philosophy #2
Page 183
Chapter 10— RuleConsequentialism
Brad Hooker
1— Introduction
Just what is the connection between moral rightness and consequences? For nearly half a century now, consequentialists have divided themselves into different camps with respect to this question. Actconsequentialists believe that the moral rightness of an act depends entirely on whether the act's consequences are at least as good as that of any alternative act. Ruleconsequentialists believe that the rightness of an act depends not on its own consequences, but rather on the consequences of a code of rules. (For prominent presentations of ruleconsequentialism, see Berkeley 1712; Austin 1832; Urmson 1953; Rawls 1955; Brandt 1959, 1967, 1979, 1988, 1989, 1996; and Harsanyi 1982.) This essay explores the prospects for ruleconsequentialism.
2— What Constitutes Benefit?
Ruleconsequentialism. holds that any code of rules is to be evaluated in terms of how much good could reasonably be expected to result from the code. By "good' here I mean whatever has noninstrumental value. What has noninstrumental value?
Utilitarians, who have been the most prominent kind of consequentialists, believe that the only thing with noninstrumental value is utility. All utilitarians have held that pleasure and the absence of pain are at least a large part of utility. Indeed, utilitarianism is often said to maintain that pleasure and the absence of pain are the only things that matter noninstrumentally. Certainly, this was the official view of the classic utilitarians Jeremy Bentham (1789), J. S. Mill (1861), and Henry Sidgwick (1874) — though in Sidgwick's case, equality seems to have independent weight as a tie breaker (Sidgwick 1907 [1874]: 417).Cop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r
us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 184
Perhaps more common over the last thirty years has been the view that utility is constituted by the fulfillment of people's desires, even if these desires are for things other than pleasure. Many people, even when fully informed and thinking carefully, persistently want things in addition to pleasure. They care, for example, about knowing important truths, about achieving valuable goals, about having deep personal relationships, about living their lives in broad accordance with their own choices rather than always in accordance with someone else's (Griffin 1986, pt. 1; Crisp 1997, chs 2, 3). The pleasure these things can bring is of course important. Still, human beings can care about these things in themselves, i.e., in addition to whatever pleasure they bring.
This view, however, can be challenged. Some desires seem to be about things too unconnected with you for them to play a direct role in determining your good. Would your desiring that a stranger recovers fully from her illness make her recovery good for you, even if you never see or hear from her again (Parfit 1984: 494)? Naturally, the fulfillment of such a desire would indirectly benefit you if it brought you pleasure or peace of mind. But this is not to say that the fulfillment of your desire that the stranger recovers herself constitutes a benefit to you. Rather if you get pleasure or peace of mind from the fulfillment of this desire, this pleasure or peace of mind constitutes a benefit to you (since you doubtless also desire pleasure and peace of mind for yourself).
The view that the fulfillment of your desires itself constitutes a benefit to you — if this view is to be at all plausible — will have to limit the desires in question. The only desires the fulfillment of which constitutes a benefit to you are your desires for states of affairs in which you are an essential constituent (Overvold 1980, 1982; Hooker 1991). You are not an essential constituent of the state of affairs in which this stranger recovers. So her recovery doesn't itself constitute a benefit to you.
There seem to be reasons for further restrictions on the desires directly relevant to personal good. Think how bizarre desires can be. When we encounter particularly bizarre ones, we might begin to wonder whether the things are good simply because they are desired. Would my desiring to count all the blades of grass in the lawns on the street make this good for me (Rawls 1971: 432; cf. Parfit 1984: 500; Crisp 1997: 56ff)? Whatever pleasure I get from the activity would be good for me. But it seems that the desirefulfillment as such is worthless in this case. Intuitively, the fulfillment of my desires constitutes a benefit to me only if these desires are for the right things (Finnis 1980, 1983; Parfit 1984, Appendix I; Hurka 1987, 1993; Brink 1989: 221–36; Scanlon 1993; Griffin 1996, ch. 2; Crisp 1997, ch. 3). Indeed, some things seem to be desired because they are perceived as valuable, not valuable merely because desired or pleasant (Brink 1989: 230–1, especially fn. 9).
Views holding that something benefits a person if and only if it increases the person's pleasure or desirefulfillment are in a sense "subjectivist' theories of personal good. For these theories make something's status as a benefit depend always on the person's subjective mental states. "Objectivist" theories claim that the contribution to personal good made by such things as important knowledge, imporCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r
us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 185
tant achievement, friendship, and autonomy is not exhausted by the extent to which these things bring people pleasure or fulfil their desires. These things can constitute benefits even when they don't increase pleasure. Likewise, they can constitute benefits even when they are not the objects of desire. Objectivist theories will typically add that pleasure is of course an objective good. These theories will also hold that ignorance, failure, friendlessness, servitude, and pain constitute harms.
For the most part, I will be neutral in this essay about which theory of personal good is best. Usually what gives people pleasure or enjoyment is also what satisfies their desires and involves the objective goods that could plausibly be listed. So usually we don't need to decide as among these theories of personal good.
But not always. Suppose the ruling elite believed that quantity of pleasure were all that matters. Then (to take a familiar leaf from Nineteen EightyFour) they might feel justified in manipulating the people and even giving them drugs that induce contentment but drain ambition and curiosity, if they thought such practices would maximize aggregate pleasure. Or suppose the ruling elite believed that the fulfillment of desire were all that matters. Again, the ruling elite might feel justified in manipulating the formation of preferences and development of desires such that these are easily satisfied. Now we can accept that — to some extent — our desires should be modified so that there is some reasonable hope of fulfilling them. But this could be pushed too far either in the name of maximizing pleasure or in the name of maximizing desirefulfillment. A life could be maximally pleasurable, have maximum desirefulfillment, and still be empty — if it lacked desires for friendship, achievement, knowledge, and autonomy.
3— Distribution
The term ''ruleutilitarianism" is usually used to refer to theories that evaluate acts in terms of rules selected for their utility — i.e., for their effects on social wellbeing. The term "ruleconsequentialism" is usually used to refer to a broader class of theories of which ruleutilitarian theories are a subclass. Ruleconsequentialist theories evaluate acts in terms of rules selected for their good consequences. Nonutilitarian versions of ruleconsequentialism say the consequences that matter are not limited to net effects on overall wellbeing. Most prominently, some versions of ruleconsequentialism say that what matters are not only how much wellbeing results but also how it is distributed, in particular the fairness of alternative distributions. 1 Table 10.1 might prove helpful.
Which version of ruleconsequentialism is best? The problem with ruleutilitarianism is that it has the potential to be unfairly inegalitarian. (For discussion of some utilitarian replies to this objection about distribution, see Hooker 1995: 30.) Consider a set of rules which leaves each member of a smaller group very badly off, and each member of a much larger group very well off (table 10.2).Cop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 186
Table 10.2
Wellbeing
First Code: per person per group for both groups
10,000 people in group A 1 10,000 100,000 people in group B 10 1,000,000
1,010,000
Now if no alternative rule would provide greater net aggregate benefit, then utilitarians would endorse this code. Yet suppose the next best rule from the point of view of utility would be one with the results set out in table 10.3.
