The Kant vs. Hume debate in Contemporary Ethics : A Different Perspective Amy Wang Junior Paper Advisor : Hans Lottenbach due Wednesday,1/5/00 0
The Kant vs. Hume debate in Contemporary Ethics : A Different Perspective The Kant vs. Hume debate in Contemporary Ethics is a debate on the source of moral requirements and their scope of influence over people. The Kantians argue moral requirements are based on categorical imperatives. 1 They consider certain actions objectively necessary of themselves regardless of one s desires. Humeans, however, argue that moral requirements are based on hypothetical imperatives. They think we cannot act morally unless we desire or could desire something that moral actions will be instrumental in obtaining. In this paper, I suggest that the source of moral requirements is neither in hypothetical nor categorical imperatives, although their scope of influence is hypothetical. Furthermore, I argue that morality is not directly related to reasons. I do this by proposing that A) moral standards are not based upon categorical imperatives, B) that moral principles are not based upon hypothetical imperatives, that C) they motivate hypothetically, i.e., by desires. Furthermore, I would like to D) explain just how a conception of reasons can be argued to have a correlation with moral imperatives and E) analyze which conceptions of reasons may or may not offer certain reasons to be moral. 1 Foot, 159 1
A. Moral Standards are not based upon Categorical Imperatives The categorical imperative would be one which presented an action as of itself objectively necessary, without regard to any other end. Kant 2 I agree with Foot that it is unlikely that acting according to a moral imperative must be of itself objectively necessary without regard to any other end. 3 Here I discuss how it is possible to interpret this necessity. I argue that none of the options (linguistic, substantive, causal, moral, and practical necessity) that support that a good action is objectively necessary of itself are both correct and what Kant is trying to say. Thus, I will argue that moral standards are not based upon categorical imperatives. Linguistic and Substantive Necessity Linguistic necessity is necessity merely by virtue of linguistic conventions. Substantive necessity is necessity not merely by virtue of linguistic conventions, but in reality. For example, a bachelor is by linguistic necessity both unmarried and male. If a bookshelf is red in reality, that is a substantive, not linguistic necessity. Thus, if morality is based on categorical imperatives, then on the first interpretation, part of being a good action is being an action that takes place for its own sake. By for its own sake, I mean that one desires it not merely as a means to something else one desires but one desires something in its nature itself. I do not mean that one must only have the end of acting morally. Whether or not there are any good actions depends upon actual facts. On the second interpretation, if good actions are substantively necessary of themselves, when they do take place in reality, it is always partly for their own sake. This is not to say that part of being a good action just is being an action that takes place for its own sake. 4 This is to say that no good action does take place in actuality unless the perpetrator of the action does care at least somewhat about doing what is good for its own sake. 2 Foot,158 3 Kant quoted in Foot,158 4 Foot,164 2
Suppose I joyfully give some children gifts. When asked what my motivation is, I may say that I am giving gifts because I love children, and loving children because it is good, and doing good partly because I like goodness for its own sake. If this is my justification, only then can my action be good, according to this view. I desire it for its goodness and not merely because it is instrumental to something else I desire. However, if I say that I do good merely because I want to do what is good and not at all because I value what is good, this is a contradiction on this view, since it is impossible to do what is good (either linguistically or substantively) unless it is done at least in part for its own sake and not merely because we need it for something else. If this is what it means for moral requirements to be categorical imperatives, then they do have a different status from etiquette standards. Etiquette, on the other hand, does permit that one can act according to etiquette rules even if one only value etiquette for its instrumentality toward getting something else one desires. However, I argue that it is ridiculous to think of moral imperatives as categorical imperatives in this way. On the linguistic necessity interpretation, it would mean that moral standards indicate that certain actions have been chosen to be set apart from other actions and for them to become good, one must not only do them, but do them with some justification that they are in that set of actions. On the substantive necessity interpretation, certain actions have been chosen to be set apart from other actions and are good, but they simply have no motivational efficacy when they are not done at least partly for their own sake. I do not deny that a morally good action can be done for its own sake but rather that it must be done partly for its own sake. Causal necessity One kind of substantive necessity would be causal necessity, necessity by virtue of causal laws. In other words, a certain action is necessary if one is bound by certain laws governing cause and effect and is to attain a certain end. Here something is necessary for a certain consequence. For example, if one says that brushing one s teeth is necessary, one assumes, it is necessary for remaining healthy, not necessary to do for its own sake. One could say, unless one 3
