CMSI Handout 4 Courtesy of Marcello Antosh 1 Clarification about imperatives To understand what imperatives are, first think of commands like these: (i) Shut the door! (ii) Clean your room! (iii) Don t kill your friends! Commands are not capable of being true or false. Rather, they are just things that people might yell at you, or things that people might command you to do. Notice that only (iii) has to do with morality. The other two are just things people yell at you. Imperatives are related to commands. Intuitively, imperatives are capable of being true or false. Kant thinks that all imperatives are expressed by ought s, but in ordinary language we also express them with other words (like should, must, need to, etc.). In this class, we are concerned with imperatives that concern morality (like (iii)), but clearly there are other, nonmoral imperatives (like (i) and (ii)). We can transform any command into an imperative by adding an ought and a reference to an agent. Let s do this for the commands above: (i) Sarah Silverman ought to shut the door. (ii) Perez Hilton ought to clean his room. (iii) David Hasselhoff ought not to kill his friends. 2 Persons as Ends version of the Categorical Imperative: ends, means, & merely as a means The Persons as Ends version of the Categorical Imperative says: each of us ought to treat all rational beings as ends in themselves, and never merely as a means to our own ends. (Treating people merely as a means would be wrong.) In his forthcoming book, Derek Parfit suggests this way of interpreting this version of the CI: The Consent Principle: each of us ought not treat people in ways to which, if these people knew the relevant facts, they could not rationally consent. We can thus make some rough definitions regarding treating people as ends, means, and merely as means: A treats B as an end if A positively respects B as a rational agent. Examples: - Hillary Duff puts herself in grave danger to save some stranger. (Extreme case) - Kanye West buys his girlfriend (who is obviously not Taylor Swift) flowers because he loves and respects her. A treats B as a means (but in a way that is NOT incompatible with treating her as an end) if A treats B in such a way that, if B knew all of the relevant facts, then B could rationally 1
consent to being treated in this way. (The rationally consent part is very important because it shows that A respects B as a rational agent - an end - who has the capacity of rationality.) Examples: - Students attend Temkin s lecture in order to learn about stuff and hear him curse. (Students using Temkin as means to knowledge and laughter.) - Stedman helps Oprah move boxes of shoes from one closet to another. (Oprah using Stedman as means to help her move stuff.) - Tougher case: a john pays a prostitute for sex. (Suppose the prostitute rationally consents. But still, is the john respecting the prostitute as a rational agent, or just taking advantage of her situation? Maybe she prostitutes only to support a drugs habit) A treats B merely as a means if A treats B in such a way that, if B knew all of the relevant facts, B could not rationally consent to being treated in this way. When A treats B in this way, A treats B as a mere object, disrespects B s rational agency, and manipulates B in order to accomplish A s own goals or aims. The important distinction is between (1) treating people merely as a means - which is basically always morally impermissible and (2) treating people in the other two ways (as a means, and as an end) - which are basically always morally permissible. Don t worry so much about the difference between treating people as an end and treating people as a means. - A sadistic villain tortures his kidnap victim for hours even while the victim protests. - Teens bully some nerd every day after school, for their own amusement, though he resists. - Tough case: Masochist Mallory wants to be tortured for hours, it s one of her deepest, long held desires. If we tortured her, would we be treating her merely as a means? It seems that she could consent, but would that consent be rational? Or would she be irrational to consent to such treatment? It s unclear. What is rational consent? Roughly: A can rationally consent to be treated in some way if A has sufficient reason to be treated in this way. A has sufficient reason to be treated in some way only if A there is some reason to be treated in that way, and no conclusive reason to be treated in another incompatible way. Note, for Kant, what we have sufficient reason to do from a moral perspective must always be determined by our PURE reason, that is, never by our particular interests, desires, or inclinations. Examples: - We all have sufficient reason and can rationally consent to be saved from drowning. - None of us have sufficient reason nor could we rationally consent to suffer in agony just because our suffering would be funny to someone else. - People who must have their legs amputated to survive some accident have sufficient reason and can rationally consent to having their legs amputated. 3 Utilitarianism: John Stuart Mill 3.1 Chapter I: General Remarks (pp. 1-5) 2