Table 10.3
Wellbeing
Second Code: per person per group for both groups
10,000 people in group A 8 80,000 100,000 people in group B 9 900,000
980,000
Let us assume that the first code leaves the people in group A less well off for some reason other than that these people opted to work less hard or imprudently took bad risks. In that case, the second code seems morally superior to, because fairer than, the first code. This is why we should reject ruleutilitarianism in favor ofCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t
la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 187
a distributionsensitive ruleconsequentialism that considers fairness as well as wellbeing.
What are the relative weights given to wellbeing and fairness by this distributionsensitive ruleconsequentialism? Clearly, wellbeing does not have overriding weight. For there can be cases in which the amount of aggregate net benefit produced would not justify rules that were unfair to some group. That was what my schematic example above was meant to show.
Does fairness have overriding weight? This is particularly unsettled territory, since even what constitutes fairness is unclear. Nevertheless, we cannot rule out the possibility that some unfair practice so greatly increases overall wellbeing that the practice is justified. But it is certainly unclear where the threshold is for fairness to trump wellbeing. Perhaps the best we can say is that, in the choice between codes, judgement will be needed in balancing fairness against wellbeing. By evaluating rules in terms of two values (wellbeing and fairness) instead of one (wellbeing), distributionsensitive ruleconsequentialism. is messier than ruleutilitarianism. Still, this seems to be a case where the more plausible theory is the messier one.
4— Criteria of Rightness versus Decision Procedures
Ruleconsequentialism is often portrayed as merely part of a broader consequentialist theory. This broader theory evaluates all things by their consequences. So it evaluates the desirability of acts by their consequences, the desirability of rules by their consequences, etc. The standard point to make along these lines is that, even if the rightness of an act depends on its consequences, better consequences will result if people do not try always to decide what to do by calculating consequences than if they try always to decide in this way. In other words, consequentialists can and should deny that
On every occasion, an agent should decide which act to do by ascertaining which act has the greatest expected good.
Consequentialists agree that our decision procedure for daytoday moral thinking should instead be as follows:
At least normally, an agent should decide how to act by referring to tried and true rules, such as "Don't harm others", "Don't steal", "Keep your promises", "Tell the truth", etc.
Why? First, we frequently lack information about the probable consequences of various acts we might do. Where we cannot even estimate the consequences, we can hardly choose on the basis of maximizing the good. Second, we often do not have the time to collect this information. Third, human limitations and biases are suchCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 188
that we are not accurate calculators of the expected overall consequences of our alternatives. For example, most of us are biased in such a way that we tend to underestimate the harm to others of acts that would benefit us.
Now if there will be greater overall good where people are largely disposed to focus and act on nonconsequentialist considerations, then consequentialism itself endorses such dispositions. 2 So consequentialists advocate firm dispositions to follow certain rules, including firm dispositions not to harm others, not to steal, not to break promises, etc. Different consequentialists thus by and large agree about how people should do their daytoday moral thinking.
What different kinds of consequentialists disagree about is what makes an act morally permissible, i.e., about the criterion for moral rightness.
Actconsequentialism claims that an act is morally right (both permissible and required) if and only if the actual (or expected) good produced by that particular act would be at least as great as that of any other act open to the agent.
In contrast,
Ruleconsequentialism claims that an act is permissible if and only if it is allowed by a code that could reasonably be expected to result in as much good as could reasonably be expected to result from any other identifiable code.3
The distinction between actconsequentialism's criterion of rightness and the dispositions it favors is important in many ways. It is important if we want to know what actconsequentialism wants from us. It is also important if actconsequentialism had better not conflict too sharply with our intuitive moral reactions. For if act consequentialism. claimed that we should always be focused on and motivated by calculations of what would maximize the good impartially conceived, many philosophers have thought it would be ridiculous. But the idea that actconsequentialism must make this ridiculous prescription is undermined by the distinction between actconsequentialism's criterion of rightness and the decision procedures it favors.
Nevertheless, the distinction is powerless to protect actconsequentialism from other objections. True, actconsequentialism's implications about focus and motivation are not as counterintuitive as might initially be thought. But this is irrelevant to objections about actconsequentialism's criterion of rightness.
5— Formulations of RuleConsequentialism
We need to augment our formulation of ruleconsequentialism. All recognizable forms of ruleconsequentialism make moral rightness depend on rules which are evaluated in terms of their consequences. But different forms of ruleconsequentialism disagree about the conditions under which rules are to be evaluated. For instance,Cop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er
U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 189
one version of ruleconsequentialism is formulated in terms of the rules the compliance with which would be optimific. Another version is formulated in terms of rules the acceptance of which would produce the most good. Should ruleconsequentialism be formulated in terms of compliance or in terms of acceptance?
Although compliance with the right rules is the first priority, it isn't the only thing of importance. We also care about people's having moral concerns. So we had better consider the costs of securing not only compliance but also adequate moral motivation. From a ruleconsequentialist point of view, "moral motivation" means acceptance of moral rules. By "acceptance of moral rules", I mean a disposition to comply with them, dispositions to feel guilt when one breaks them and to resent others' breaking them, and a belief that the rules and these dispositions are justified (Brandt 1967 §8 [1992: 120–1], 1979: 164–76, 1996: 67, 69, 145, 156, 201, 266–8, 289).
The focus on acceptance of rules, i.e., dispositions, is crucial because the acceptance of a rule — or perhaps at this point it would be better to say the internalization of a rule — can have consequences over and above compliance with the rule (Lyons 1965: 137ff; Williams 1973: 119–20, 122, 129–30; Adams 1976, especially p. 470; Blackburn 1985: 21 n. 12).
The most obvious example of this involves rules that deter perfectly. Suppose you accept a rule prescribing that you retaliate against attackers. Suppose also that you are totally transparent, in the sense that people can see exactly what your dispositions are. So everyone knows about your disposition to retaliate, and therefore never attacks you. Thus, your accepting that rule is so successful at deterring attack that you never have an opportunity to comply with the rule. Your accepting the rule thus obviously has important consequences that simply cannot come from your acting on the rule, since you in fact never do (Kavka 1978).
Now suppose everyone internalized rules such as "Don't kill except when killing will maximize the aggregate good," "Don't steal except when stealing will maximize the aggregate good," "Don't break your promises except when breaking them will maximize the good3," etc. Presumably, if everyone had internalized these rules, sooner or later awareness of this would become widespread. And people's becoming aware of this would undermine their ability to rely confidently on others to behave in agreedupon ways. Trust would break down. The consequences would be terrible. And these terrible consequences would result, not from individual acts of complying with these rules, but from public awareness that the rules's exception clauses — the ones prescribing killing, stealing, and so on when such acts would maximize the good — were too available (Brandt 1979: 271–7; Harsanyi 1982: 56–61, 1993: 116–18; Johnson 1991, especially chs 3, 4, 9).
