Download 6 Do the Right Thing and more Slides Ethics in PDF only on Docsity!
Do the Right Thing
An Account of Subjective Obligation
ELINOR MASON
Early in Spike Lee’s 1989 movie, the character Da Mayor, who might be seen as the moral compass of the movie, tells the protagonist, Mookie, “Always do the right thing.”^1 The advice seems useless. What is the right thing to do? Mookie brushes Da Mayor off and goes back to delivering pizza. It seems that in complex moral situations, it is often the case that we just don’t know what the right thing to do is. If we don’t know what the right thing to do is, being told to ‘do the right thing’ is not helpful. Perhaps, however, what Da Mayor is saying to Mookie is not that Mookie should do ‘the right thing’, where ‘the right thing’ is defined independently of the agent. Rather, Da Mayor is telling Mookie to do the right thing from Mookie’s own point of view. As moral theorists would put it, Da Mayor is telling Mookie to do the ‘subjectively right’ thing. But that seems useless too. In what way is this advice helpful? Is there a way to derive something helpful here? Should we be concerned with helpfulness? Moral theorists have put this issue in terms of action guidance. It seems that there must be a sense of ‘subjective rightness’ that is action guiding, and that correlates at least somewhat with what the agent is morally responsible for. However, it is not clear how this subjective sense of rightness should be defined, and how it relates to more objec- tive senses of rightness. In this paper I elucidate the notion of subjective rightness.
(^1) Do the Right Thing. Dir. Spike Lee.
1. Actualism, prospectivism, and subjectivism
It is common to distinguish between different senses of rightness, or of ‘ought’. In particular, the actualist (sometimes called ‘objective’) sense of rightness or ‘ought’, the prospective sense, and the subjective sense. 2 The first two are both defined independently of the agent, though they differ in the way that they specify the conditions that fix rightness. The prospective ‘ought’ picks out (roughly) ‘what a rational (or reasonable) agent would do in the situation’; and the actualist ‘ought’ refers to ‘what (in some sense to be specified by particular normative theories) would actually be best—what the agent would ideally do’.^3 The difference between the actualist and the prospective senses of rightness is illustrated by a well-known example. Both Donald Regan (1980) and Frank Jackson (1991) present a version:
Jill is a physician who has to decide on the correct treatment for her patient, John, who has a minor but not trivial skin complaint. She has three drugs to choose from: drug A, drug B, and drug C. Careful consideration of the literature has led her to the following opinions. Drug A is very likely to relieve the condition but will not completely cure it. One of drugs B and C will completely cure the skin condition; the other though will kill the patient, and there is no way that she can tell which of the two is the perfect cure and which the killer drug. What should Jill do? (Jackson, 1991, 462–3)
In most contexts, we should not risk the very bad outcome. In other words, we ought to do what is prospectively best, even though we know for sure that that will not be right according to actualism. 4 The example shows that the prospective ‘ought’ does not try to approximate the actualist ‘ought’. The point of the prospective ‘ought’ is not to point to
(^2) Correspondingly, we are familiar with talk of different sorts of obligation: subjective obligation, prospective obligation and objective obligation. In this paper I assume that these three terms (ought, rightness, obligation) are simply corollaries of each other. 3 Michael Zimmerman gives us a thorough account of the issues (2008, 2014). See Mason (2013) for my account of and argument for a prospective 4 ‘ought’. Fred Feldman discusses a version of this example in Feldman (1986, 46–7), but Feldman’s conclusion remains that what you ought to do is prescribe the drug that would actually be best. The most thorough defense of prospectivism is due to Zimmerman (see 2006, 2008, 2014). See also Bentham (1789); Mill (1861); Smart (1973); Gruzalski (1981); Jackson (1991); Oddie and Menzies (1992); Howard-Snyder (1997); Timmons (2012); Mason (2013). I will not say more here about the varieties of prospectivism.