The Basics: This chapter mainly deals with methodological considerations about how to discover the foundations of morality, that is, the first principles of morality. (By first principles Mill means something like axioms - like the axioms of different branches of logic and mathematics.) Mill s Question: What is the foundation of morality? What are the first principles of morality, from which all other moral principles can be derived? What is the most basic element of the correct theory of morality? (p. 1) The Principle of Utility/The Greatest Happiness Principle: Mill thinks the principle of utility (i.e. the greatest happiness principle) has played and should play a fundamental role in any theory of morality. (p. 3) Mill s attack on Kant: Kant s Categorical Imperative isn t a good foundation of morality. Why? Among other things it permits the most outrageously immoral rules of conduct. And when it get things right, that s only because it has shown that the consequences of certain maxims would be very, very bad (p. 4). 3.2 Chapter II: What Utilitarianism Is (pp. 6-26) important Some clarifications: Mill clarifies what he means by utility. It does not mean: the property of being useful, where useful is opposed to pleasurable (p. 6). Utilitarianism (for short, U): The Principle of Utility/ The Greatest Happiness Principle = actions are right in proportion as they tend to promote happiness; wrong as they tend to produce the reverse of happiness (p. 7). Two important definitions: (1) Happiness =df pleasure and absence of pain. (2) Unhappiness/reverse of happiness =df pain and absence of pleasure. Happiness is the only thing desirable as an end in itself. All desirable things are desirable because they either (i) lead to happiness inherently, or (ii) are a means to something that leads to happiness inherently (p. 7). U is not a self-indulgent form of hedonism. E.g. U does not demand that we should all do ecstasy and have orgies (p. 7). (No, that isn t a good (enough!) reason to reject U.) Mill s distinction between higher and lower pleasures: roughly, a distinction between pleasures of mere sensation (taste, touch, sound, sight, etc.) and pleasures of the intellect (moral sentiment, imagination, creativity, understanding) (p. 8). Importantly, Mill thinks there is a qualitative difference between the higher and lower pleasures, while others have thought the difference is merely quantitative (pp. 8-11). Informed Preference Test; the Doctrine of Competent Judges: To determine which of any number of pleasures is best, or better than the others, we must consult the opinions of those 3
who are competently acquainted with all of these pleasures (pp. 8-11). If there is disagreement, we have to go with the opinions of the majority of the competent judges. Famous quote: It is better to be a human being dissatisfied than a pig satisfied; better to be Socrates dissatisfied than a fool satisfied. And if the fool, or the pig, are of a different opinion, it is because they only know their own side of the question. The other party to the comparison knows both sides (p. 9). This is VERY important for Mill, also dubious. Important features of Utilitarianism: Utility Maximization: not the agent s own greatest happiness, but the greatest amount of happiness altogether (p. 11). Impartiality: the happiness which forms the utilitarian standard of what is right in conduct is not the agent s own happiness but that of all concerned. As between his own happiness and that of others, utilitarianism requires him to be as strictly impartial as a disinterested and benevolent spectator (p. 17) Talk about impartiality Rule vs. Act U, p. 19, 23, 24 Admits and EXPLAINS exceptions when duties conflict: E.g. it s permissible to tell lies to save lives (p. 23). Distribution?: Mill says little about the distribution of happiness! Potential problem! Act U vs. Rule U: Acts that would maximize happiness, vs. Rules, which, if generally followed would maximize happiness. Mill doesn t discuss explicitly discuss the difference. (But see: pp. 19, 23, 24) It mainly comes up for Mill in section V, which we don t read. Objection: We can t promote or attain happiness. Hence, doing so can t be what we are morally or rationally obligated to do (p. 12). Reply: Fiddlesticks! Of course we can (pp. 12-13). Objection: We can and should do without happiness since doing so is more virtuous and noble. So, promoting happiness can t be what we re morally obligated to do (p. 12). Reply: True, it s possible to do without happiness, but this doesn t imply anything about what we ought to do morally. Also, what s often so good about doing without happiness is that this sacrifice is a means to the happiness of others (hero, martyr). Concession: Given the imperfect state of the world, it is often true that the best way to promote overall happiness is to do without so much of your own and promote that of those who have much less of it (pp. 15-17). Objection: U is too idealistic, too demanding (p. 18). Reply: The is/ought distinction (originally from Hume). It may be true that people don t or wouldn t do what U says they ought morally to do, because they think it s too demanding. But it doesn t follow that U isn t the correct moral theory. 4