I am aware that there has been some controversy over the argument just outlined. But there is another way in which a costbenefit analysis of internalization is richer than a costbenefit analysis of compliance. Getting one code of rules internalized might involve greater costs than getting another code internalized. These costs are immensely important. For example, one possible objection to a code might be that it is so complicated or calls for so much selfsacrifice that too much of humaniCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 190
ty's resources would have to be devoted to getting it widely internalized. The internalization costs would be so high that internalizing this code would not, on balance, be optimal. When this is the case, ruleconsequentialists hold that the code isn't justified, and complying with it isn't required.
These points about internalization costs beg to be deployed at a number of places in the discussion of ruleconsequentialism. One such place I explore in the next section.
6— Collapse
If we formulate ruleconsequentialism in terms of compliance, we risk having ruleconsequentialism collapse into actconsequentialism. The objection that rule consequentialism collapses into extensional equivalence with actconsequentialism assumes rules are to be evaluated in terms of only the effects of compliance. While compliance can be one effect of internalizing rules, we have seen that there are also other effects. We must consider not only the benefits of compliance but also the other effects of rule internalization. With these effects factored into the evaluation of rules, the costbenefit analysis will not favor rules extensionally equivalent to act consequentialism.
One version of the objection that ruleconsequentialism collapses into actconsequentialism claims that ruleconsequentialism must favor just the one simple rule that one must always do what will maximize the good (Smart 1973: 11–12). The objection assumes that, if each person successfully complies with a rule requiring the maximization of the good, then the good would be maximized. That the good would be maximized under these conditions has been challenged (see Hodgson 1967, ch. 2; Regan 1980, ch. 5). But whether or not everyone's complying with the actconsequentialist principle would maximize the good, we should again consider the wider costs and benefits of rule internalization. The impartial good would not in fact be maximized by the internalization of just this one actconsequentialist rule. To internalize just the one actconsequentialist rule is to have just one moral disposition, the disposition to try to comply with actconsequentialism. To have just this one moral disposition is to have actconsequentialism as one's moral decision procedure. But we've already seen why actconsequentialism is not a good decision procedure.
In addition, the costs of getting a disposition to try to comply with actconsequentialism internalized would be extremely high. For getting that one rule internalized amounts to getting people to be disposed always to do what would be impartially best. Such a disposition would have to overcome people's immensely powerful natural biases towards themselves and their loved ones. To be sure, there are great benefits to be gained from getting people to care about others, and to be willing to make sacrifices for strangers. But think how much time, energy, attention, and psychological conflict that would be required to get people to internalizeCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 191
an overriding completely impartial altruism (if this is even possible at all). The costs of trying to make humans into saints would be too great.
That may seem like a paradoxical thing to assert. Wouldn't a world full of people each with an overriding disposition to maximize the impartial good be so ideal as to be worth any costs of getting from here to there? I think not. Bear in mind that the costs would hardly be a onceandforalltime sacrifice. Rather, getting this overriding impartiality internalized would have to be done for every new generation. We are contemplating here a radical reshaping of something deep in human nature. It is not as if the impartiality internalized by one generation will be reflected in the genes of their children. Rather, there will be the high cost of getting the overriding impartiality internalized in their children, just as there was when the parents were children themselves. (I am ignoring here the possibility of genetic engineering to create more altruistic humans.) The internalization costs will be incurred for each new generation of humans.
I have been arguing as if getting overriding impartiality internalized by the vast majority is a serious possibility, though one with prohibitive transition costs. It may not, however, be a serious possibility. In any case, the only realistic way to make humans totally and always impartial would be to reduce their special concern for themselves and those with whom they have special attachments. What would be left might be merely a life of insipid impartiality, devoid of deep personal attachments and inimical to great enthusiasm and joy. Strong concern and commitment focused on particular projects and individuals play an ineliminable role in a rewarding human life. But these features would have to be eliminated if human beings are to internalize an overriding motivation to maximize the impartial good. (This paragraph borrows heavily from Sidgwick 1907: 434; Williams 1973: 129–31; Adams 1976; Parfit 1984: 27–8, 30, 32–5; Dworkin 1986: 215; Griffin 1986, chs 2, 4, 1996: 77, 104; Crisp 1997: 106.)
So in the light of the transition and permanent costs of getting internalized an overriding impartiality, I hold that there must be some point short of this where the costs of going further outweigh the benefits. Remember why this matters here. Getting internalized an overriding impartiality would be part of getting internalized an overriding disposition to do what will maximize the impartial good. So if there is a compelling ruleconsequentialist reason against getting internalized an overriding impartiality, there is a compelling ruleconsequentialist reason against getting internalized an overriding disposition to do what will maximize the impartial good. I have just argued that there is a compelling ruleconsequentialist reason against getting internalized an overriding impartiality. Such a disposition would not find favor with rule consequentialism. So there is a compelling ruleconsequentialist reason against getting internalized an overriding disposition to do what will maximize the impartial good. This kills the first way of developing the collapse objection.
The other way of developing the collapse objection starts by admitting that internalization of just the one actconsequentialist rule would lead to bad consequences. But this way of developing the collapse objection maintains that utility could beCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 192
gained from the provision of specific exception clauses to moral rules against harming others, breaking promises, etc. If this is right, then ruleconsequentialists are forced by their own criterion for rule selection to embrace rules with these exception clauses. The same sort of reasoning will militate in favor of adding specific exceptions aimed at each situation in which following some rule would not bring about the best consequences. Once all the exception clauses are added, rule consequentialism will have the same implications for action that actconsequentialism has. This would be a fatal collapse.
To this way of developing the collapse objection, ruleconsequentialists will reply by returning to the points about trust and expectations that I alluded to earlier. How much confidence would you have in others if you knew they accepted such highly qualified rules? How much mutual trust would there be in a society of agents who accepted endless exceptions to rules against harming others, breaking promises, lying, etc.?
Furthermore, the point about internalization costs is again relevant. The more plentiful and more complicated the rules to be learned, the higher the costs of learning them would be. At some point the costs of having to learn more rules, or more complications, would outweigh the benefits. Hence, the rules whose teaching and internalization would have the best results are limited in number and complexity. These limitations will keep the code from being extensionally equivalent with act consequentialism. So this kind of ruleconsequentialism does not collapse into actconsequentialism.
7— RuleConsequentialism and the Distribution of Acceptance
A relatively simple form of rule consequentialism selects rules by their consequences given internalization of them by 100 percent of the population. But I think the theory should be formulated in terms of internalization by less than 100 percent of the population. Ruleconsequentialism needs to be formulated this way in order to make room for rules about what to do when others have no moral conscience at all. Let us refer to such people as unmitigated amoralists.
