Literature Review

IS IT I M P O S S I B L E T O R E L I E V E S U F F E R I N G ? *

MICHAEL ALMEIDA AND MARK BERNSTEIN

1. I n t r o d u c t i o n

I m a g i n e a situation in which three victims, A, B, and C, are attached

to a m a c h i n e that causes pain in adjustable increments. S u p p o s e the m a c h i n e has been adjusted to three increments f o r A, six increments for

B and nine increments for C, and each o f the victims is in pain. There

is no perceptible difference between experiences o f A and B and no

perceptible difference between the experiences o f B and C. But C is

noticeably worse o f f than A. N o w suppose that we can relieve entirely

the pain o f A, B, and C and that there are no other morally relevant

features o f the situation.

It seems an indisputable duty o f b e n e f i c e n c e - - c e r t a i n l y for s o m e o n e

w h o can do so at no moral cost to a n y o n e - - t o completely relieve the

pain o f A, B, and C. In fact we k n o w that c o m p l e t e l y relieving that pain

is m a x i m a l l y beneficial in this situation since (by hypothesis) no other

feature o f the situation is morally relevant. But there is an intriguing new argument that presents a serious challenge to any principle o f beneficence

that requires us to relieve the pain o f A, B, and C. The a r g u m e n t - - d u e

to Stephen S c h w a r t z - - e n t a i l s that it is impossible to relieve the pain o f A, B and C without doing some moral wrong. ~ S c h w a r t z advances two

basic principles in what we call his Impossibility Proof. the Principle of Equality and the Principle of Differential Treatment. T h e principles have strong credentials and w e are p r e p a r e d to c o n c e d e that certain

versions o f these are a priori metaethical principles. It does seem clear that suitably qualified versions o f these principles incorporate rational

constraints on moral deliberation. But these principles together with the

313

MICHAEL ALMEIDA AND MARK BERNSTEIN

plausible intransitivity o f " a s painful as" are the basic assumptions in the

proof. A c c o r d i n g to S c h w a r t z these credible assumptions alone entail

the incredible conclusion that it is impossible to relieve the pain o f A,

B, and C and d o nothing wrong.

In s e c t i o n 2 w e d i s c u s s s u i t a b l e f o r m u l a t i o n s o f e a c h b a s i c

assumption in the a r g u m e n t and provide the original (informal) version

o f the Impossibility Proof In section 3 we offer a formal version o f the Impossibility Proof We observe that the p r o o f is invalid unless we assume two additional metaethical principles: Irrelevance oflmperceptible Pain and Transitivity of Equal Moral Treatment. We s h o w that these additional assumptions cannot both be true and we c o n c l u d e that the Impossibility Proof is therefore unsound. We argue further that there is no reason to expect that a sound version the p r o o f is f o r t h c o m i n g . We offer s o m e

c o n c l u d i n g remarks in section 4.

2. W h y it is Impossible to Relieve S u f f e r i n g T h e Principle of Equality or PE expresses the proposition that two

individuals that are the same in every morally relevant respect must be

treated in the same way. Since the principle imposes no restrictions on

what constitutes a morally relevant feature o f individuals it is consistent

with Libertarian principles o f justice allotting g o o d s a c c o r d i n g to free

c h o i c e as well as Marxist principles o f justice allotting g o o d s a c c o r d i n g

to need as well as Utilitarian principles allotting goods according to value

maximization. I f we suppose that two items P and Q are distributed to

two individuals A and B, William Frankena urges that any acceptable

distribution must meet the formal restrictions o f PE.

It is a requirement o f both reason and o f c o m m o n thinking

about justice that similar cases be treated similarly. This

means that if A and B are similar then P and Q must be

similar. But, if this is so, tben it would a p p e a r that justice

also d e m a n d s that if A and B are dissimilar, then P and Q

m u s t be dissimilar. 2

B u t o f c o u r s e agents that are similar in e v e r y m o r a l l y relevant

r e s p e c t - - as Frankena w o u l d surely c o n c e d e - - m a y sometimes be treated

314

IS IT IMPOSSIBLE TO RELIEVE SUFFERING?

in ways that are dissimilar in several respects. The expression of gratitude,

for instance, can be permissibly dissimilar even in situations differing

in no moral respect. The Principle of Equality demands only that the expression o f gratitude differ in no respect that has moral relevance in such situations. It is likely to be morally unimportant precisely how that

gratitude is expressed. So PE allows that treatment that is morally similar may be otherwise significantly dissimilar.