118 Elinor Mason
actual value system, and it doesn’t matter what it is. However, acting subjectively rightly suggests some moral content, and so if one is acting subjectively rightly there must be some account of rightness in play. In order to know what acting subjectively rightly means, we need to know what it is indexed to. The most common strategy is to say that the agent is acting subject- ively rightly when she does what she believes is the right thing to do according to a particular moral theory. So one acts subjectively rightly ‘sub utilitarianism’, or ‘sub Kantianism’, and so on. Call this the ‘theory- relative approach to subjective obligation’.^6 Alternatively, one might say that the agent acts subjectively rightly when she does what she takes to be the right thing according to The True and The Good, as Susan Wolf puts it (1990) where we boldly take the true value system itself as given. We might call such an approach the ‘morality relative approach to subjective obligation’. In this paper I am neutral between the theory- relative and the morality-relative approaches. Either way, the possibility of acting subjectively rightly only arises within the context of a value system that the agent accepts. Subjective rightness has an objective, or external element: in order for an agent to be acting subjectively rightly she must be in the right ballpark to begin with. Some agents are not. Some agents have no views, or radically false views about morality, or are seriously normatively uncertain, and such agents simply do not act subjectively rightly, though they may of course act according to conscience. There are various things we might say about agents outside the ballpark, but I leave that discussion aside here.^7 It is worth noting, however, that both the theory-relative and the morality-relative accounts of the subjective ‘ought’ do apply when an agent is uncertain about the right thing to do within her moral theory. She can be uncertain or mistaken about something within morality, so long as she is not in a different ballpark. To extend the metaphor: a baseball player can make mistakes or be uncertain about the rules or strategy while playing baseball, and we can talk sensibly about what they
(^6) See Hudson (1989) for an argument for this position. Both Holly Smith (2011) and Fred Feldman (2012) take the theory relative approach. (^7) I say more about this distinction and argue for morality-relative subjective obligation in my manuscript, ‘Ways to Be Blameworthy’. I also discuss agents who are outside of subjective obligation there.
120 Elinor Mason
subjectively ought to do (sub baseball). But if the player is actually playing a different game, the notion of what they subjectively ought to do (sub baseball) makes no sense.
3. Action guidance
The idea that a moral theory should be action guiding is widely espoused, and it is often suggested that subjective rightness is the action guiding sense of rightness. 8 But it is not completely clear what action guidingness is. Frank Jackson says: “the fact that a course of action would have the best results is not in itself a guide to action, for a guide to action must in some appropriate sense be present to the agent’s mind. We need, if you like, a story from the inside of an agent to be part of any theory which is properly a theory in ethics” (1991, 466–7). It seems that we somehow have to shift the focus of the moral instruction from hitting the target to something that the agent is more in control of, something that the agent is able to do. We might simply translate the actualist instruction, which mentions a target that is external to the agent, into an instruction that mentions only the agent’s beliefs about that target. Instead of saying, ‘do the right thing’, we should say, perhaps, ‘do what you believe the right thing to do is’.^9 However, it seems that if we want action guidance, this will not do. An instruction that simply shifts focus from the target to the agent’s beliefs regarding the target may not be helpful. If an agent has no clue what is right, an instruction to do what she believes is right will not help. Consider Ross’s proposal, that the (subjectively) right thing for some- one to set himself to do is what “he thinks to be morally most suitable in the circumstances as he takes them to be.”^10 Imagine someone sincerely asking for advice in a difficult moral situation. Such a person might say, “I don’t know how things will turn out, though I have a good guess.
(^8) For arguments that theories should be action guiding see Gruzalski, (1981); Hudson (1989); Jackson (1991); Lockhart (2000); Mason (2003); Sepielli (2009, 2012); Smith (2010, 2011, 2012); Feldman (2012); Timmons (2012). (^9) See Prichard (1932) and Ross (1939) (Ross changed his view from an objectivist view to a subjectivist one after being convinced by Prichard). Doug Portmore notes that pure subjectivism is not action guiding (2011, 22), as does Andrew Sepielli (2012, 60). (^10) Ross (1939, 161). I ignore the complication that Ross is fully subjectivist about the value system.