Suppose we assume internalization of the rules by 100 percent of the population. We might still need rules for dealing with noncompliance, since internalization by 100 percent of the people does not guarantee 100 percent compliance. Some people might fully accept the best rules and yet sometimes, seduced by temptation, act wrongly. Thus there is need for rules dealing with noncompliance. These rules might specify, for example, what penalties apply for what crimes. They might also specify what to do when those around you accept that they should be helping to save others but aren't.
Contrast what is needed to deter or rehabilitate someone with a moral conscience too weak to ensure good behavior in some circumstances, with what is needed to deal with unmitigated amoralists (people who have no moral conscience at all). IfCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 193
we imagine a world with acceptance of the best code by 100 percent of the population, we have simply imagined unmitigated amoralists out of existence. Hence, we have imagined out of existence any ruleconsequentialist rationale for having rules for deterring and dealing with unmitigated amoralists.
Here is why. On the ruleconsequentialist view, there is always at least some cost associated with every additional rule added to the code. Every additional rule takes at least a little time to learn and at least a little memory to store. Then the question is whether there is some benefit from internalization of the rule that outweighs the cost. We can of course frame rules applying to nonexistent situations. For example, "be kind to any rational nonhumans living on the moon." But, if the situation envisaged really is nonexistent, where is the benefit of including such a rule in the code to be internalized? Presumably there are no benefits from such nevertobe applied rules. These rules, which have some costs and no benefits, fail a costbenefit analysis.
The reasoning seems to me to generate the following important conclusion. Ruleconsequentialism cannot generate or justify rules about how to deter murder, rape, robbery, fraud, etc. by unmitigated amoralists, unless ruleconsequentialism picks its rules with reference to an imagined world where there is internalization of the envisaged rules by less than 100 percent of the population. So ruleconsequentialism should evaluate rules in terms of internalization by less than 100 percent of the population.
But should we assume internalization by 99, or 90, or 80 percent, or even less? 4 Any precise number will of course be somewhat arbitrary, but we do have some relevant factors to consider. On one hand, we want a percentage close enough to 100 percent to hold on to the idea that moral rules are for acceptance by the whole society of human beings. On the other hand, we want a percentage far enough short of 100 percent to make salient the problems about noncompliance — such problems should not be thought of as incidental. Acknowledging that any one percentage will nevertheless be somewhat arbitrary, I propose we take internalization by 90 percent of people in each future generation as the condition under which rules have to be optimal. Let me just add that this distinction between the 90 percent who are moral and the 10 percent who are amoral is supposed to cut across all other distinctions, such as distinctions in nationality and financial status.
8— Arguments for RuleConsequentialism
One argument for ruleconsequentialism is that general internalization of ruleconsequentialism would actually maximize the impartial good. The idea is that from a purely consequentialist point of view ruleconsequentialism seems better than actconsequentialism and all other theories.
Many actconsequentialists reply by invoking their distinction between their criterion of rightness and the decision procedure for daytoday moral decisions. TheyCop yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 194
admit actconsequentialism is not a good procedure for agents to use when deciding what to do. But they think this does not invalidate actconsequentialism's criterion of rightness. They would add that, even if ruleconsequentialism is an optimal decision procedure, this would not entail that ruleconsequentialism correctly identifies what makes right acts right and wrong acts wrong.
Let us turn, then, to arguments for ruleconsequentialism other than the one that internalizing ruleconsequentialism would maximize the good. 5 Consider the moral code whose acceptance by society would be best, i.e., would maximize net good, impartially calculated. Shouldn't we try to follow that code? Isn't the code best for general adoption by the group of which we are members the one we should try to follow? These general thoughts about morality seem intuitively attractive and broadly ruleconsequentialist.
And consider the related question "What if everyone felt free to do what you're doing?" This question may in the end prove to be an inadequate test of moral rightness. But there is no denying its initial appeal. And there is no denying that ruleconsequentialism is an (at least initially) appealing interpretation of the test.
Ruleconsequentialism thus taps into and develops familiar and intuitively plausible ideas about morality. Morality is to be understood as a social code, a collective enterprise, something people are to pursue together. And the elements of this code are to be evaluated in terms of both fairness and the overall effects on the well being of individuals, impartially considered.
But ruleconsequentialism's leading rivals all likewise emerge from attractive general ideas about morality. For example, actutilitarianism can be seen as emerging from the idea that all that ultimately matters from the moral point of view is whether individuals are benefitted or harmed, that everything else is only instrumentally important (Scanlon 1982: 108). And actconsequentialism, the broader theory than actutilitarianism, can be seen as emerging from the intuition that it can't be wrong to do what produces the most good (Foot 1985).
Now consider moral contractualism, the theory that an act is right if and only if allowed by rules which could not be reasonably rejected by anyone motivated to find rules that no one with this same motivation could reject. Contractualism develops from the idea that morality consists of rules to which everyone would consent under appropriate conditions. This seems a very appealing general idea — moral rules grounded in reasonable agreement.
Consider yet another theory. The moral particularism of Jonathan Dancy (1981, 1983, 1993) builds on the idea that moral truth is found not in cold inflexible principles but rather through a finely tuned sensitivity to particular cases in all their rich complexity. Actually, to be distinct, moral particularism must go beyond the claim that there are some conflicts between competing moral considerations which are so difficult that agents would have to have fine moral sensitivity and judgement to resolve them correctly. To be distinct, moral particularism must be the view that what counts as a consideration at all can be decided only on a case by case basis. This is just how Dancy frames his theory: the very same consideration can countCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed
un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 195
morally in favor of doing an action in one case, and against in another, and there are few if any considerations that must always count on the same side morally. (Dancy points out that such properties as moral rightness itself do always count morally in favor of an act.)
Finally, as I understand what has come to be called virtue ethics, this approach grows from the thought that right and wrong actions can be understood only in terms of choices that a fully virtuous person would make. This thought then suggests that we take the nature of and rationale for the virtues as the primary focal points for our moral philosophy.
Thus all these moral theories — ruleconsequentialism, actutilitarianism, actconsequentialism, contractualism, particularism, virtue ethics — tap into familiar and intuitively attractive general ideas about morality, though different ones. So no one could claim that any one theory is the only one with this feature. The conclusion to draw from this is simple. The fact that a theory arises from and develops attractive general ideas about morality is hardly enough to show that it is superior to all its rivals.
Now among the questions we can go on to ask about competing moral theories are (1) whether they are coherent and develop from initially attractive ideas about morality, and (2) whether the claims they end up making about right and wrong in various circumstances are intuitively plausible. 6 I have already argued that rule consequentialism develops from attractive ideas about morality. But I shall not fully discuss here the objection that ruleconsequentialism incoherently claims that maximizing the good is the overarching goal and then that following certain rules can be right even when breaking them would produce more good.7 I admit that if we start from an overarching commitment to maximize overall good, then our ruleconsequentialism might be an incoherent account of moral rightness. But I propose our route to ruleconsequentialism starts elsewhere: we don't start from, and indeed don't have, an overarching commitment to maximize overall good. If I am right about that, then this objection falls apart (Hooker 1995: 27–9; and 1996: 538–9).