E.g. Suppose U says Lindsay Lohan should drive sober. Although she always has and always will drive drunk, it doesn t follow that she should not drive sober (p. 18). Objection: U requires that we be motivated to do our duty all the time. Even if we take into accounts Hume s is/ought stuff, such a requirement really is too demanding - so demanding that no true moral theory could require it (p. 18). Reply: no [plausible] system of ethics requires that the sole motive of all we do shall be a feeling of duty the motive has nothing to do with the morality of the action, though much with the worth of the agent (p. 18-19). So, U. tells us we have to bring about the best outcome, it doesn t tell us what set of motivations and dispositions would do that best. If acting like a Kantian would actually bring about the best outcome, then U. would say we should act like a Kantian! Objection: according to U, only the consequences matter, and each person s moral character, which causes the person to act how he or she does, doesn t matter at all (p. 20). Reply: No plausible moral theory holds that an act is right (or wrong) because it is done by a good (or bad) person. True, good people tend to do the right thing, but that s not always the case. There s a morally relevant difference between right (and wrong) acts and good (and bad) persons (p. 20). Concession: it s true that U basically ignores beauties of character and focuses on consequences of actions. But doing so allow U to be a better theory (p. 21). Objection: U is somehow contrary to religion (p. 22). Reply: Fiddlesticks (p. 22)! Important Objection: Sometimes, what will best promote my happiness is expedient - it s a quick fix that suits my purposes and isn t, intuitively, a morally right act at all. Doesn t U imply that these expedient immoral acts are really morally required and right? If so, U is an implausible moral theory. E.g. Telling lies often gets me out of tough situations, but doing so is, intuitively, wrong (pp. 22-23). Important Reply: We can distinguish between acts and rules. That is, we can distinguish between acts that would maximize happiness in some particular circumstance from acts that, if practiced generally as rules, would maximize happiness. Sometimes the morally right act is the one that doesn t maximize happiness in a particular circumstance, but rather is that act that, if practiced generally as a rule, would maximize happiness (pp. 22-23). E.g. If Justin Timberlake cheated on Jessica Beal just this once, he might maximize happiness - he d have fun, she d never find out. But, if he cheated on her regularly, that might not maximize happiness - he d eventually catch (another) STI, he d be paranoid about sneaking around all the time, she d find out, they d fight, etc. Also, consider what would happen not just if Justin cheated regularly, but if everyone in relationships did. (Notice the might s in this reply. What if regular cheating didn t have bad effects?) Important Objection: It s much too complicated to determine what would maximize happiness, whether in a particular circumstance, or generally. This is because our acts often have unpredictable consequences. Maybe, if I shoot this gun into a crowd, I d kill the next Adolf Hilter. Maybe not. So what should I do (p. 23)? Important Reply: We can distinguish between acts that would in fact maximize happiness from acts that we can reasonably believe to maximize happiness. We often do not know what would in fact maximize happiness. But, we 5
know and can better learn what things tend to maximize happiness, and we therefore can form reasonable beliefs and expectations about what would maximize happiness (pp. 23-25). E.g. If Britney Spears neglects her children, that might turn out to maximize happiness. Why? Suppose that if she didn t neglect them, they would in fact become the next Adolf Hitler and Pol Pot. If so, it would be better if she neglected them and they died from starvation. Yet, we know from experience that, when parents neglect their children, that doesn t tend to promote the maximization of happiness. So, it is reasonable for us to believe that neglecting children isn t something that will maximize happiness, even if we don t know exactly what will happen in any given case of child neglect. The Principle of Utility as a common umpire: U allows for there to be exceptions to moral obligations, especially when different obligations conflict (e.g. conflict between telling the truth and protecting friends). When obligations conflict, apply the Principle of Utility to determine what you ought to do (pp. 25-26). 3.3 Chapter III: Of the Ultimate Sanction Of The Principle Of Utility (pp. 27-34) A sanction is, roughly, something that makes a principle or rule of conduct binding, or something that gives binding force to this principle or rule of conduct. Sanctions include motives and reasons for adhering to these principles and rules. Mill s Ultimate Sanction Question: Consider the Principle of Utility: What is its sanction? What are the motives to obey [it]? Or, more specifically, what is the source of its obligation? Whence does it derive its binding force? (p. 27). Think of Mill s Ultimate Sanction Question this way. Sometimes you feel that there are just some things you morally ought to do (have to do, must do, should do, etc.). E.g. You may feel that you ought to help out your loved ones if they are in grave danger. Your obligation is thus: I ought to help my loved ones if they are in grave danger. But what makes this obligation binding upon you, or what gives it its force? Perhaps nothing other than the fact that it is your feeling - a feeling that you have as an agent who can respond to such a feeling by acting on it. Perhaps something else gives this obligation its force, like your fear of punishment or of guilt if you didn t follow this obligation. But now consider the Principle of Utility. This principle says that actions are right in proportion as they tend to promote happiness; wrong as they tend to produce the reverse of happiness. And, as an obligation this principle says: you ought to promote happiness and minimize unhappiness. In this chapter Mill seems to be wondering what could possibly make this obligation (if indeed it is an obligation) binding upon you, or what could possibly give this obligation its force. Two kinds of sanctions of obligations: External and Internal 6