Do the Right Thing 121
Let’s take a paradigmatic case of helpful action guidance, a cake recipe. A cake recipe tells you what ingredients you need, in what quantities, and what to do with them. Is the cake recipe a good model of what we are looking for from subjective obligation? It might seem that it is: if you know what the right thing to do is, you don’t need to worry about subjective obligation, you can just do the right thing directly. So, similarly, if you know how to make a cake, you don’t need a recipe. You just make a cake. A recipe is a sort of crutch for those lacking information. This is the picture that Holly Smith presents us with.^11 On her view, we can think of secondary principles as constituting our subjective obligation. As Smith puts it, “The principle of subjective rightness lays out the evaluative status of actions, relative to the principle of objective rightness, for agents who are mistaken or uncertain about whether those actions have the right-making features specified by the principle of objective rightness” (2010, 72).^12 As Eugene Bales says, a moral theory like utilitarianism tells us what to do in one sense: it tells us to maximize the good. But that is not an “immediately helpful description” (1971, 261). Smith aims to provide an immediately helpful description that can be derived from the right-making characteristics of a theory. So, is this how we should understand the subjective ‘ought’? I will argue that this picture is problematic. The first problem is that it is very difficult to pin down what ‘usability’ is. Does it have to be usable by the person actually trying to use it? Or by someone reasonable? Actual recipe books vary according to the intended audience. Recipe books for students assume more know- ledge than recipe books for children, but much less knowledge than recipe books for serious cooks. A recipe book aimed at students might say ‘cream the butter and sugar’, but it would not tell you to ‘make a roux’. The principles that would be usable for one person might not be usable for another.^13 If we settle on a particular level at which to pitch the subjective ‘ought’, we have effectively produced another iteration of the actualist ‘ought’.
(^11) See also Feldman (2012, 158). (^12) Note that Smith’s position is not that subjective rightness approximates to actualist rightness, it is rather that both appeal to the same right making features. (^13) Additionally, any account of usability will have to address the issue of proper usage. I might use a recipe book to make a cake by shredding it up and molding it into a cake. See Smith (2012) for an account of usability.
Do the Right Thing 123
How action guiding the instruction is depends on how much the agent knows already. As soon as we settle on something that would be helpful for one person, we have excluded people in different situations. The second problem is that what we ought to do varies over time, and as a result of previous decisions, including bad decisions. With every decision that we make, the options open to us change. If I add too much sugar, I have to add more of the other ingredients to even things up, and then I need a bigger baking tin, and so on. A decision procedure for behaving morally would have to be enormously flexible and complex. If I take a wrong turn, the options before me may not be as good as they would have been, but I still have to make choices. 14 Smith recognizes these problems. She suggests a hierarchy of second- ary decision principles, and defines the subjectively right act for a particular moral theory as the one that is prescribed by the highest principle of subjective rightness relative to that theory. The order of principles within the hierarchy is a substantive issue, and not one that Smith goes into. The agent must act on the highest usable principle. If an agent cannot use a principle because she does not have the relevant beliefs (or does not know that she has), then she should have a look at the next one down the hierarchy, and so on until she finds one that she can use. She will never be in a position where she cannot find a usable guide to action. However, this is unsatisfactory. First, it is not at all clear that it will be possible to specify a hierarchy of principles in a way that is not just ad hoc. Any particular decision procedures, such as maxi-min (maximizing the value of the worst possible outcome) or maximizing expected utility are controversial, and whether they are the appropriate response to uncertainty will vary according to the facts of the case. It seems that it would be very hard for this to be specified in a quite general moral theory. Second, just as it is possible that an agent could lack knowledge necessary to apply a principle, she could lack knowledge of what the principles are, or what the order in the hierarchy is. The prospects for an immediately helpful instruction that always applies are bleak. There is simply too much variation in the circumstances
(^14) This relates to the debate about blameless wrongdoing, and the issue of which comparison class is relevant to assessing the rightness of an act. See Parfit (1984); Feldman (1986); Mason (2002).