But agents that are similar in every morally relevant respect--as

again Frankena would certainly c o n c e d e - - m a y sometimes be treated

in ways that are morally dissimilar. Smith might permissibly agree to rescue Jones and permissibly refuse to rescue Brown, for instance, even

if Brown is similar to Jones in every morally relevant respect. Smith's

rescue o f Jones might be supererogatory and so something that Smith

need not do in every similar situation. Or Smith might be able to rescue

each, Jones and Brown, but simply lack the ability or resources to rescue

both. Or Smith might be able to rescue both but lack the knowledge to do

so, and so on. Certainly most would agree that attention to these and other

qualifications is important in the application of PE to particular cases. But given the appropriate qualifications we are prepared to concede the

version of PE that is assumed in the Impossibility Proof Let's assume then that the principle o f equality leaves these qualifications implicit.

PE. l f A and B are the same in every morally relevant respect then A

and B must receive the same moral treatment. 3

The Principle o f Differential Treatment or PDT expresses the proposition that individuals that are dissimilar in some moral respect

must be treated in ways that are morally dissimilar. 4 PDT entails, for instance, that it is unjust to distribute benefits equally between A and B,

i f A is different from B in ways that are relevant to that distribution. The

principle is implicit in views about justice as divergent as Liberal Equality

and Utilitarianism. It might be, for instance, that merit is relevant to the distribution of benefits and that A has merited more than B. If so then

PDTwould ensure that A receives the greater benefits she has merited. Or it might be that the maximization of value is relevant to the distribution

of benefits and that a distribution favoring A would maximize value. If

so then PDT would again ensure that A receives the greater benefits.

315

MICHAEL ALMEIDA AND MARK BERNSTEIN

T h e r e are a n a l o g o u s r e s t r i c t i o n s on P D T c o n c e r n i n g ability, knowledge, resources, and supererogation. But given the appropriate qualifications we are prepared to concede the version of PDTassumed in the Impossibility Proof Let's assume then that the principle o f differential treatment leaves these qualifications implicit. PDT. I f A and B differ in some morally relevant respect then A and B

must receive different moral treatment.

The formal principles PE and PDTare otherwise well-known and well- received metaethical principles. But these credible p r i n c i p l e s - - t o g e t h e r with the intransitivity o f "as painful a s " - - a r e the basic assumptions in

an intriguing proof that it is sometimes impossible to be m o r a l )

In the original Impossibility Proof we are offered a simple three- person model designed to make clear the logic o f the situation. We are given the following informal presentation o f the argument.

The Principle o f Equality, the Principle o f Differential Treatment and [the intransitivity o f m o r a l l y r e l e v a n t characteristic R = "as painful as"l are sufficient to generate contradictions. Consider the tbllowing simple model. It is morally required to treat items indistinguishable with respect to morally relevant characteristic R similarly. It is morally required to treat items distinguishable with respect to R differently. Now we have the situation that A is indistinguishable from B in respect R, B is indistinguishable from C, but A is distinguishable from C. Thus it is morally

required to treat A and C similarly, because A and B must be treated similarly and B and C must be treated similarly, but it is also morally required to treat A and C differently. This situation will arise whenever A and C are required to be treated differently and [R is a vague characteristic], but this is just the situation with all morally relevant characteristics.

Thus it is impossible to be moral. 6

The Impossibility Proof concludes that it is impossible to be moral since anyone who completely relieves the pain o f A, B, and C must do

316

IS IT IMPOSSIBLE TO RELIEVE SUFFERING?

something morally wrong. Suppose that we relieve the pain o f A, B, and

C and treat A and C the same way. In that case the pain is completely

relieved but we have violated the moral requirement to treat A and

C differently. Suppose then that we relieve the pain o f A, B, and C and

treat A and C differently. In that case the pain is completely relieved but

we have violated the moral requirement to treat A and C the same way.

So, no matter how the pain is relieved there is a violation o f some moral

requirement. It is therefore impossible to relieve the pain o f A, B, and C

without doing something wrong.

The conclusion that it is impossible to relieve the pain o f A, B, and C

without wrongdoing is incredible. In section (3) we consider a formal

version of the proof and show that the Impossibility Proof is unsound. We argue further that there is no reason to expect that a sound version of the

proof is forthcoming. We offer some closing comments in section 4.