External sanctions give your obligations their force/make them binding upon you because of something that is ultimately external to you (p. 28). E.g. Kant s prudent shopkeeper believes: I should treat my customers fairly. What gives this obligation, this should, its binding force upon him is his awareness of what others might do to him if they discover that he is unfair. Things that others might do to him are external sanctions of his obligation. To get rid of the binding force of this obligation, we only need to remove these external sanctions: If he knew he could treat his customers unfairly and get away with it, he would not feel bound to the obligation I should treat my customers fairly since there would be no external sanctions. Internal sanctions give your obligations their force/make them binding upon you because of something that is ultimately internal to you, like your conscience (pp. 28-29). E.g. Jack Bauer believes: I should protect my country. What gives this obligation its binding force is Jack s deep feeling in his own mind that abiding by this obligation is something he must do. To get rid of the binding force of this obligation, we would need to alter Jack s deep feelings somehow. The ultimate sanction of all morality (including the Principle of Utility) = a subjective feeling in our own minds the conscientious feelings (p. 29). Objection: People who think the ultimate sanction of morality isn t subjective/inside our own minds/mind-dependent but rather that it is objective/outside our own minds/mind-independent are more likely to follow their moral obligations. Since moral theories should tell us to obey these obligations, no plausible theory can hold that the ultimate sanction of morality is subjective (pp.29-30). Reply: Obviously the ultimate sanction is in our own minds/minddependent/subjective - without that deep feeling that you morally ought to do something, there s no binding force to do that thing (except for any external sanctions, obviously). It may be true that people who falsely believe that the sanction is not mind-dependent are more likely to do what they think they ought to do morally, but this is no objection to U as a theory of morality. If we want to get people to follow what U says, we should get them to feel the force of the Principle of Utility, through education, etc. (p. 30). Is morality innate or acquired?: We don t need to settle this. If it s innate (think a priori, think Kant), there s no reason to think that morality shouldn t be concerned with promoting the happiness of ourselves and others. If it s implanted (think a posteriori, think Mill), so what? It still matters to us we in fact care about the promotion of our own happiness and that of others. Either way, U as a theory is on the right track (p. 30-31). Empirical evidence in support of people feeling bound to the Principle of Utility: Obviously, most of us care about promoting our own happiness. Also, humans are social beings. We want to live in harmony with our fellow humans (if for no other reason than that this will make it more likely that our own happiness will be promoted). We recognize that the only way this is feasible is if we all respect each other, and if we all have a say in how our society works. To do this, we all know we must take into account the considerations of others. We all know that, the more we do this, the better off we all will be cooperation helps each of us, while free-riding and 7
inequality ultimately harms us all. As we cooperate more and more, we start to become more impartial we start to regard the interests of our peers as our own interests. Thus, we begin to want not only to maximize our own happiness, but the happiness of our society as a whole. Thus, we all begin to feel the force of or feel bound to abide by the Principle of Utility (pp. 31-34). Objection: Look, Mill, I don t feel bound to the Principle of Utility because I don t really feel that compelled to help out others (pp. 33-34). Reply: Think about your fellow humans for a moment. Don t you realize they re just like you in many important ways? They are! Don t you feel like you should help them if you can? Of course you do of course you feel some kind of sympathy for them, deep down, some kind of concern for their happiness. Sure, that feeling that you should promote the happiness of all sentient beings may be weak. It may be weaker than your self-interest, for instance. Nevertheless, it is the foundation of all morality. It is the ultimate sanction of the Principle of Utility (p. 34) and it can be strengthened through proper education (social conditioning?). 