124 Elinor Mason
be guided by the actualist ought. Paradoxically, we ought not to do what we actualism-ought to do. Holly Smith has produced an example that echoes the Regan/Jackson choice situation (Smith 2011). An agent has to choose between three options, where she knows that one of A or B is the prospectively right action, but the other is terrible, and she knows that C is not great but certainly acceptable. Obviously, the agent should choose the course of action that she knows is not prospectively right: she should choose the safe option. So we ought not to do what we prospectivism-ought to do. Smith’s argument shows that the subjectively right act is not simply ‘the act most likely to be the prospectively right act’. This, of course, is parallel to the point of the original Regan/Jackson example, which is that the prospectively best act is not ‘the act most likely to be the actualism- right act’. Both arguments illustrate the important point that the various ‘ought’s are not hierarchical. But they show something else too. They show that there is no general requirement that we try to do what is ‘right’ in either of the objective senses of rightness. This may sound odd, but it is not. Actualist and prospective right- ness are ways of talking about the ideal thing to do, and it is very useful (for moral learning, for advice, and so on) to have a sense like that. But these senses of rightness are essentially subjunctive. They are not actually instructions. Morality does not say, ‘do what is actualism-right!’, or ‘do what is prospectivism-right!’ Rather, it says something subjunctive: ‘this is what it would be best to do’, or ‘this is what a reasonable person would do’. Subjective obligation, by contrast, involves a direct instruc- tion. That is the sense in which it is action guiding. The subjective ‘ought’ says, ‘you ought to do φ!’, and that is more than just a way of saying, ‘φ would be good’. Ross and Prichard’s account of subjective obligation is action guiding in the accessibility sense. As Ross puts it, the agent ought to set himself to do what “he thinks to be morally most suitable in the circumstances as he takes them to be” (1939, 161). This is certainly accessible to the agent, because an agent can always introspect and come up with what she takes to be morally most suitable, so long as we allow a fairly weak reading of ‘most suitable’ such that it might involve plumping between apparently equally good options, as I stressed earlier. If the agent has no idea at all about what is morally most suitable, she counts as doing what she takes to be morally most suitable when she plumps between all the options.
126 Elinor Mason
In sum, if we appeal to something internal to an agent, we are giving an instruction that is accessible. It is not action guiding in the sense that we might have hoped for action guidance: it doesn’t actually give us any help in figuring out what to do. It is too general for that. But, as I have argued, moral theory cannot provide immediately helpful instructions.
4. Formulating the subjective ‘ought’: beliefs and tryings
So one might now think that an agent’s subjective obligation is to do what she believes is most morally suitable. I will refer to this as the belief formulation of subjective obligation. At first glance, it seems that the belief formulation obviously meets the accessibility requirement. But it is not so obvious on reflection. I sometimes have beliefs that are not accessible to me (this is often the case with implicit bias and related phenomena, where I may have beliefs I do not know about, or may think I have beliefs that I do not really have). It is possible that our relevant beliefs about the moral situation are not accessible. 16 Our beliefs can be buried in all sorts of ways. The worry here is not about tacit beliefs, or about beliefs that are not before the mind at any particular moment. Many of our beliefs are tacit, in that we have never consciously formulated them, but would assent to them if asked. As Dennett pointed out, it comes as no surprise to us that zebras do not wear overcoats in the wild (1978, 104). There are also beliefs that are not occurrent (not ‘before the mind’) but that we have stored somewhere in our mind. Again, such beliefs are not usually inaccessible in the relevant sense, though we may on occasion have trouble recalling them (‘what is this person’s name?’). Rather, the worry is about beliefs that we have unconsciously , that we do not have any voluntary access to. As I say, our implicit biases might be thought of as unconscious beliefs, though this is controversial (biases seem to involve both cognitive and affective elements, for one thing). Another sort of example involves motivated burying of our beliefs: we sometimes engage in self-deception and motivated ignorance. And per- haps there are cases where for no particular reason a belief gets buried—we
(^16) Harry Gensler discusses problems for a belief-based account of subjective obligation in Gensler (1987). Holly Smith raises some of these problems in her account of subjective obligation (2010).