does an action partly for goodness sake, one will not be able to be motivated to do what is good. In other words, it is simply by cause-effect laws governing the universe that one cannot do a good action if one cares nothing for its goodness, although by linguistic conventions alone, one can do a good action without doing it for goodness sake alone. Nevertheless, any causal laws of the kind that would make this necessary do not seem to exist in our universe. We are able to will actions we do not at all desire for other things that we do desire. Moral necessity Another interpretation of being objectively necessary of itself would be necessity by virtue of moral laws. In other words, if one were bound to act according to moral laws, one would necessarily do certain things. Doing what is good would have a unique status from the point of view of moral necessity, but not doing what is, say, according to etiquette. The problem with this view, however, is why we should interpret Kant s usage of necessity as moral necessity and not etiquette necessity or some other necessity. 5 Practical necessity Practical necessity is necessity by virtue of being bound to considerations of reasons. 6 However, I have doubts that moral action is practically necessary. At the end of this paper I discuss different kinds of practical reasons and how they relate to morality. Because of this exploration of what reasons can be, I do not think an action good in itself is always necessary to a will conforming to reason. From Kant s writing, it appears that he means by necessitation what happens when a will is bound by objective laws. 7 But it is unclear the content of these objective laws or that conformity with a reason must necessarily cause one to act morally. Objections/Counterobjections (Korsgaard) The only other possible way that I can conceive this idea of necessity working is if one thinks of the term necessary as meaning intrinsic and final good in this context. Korsgaard, in 5 Foot,162 6 Kant,25 4
The authority of reflection, talks about two distinctions in goodness: the distinction between final and instrumental value which concern reasons for valuing something, and the distinction between intrinsic and extrinsic value which concern the source of the value. 8 She writes, Both final value and intrinsic value may seem to be in a certain way ultimate, or foundational. 9 if we are going to get categorical duties out of it [intrinsic value], the value in terms of which we justify action must be independent of people s particular desires and interests. 10 the intrinsically normative entity, the entity that brings a regress of justification to a satisfactory end, must combine these two conceptions[intrinsic and final value]. 11 However, I object to this conception of necessity on the grounds that it fails to capture the Kantian idea that categorical imperatives specify actions of themselves objectively necessary, without regard to any other end. 12 Korsgaard goes on to write, Values are not discovered by intuition to be out there in the world. Good maxims are intrinsically normative entitites, but they are also the products of our own legislative wills. In that sense, values are created by human beings. Of course we discover that the maxim is fit to be a law; but the maxim isn t a law until we will it, and in that sense create the resulting value values are constructed by a procedure, the procedure of making laws for ourselves. 13 She argues that values come from our own legislative procedure. However, it is very unclear why people would legislate something that was independent of their particular desires and interests. Values are a type of desire. Thus, I conclude that moral imperatives are not categorical imperatives. I think that one is able to do a good action and with a good motivation which does not necessarily include being good for its own sake. It all depends upon exactly what is specified as being good. I do not see why action specified as good necessarily must be accompanied by the intention of doing it for its own sake. Like Foot, I think it is possible to do good action without acting according to categorical imperatives. 7 Kant,24 8 Korsgaard,111 9,111 10,111 11,111 12 Foot,158 5
