Moral Value Monism

Chapter 4

Moral value monism

4.1 Intro duction

So far, I dealt with two of the three main traditional background assump-

tions of the ideal of equality: moral universalism and volitional individu-

alism. Two alternatives were intro duced: moral realism and realistic in-

dividualism. Both are based on the Wittgenstein-Davidson approachto

language and interpretation. In this chapter, the third assumption, moral

value monism, is discussed. Value monism was held to b e one of the tradi-

tional background assumptions of the ideal of equality b ecause of the idea

of a contract. In contracts the go o ds for p eople are compared to each other

and in contracts it is agreed that the loss of one go o d is comp ensated by

some other go o d. If you giveupyour right to that which will b e advan-

tageous for me, then I will giveupmy right to something else in return.

Such a trade-o suggests the idea that there is one value with resp ect to

which all go o ds have to b e compared. This moral value monism can also

b e discerned as the assumption b ehind the discussion on the prop er equal-

isandum; it seems that it is assumed that there is just one equalisandum.

In this chapter moral value pluralism is intro duced as an alternativeto

moral value monism. It leads to the idea that there are several equalisanda

instead of just one. This will have in its turn as a consequence that the

articulation of the ideal of equality has to b e indep endent of a particular

equalisandum.

This chapter consists of two parts. In the rst part, the reasons for

accepting moral value monism are examined and moral value pluralism is

intro duced. I follow mainly Sto cker's line of argument against monism

[Sto cker, 1990]. Two main reasons for value monism are discussed. One is 115

116 CHAPTER 4. MORAL VALUE MONISM

based on the idea that the main concern of morality is that it should b e

action-guiding and the other on the idea of comp ensation. Of b oth it is

shown that they are not valid. Subsequently I present an argument against

monism that is based on a discussion on the phenomenon of akrasia or

weakness of will. It is shown that although in spite of what is commonly

held, monism can account for akrasia. It is argued that monism cannot

account for all common sorts of con icts and genuine choice. At rst sight

it seems that an explanation for moral con icts and genuine choices calls

for moral universalism, but as is shown, moral particularism within a moral

realistic framework can account for these con icts and genuine choices to o.

In the second part, it is argued that there are several equalisanda instead

of just one. I arrive at equalisanda by discussing an issue touched up on in

the previous chapter, namely the role of resp onsibility in the ideal of equal-

ity and the prop er equalisandum. As was argued in the previous chapter,

incorp orating resp onsibility in the equalisandum threatens the pro ject of

this study b ecause evaluations of distributions without taking their history

into account b ecome imp ossible. Several recent prop osals for an equal-

isandum, notably those of Rawls' primary go o ds, Dworkin's resources and

Cohen's advantages, are discussed. Scanlon's idea on the equalisandum

exp osed in his Preferences and urgency [Scanlon, 1975] and his idea on

resp onsibility explained in The sigini cance of choice [Scanlon, 1988] are

presented. They are of help in arguing that there are several equalisanda

and that resp onsibility is not to b e incorp orated in the ideal of equality.

Thus the threat to the aim of this study, the developmentofanevaluation

of distributions that is pattern like, is neutralised. In this discussion on

the prop er equalisandum and resp onsibility, I arrive at a plurality of equal-

isanda. They can b e subsumed under the name of `lib erties', referring to

real p ossibilities for individuals to act or enjoy situations that are valuable,

without su ering from disadvantages that others do not su er from.

4.2 Monism versus pluralism

4.2.1 Reasons for value monism

One of the main reasons for accepting monism is that it is held to b e ne-

cessary for a de nite answer to the question: `What to do?', to which the

answer is of course: `Cho ose the b est option!'. A complete ordering of avail-

able options is a guarantee that there is a b est option that should b e chosen.

It is held that such a complete ordering assumes just one value, whichis

the value according to which the options are evaluated. If more values

1 Chapter 3 p. 89.

4.2. MONISM VERSUS PLURALISM 117

havetobetaken into account then a b est option is not any longer guar-

anteed and some arbitrariness is held to b e inevitable and morality lo oses

its de nite action-guiding character. Once taking morality as essentially

action-guiding, one value has to b e assumed otherwise it could happ en that

morality ended up with arbitrary choices, thus the argument. This value

is taken to b e the central value of morality. In history it can b e seen that

several suggestions have b een made for this value, happiness, eudaimonia,

welfare, to mention only a few. Action-guidingness as the central idea of

morality is given as one reason for monism.

The second reason for value monism is more or less particular for the

discussion related to equality. It is based essentially on the same assumption

ab out comparing options and go o ds as the one of the rst reason. In using

the ideal of equality the idea of comp ensation is a central one. It is held

that it has to b e p ossible that lack of some sort of go o d can b e comp ensated

by some other go o d in order to arrive at equality. Comparing several go o ds

in order to comp ensate, assumes one value with resp ect to which the go o ds

and lacks are compared. If there existed more values then there would b e

no guarantee that the lack can b e comp ensated prop erly by some go o d.