What other route to ruleconsequentialism might there be? In the next few sections, I will show that ruleconsequentialism's implications about what is right or wrong in particular circumstances match our confident moral convictions quite well. But let me immediately address the familiar challenge to the idea that moral theories are to be tested by their match with intuitions. The familiar challenge is that moral convictions are merely inherited prejudices and as such cannot provide good reason for anything.
In reply to this challenge, let me say I of course recognize that people from different cultures have different moral intuitions, as do people even from the same culture. We must always be willing to reconsider our moral intuitions. They are scarcely infallible.
But, while they are not infallible, they can be crucial. Suppose we have two moral theories which are each coherent developments of appealing general ideas about morality. Suppose one of these theories has implications that match our convictions quite closely, and the other has implications that conflict with many of our mostCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed
un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 196
confidently held moral convictions. In this case, I cannot see what could reasonably keep us from thinking better of the theory with the more intuitively plausible implications. Indeed, it seems to me that we are at least as confident about what is right in some specific kinds of situation as we are about any of the general ideas about morality that get developed into different moral theories such as Aristotelianism, Kantianism, contractualism, and actconsequentialism. This is why almost all moral philosophers are unable to resist ''testing" these theories by comparing the judgments that follow from them with our confident convictions about right and wrong in various kinds of situations.
Let me take stock. I've suggested three different ways of arguing for ruleconsequentialism.
One is that ruleconsequentialism is, from a purely consequentialist point of view, best. I myself am not relying on this argument.
The second is that ruleconsequentialism develops from some very attractive general ideas about morality. Though this is an important feature of ruleconsequentialism, I acknowledge ruleconsequentialism is hardly the only theory that plugs into or develops from attractive general ideas about morality. So the fact that a theory is a coherent development of some initially very attractive ideas is not enough to make it superior to all its rivals.
The third argument for ruleconsequentialism is that we can reach a reflective equilibrium between ruleconsequentialism and our confident moral convictions. At least some moral convictions seem more secure than any theory that could oppose them. If this is right, then appeal to reflective equilibrium between abstract theory and moral conviction must be part of the defense of ruleconsequentialism.
9— RuleConsequentialism on Prohibitions
Whatever actconsequentialism says about daytoday moral thinking, actconsequentialism's criterion of moral rightness entails that whenever killing an innocent person, or stealing, or breaking a promise, etc., would maximize the good, such acts would be morally right. W. D. Ross put forward the following example (table 10.4) to illustrate that keeping one's promises can be right even when this would produce slightly less good (Ross 1930: 34–5):
Table 10.4
Numbers below represent units of good
Effects on person A Effects on person B Total good
Keeping promise to A 1000 0 1,000
Breaking promise to A 0 1,001 1,001Co py ri
gh t @ 20 00 . Jo hn W il ey a nd S on s, I nc .
Al l ri
gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er
U. S. o
r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 197
Most of us would agree with Ross that keeping the promise would be morally right in this case. Actconsequentialism, of course, favors breaking the promise in this case, since that is the alternative with the most good. So, if we agree with Ross about this case, we must reject actconsequentialism.
Most of us also believe (as Ross went on to observe) that, if breaking the promise would produce much greater good than keeping it, breaking the promise could be right. We believe parallel things about inflicting harm on innocent people, stealing, lying, etc. Thus most of us reject what is sometimes called "absolutism" in ethics. Absolutists hold that certain acts (e.g., physical attack on the innocent, promisebreaking, stealing, lying) are always wrong, even when they would prevent the most extreme disasters.
Absolutism and actconsequentialism are, we might say, two ends of a spectrum. Whereas absolutism never permits certain kinds of act, even when necessary to prevent extreme disaster, actconsequentialism insists such act are right not only when a great disaster is at stake but also when a marginal gain in net good is in the offing. Actconsequentialists seem mistaken about these cases of marginal gain, just as absolutists seem mistaken about the disaster cases. Thus, absolutism seems to go too far in one direction, actconsequentialism in the other.
Ruleconsequentialism, on the other hand, concurs with our beliefs both about when we can, and when we cannot, do normally forbidden acts for the sake of the overall good. It claims that individual acts of murder, torture, promisebreaking, and so on, can be wrong even when they result in somewhat more good than not doing them would. The ruleconsequentialist reason for this is that the general internalization of a code prohibiting murder, torture, promisebreaking, and so on would clearly result in more good than general internalization of a code with no prohibitions on such acts.
Another rule whose general internalization would be optimal is a rule telling us to do what is necessary to prevent disasters. This rule is relevant when the only way to prevent a disaster is to break a promise or do some other normally prohibited act. In such cases, ruleconsequentialism holds that the normally prohibited act should be done. I mention this rule about preventing disaster because its existence undermines the objection that ruleconsequentialism would, in a counterintuitive way, prescribe sticking to rules even when this would result in disaster.
10— Doing Good for Others
Morality paradigmatically requires us to be willing to make sacrifices for others. Yet actconsequentialism is widely accused of going too far here too. Utility, impartially calculated, would be maximized if I gave away most of my material goods to the appropriate charities. Giving away most of my material goods is therefore required of me by (most versions of) actconsequentialism. I should probably even change to some more lucrative employment so that I would then have more money to give toCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed
un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 198
charity (Unger 1996: 151). I could make much more money as a corporate lawyer, banker, stockbroker, accountant, gossipcolumnist, or bountyhunter than as an employee of a philosophy department. If people should be willing to make any sacrifices that are smaller than the benefits thereby secured for others, then I should move to the better paying job so that I will have a bigger salary to contribute to the needy. With a bigger salary, I would then have to give an even larger percentage of my earnings to aid agencies. The result would be a life of devoted moneymaking — only then to deny myself virtually all the rewards I could buy for myself with the money. After all, from an actconsequentialist perspective, my own enjoyment is insignificant compared to the very lives of those who would be saved by my additional contributions. Such reflections give special poignancy to Shelly Kagan's remark: "Given the parameters of the actual world, there is no question that [maximally] promoting the good would require a life of hardship, selfdenial, and austerity" (1989: 360).
But many of us may on reflection think that it would be morally unreasonable to demand this level of selfsacrifice for the sake of others. 8 However praiseworthy such selfsacrifice may be, most of us are quite confident that perpetual selfimpoverishment for the sake of strangers is above and beyond what morality requires of us.9
I have been discussing the objection that actconsequentialism requires us to make huge sacrifices in order to maximize our contribution to famine relief. Act consequentialism also requires selfsacrifice even when the benefit to the other person is only slightly larger than the cost to the agent. Consider, for example, the corner office in our building. Offices are allotted on the basis of seniority. Suppose you are the most senior person who might want this corner office. But if you do not take it, it will go to an acquaintance who spends ten percent more time in her office than you do in yours. Suppose we therefore reasonably guess that she would benefit a bit more from moving into this office than you would. This is not a life and death matter. Nor will she be so depressed by not getting the corner office that her work or domestic life will be seriously compromised. Nevertheless, she would get a bit more enjoyment out of the better office than you would. But you still take it for yourself. No one would think you unreasonable or immoral for doing so. Except in special circumstances, morality does not, we think, really require you to sacrifice your own good for the sake of slightly larger gains to others.