B) Moral Requirements are not based upon hypothetical imperatives. The former[hypothetical imperatives] present the practical necessity of a possible action as a means to achieving something else which one desires (or which one may possibly desire). Kant 14 I suggest here that moral requirements main function is merely explicative and thus that moral requirements are not based upon hypothetical imperatives. I would revise the hypothetical imperatives for a moral imperative, which I would express as the following : something that represents the necessity that an action be done in order to be good. In other words, what I previously called moral necessity. As I argued in part A, moral good is something which does not necessarily have to be done for its own sake. However, I do not think it proper to call moral requirements as based upon hypothetical imperatives either because I do not think they are directly related to what one wills or ends/means considerations. As long as it is possible to will anything, then the means to satisfying that will is represented as practically necessary by hypothetical imperatives. My main point is that for a moral requirement to be a hypothetical imperative, this requires only that it is possible for one to will to do what is good or one wills anything which doing what is good is necessary to achieving. I think it is possible to will to do many things. Another distinction that I want to make is that what it is good to do on the most general level does not necessarily change with one s desires. At the top level of abstraction of morality, we are given very general specifications about moral principles, which I call form. We apply them to actual situations to get moral imperatives. For example, suppose the basic form I am given is a circle. Then, to be in accordance with the form, I must shape the matter in such a way that it conforms. I have a general principle, like a general rule of thumb, and it must be applicable whatever materials I am given no matter how flexible they are. In some cases even if I am not given material with the capacity to have the desired form, being in accordance with the 13 Korsgaard,112 14 Foot,158 6
form will not be within the capability of the material, but this does not mean that the form changes. In this sense, I think general moral imperatives are nonhypothetical. They tell you what you need to do whatever your desires are in order to be moral. They offer rules of form that may take in account a wide range of factors, including desires but these most general rules do not change because of one s desires. ANALOGOUS MODEL moral rule of form : an ideal (example: circle) material in given form (example, ink in triangle form) : material in form compliant to rule of form : O An example of such a moral rule of form would be that to be good, you must love your neighbor as yourself. Moral imperatives do not necessarily have anything to do with what one desires, even if they do represent something that happens to be what it is possible to desire. A moral imperative would be like a hypothetical imperative in the sense that it is practically necessary to do for something. But unlike a hypothetical imperative, what it is for does not have to be related in the least to what one desires or what it is possible to desire. A moral imperative would apply to a rock even though a rock cannot possibly desire anything! I think that a moral imperative is practically necessary for something, but that it is not necessary to specify that it has to be related to something one desires or could desire. However, this does not mean it is a categorical imperative either, because one does not necessarily have to value what is good in any way at all to do what is good. For example, one could love one s neighbor as oneself because one values one s neighbor because of who the neighbor is, without at all valuing what is good for what goodness is. Yet, one would be acting morally in loving one s neighbor as oneself. Only if there was a moral requirement that one value what was good in order to be good, would that omission be immoral. However, that omission would not make the commission of another action immoral as well. One would merely be doing something morally and another thing immorally. However, moral action would still be possible. Thus I conclude that a moral rule of form does not necessarily have to be a hypothetical imperative or a categorical imperative. 7
Objections/Counterobjections (Railton) Railton gives a good analogy in his paper, On the Hypothetical and Non-Hypothetical. A deductive argument owes its conclusion to both the premisses and rules of inference. Thus, he suggests that practical reason also owes its conclusion to both hypothetical and non-hypothetical considerations, which has what he calls the elements of end-setting and ends/means-adjusting 15 However, Railton is talking about the Hypothetical and Non-Hypothetical in Reasoning about Belief and Action and not about moral action. In this case, however, I take moral rules of form to be merely a set of guidelines for moral actions. I somewhat distinguish moral rules from end-setting. Moral principles merely suggest a standard of evaluation and do not give an answer of what one chooses to do. There is no demand of ends/means-adjusting, only rules which can help one determine what ends it is good to have and the means which are good to take toward those ends. It is thus only appropriate to understand moral imperatives as not hypothetical. Even if the form of moral imperatives was a rule like Do what you want to do, it would still be non-hypothetical in the sense that the rule is unchanging. In other words, if that moral principle is true, then it will be good to do what you want to do, whether or not you want to do anything at all. Moral principles do help you answer the question of what to do to satisfy ends which are to do what is good but that is different from saying moral principles or any other principles have hypothetical imperatives as their source. 15 Railton, 79 8
