrast with what he calls ��welfarist contractualism��, does not have (i) any specification of the grounds for reasonable rejection; and (ii) any specification of the method for determination of relative strength of these grounds. (217) Unlike the Complaint Model, the Generic Reasons Model employs reasons from various grounds in rejecting principles. It may be challenged that to determine whether a certain consideration other than well-being is taken as a relevant generic reason in a particular situation involves a substantive judgment about the status of that consideration. It may also be challenged that in case of conflicting generic reasons, the appeal to judgment also commits us to substantive moral judgments. For example, if we say a person can reject a principle because that principle is unfair to him, then critics may question why unfairness constitutes a generic reason against that principle if it does not involve any loss in well-being of that person? It may also be questioned that why unfairness may sometimes outweigh other generic reasons in case of conflicts. Why is it not reasonable and not justifiable to the person because it is unfair?

Scanlon defends his Generic Reasons Model by appealing to what he calls ��holism in moral justification��. ��Contractualism is not based on the idea that there is a ��fundamental level�� of justification�� and ��in assessing one principle we must hold many others fixed.�� (214) It draws an analogy from holists in epistemology who hold that justification of knowledge is just like rebuilding our boat while at sea. Scanlon rejects the idea that only ��well-being�� matters in assessing principles and reasonable rejection depends only on the magnitude of well-being affected. First, he holds that it is impossible to give up other moral claims to provide a context for assessing principles. For example, in assessing a principle defining our duty to aid others, it is assumed that any person in the position to aid others must be one that is entitled to dispose his resources freely, and any person in the position of being aided must be one that is not entitled to take what one needs whenever he wants, perhaps simply by force. ��Many other moral claims must be presupposed in order to provide a context in which that principle can be understood.�� (214)

Second, Scanlon does not consider well-being as a notion without moral content. The appeal to ��well-being�� in rejecting principle is itself a substantive claim. In fact Scanlon holds that ��the judgment that any consideration constitutes a relevant, possibly conclusive, reason for rejecting a principle in the context of contractualist moral thinking as I am describing it is a judgment with moral content.�� (217) Usually people have strong reasons to care about their well-being, and this can be generalized in terms of ��how one��s life goes��, however, the claim that effects on well-being is a generic reasons is itself a substantive judgment too. Also, ��well-being�� is not a well-defined notion; therefore substantive moral choices are involved not only in giving this notion moral significance but also in defining its boundaries. (217)

Third, ��well-being�� cannot explain all other moral notions in itself. Scanlon even thinks that an important strength of contractualism is that ��it can account for the significance of different moral notions, within a unified moral framework, without reducing all of them to a single idea.�� (216) In many cases, gains and losses in well-being are obviously a relevant factor in deciding rejectability of principles, but sometimes we have to take into considerations of other factors. For examp

?lt;/P>

previous - next

?lt;/P>