3.4 Chapter IV: Of What Sort Of Proof The Principle Of Utility Is Susceptible (pp. 35-41) Mill s Goal: to offer a proof that the Principle of Utility is the only criterion (i.e. standard) of morality. What he seems to mean by proof is not what logicians and mathematicians mean by proof. Rather he seems to mean: reasons, evidence, or considerations in favor of the idea that the Principle of Utility is the only criterion of morality. Problem: like the first principles of any field (e.g. math, logic), we cannot, by logical reasoning alone, establish that they are true (p. 35). But we can offer reasons, evidence, and considerations sufficient to determine the intellect (pp. 4-5). Proof Sketch: There seem to be three steps in Mill s proof that the Principle of Utility is the only criterion of morality. Step (1) The happiness of each person is a good to that person: since each person desires his/her own happiness, this person s happiness is a good to this person (p. 35). Step (2) The general happiness is a good for the aggregate of all persons: since each person s happiness is a good to that person, the general happiness is a good to the aggregate of all persons (pp. 35-36). Step (3) Nothing other than happiness is good in itself: insofar as things other than happiness are good, this is only because these things are conducive to or parts of happiness (pp. 36-41). Comment: Everyone is confused about Mill s proof. At times he seems to say things that are inconsistent with his theory. At other times he seems to make extremely bad non sequiturs - i.e. conclusions that don t follow logically from his prior claims. He even seems to make mistakes that he himself warns against both in this book and in his book on symbolic logic (namely, mistakes related to the is/ought distinction). But, Mill was a genuine genius, so try to get a rough idea of his argument even if (even though?) it totally sucks. 8
Impartial reconstruction of Mill s Proof: Step (1) From the impartial perspective, it is a good if each person is happy. (2) So, from the impartial perspective, the happiness of all these people is a good. (3) Same as above. (4) Therefore, what we ought to do morally is promote the happiness of the set of all persons - that is, we ought to maximize happiness from the impartial perspective. 4 Extra: Goodman s new riddle of induction / the grue paradox This is a puzzle about inductive arguments. Inductive arguments try to conclude that some hypothesis is true, or likely to be true given some set of data. Consider a simple example of an inductive argument. Let our hypothesis be: The next raven anyone sees will be black. We might try to conclude that this hypothesis is true or likely to be true by arguing like this: All the ravens I ve ever seen are black. All the ravens you ve ever seen are black. My mom s seen lots of ravens, and they are all black. I have friends in other countries, and all the ravens they ve ever seen are black. Together, we must have seen at least five hundred ravens. So, the fact that all of us have seen so many black ravens in so many different parts of the world gives us good reason to believe that the next raven anyone sees will be black. Now that we have a basic idea of what inductive arguments are, let s consider Goodman s riddle. Goodman s riddle is basically that, for every intuitively plausible inductive argument, there are (infinitely) many other intuitively implausible inductive arguments that use the same set of data to reach different conclusions - conclusions that are inconsistent with the intuitively plausible conclusion. Let s look at Goodman s riddle using the case that Temkin discussed -- the case involving emeralds. Suppose we are geologists studying emeralds. So far, our set of data is this: Emerald Data: We ve observed thousands of emeralds so far, and they have all been completely green. We have not yet observed any emeralds that are not completely green. Now consider the following hypothesis: Green Hypothesis: All emeralds are completely green We might think that this set of data to gives us reason to believe that the Green Hypothesis is true, or likely to be true. One reason for thinking this depends on the following inductive argument which appeals to this set of data to support this hypothesis. Inductive Argument For The Green Hypothesis: (1) We ve observed thousands of emeralds so far, and they have all been completely green. We have not yet observed any emeralds that are not completely green. (2) Green Conclusion: So, the Green Hypothesis is either true, or at least likely to be true that is, either all emeralds are completely green, or it is likely that all emeralds are completely green. 9
Now, notice the following. We can appeal to the very same set of data the Emerald Data and offer an inductive argument that is analogous to the one above, but which supports a hypothesis that is both implausible and which makes claims that can be logically inconsistent with the claims made by the Green Hypothesis. In other words, we can offer an argument just like the one above but that reaches a different conclusion. And this different conclusion can conflict with the conclusion above in a very direct way. To offer such an argument, let us first stipulate the following definition. Definition of grue : Something, x, is grue if and only if: either (i) x is first observed before 2050 and x is completely green, or (ii) x is first observed during or after 2050 and x is completely blue. Another way of stating the definition of grue is this: A grue thing is something that either (i) is first observed before 2050 and is completely green, or (ii) is first observed during or after 2050 and is completely blue. Notice that, given the definition of grue, the Emerald Data are consistent not only with the Green Hypothesis, but also with the following hypothesis: Grue Hypothesis: All emeralds are grue The Emerald Data are consistent with the Green Hypothesis because all observed emeralds have been completely green. And if all observed emeralds have been completely green, then it is at least possible that all emeralds are completely green. That is, given the Emerald Data, it is at least possible that the Green Hypothesis is true (and this is why these data are consistent with the Green Hypothesis). The Emerald Data are ALSO consistent with the Grue Hypothesis in the EXACT SAME WAY! For all