Wiggins for example formulates this idea as the principle of comp ensation

in kind:

..., if course x is b etter in resp ect of eudaimonia than course

y, then there is no imp ortant disadvantage that x has in com-

parison with y, or no desirable feature that y o ers that x do es

not o er to o, byway of an equal or greater degree of that very

feature [Wiggins, 1982b, p. 259]

Such a principle can b e seen as a reason to o for value monism. Both reasons

are discussed subsequently.

De niteness of morality

As mentioned ab ove, one reason for monism is the idea that morality should

give a de nite answer to the question: `What to do?' A complete ordering

of options will secure a de nite answer, b ecause a complete ordering always

has a b est element. This element, the b est option, is of course the answer.

But it is not clear at all that the only concern of moralityistogive de nite

answers to the question: `What to do?' And even if it is, it is not clear that

the b est has always to b e chosen and even if the b est has to b e chosen it

is not clear why a complete ordering on one scale means value monism. It

could b e p ossible that there is a complete ordering without there b eing one

sup ervalue which is resp onsible for this ordering.

118 CHAPTER 4. MORAL VALUE MONISM

Let me rst turn to the idea that ethics is concerned with de nite an-

swers to the question: `What to do?'. It can b e admitted that ethics is

concerned with this question, but it is not its only fo cus. Morality is con-

cerned also with judging p ersons, virtues, or asp ects of actions instead of

the action itself, or that it is concerned with what is worthwhile to strive

at. It is not merely the particular action which is of concern to morality

but also the question: `What if the circumstances were di erent?'

For example supp ose as a derivativeofRawls' theory that the one and

only value to b e taken into account is the well-b eing of the worst-o p erson.

Now it can b e argued, and in fact some do, by p ointing to incentives,

that we should divide go o ds not equally, b ecause then the worst-o would

be even worse o . It is b etter even for the worst-o that go o ds are not

divided equally, b ecause equalitywould lead to a decrease of the incentive

to work and so to a diminished total amount of go o ds to divide, resulting

in the worst-o receiving less. Thus is argued that there would b e less to

divide b ecause the incentive for pro duction would b e far less with an equal

distribution. Against this reasoning, one can hold that p ersons should not

be moved by their own income but they should b e moved by the well-b eing

of the worst-o . If that value were the incentive one could and should

distribute go o ds equally. The incentive argumentwould not b e any longer

a reason for di erences. On the other hand, one knows that p eople should,

but in fact are not motivated to work for the well-b eing of the worst-o

and so in fact the worst-o will actually end up worse o in case go o ds are

distributed equally and the distribution should not b e equal.

In this example one could say that you should divide the go o ds equally

and you could say that you should not divide the go o ds equally. The

former gives the answer to the question: `What has to b e done idealiter?',

the latter: `What has to b e done given that p eople inhabiting this world

are not p erfect?' But even if one admits that given the feasible options,

go o ds should not b e divided equally, it still makes sense to remark that

idealiter go o ds should b e distributed equally. An equal distribution is not

simply worse than an unequal distribution. It is worse b ecause we are not

p erfect. That is an imp ortant supplement to the judgement that an unequal

distribution is b etter. The judgement that it is b etter not to divide go o ds

equally, is not complete. One asp ect is missing by neglecting namely the

judgement that idealiter equalitywould b e b etter.

Not all moral judgements are action-guiding, they also concern judge-

ments. Admitting this, one reason for the requirement of monism in ethics

is undermined, although it is not yet shown to b e false.

This example is derived from Cohen's interpretationofRawls' di erence principle

in which the role of incentives cannot b e used as a phenomenon to which the di erence

principle can b e applied if it is cited by those whose incentives it concerns [Cohen, 1993].

4.2. MONISM VERSUS PLURALISM 119

Let me for the sake of the argument, admit that morality is concerned

exclusively with the question: `What to do?'. After all in this study I

am concerned with evaluating options i.e. distributions, with resp ect to

the ideal of equality. The next question then will b e: `Is monism the

appropriate way of arriving at de nite answers?'

It is clear that if one has a complete ordering of available options there

is at least one maximal element that is b est. A common de nite answer to

the question: `What to do?' is: `The b est available option.'

Although maximisation of some go o d and striving for the b est is seem-

ingly obviously the right metho d, in fact it is not the only right one. It

is argued against the maximisation view, according to which the option

with most of the value should b e chosen, that it can lead to worse results.

For example, p erfectionism leading to sp ending an in nite amount of time

on some task in order to ful l it p erfectly well, will lead in the end to

nothing. Maximisation is, to b orrow the concept from Par t, indirectly

self-defeating. It learns that in order to arrive at the b est we should not

strive for it [Par t, 1984, p. 5]. It would disturb in a sense the relationship

with ourselves. We could not describ e directly to ourselves what we care

ab out i.e. the b est. The relationship with other p ersons would b e disturb ed

to o. For example, giving up friendship just b ecause one can develop a b et-

ter one undermines the whole idea of friendship. This is at least as serious

as indirect self-defeatingness. Best can b e the enemy of good. Beyond these

problems for the maximisation view, sup ererogation is mentioned against

this view. With sup ererogatory acts are meant acts that are not obligatory

but nevertheless very go o d. Such acts are p erformed by moral saints. The

argument states that the distinction b etween what is very go o d and what

should b e chosen cannot b e made in the maximisation view.