Do the Right Thing 127
prospectively ought to do is relative to her evidence. Yet, when the ignorance is culpable we want to be able to criticize her for that (of course it may nonetheless be the case that she has to proceed on the basis of that ignorance). Jackson argues that culpable ignorance can be made sense of in terms of actual beliefs (1991, 464). Jackson says that one of the options is always to get more information, and that getting more information has an expected utility just as any other option does. So if the agent forms a belief without getting the relevant information, in a case where the expected utility of getting more information is high enough to outweigh the costs, then she has acted wrongly. Jackson’s conclusion is obviously right, but his reasoning illustrates an important point that Jackson does not acknowledge. The aim that guides whether or not you should seek more information is not itself an epistemic aim. In thinking about whether or not I should seek more information I must think about my practical aim. Imagine that I must decide on where to invest my money. If I am only concerned to make as much money as possible, the only thing I need to know about is likely returns. If, on the other hand, I am concerned to be moral, I should find out more about the investment strategies of the various companies. But these are practical reasons, pointing in different directions according to the practical aim. There can be no purely epistemic reason to seek more information. Richard Feldman makes this point when he argues that epistemic justification concerns only the relationship between actual possessed evidence and belief (2004). As Feldman says, there may be further reasons to look for more evidence, but they are not epistemic reasons. The same applies to beliefs that are beliefs about value. An agent might believe that P would be morally good, but in order to judge whether she should seek more information about the situation, we have to know what her practical aim is. An agent with the perverse aim of doing as little good as possible will have different reasons to seek more information than an agent who has the aim of being good. Thus formulating subjective obligation in terms of beliefs alone does not capture the requirement to seek more information. 19 We have to add the agent’s goal to the picture.
(^19) H. A. Prichard (1932) recognizes this issue and discusses it on pp. 90–1. Prichard’s account of the issue is not completely clear. Julia Driver criticizes Frank Jackson on the same point (Driver, 2012, 115–16).
Do the Right Thing 129
It might be suggested that once we acknowledge the goal of behaving morally permissibly, we can still formulate subjective obligation in terms of beliefs. Either an agent has a belief (perhaps a suspicion) that her other beliefs are flawed, or she does not. In the former case, she has all it takes to know that she should find out more, and so her subjective obligation is to find out more. If the latter, if she has no suspicion that her beliefs are flawed, then how could it be the case that she subjectively ought to seek more evidence? Subjective obligation is supposed to be accessible. But if she has no suspicion that her beliefs are flawed, then she has no subjective reason to seek more evidence. But this misses something out. What we ought to be doing is much more general and stretches over time. We ought to have a general attitude of alertness, a readiness to examine our beliefs, a disposition to double check our evidence in the light of our aim. This cannot be expressed in terms of beliefs, which are just one sort of mental state, and at any one moment a belief set may lack a relevant belief about possible flaws in the current belief set. Yet having this attitude of alertness seems an essential part of what it would be subjectively right to do. So formulating subjective obligation in terms of beliefs is too narrow. We need some- thing that captures our ongoing duty to weigh the costs and benefits of seeking more evidence, and to be generally alert and sensitive to moral features of our environment. I propose that we can capture the essential requirements by formu- lating the subjective ‘ought’ in terms of trying.^20 I am using the concept of ‘trying’ in a broad sense, so that it applies to doings as well as ‘setting oneself ’. Sometimes trying to bring about a morally acceptable state of affairs just is figuring out what would happen, seeking more information, prescribing a drug, and so on. The sort of trying that I am interested in crucially involves a goal: a standing intention to be alert to the goal, and taking steps to achieve the goal as appropriate. But trying is not neces- sarily intentional in the moment; it does not necessarily involve con- scious pursuit of the goal at any particular moment; it does not involve believing that you will certainly succeed, and it does not necessarily
(^20) I argue for formulating subjective obligation in terms of ‘trying’ in my 2003. Andrew Sepielli (2012) also formulates subjective ought in terms of trying, and though his approach to the question is rather different to mine I think we end up in the same place for more or less complementary reasons.