3. Is it Impossible to Relieve Suffering?

We have assumed that three victims, A, B, and C, have been attached

to a machine that causes pain in adjustable increments. The machine has

been adjusted to three increments for A, six increments for B and nine

increments for C. There is no perceptible difference between experiences

of A and B and no perceptible difference between the experiences of B

and C. But C is noticeably worse off than A. We can relieve entirely the

pain of A, B, and C and there are no other morally relevant features of

the situation. Among the options we have available in this situation is turning each

dial to zero and completely relieving the pain of A, B, and C. Since there

are no other morally relevant features of the situation--there is not, for

instance, the option to produce pleasurable mental states or otherwise

improve the well-being of A, B and C - - t h e most beneficent option we

have is to fully relieve the suffering of A, B, and C. Therefore we ought

to fully relieve that suffering.

The first five assumptions in the Impossibility Proof describe the situation we have been discussing. Since we agree that such a situation

is possible we have no quarrel with stipulating premises (1)-(5).

I. A perceives the same degree of pain as B.

317

MICHAEL ALMEIDA AND MARK BERNSTEIN

Assumption

2. B perceives the same degree o f pain as C.

Assumption

3. C perceives more pain than A.

Assumption.

4. Other than the experience o f pain there are no other features o f the

situation that have moral relevance.

Assumption.

5. We ought to relieve entirely the pain o f A, B and C.

Assumption

O f course the Impossibility Proof is unsound unless it can plausibly establish both that A is the same as B in every morally relevant respect

and that B is the same as C in every morally relevant respect. We know

that the experience o f A is indistinguishable from the experience o f B,

and that the experience of B is indistinguishable from the experience o f

C. We also know that other than the experience o f pain no other feature

of the situation has moral relevance. But we still cannot conclude that

therefore A is the same as B in every morally relevant respect or that therefore B is the same as C in every morally relevant respect. After all

we also know that there are differences in pain between A and B and

between C and B that are too small to be noticed. So among the additional

assumptions necessary to the Impossibility Proof is the moral irrelevance of imperceptible pains or liP.

liP. If A and B differ only with respect to the pain each

experiences and the experience o f B is not perceptibly

more painful than the experience of A, then B is not worse

off than A.

The conditions in liP ensure that no experience of pain that is below the threshold of perception--no pains that are not "just noticeable" to

use the description o f Torbjorn Tannsjo--matter to the well-being o f

A, B or C. The assumption of liP together with premises (1)-(5) entail (6)-(8). 7

6. A is the same as B in every morally relevant respect.

From (1), (4) & liP

318

IS IT IMPOSSIBLE TO RELIEVE SUFFERING?

7. B isthe same as C i n every morally relevant respect.

F r o m ( 2 ) , ( 4 ) & I I P 8. A is different from C i s some morally relevant respect.

F r o m ( 3 ) & ( 4 )

We should note that even proponents o f the Impossibility Proof have good reason to doubt liP. The metaethical position that 'as good as' and 'as bad as' are transitive relations is at least as plausible as liP. But the principle l i P is not in general consistent with the transitivity o f 'as good as' or 'as bad as'. Given l i P we know that the experience o f C is as good as the experience of B and the experience of B is as good as the experience o f A. But we also find--contrary to the transitivity of 'as

good a s ' - - t h a t the experience of C is not as good as the experience o f

A. l i p therefore entails that 'as good as' and 'as bad as' are intransitive

relations. 8

It is worth mentioning too that there are several other counterintuitive

implications of liP. Briefly reconsider the experiences o f A, B and C. It might have been true that each imperceptible increment o f pain was

caused by a different moral agent. Since no moral agent causes any

perceptible pain HP entails that no agent lessens the well-being A, B or C. There can be no duty to prevent these agents from causing pain toA, B

and C, since these agents are doing nothing that makes a moral difference

to anyone. But it is difficult to see how we might have the obligation

described in premise ( 5 ) - - t h e obligation to relieve entirely the pain o f

A, B and C - - a n d also have no obligation to prevent these agents from

causing pain to A, B, and C. 9

But suppose we concede without further discussion that there is a

more convincing argument for liP forthcoming. The Impossibility Proof appeals next to the Principle of Differential Treatment and premise (8) to derive (9). The inference is uncontroversial.