I have offered two objections about the demands of actconsequentialism. (1) Actconsequentialism requires huge sacrifices from you. (2) Actconsequentialism requires you to sacrifice your own good even when the aggregate good will be only slightly increased by the your sacrifice. In both ways, actconsequentialism is unreasonably demanding.
In contrast, ruleconsequentialism would not require you to pass up the corner office and let your colleague have it. You are certainly permitted to do that if you want, but ruleconsequentialism would not require such impartiality in your decisions about what to do with your own time, energy, money, or place in line. The rules the internalization of which could reasonably be thought to produce the mostCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 199
good would allow each person considerable partiality towards self (and even require partiality towards friends and family — see Brandt 1989: fn. 22.). For, as I noted earlier, the costs of getting a complete impartiality internalized by each new generation would be prohibitive.
Likewise, whereas actconsequentialism requires huge sacrifices for the sake of maximizing the good, ruleconsequentialism seems not to require more than a reasonable amount of sacrifice for this purpose. Why? A ruleconsequentialist might point out that, if everyone relatively well off in the world were to contribute quite modest amounts to the best aid agencies, the worst elements of poverty could be overcome.
The World Bank has been calling for contributions from the rich countries of 0.7 percent of GDP, the current average being less than half that. Much of this aid does not go to the most needy, but instead to countries that offer business for, or military alliances with, the donor country. The UN estimates that if merely 60 percent of the aid that the rich countries now give (i.e., 60 percent of about $57 billion) were intelligently spent on providing basic health services and clean water and on eliminating illiteracy, these problems could be fixed (The Economist, June 22, 1996: 64).
A ruleconsequentialist will be interested in redistribution beyond what is required to secure the very basic necessities. But even after including these other potential benefits in the costbenefit analysis, we might well conclude that the amount the world's relatively well off would each be required to give would not be unreasonably severe. (For challenges to this view, see Carson 1991; Mulgan 1994 and 1997.)
Consider the following example. Walking along a deserted road on your way to the airport for a flight to the other side of the world, you see a child drowning in a shallow pool beside the road. You could easily save the child, at no risk to yourself. But if you do save the child, you will miss your flight and lose the cost of the nonrefundable ticket. 10
Everyone agrees you are obligated to save the child. This is true even if you are not terribly rich. Suppose the ticket costs as much as a tenth of your annual income. You would still be morally wrong not to make the sacrifice and save the child. And even if the probability of the child's drowning without your rescue is less than 100 percent — suppose, for example, it is 80 percent — you are obligated to sacrifice your ticket to save the child.
Now consider a variant of the example (Singer 1972: 233; Kagan 1991: 924–5; and Murphy 1993: 291. See also Hooker 1995: 25–6). You and I are walking to the airport when we see two small children drowning in a lake. You and I could each easily save the children, at no risk to ourselves. The two children are positioned in the lake in such a way that you and I could each save one and still get to our flights. But if one of us saves both children, he will miss his flight. Suppose you save one child, but I do nothing. Surely, you should now save the other.
Yet, were ruleconsequentialism framed in terms of 100 percent compliance, how could it tell you to save the other? With 100 percent compliance, there would be noCop yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed
un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 200
need for you to save the second child. With 100 percent compliance, once you'd done your share, you'd have done all that was needed. The rule that would be best given 100 percent compliance would presumably not require you to sacrifice more than you would have to sacrifice if everyone did their part. But if this rule is applied to our case, where I am in fact not coming to the rescue, you are not obligated to save this child. This is clearly an implausible implication.
But I argued that ruleconsequentialism should be framed in terms of less that 100 percent compliance. If ruleconsequentialism is framed in terms of 90 percent compliance, we can envisage that there is a need for rules about how to act when others around you aren't doing their part. The rule might be, "When you happen to be surrounded by others who are not helping, then prevent disaster even if this involves doing more than you would have to do if the others helped." This rule would require you to save the second child from the shallow pond.
But if the world we live in — the real world — is one where partial compliance is ubiquitous, then a rule requiring you to make up for the noncompliance of others could become unreasonably demanding. Just how much would ruleconsequentialism require you to make up for noncompliance by other people in a position to help? In earlier work, I assumed that ruleconsequentialism would formulate a rule about aiding the needy in terms of a fairly precise level of contribution or sacrifice to the reduction of world poverty.
I now think this approach is hopeless. Consider a concrete moral code that could reasonably be expected to produce at least as much good as any other we can identify. It would contain rules requiring us not to injure others physically, not to steal, not to break promises, not to lie, etc. These rules might have some exceptions built into them (though not a general breaktherulewheneveryoucouldtherebyproducemoregood exception, nor an unlimited set of much more specific exceptions). Nonetheless, there is pressure to have fairly general rules that can be applied to a wide array of situations. Oxfam's petitioning the rich to help the very poor is hardly the only situation where some people have an opportunity to help others at relatively little cost to themselves. There will be situations where the rich can help other rich, situations where poor can help other poor, even situations where the poor can help the rich. And there will be situations where the help needs to be in the form of physical effort, other situations where the help needs to be in the form of money or time.
Given all this, perhaps the optimific rule for such a world would not be "the rich should give the very needy at least precisely n percent of their annual income", but rather "people should help others in great need when they can do so at modest to themselves, cost being assessed aggregatively, not iteratively" (Cullity 1995: 293–5). Such a rule would apply in a wide array of situations — indeed, whenever some person can help another in great need. It is not limited to what the rich should do nor to what should be done concerning world poverty.
But because cost to the agent is to be assessed aggregatively rather than iteratively, the rule does not require one to help another in great need whenever the cost of helping on that particular occasion is modest. Having to help others wheneverCop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es
pe rm it te d un de r U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 201
doing so on that occasion involves modest cost could easily be very costly. For each of us faces an indefinitely long string of such occasions, because any day on which we could give money to UNICEF or Oxfam counts as such an occasion. But many small sacrifices added together can amount to a huge sacrifice. The end of that road is selfimpoverishment. If I am right, ruleconsequentialism instead endorses a rule requiring sacrifices over the course of your life that add up to something significant. It allows but does not require personal sacrifice beyond this point.
I propose that this rule would have good consequences even in possible worlds that are either much poorer or much richer than ours. I don't have space here to argue either that ruleconsequentialism would indeed end up with this rule in all possible worlds, or that this rule always has intuitively acceptable consequences. I mention this rule only in order to sketch one way in which a defense of ruleconsequentialism might go (cf. Hooker 1998).