130 Elinor Mason
for or against the value to be realized in the future. This is the argument that I gave above, that we need an aim to regulate how and when to seek more evidence, because we need something to balance the costs of seeking more evidence against. Thus, on my view, trying to do well by morality involves weighing the costs of seeking more evidence against the values that might be achieved. Smith throws a spanner in the works by suggesting that in the case of deontological principles, an account of subjective obligation cannot make sense of the duty to seek more information. First, Smith introduces the very useful notion of deontic value, which aggregates the value of fulfilling different duties, so that in thinking about whether to seek information, we should think about the comparative weights of the different duties that might be fulfilled. Smith argues that although we have an objective duty to seek more information when doing so would maximize deontic value (i.e. it would be best to do so), we cannot make sense of a subjective duty to do that. Smith argues by analogy with making promises and then keeping them, that there is no deontic value in fulfilling created duties. She imagines an agent, Devon, who has a choice between making a less weighty and a more weighty promise (2014, 27–8). Smith argues convincingly that although Devon will have a duty to fulfill either promise if he makes it, and although breaking the less weighty promise would be less bad than breaking the more weighty promise (and thus more deontic value would be created by making the weightier promise), Devon has no duty to make the weightier promise. So, Smith concludes, there is no deontic value in fulfilling duties that we create. From there, she concludes that as subjective obligations are created by seeking information, there is no value in fulfilling them, and hence (often) no reason to seek more information. The deep issue here is over the extent to which subjective obligation involves optional creating rather than discovering what ought to be done. Smith says, “according to a subjective deontic moral code, gathering information or failing to do so can create subsequent duties for the agent because gathering information produces belief-states that generate new duties. These created duties must be handled in the same way that the duties created by making promises are: fulfilling such prima facie duties must be accorded no positive value, even though violating these duties is accorded negative value” (2014, 29). But although promises are a
132 Elinor Mason
sometimes optional way of creating duties, subjective obligation does not involve the optional creation of duties. Rather, in finding out about the world, we find out what doing well by morality would involve. We decide whether to find out about the world by thinking about what moral issues might arise, might already be lurking. If it seems that there are morally relevant issues, then we should seek that information, and in doing so, we discover rather than create our duties. This might make it sound as if subjective rightness is a matter of ‘trying to do what is right’. In which case it seems that the question arises, ‘what sort of rightness should the agent be aiming for?’ Clearly it can’t be subjective rightness, because that would be circular. So it must be actualist or prospective rightness. But that would make subjective rightness dependent on those other notions. Subjective rightness is not a matter of trying to do what is right. As I argued earlier, subjective rightness is not parasitic on prospective or actualist rightness. The agent should not be aiming to do any particular set of acts, defined in terms of what ideal agents or semi-ideal agents would do: that might be counterproductive, as Regan/Jackson style examples, show.^22 There is nothing special about the actualism-right or prospectively right acts from a practical point of view. As I said, they are subjunctive accounts of rightness, they pick out what an idealized agent would do. Rather, the subjective ‘ought’ tells the agent to try to do well by the right and good making features that are in the world. This does not appeal to any prior account of rightness. It does not tell the agent to aim for the balance of value and risk that the reasonable agent would aim for. Instead, it tells her to try to balance value and risk as best she can. Obviously, if she is sensible, she will not aim for the best possible outcome if that is very risky. Equally, she will not aim for the prospect- ively best option if that would be risky.^23 She just has to balance the
(^22) Often, doing what is prospectively right and what is subjectively right coincide, but this is not necessarily so. Sometimes we do not know which action is prospectively right. So we might try to do the action that is prospectively right. But this is not quite right. Trying to do the action that is prospectively right fetishizes the prospectively right action in cases like Smith’s, where we do not know which that action is and risk something very bad. (^23) This refers to Smith’s argument, discussed earlier: the point is that the prospectively best act is the one the reasonable agent would do, and that is not necessarily accessible to the (not necessarily completely reasonable) agent.