9. The treatment of A must be morally different from the treatment of C.

From (8) & PDT There is a genuine problem presented in the Impossibility Proof only

if we can derive a conclusion inconsistent with premise (9). Indeed that

derivation concludes the proof and it proceeds in two simple steps. The

319

MICHAEL ALMEIDA AND MARK BERNSTEIN

Principle of Equality together with (1)-(8) entail premises (10) and (11). The inference in this first step seems straightforward and we have no

interesting objection to it.

10. The treatment o f A must be morally the same as the treatment o f

B.

From (5) & PE. 11. The treatment o f B must be morally the same as the treatment of

C.

From (6) & PE For the second step we are offered an inference from premises (10)

and (11) to premise (12). And we have quickly arrived at a premise that

is inconsistent with premise (9).

12. The treatment o f A must be morally the same as the treatment o f

C.

From (10) & (11)

Premise (12) completes the simple two-step argument. We have

therefore derived the impossible! The closing propositions in the

Impossibility Proof summarize the most important consequences o f the argument.

13. Therefore the treatment o f A must be morally different from the

treatment o f C and the treatment of A must be morally the same as the treatment of C.

From (9) & (12) Contradiction! 14. Therefore it is impossible to relieve entirely the pain of A, B, and C

without doing something morally wrong.

From (9), (12) & (13)

The conclusions of the Impossibility Proof are unwelcome news. It is an extraordinary result to arrive at the conclusion that we cannot

relieve the pain o f A, B, and C without engaging in some wrongdoing.

But we cannot arrive at that conclusion unless the closing two-step

argument establishes premise (12). And we can show that that inference

is unsound.

Suppose we concede without further argument premises (I)-(11).

We will find there is still no plausible way to arrive at the conclusion in

(12). It should be obvious that the direct inference to (I 2) from premises

320

IS IT IMPOSSIBLE TO RELIEVE SUFFERING?

(10) a n d (11) is i n v a l i d . T h e r e is no r u l e o f i n f e r e n c e that w i l l t a k e us

f r o m (10) a n d (11) i m m e d i a t e l y to p r e m i s e (12). So it is p e r h a p s b e s t

to a s s u m e that this s m a l l i n f e r e n c e is e n t h y m e m a t i c . T h e m e t a e t h i c a l

p r i n c i p l e that s e e m s i m p l i c i t in the i n f e r e n c e to p r e m i s e (12) asserts

that the r e l a t i o n o f ' e q u a l m o r a l t r e a t m e n t ' is t r a n s i t i v e . C o n s i d e r t h e n a

s i m p l e f o r m u l a t i o n o f the t r a n s i t i v i t y o f e q u a l m o r a l t r e a t m e n t in TMT.

T h i s p r i n c i p l e t o g e t h e r w i t h (10) a n d (11) d o e s e n t a i l (12).

T M T . I f the t r e a t m e n t o f A m u s t b e m o r a l l y the s a m e as t h e

t r e a t m e n t o f B a n d the t r e a t m e n t o f B m u s t b e m o r a l l y the

s a m e as the t r e a t m e n t o f C t h e n the t r e a t m e n t o f A m u s t b e

m o r a l l y the s a m e as the t r e a t m e n t o f C.

L e t ' s n o w s h o w that t h e r e is no r e a s o n to b e l i e v e that TMT is true

u n l e s s w e can be sure that the r e l a t i o n ' s a m e in e v e r y m o r a l r e s p e c t ' is

transitive.

R e c a l l that the Principle of Equality g u a r a n t e e s that m o r a l a g e n t s w h o a r e the s a m e in e v e r y m o r a l r e s p e c t m u s t r e c e i v e the s a m e m o r a l

t r e a t m e n t . N o w s u p p o s e A is the s a m e as B in e v e r y m o r a l r e s p e c t a n d

B is the s a m e as C in e v e r y m o r a l r e s p e c t a n d the r e l a t i o n ' s a m e in

e v e r y m o r a l r e s p e c t ' is transitive. It then f o l l o w s that A is the s a m e as

C in e v e r y m o r a l respect. I f A, B and C are the s a m e in e v e r y m o r a l

r e s p e c t then PE g u a r a n t e e s that A, B, and C all r e c e i v e the s a m e moral treatment. So PE t o g e t h e r with the t r a n s i t i v i t y o f ' s a m e in e v e r y m o r a l r e s p e c t ' entails TMT.