C) Ways in which people can be moved to act according to Moral Imperatives I argue here that there are many ways in which people can be moved to act according to moral imperatives, and it is necessarily because of desires. In other words, moral imperatives can motivate only by hypothetical considerations. I differentiate this discussion from my previous discussion in parts A and B. They discuss the source of moral requirements. Here I discuss the scope of influence of moral requirements over people. I take into account that people have wills with different capacities (or perhaps no capacity at all) to act according to laws for the laws own sake. My argument involves a conception of the subject of this motivation and the basic requirements for action. The subject, I propose, is very complex, like a government system, which may include several roles, like executive, and legislative, 16 and which must have some desire(s), and do not merely represent desires. However, I am going to limit the subject to that contained within one human body/soul/spirit. The basic requirements for action for this self, I will argue, is (1) a minimal conception of options and (2) some free will 17, or space to control oneself untouched by his environment and (3) some desires. First, I will clarify what I mean by desire. By desire, I mean not only what one is attracted to, but also what repulses one. This would include spiritual desires like things that one values and loves, and not just fleshly desires. There are many different specific levels of desires that also complicate the issue. Most generally, there are narrow-minded and broadminded desires. One can be attracted to a picture as a whole but repulsed by its pieces. These may battle for attention in the legislative self. Or one may desire certain consequences but not the effort that is necessary to accomplish them. One may choose to act upon any of these, but not upon anything outside of this. In addition, it is unnecessary to act to fulfill all desires. If one has no desires, I fail to see how one can qualify as a subject of action. One can still be a subject, but one 16 Korsgaard,107 (on acting, governing self) 17 Korsgaard,94 9
can only be a dormant subject to be likened to a robot. It is in motion but without a functional legislative self and hence not a functional executive self. There is no one way to be moved to act according to moral imperatives. One may be motivated because one (a) desires to do that which is good or (b) desires to get something that requires (indirectly) one to do that which is good and one (d) believes that a certain action involves doing that which is intended. These are all specifications that one can consider when legislating for oneself. The advantage of this conception that moral requirements command hypothetically is that one can grant that a moral requirement may be recognized but not desired. Or, alternatively, a moral requirement may be recognized and desired and a certain action may be believed to be instrumental in attaining that which one desires, but one may still act according to some other desire that conflicts with this desire. The consequence of saying that moral requirements necessarily have motivational efficacy for all people has unsatisfactory consequences because it is not necessary in principle for one to desire anything that morality is instrumental in satisfying and it is furthermore possible to desire things that immorality is instrumental in satisfying. Objection/Counterobjection (McDowell/Korsgaard) Against Psychologism McDowell might call this picture of motivational efficacy one that relies too much upon a psychologistic view, relying too much upon the facts of individual psychology to explain action when there are things beyond desires, like transcendental value. 18 I would argue, however, that value is part of one s desires. I think this transcendental idea of value really is a weird metaphysic. Furthermore, I have taken desires to include value as well, but by this I only mean one s personal value, not some intersubjective value or someone else s values. Surely, what motivates one is what oneself desires and not what someone else desires (which is what motivates that someone else). 10