observed emeralds have also been grue. Why is this? Look at the definition of grue : to be grue it is enough to be observed before 2050 and be completely green. But this is true of all the emeralds we ve observed so far. So, all the observed emeralds have been grue. Since all the observed emeralds have been grue, it is at least possible that all emeralds are grue. That is, given the Emerald Data, it is at least possible that the Grue Hypothesis is true (and this is why these data are consistent with the Grue Hypothesis). Since the Emerald Data are consistent with the Grue Hypothesis in this way, notice that we can offer an inductive argument in favor of the Grue Hypothesis that seems to be analogous to the argument we offered in favor of the Green Hypothesis. Inductive Argument For The Grue Hypothesis: (1) We ve observed thousands of emeralds so far, and they have all been grue. We have not yet observed any emeralds that are not grue. 10
(2) Grue Conclusion: So, the Grue Hypothesis is either true, or at least likely to be true that is, either all emeralds are grue, or it is likely that all emeralds are grue. Notice, however, that these two hypotheses make claims which can be logically inconsistent. This is because they draw different conclusions about the properties that any emerald discovered during or after 2050 would have. To see this, suppose some emerald is first observed during or after 2050. Now consider what properties this emerald would have to have if both hypotheses were true. If the Grue Hypothesis were true, then this emerald would be grue. But, more specifically, this emerald would be completely blue, since it was discovered during 2050 (recall the definition of grue ). Now, if the Green Hypothesis were also true, then this emerald would also be completely green. But that is impossible: this emerald cannot be both completely green and completely blue. So, since this is impossible, at least one of our hypotheses must be false if some emerald is discovered during or after 2050. In this way, these two hypotheses can make claims that are logically inconsistent. (We could also offer a similar argument for the conclusion that, if one of these hypotheses is likely to be true, then the other cannot also be likely to be true. It would be similar to the one just presented.) In fact, things get worse. For it seems that, by appealing to the Emerald Data, we can offer inductive arguments not just in favor of the Green Hypothesis and the Grue Hypothesis, but for any number of hypotheses. For instance, suppose we define the word gron as follows: A gron thing is something that either (i) is first observed before 2050 and is completely green, or (ii) is first observed during or after 2050, weighs at least one ton, and is not completely green. And suppose we define the word gradioactive as follows: A gradioactive thing is something that either (i) is first observed before 2050 and is green, or (ii) is first observed during or after 2050, is radioactive, and is not completely green. So, the Emerald Data are consistent with the following hypotheses and, because of this, it seems we can appeal to this set of data to offer inductive arguments in favor of these hypotheses as well: Gron Hypothesis: All emeralds are gron. Gradioactive Hypothesis: All emeralds are gradioactive. So, there seems to be a problem. Because the same set of data allows for the possibility that conflicting hypotheses are true, we can offer analogous inductive arguments in favor of these hypotheses. But, intuitively, it seems that the data do not really support all of these hypotheses. Instead, intuitively, it seems that the data only support the Green Hypothesis. The general problem, then, is this: The General Problem Presented By The New Riddle Of Induction: The problem is to explain why only some inductive argument(s), which appeal to a given set of data, is (or are) any good. In other words, the problem is to explain why a given set of data actually only supports some hypotheses or conclusions and not others, even though all these hypotheses and conclusions seem to appeal to the same set of data in the same way. 11
A general point of relevance to this class (according to Temkin): when confronted with powerful skeptical arguments like Hume s Riddle of Induction or the New Riddle of Induction, we may find them intellectually interesting and challenging, but we don t really BELIEVE them. As Temkin says, we don t start drinking gasoline rather than milk, or taking the shortcut out the seventh story window rather than the elevator, just because it seems that there are powerful ARGUMENTS in support of the conclusion that we have no more REASON to do the one rather than the other. That is, even if we believe we don t, or even CAN T have an answer to the deepest skeptical worries that are raised in the domain of science or the ordinary domain of everyday life, we don t all become skeptics in those domains. Nor SHOULD we, most of us think. But, for many, as soon as they confront skeptical arguments in the MORAL domain, they become skeptics about morality. They are now suddenly prepared to assert that there is no REASON to believe that there is REALLY anything WRONG, say, with tearing the skin off of little children just because one like s to hear high pitch squeals. But can this be right? Is it more plausible to think that such a moral claim might be false, than that most of the basic claims of science might be false? Why do we seemingly hold ethics to a HIGHER standard than science when it comes to skeptical arguments? 12