11— Conclusion
Ruleconsequentialism has an uncertain future. It needs to be carefully formulated if it is to avoid being a sitting target. In this essay, I have tried to improve its defenses by finetuning its formulation. I have also argued here that the theory develops from appealing general beliefs about morality, that it does not collapse into act consequentialism, and that it coheres well with our intuitions about moral prohibitions and permissible partiality. As I see things, the theory is healthy now. But it is hardly invulnerable. Like someone walking through a dangerous city who has so far managed to fight off muggers emerging from behind every corner, the theory might meet an attack it cannot survive. I am curious to see whether that happens. 11
Notes
1. The use of "consequentialism" and "utilitarianism" so that consequentialism allows for a concern for distribution in a way that utilitarianism does not is very common. For some examples, see Mackie 1977: 129, 149; Scanlon 1978, especially sect. 2; Scheffler 1982: 26–34, 70–9; Parfit 1984: 26; Griffin 1992: 126; 1996: 165. Examples of writers including distributive considerations within utilitarianism include Brandt 1959: 404, 426, 429–31; Raphael 1994: 47; Skorupski 1995: 54; and arguably Mill 1861.
2. See Mill 1861, ch. II; Sidgwick 1907: 405–6, 413, 489–90; Moore 1903: 162–4; Bales 1971; Hare 1981; Railton 1984: 140–6, 152–3; Parfit 1984: 24–9; Brink 1989: 216, 256–61. For criticism of the mileage some have tried to get from this, see Johnson 1989; Griffin 1992: 123–4; 1996: 104–5. See also Williams 1973: 123.
3. I have taken the liberty here of formulating the theory in what seems to me the most attractive relatively succinct way. I shall have to add complications to this formulation later.Cop
yr ig ht @ 2 00 0. J oh n Wi le y an d So ns , In c.
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er
U. S. o r ap pl ic ab le c op yr ig ht l aw .
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 202
4. Earlier ruleconsequentialists have addressed this question. See, e.g., Brandt: 1967 §8; and 1988 [1992: 149–54].
5. I am grateful to Dale Miller for pressing me to provide the argument developed in the rest of this section.
6. There are other questions to ask as well. For a discussion of the importance of questions about helpfulness and about parsimony at the level of first principles, see Hooker 1996.
7. See Lyons 1965, ch. IV; Williams 1972: 99–102, 105–8; Slote 1992: 59; Raphael 1994: 52; Scarre 1996: 125–6. Relatedly, Regan (1980: 209) complains that ruleconsequentialists "are only halfhearted consequentialists".
8. See Crisp 1992; and Quinn 1993: 171: "We think there is something morally amiss when people are forced to be farmers or flute players just because the balance of social needs tips in that direction. Barring great emergencies, we think people's lives must be theirs to lead."
9. Even Kagan, whose book goes on to defend actconsequentialism for 400 pages, starts by acknowledging that it "strikes us as outrageously extreme in its demands" (1989: 2.).
10. This example has been central to the contemporary philosophical debate about beneficence. The example and the debate owe their prominence to Singer 1972 (see his restatement in ch. 8 of his 1993). Some important contributions to this discussion are Fishkin 1982; Scheffler 1982; Kagan 1989, especially pp. 3–4, 231–2; Nagel 1991; Murphy 1993; Cullity 1994; and Unger, 1996.
11. I have had help with this essay from Roger Crisp, Jonathan Dancy, Max de Gaynesford, Hanjo Glock, Hugh LaFollette, Andrew Mason, Elinor Mason, Dale Miller, David Oderberg, and Derek Parfit, and the audience at the 1997 International Society for Utilitarian Studies Conference.
References
Adams, R. M.: "Motive Utilitarianism," The Journal of Philosophy, 73 (1976): 467–81.
Austin, J.: The Province of Jurisprudence Determined (1832).
Bales, R. E.: "Actutilitarianism: Account of Rightmaking Characteristics or Decisionmaking Procedure?" American Philosophical Quarterly, 8 (1971): 257–65.
Bentham, Jeremy: An Introduction to the Principles of Morals and Legislation (1789).
Berkeley, G.: Passive Obedience, or the Christian Doctrine of Not Resisting the Supreme Power, Proved and Vindicated upon the Principles of the Law of Nature (1712).
Blackburn, Simon: "Errors and the Phenomenology of Value," Morality and Objectivity, A Tribute to J. L. Mackie, ed. T. Honderich (London: Routledge and Kegan Paul, 1985).
Brandt, R. B.: Ethical Theory (Englewood Cliffs, NJ: PrenticeHall, 1959).
———: "Some Merits of One Form of Ruleutilitarianism," University of Colorado Studies in Philosophy, 3 (1967): 39–65. Reprinted in Brandt 1992, pp. 111– 36.
———: A Theory of the Good and the Right (Oxford: Clarendon Press, 1979).
———: "Fairness to Indirect Optimific Theories in Ethics," Ethics, 98 (1988): 341–60. Reprinted in Brandt 1992, pp. 137–57.
———: "Morality and its Critics," American Philosophical Quarterly, 26 (1989): 89–100. Reprinted in Brandt 1992, pp. 73–92.
———: Morality, Utilitarianism, and Rights (New York: Cambridge University Press, 1992).
———: Facts, Values, and Morality (New York: Cambridge University Press, 1996).Co py ri gh t @ 20 00 . Jo hn W il ey a nd S on s, I nc .
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t
la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 203
Brink, David O.: Moral Realism and the Foundations of Ethics (New York: Cambridge University Press, 1989).
Carson, Thomas: "A Note on Hooker's Rule Consequentialism," Mind, 100 (1991): 117–21.
Crisp, Roger: "Utilitarianism and the Life of Virtue," Philosophical Quarterly, 42 (1992): 139–60.
———: Mill on Utilitarianism (London: Routledge, 1997).
Cullity, Garrett: "International Aid and the Scope of Kindness," Ethics, 105 (1994): 99–127.
———: "Moral Character and the Iteration Problem," Utilitas, 7 (1995): 289–99.
Dancy, Jonathan: "On Moral Properties," Mind, 90 (1981): 367–85.
———: "Ethical Particularism and Morally Relevant Properties," Mind, 92 (1983): 530–47.
———: Moral Reasons (Oxford: Blackwell, 1993).
Dworkin, Ronald: Law's Empire (Cambridge, MA: Harvard University Press, 1986).
Finnis, John: Natural Law and Natural Rights (Oxford: Clarendon Press, 1980).
———: Fundamentals of Ethics (New York: Oxford University Press, 1983).
Fishkin, James: The Limits of Obligation (New Haven: Yale University Press, 1982).
Foot, Philippa: "Utilitarianism and the Virtues," Mind, 94 (1985): 196–209.
Griffin, James: WellBeing: Its Meaning, Method and Moral Importance (Oxford: Clarendon Press, 1986).