McDowell might further object that we ascribe the desire because something moves one when it may not be the desire at all which motivates one. He argues that one s conception alone can move one and that we ascribe the desire on afterthought because we see that we were moved to do the action. 19 What is problematic is whether it is one s desires coupled with a conception of what is necessary to satisfy the desire that is the correct picture for motivational efficacy, or rather what we believe we desire (although we may really desire otherwise) which motivates us. In other words, it might be objected that our representative self can lie to the legislative self or our acting self may disobey the legislative self. Thus, it is possible in some sense for the representative self not to represent one s desires correctly. However, this is beyond the point. It is unclear that these are three selves in one when it could as likely be three roles of one self. In this case, the truth would be that there really is no representative self or legislative self or executive self. It may be that the same self just represents desires as it desires, legislates as it desires, and acts upon whichever legislation it is most (net) attracted to in a given moment. McDowell might argue that we are simply begging the question, defining the subject of action so that it must act upon desires. But it is just the case that through our conception we are informed of our options but our desires specify which ones we act upon. Without them, we must be asleep or dead. These are merely what I take to be linguistic necessities. Someone of course could act on a moral requirement if it is legislated, but only because one was more attracted to it than the other options, i.e. one desires it the most. We can liken our motivations to a force diagram where the net force results in action. This way we can also account for the distinction McDowell makes between the continent and the virtuous man. The virtuous man only legislates a few options which are good whereas the continent man legislates many options which include both the good and the bad actions. Thus, the continent man weighs more options. 20 18 McDowell,105 19 McDowell,79 11
Accounting for Freedom Korsgaard might object that this interpretation does not sufficiently account for freedom since it does not make it necessary for reason to endorse a desire but rather for desire to give a considered choice motivational force. However, my interpretation does not take desiring as an external constraint, but something that is a part of personal identity. She writes, If the bidding from outside is desire, then the point is that the reflective mind must endorse the desire before it can act on it, it must say to itself that the desire is a reason. As Kant puts it, we must make it our maxim to act on the desire. Then although we may do what desire bids us, we do it freely. 21 Korsgaard treats desire as if it were external to the will, which she says itself is practical reason. 22 She explains that we can be free because of the reflective character of the mind that allows us to approve of a desire before acting on it. 23 However, it is unclear how the reflective mind can endorse the desire unless the individual desires to do so or desires it for some other purpose. The reflective mind is not above the individual or the total individual, nor is desire. My interpretation still allows a certain kind of freedom, to do that which one desires. It is unclear why this freedom must be freedom to do what one endorses with one s reflective mind rather than what one desires in one s heart in the case that they conflict. In this section, I have insisted that moral requirements motivate only hypothetically, and thus I have somewhat adopted a Humean answer to the question of the scope of influence of moral requirements on individuals. Although we may endorse different desires with our reflective mind, it is unclear that this endorsement itself is capable of motivating or that the will is this endorsement itself. In the next section, I intend to finish up a discussion I started previously in which I suggested that moral requirements were not directly related to reasons. 20 McDowell,91-3 21,94 22 Korsgaard,97-98 23 Korsgaard,94 12
D. How to Correlate Morality with Reasons There are several ways that one can argue that practical reason is correlated with morality. In the following list, I make some proposals of how they can be correlated, which I admit allow a good deal of flexibility in the definition/possible kinds of practical reason which is what a large part of the debate is on. (1) One may argue that morality offers a normative system and practical reason is just any solution to the problem of what to do which a normative system provides. There can be better or worse answers, but one has practical reason to do just about anything that one chooses. (2) Secondly, one might argue that what one has reason to do means what it is good to do. (3) Thirdly, one might argue that what one has reason to do means what one believes it is good to do. (4) One may additionally argue that it means to do what one desires, or purposes to do, which whether or not one believes it or not, happens to be good, although one may deny it. (5) Or alternatively, one may think it means someone or other desires or purposes for you to do something that is good. As