Do the Right Thing 133
5. Conclusion
I have argued for an account of subjective obligation. I argued that subjective rightness is action guiding only in a weak sense. We cannot give a general account of subjective obligation, that would apply across all circumstances, and would also be a genuinely helpful instruction to an agent. However, subjective obligation should be action guiding in a different but important sense: the subjective ‘ought’ is a genuine impera- tive: it tells the agent what to do, rather than what it would be ideal, or good to do. It must be accessible to the agent, she must be able to follow the instruction. Thus subjective obligation must be formulated in terms of some- thing in the agent’s own psychology. Rather than specifying a target, as actualist and prospective forms of rightness do, subjective rightness must tell the agent to do something she can identify. Usually, sub- jective obligation is formulated in terms of the agent’s beliefs: an agent should do what she believes is morally appropriate in the circum- stances. I argue against the belief-based account of subjective obliga- tion, on the grounds that it cannot make sense of the agent’s continuing obligation to seek information and be alert to the possi- bility that her beliefs are flawed. On my account, an agent’s subjective obligation is to try to do well by morality. Putting subjective obliga- tion in terms of trying makes sense of the dynamic and diachronic nature of subjective obligation. Our subjective obligation cannot be captured in a snapshot, ‘do what seems best right now’. Rather, our subjective obligation is to be committed to morality, to have a general attitude of alertness and awareness to relevant changes in the situation to red flags, to opportunities. Subjective obligation is not a set of atomic duties, it is a commitment. Let’s return finally to Mookie. Da Mayor tells him, “Always do the right thing.” Da Mayor is not telling Mookie to do what is objectively or prospectively right. Rather, Da Mayor is reminding Mookie to try to do well by morality. At the end of the movie the police kill an unarmed Black man, and Mookie picks up a trash can and throws it through a window, starting a riot. The point of the movie (arguably) is that Mookie is trying to do well by morality in doing that. He cannot let another racist killing by the police stand. He is not thinking it all through, or calculating utilities, when he picks up that trash can. He is
Do the Right Thing 135
acting out of moral outrage. He is following Da Mayor’s advice, which was perfectly good advice (though not perfectly formulated). 25
Bibliography
Bales, Eugene (1971). ‘Act-Utilitarianism: Account of Right-Making Characteristics or Decision-Making Procedure?’, American Philosophical Quarterly 8: 257–65. Bentham, Jeremy (1789). An Introduction to the Principles of Morals and Legislation. Oxford: Clarendon Press. Dennett, Daniel (1978). Brainstorms. Cambridge, MA: MIT Press. Driver, Julia (2001). Uneasy Virtue. New York: Cambridge University Press. Driver, Julia (2012). Consequentialism. New York: Routledge. Feldman, Fred (1986). Doing the Best We Can. Dordrecht: Reidel. Feldman, Fred (2006). ‘Actual Utility, the Objection from Impracticality, and the Move to Expected Utility’, Philosophical Studies 129: 49–79. Feldman, Fred (2012). ‘True and Useful: On the Structure of a Two Level Normative Theory’, Utilitas 24: 151–71. Feldman, Richard (2004). ‘The Ethics of Belief ’. In E. Conee and R. Feldman (eds.), Evidentialism. Oxford: Oxford University Press, 166–95. Gensler, Harry J. (1987). ‘Paradoxes of Subjective Obligation’, Metaphilosophy 18 (3–4): 208 – 13. Graham, Peter A. (2010). ‘In Defense of Objectivism about Moral Obligation’, Ethics 121: 88 – 115. Gruzalski, Bart (1981). ‘Foreseeable Consequence Utilitarianism’, Australasian Journal of Philosophy 59: 163–76. Howard-Snyder, Frances (1997). ‘The Rejection of Objective Consequentialism’, Utilitas 9(2): 241 – 8. Howard-Snyder, Frances (2005). ‘It’s the Thought that Counts’, Utilitas 17(3): 265–81. Hudson, James (1989). ‘Subjectivization in Ethics’, American Philosophical Quarterly 26: 221 – 9. Jackson, Frank (1986). ‘A Probabilistic Approach to Moral Responsibility’, Studies in Logic and the Foundations of Mathematics 114: 351–65. Jackson, Frank (1991). ‘Decision-Theoretic Consequentialism and the Nearest and Dearest Objection’, Ethics 101(3): 461–82. Lee, Spike (1989). Do the Right Thing. Universal City Studios, Inc. DVD. Lockhart, Ted (2000). Moral Uncertainty and Its Consequences. Oxford: Oxford University Press. McConnell, Terrance (1988). ‘Ross on Duty and Ignorance’, History of Philosophy Quarterly 5(1): 79–95.
(^25) I am grateful to audiences at the 2014 Bled Philosophy conference, the 2016 Arizona Workshop on Normative Ethics, and the Rutgers Value Theory Group where earlier versions of this paper were presented. A much earlier version was presented at the Hebrew University of Jerusalem Conference on Uncertainty, in 2011, and I owe thanks to the organizers and participants. I am also very grateful to Fred Feldman and Stewart Cohen for discussions on these topics over the course of several years. Thanks are also due to two anonymous referees for this volume.
136 Elinor Mason