———: "The Human Good and the Ambitions of Consequentialism," Social Philosophy and Policy, 9 (1992): 118–32.
———: Value Judgement: Improving Our Ethical Beliefs (Oxford: Clarendon Press, 1996).
Hare, R. M.: Moral Think: Its Method, Levels, and Point (Oxford: Clarendon Press, 1981).
Harsanyi, John: "Morality and the Theory of Rational Behavior," Utilitarianism and Beyond, eds., A. Sen and B. Williams (Cambridge: Cambridge University Press, 1982), pp. 39–62. Reprinted from Social Research, 44 (1977).
———: "Expectation Effects, Individual Utilities, and Rational Desires," Rationality, Rules, and Utility: New Essays on the Moral Philosophy of Richard Brandt, ed. B. Hooker (Boulder: Westview Press, 1993) pp. 115–26.
Hodgson, D. H.: Consequences of Utilitarianism (Oxford: Clarendon Press, 1967).
Hooker, Brad: "Mark Overvold's Contribution to Philosophy," Journal of Philosophical Research, 26 (1991): 333–44.
———: "Ruleconsequentialism, Incoherence, Fairness," Proceedings of the Aristotelian Society, 95 (1995): 19–35.
———: "Rossstyle Pluralism Versus Ruleconsequentialism," Mind, 105 (1996): 531–52.
———: "Ruleconsequentialism and Obligations Toward the Needy," Pacific Philosophical Quarterly, 79 (1998): 19–33.
Hurka, Thomas: "The Wellrounded Life," Journal of Philosophy, 84 (1987): 707–26.
———: Perfectionism (New York: Oxford University Press, 1993).
Johnson, Conrad: "Character Traits and Objectively Right Action," Social Theory and Practice, 15 (1989): 67–88.
———: Moral Legislation (New York: Cambridge University Press, 1991).
Kagan, Shelly: The Limits of Morality (Oxford: Clarendon Press, 1989).
———: "Replies to My Critics," Philosophy and Phenomenological Research, 51 (1991): 924–5.
Kavka, Gregory: "Some Paradoxes of Deterrence," Journal of Philosophy, 75 (1978): 285–302.
Lyons, David: Forms and Limits of Utilitarianism (Oxford: Clarendon Press, 1965).
Mackie, J. L.: Ethics: Inventing Right and Wrong (Hammondsworth: Penguin, 1977).
Mill, J. S.: Utilitarianism (1861).Co py ri gh t @ 20 00 . Jo hn
W il ey a nd S on s, I nc .
Al l ri gh ts r es er ve d. M
ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd
er U .S . or a pp li ca bl e co py ri gh t la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost
Page 204
Moore, G. E.: Principia Ethica (Cambridge: Cambridge University Press, 1903).
Mulgan, T.: "Rule Consequentialism and Famine," Analysis, 54 (1994): 187–92.
———: "One False Virtue of Rule Consequentialism and One New Vice," Pacific Philosophical Quarterly, 77 (1997): 362–73.
Murphy, Liam: "The Demands of Beneficence," Philosophy and Public Affairs, 22 (1993): 267–92.
———: "A Relatively Plausible Principle of Benevolence: A Reply to Mulgan," Philosophy and Public Affairs, 26 (1997): 80–6.
Nagel, Thomas: Equality and Partiality (New York: Oxford University Press, 1991).
Overvold, Mark: "Selfinterest and the Concept of Selfsacrifice," Canadian Journal of Philosophy, 10 (1980): 105–18.
———: "Selfinterest and Getting What You Want," The Limits of Utilitarianism, eds. H. B. Miller and W. H. Williams (Minneapolis: University of Minnesota Press, 1982), pp. 186–94.
Parfit, Derek: Reasons and Persons (Oxford: Clarendon Press, 1984).
Quinn, Warren: Morality and Action (New York: Cambridge University Press, 1993).
Railton, Peter: "Alienation, Consequentialism, and the Demands of Morality," Philosophy and Public Affairs, 13 (1984): 174–31.
Raphael, D. D.: Moral Philosophy, 2nd edn. (Oxford: Oxford University Press, 1994).
Rawls, John: "Two Concepts of Rules," Philosophical Review, 64 (1955): 3–32
———: A Theory of Justice (Cambridge, MA: Harvard University Press, 1971).
Regan, D.: Utilitarianism and Cooperation (Oxford: Clarendon Press, 1980).
Ross, W. D.: The Right and the Good (Oxford: Clarendon Press, 1930).
Scanlon, T. M.: "Rights, Goals and Fairness," Public and Private Morality, ed. S. Hampshire (Cambridge, UK: Cambridge University Press, 1978) pp. 93–111.
———: "Contractualism and Utilitarianism," Utilitarianism and Beyond, eds. A. Sen and B. Williams (Cambridge: Cambridge University Press, 1982), pp. 103–28.
———: "Value, Desire, and Quality of Life," The Quality of Life, eds. M. Nussbaum and A. Sen (Oxford: Clarendon Press, 1993), pp. 185–200.
Scarre, Geoffrey: Utilitarianism (London: Routledge, 1996).
Scheffler, S.: The Rejection of Consequentialism (Oxford: Clarendon Press, 1982).
Sidgwick, Henry: Methods of Ethics, seventh edn. (London: Macmillan, 1907), first edn., 1874.
Singer, Peter: "Famine, Affluence, and Morality," Philosophy and Public Affairs, I (1972): 229–43.
———: Practical Ethics, second edn. (Cambridge, UK: Cambridge University Press, 1993).
Skorupski, John: "AgentNeutrality, Consequentialism, Utilitarianism . . . A Terminological Note," Utilitas, 7 (1995): 49–54.
Slote, Michael: From Morality to Virtue (New York: Oxford University Press, 1992).
Smart, J. J. C.: "Outline of a System of Utilitarian Ethics," Utilitarianism: For & Against, eds. J. J. C. Smart and Bernard Williams (Cambridge: Cambridge University Press, 1973), pp. 3–74.
Unger, Peter: Living High and Letting Die: Our Illusion of Innocence (New York: Oxford University Press, 1996).
Urmson, J. O.: "The Interpretation of the Philosophy of J. S. Mill," Philosophical Quarterly, 3 (1953): 33–9.
Williams, Bernard: Morality: An Introduction to Ethics (New York: Harper and Row, 1972).
———: "A Critique of Utilitarianism," Utilitarianism: For & Against, eds. J. J. C. Smart and B. Williams (Cambridge, UK: Cambridge University Press, 1973) pp. 77–150.Co
py ri gh t @ 20 00 . Jo hn W il ey a nd S on s, I nc .
Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh
er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.
EBSCO : eBook Collection (EBSCOhost) - printed on 5/24/2019 3:06 PM via LOYOLA MARYMOUNT UNIV AN: 44701 ; LaFollette, Hugh.; The Blackwell Guide to Ethical Theory Account: loyola.main.ehost