silly as these may sound, they indicate that practical reason is a very vague term that can be used in philosophical context to have multiple meanings. I thus argue that the best way to attain clarification of the matter is to subdivide reasons into a variety of classes which all share one common thing so that there is no more room for argumentation. There is no one solution about what to choose, or one point of view by which to analyze them. Williams gives this model of the internal interpretation as the following : A has a reason to φ iff A has some desire the satisfaction of which will be served by his φ-ing. Alternatively, we might say some desire, the satisfaction of which A believes will be served by his φ-ing 24 Williams also gives a model of what he calls the external reasons which he thinks does not give reasons, that claims there are reasons independent of one s desires. There is a way in which Williams says one can have a reason which one is ignorant of. That is the case Williams mentions of which could be due to incomplete deliberation. Let us distinguish reasons with 24 McDowell,101 13
respect to objective reasons and subjective reasons. By objective reasons, I mean that action will in fact serve one s intended purpose. By subjective reasons, I mean that action that one believes will serve one s intended purpose. Thus, there are objective and subjective internal reasons and objective and subjective external reasons. This distinction, I hope, will clear some confusion. I propose that to deal with this confusion, we also divide our concept of rationality into the following categories: internal subjective, internal objective, external subjective, external objective. Is this what we really mean? If so, we can without difficulty say that it is internally subjectively irrational to act against what one believes to be what one has internal reason to do, internally objectively irrational to act against what one does have internal reason to do, externally subjectively irrational to act against what one believes one has external reason to do, and externally objectively irrational to act against what one does have external reason to do. This way we know exactly what we mean and the truths we arrive at seem merely tautologies. In this case, then, I am speaking of the possibility of morality to supply internal objective reasons. But whether there are necessarily internal objective reasons to be moral is a question that is left open to speculation. It is not a linguistic necessity. One might argue that with this conception, we allow one to say that one can be internally subjectively rational and at the same time internally subjectively irrational. In other words, one may desire two different things at once which cannot both be fulfilled by any one action alone. Then with respect to one desire one would be rational and with respect to another desire irrational. But all this shows is that our conception is still at some level of generality. If we wanted, we could go one step further, with an additional category, that is moral/immoral. Thus we can be motivated by moral external objective reasons or immoral external objective reasons, moral internal subjective reasons, etc. By being morally externally objectively rational, I mean acting in accordance to someone else s end for one to be moral. By being internally subjectively rational, I mean acting in accordance to one s good desires. This 14
may seem a bit tedious, but in this way, one may also more clearly see morality and rationality as two independent qualifiers of types of action. Objections and Counterobjections The main objection to this view would be that the multiplication of terms in a pattern like Williams s still does not capture what is meant by reason as a single term, and that there is nothing that the different types of reason necessarily share. But my main intent is to demonstrate how the term is used with multiple meanings, and this vagueness is the cause of a multitude of misunderstandings. For example, some may construe reasons as reflective success like Korsgaard 25 or a kind of consistency between intention and attitude or maximization, like Gauthier 26, or even as moral action itself. Much of the debate ends up in linguistic difficulty. Another objection might be that in order to subdivide the categories, each subsequent level must truly be a subclass of the general level it is under. Williams argues that the only reasons are internal reasons. But this is because he takes reasons to have a meaning that only allows internal reasons. Someone who takes reasons to be inclusive of internal and external reasons would share my conception. What internal and external reasons share in common is not just that they are both either internal or external reasons, but that they both are based on endsmeans considerations, whether believed, actual, someone else s ends, or one s own. In addition, by reasons, I do not mean they are necessarily motivational. All I take is necessary is that some one desire factor into some one motivation. Additionally, this is not necessarily the conception Williams has of external reasons, that they are related to someone else s ends. For example, Williams takes external reasons to include considerations, like for example, of family pride. 27 However, it would work in his conception for an external reason statement to be related to someone else s desires. This would explain both that 25 Korsgaard,93 26 Gauthier,213 27 Williams,106 15