But s u p p o s e t h a t A is the s a m e as B in e v e r y m o r a l r e s p e c t and B is the

s a m e as C in e v e r y m o r a l r e s p e c t and the r e l a t i o n ' s a m e in e v e r y m o r a l

r e s p e c t ' is not transitive. In that c a s e w e can c o n s i s t e n t l y a s s u m e that A

is n o t the s a m e as C in e v e r y m o r a l r e s p e c t . But t h e n PE g u a r a n t e e s that A a n d B r e c e i v e the s a m e m o r a l t r e a t m e n t and that B and C r e c e i v e the

s a m e m o r a l treatment. But PDT g u a r a n t e e s that A a n d C r e c e i v e different m o r a l t r e a t m e n t . R e c a l l t h a t the Principle of Differential Treatment g u a r a n t e e s that m o r a l a g e n t s w h o d i f f e r in s o m e m o r a l r e s p e c t m u s t

r e c e i v e d i f f e r e n t m o r a l t r e a t m e n t . But then the a n t e c e d e n t o f TMT is true a n d the c o n s e q u e n t is false. T h e r e f o r e the Principle of Equality and the Principle of Differential Treatment t o g e t h e r w i t h the i n t r a n s i t i v i t y o f

321

MICHAEL ALMEIDA AND MARK BERNSTEIN

' s a m e in every moral respect' entail that TMT is false.

The Impossibility Proof of c o u r s e assumes PE and PDT. I f the p r o o f also assumes that ' s a m e in every moral respect' is intransitive, then it c a n n o t assume TMT. We k n o w that A is the same as B in every moral respect f r o m premise (6) and that B is the same as C in every moral respect f r o m premise (7). But w e also k n o w that C is not the same as A in every moral respect f r o m premise (8). So the p r o o f clearly assumes

that ' s a m e in e v e r y moral respect' is intransitive. T h e r e f o r e the p r o o f

cannot also a s s u m e TMT. A n d so w e arrive at the c o n c l u s i o n that the

Impossibility Proof is unsound. E v e n if w e c o n c e d e premises (1)-(11) w e have no w a y to derive (12). The Impossibility Proof therefore presents no problem f o r any principle o f beneficence that requires that we relieve entirely the pain o f A, B, and C.

4. C o n c l u d i n g R e m a r k s

We have c o n c e d e d about as m u c h as possible to the Impossibility Proof We have been willing to read the principles PE and PDT as implicitly qualified. These qualifications were necessary to a c c o m m o d a t e

s u p e r e r o g a t o r y a c t i o n s , and o u r l i m i t a t i o n s in k n o w l e d g e , ability,

resources, etc. We have agreed that perhaps a more plausible a r g u m e n t

f o r liP is forthcoming. A n d we h a v e been willing to take the closing two-step argument from ( I 0) and (11) to (12) as e n t h y m e m a t i c . But even

if it is c o n c e d e d that (1)-(11) are true, w e have shown that there is no

w a y to arrive at (12). We have therefore c o n c l u d e d that the a r g u m e n t

is unsound. But further, given the c o n c e s s i o n s w e have made to the

argument, it seems to us unlikely that a sound version o f this intriguing

p r o o f is forthcoming.

THE UNIVERSITY OF TEXAS SAN ANTONIO, TEXAS 78249-0643

USA michael.almeida @ utsa.edu [email protected]

322

IS IT I M P O S S I B L E T O R E L I E V E S U F F E R I N G ?

N O T E S

* T h a n k s to S t e p h e n S c h w a r t z for c o m m e n t s o n an earlier draft.

See S t e p h e n P. Schwartz, ' W h y it is I m p o s s i b l e to B e M o r a l ' American Philosophical Quarterly Vol. 36, No. 4 (1999) 3 5 1 - 3 6 0

2 Cf. W i l l i a m F r a n k e n a , ' T h e C o n c e p t o f Social J u s t i c e ' in R i c h a r d B. B r a n d t

(ed.) Social Justice ( E n g l e w o o d Cliffs, N.J.: P r e n t i c e - H a l l 1962) p. 9 ff. 3 Since these are m e t a e t h i c a l p r i n c i p l e s the ' m u s t ' in PE ( a n d in PDT t h a t

follows) is (at least) a rational ' m u s t ' . B u t it s e e m s fair to c o n c l u d e t h a t

violations o f these principles are b o t h irrational a n d unjust.

4 In his Impossibility Proof S c h w a r t z a d v a n c e s a version o f the Principle of Differential Treatment that is too w e a k to yield the c o n c l u s i o n t h a t it is s o m e t i m e s i m p o s s i b l e to be moral. His version states t h a t i f A a n d B d i f f e r

in some m o r a l l y relevant respect t h e n A a n d B m u s t (or at least may) r e c e i v e different m o r a l treatment. We will t h e r e f o r e find t h a t it is perfectly p o s s i b l e

to relieve the pain o f A, B, and C and not treat A and C differently. A n d so it is not impossible to b e moral. See Stephen E Schwartz, ' W h y it is Impossible

to Be M o r a l ' op. tit. But we are p r e p a r e d to c o n c e d e the stronger version o f the principle n e e d e d to yield the c o n c l u s i o n that it is i m p o s s i b l e to r e l i e v e

the pain o f A. B, a n d C and do n o t h i n g wrong.

5 T h e relation "as painful as" is intransitive ill. (i) E o is as painful as E t a n d

(ii) E~ is as painful as E 2 does not entail that (iii) E o is as painful as E 2. We say that it is " s o m e t i m e s " i m p o s s i b l e to b e moral since there are s o m e

moral s i t u a t i o n s - - f o r instance cases w h e r e one is s i m p l y required to keep

a p r o m i s e - - i n w h i c h the a s s u m p t i o n s o f the Impossibility P r o o f are not all true.

6 See S t e p h e n P. Schwartz, ' W h y it is I m p o s s i b l e to Be M o r a l ' , op. tit. v See Torbjorn Tannsjo, 'Classical H e d o n i s t i c U t i l i t a r i a n i s m ' , Philosophical

Studies 81 (1996): 97-115. 8 Derek Parfit was a m o n g the first to note that principles such as l i p entail

the intransitivity o f ' a s good a s ' a n d ' a s bad as'. N o t i c e that HP also e n t a i l s that 'as well o f f a s ' and ' w o r s e o f f t h a n ' are also intransitive. Parfit rejects

such p r i n c i p l e s for that reason. See Derek Parfit Reasons and Persons (Oxford: O x f o r d University Press, 1984) esp. p 78 ff. A m o n g others w h o

reject principles such as HP see Torbjorn Tannsjo, Hedonistic Utilitarianism (Edinburgh: E d i n b u r g h U n i v e r s i t y Press, 1998) esp. p. 70 ff., For additional

worries c o n c e r n i n g the intransitivity e n t a i l e d by p r i n c i p l e s s u c h as lIP see Warren S. Q u i n n , ' T h e Puzzle o f the S e l f - T o r t u r e r ' , Philosophical Studies 59 (1990) 79-90. Q u i n n there d e v e l o p s a " u t i l i t y - p u m p " a r g u m e n t t h a t

3 2 3

M I C H A E L A L M E I D A A N D M A R K B E R N S T E I N

shows how moral agents reasoning according to principles such as l i P leave

themselves open to exploitation. Suppose mental state M] is not perceptibly

worse than M o. An agent who places even the slightest utility on m o n e y

would exchange M o for M 1 and a very small amount 9. Offered the chance to exchange M~ + 9 f o r M 2 + 29, the agent reasoning according to IIP w o u l d

again make the exchange. As the agent approaches the exchange up to Mj00o + 10009 he might well be in excruciating pain and only slightly richer. Still IIP would have him continue to exchange upward since no e x c h a n g e offers

any cost in utility and each exchange offers a nominal gain. But the agent's

willingness to exchange M ~000 + 10009 for the complete relief in M 0 shows the error in putting the value o f each imperceptible increment o f pain at

zero (or less than he value ofg).

It might be urged that a generalized utilitarianism could solve this problem.

Preventing " a l l " o f these utilitarian agents from causing pain to A , B, and C is better than preventing none. But o f course preventing 'all except o n e ' is

no different from preventing ' a l l ' . So generalize utilitarians have no better

reason to prevent all from causing pain than they have to prevent all except one. But then they have no better reason to prevent all except one than they

do to prevent all except two, And so on. So generalized utilitarianism offers no obvious solution to such imperceptibility problems.

3 2 4