external reasons do not necessarily relate to morality 28 and that they can be true independently of the agent s motivations 29 while still relating them in some way to someone s motivations, and thus having it share something with internal reasons. Williams writes, Now no external reason statement could by itself offer an explanation of anyone s action. 30 I suppose I have a somewhat different conception of reasons. I do not think they must necessarily explain someone s action. Here I have shown that morality is only indirectly related with reasons. The former shows what one must do to be good. The latter, at least in my own conception, is focused on endsmeans reasoning given a certain end. of one s own, another s, believed, or actual. Thus, this reinforces my earlier discussion that moral requirements have no source from categorical imperatives or hypothetical imperatives. Whether one has reason to be moral depends on one s ends and ends/means considerations. 28 Williams,106 29 Williams,107 30 Williams,106 16
E. Which conceptions of reasons may or may not offer certain reasons to be moral. Whether one has internal or external subjective reasons to be moral all depends upon one s personal beliefs about one s desires, others desires, and the things that will help satisfy them. Subjective reasons to be moral thus vary from person to person. Internal objective reasons to be moral depend upon whether being moral will satisfy any one of one s desires. I believe this to be the case for most people because most people do have some desires for consequences of moral actions, although how much motivational force they may have might be small. As for external objective reasons to be moral, I believe this is even more certain. Surely, there is at least one person who appreciates it when other people act morally. Someone might remark that such an analysis is pointless because it is of little guidance to the legislative self because one can simultaneously have reasons for and against something. They may ask for some means to evaluate these reasons and propose something like the weightiest reasons, (but not weightiest because of motivational force itself) but by some evaluative principle. Let them choose for themselves what criteria they wish to measure by and measure their criteria with itself. 17
Conclusion I chose to hold a position in the Humean vs. Kantian Ethics debate in which moral requirements do not have their source in either categorical or hypothetical imperatives, that they motivate hypothetically, and that they are weakly correlated with reasons. I showed to some extent how linguistic troubles arise in the debate and introduced some debate on what role reason and desire play. I personally think the source of moral requirements is possibly from God who gives a moral rule of form and that this always implies an external reason to internalize external reasons. What reason do we have to obey such an imperative? To this, I would answer that we have an external reason. It is thus not necessarily hypothetical on this view, because we do not have to make somebody else s ends our own. It is also not categorical on this view unless God specifies that it must be done with some desire to do it simply because its goodness is desired for its own sake. 18
Bibliography 1. Darwall, Stephen L. Impartial Reason. Ithaca: Cornell University Press, 1983, Part IV. 2. Foot, Philippa, Morality as a System of Hypothetical Imperatives, Virtues and Vices. Berkeley: University of California Press, 1978, 157-73. 3. Gauthier, David, Reason and Maximization, Moral Dealing. Ithaca: Cornell University Press, 1990, 209-233. 4. Hampton, Jean, The Authority of Reason. Cambridge: Cambridge University Press, 1998, ch.4. 5. Kant, Groundwork of the Metaphysics of Morals. Cambridge: Cambridge University Press, 1997. Mary Gregor, (ed.). 6. Korsgaard, Christine, Skepticism about Practical Reason, Creating the Kingdom of Ends. Cambridge: Cambridge University Press, 1996, 311-34. 7. Korsgaard, Christine, The Sources of Normativity. Cambridge: Cambridge University Press, 1996, Lecture 3. 8. McDowell, John, Are Moral Requirements Hypothetical Imperatives?, Mind, Value, and Reality. Cambridge/MA: Harvard University Press, 1998, 77-94. 9. McDowell, John, Might There Be External Reasons?, Mind, Value, and Reality. Cambridge/MA: Harvard University Press, 1998, 95-111. 10. Railton, Peter, On the Hypothetical and Non-Hypothetical in Reasoning about Belief and Action, Garrett Cullity and Berys Gauts (ed.s), Ethics and Practical Reason. Oxford: Clarendon Press, 1997, 53-79. 11. Williams, Bernard. Internal and External Reasons, Moral Luck. Cambridge: Cambridge University Press, 1981, 101-113. This paper represents my own work in accordance with University regulations. Amy Wang. 19