The phenomena cited ab ove could b e accounted for if not only the b est

should b e striven at but `go o d enough' would b e enough and should b e

aimed at, i.e. some sort of satis cing theory. In this view p erfectionism is

not required, it do es not require to give up some friendship for a slightly

b etter one and sup ererogatory actions can b e seen as those actions b eing

b etter than just go o d enough.

The maximisation view can o er two arguments to rebut these attacks.

It can argue that maximisation is still not to b e rejected b ecause of the

examples cited to show maximisation is wrong, b ecause the examples can

b e accounted for in the maximisation view. The other argument takes the

satis cing theory to fo cus on the wrong value that is assumed to b e central.

These arguments are discussed subsequently.

The examples cited ab ove can b e accounted for in the maximisation

view by considering the actions whichhavetobechosen according to the

satis cing theory, namely to b e the b est of the available options. Not b eing

120 CHAPTER 4. MORAL VALUE MONISM

a p erfectionist is the b est option available to us. Not changing friends just

b ecause one p erson would b e a slightly b etter friend than another, is the

b est available option. Similarly, the alternative which is not sup ererogatory

is the b est available option given the fact we are not saints, although of

course it would b e b etter if wewere saints; of course we can b e ashamed of

not meeting these high standards. Sup ererogatory acts can b e accounted for

in the maximisation view as actions of whichitisapitywe cannot p erform

them b ecause we are not p erfect, just as in the case of admitting that the

option of distributing go o ds equally should not b e chosen, b ecause it is not

the b est option given the fact we are not motivated as we ideally should b e.

The conclusion is that the examples cited to show the maximisation view

is not covering moral phenomena, are not conclusive, the phenomena can

b e accounted for in the maximisation view. But is this answer sucientto

save the maximisation view?

It can b e argued that the maximisation view is still not saved, b ecause

this view, in which the option which should or could b e chosen is considered

to b e the b est option, is still defective in a certain sense. The satis cing

theory can make more and ner discriminations than the maximisation

view. If for example, an option which has enough of some value F in order

to b e chosen, then according to the maximisation view this option would b e

maximal, and it would not makeany sense to say that another option has

more value. In the satis cing view one can say that this option has enough,

but some other option also having enough value in order to b e chosen, has

even more value. In the maximisation view, these two options would have

an equal amountofvalue, one could not discriminate b etween them. In the

satis cing theory one can. It allows for more and ner discriminations and

makes this view preferable, b ecause it ts b etter in our ways of reasoning,

in whichwe make these ner descriptions.

It could b e argued that these ner discriminations could b e made also

by the maximisation view. It could b e held that b eyond what is b est

for doing, an option can have more of some other value than the value

`to b e p erformedness' than the other option. But by this resp onse value

monism is left and value pluralism is turned to and the motivation for the

maximisation view is undermined.

Maximisation is confronted with some problems even if the examples of

self-defeatingness of p erfectionism, friendship and sup ererogatory acts can

b e accounted for. Let me turn to the other argument that is mentioned in

defence of the maximisation view.

Against the non-maximisation view cited to question the maximisation

view, it is argued that if some option with less of the go o d than the maxi-

mum is b etter than the option with the maximum, then there is some other

value with regard to which options are judged and not just the go o d which

4.2. MONISM VERSUS PLURALISM 121

was meant to ful l that role. To b e more formal, if an option with a certain

amountsay n of the value F is b etter than the option with the most amount

of this value say m, then this b etterness of that option assumes some other

value V so that V nF V mF even if m> n. So the wrong value was

taken to b e the only one, it should b e V instead of F . By taking the right

value V instead of F the maximisation theory app ears to b e the right one.

Although the reasoning of the maximisers seems to b e quite convincing,

there is still a problem. This reasoning shows to o that maximisation is not

conceptually true either as is claimed by the maximisers, b ecause saying

that some option is b etter than some other option b ecause it has some more

value V , also assumes some other value W in which this `b etterness' of the

option is captured. The maximisation view leads to an in nite regress.

However contrary to Sto cker's opinion, this is not the main problem, for

the regress argument can b e rebutted by reinterpreting the value of the

maximisers. The value meantby the maximisers can b e taken such that it

is necessarily maximised. Its maximisation is appropriate by de nition, and

thus the in nite regress is stopp ed. The main problem is that maximising

this value will not help us in decisions, it represents the end of delib eration

ab out what to do.

The p ossibility of a satis cing theory, just b e happy with enough and

not strive always to the maximum, can b e interpreted in suchaway that the

maximisers can stick to their idea that there is just one value which istobe

maximised. But this will sacri ce the meaning of this value for choices, it

will represent the outcome of delib eration and monism is no longer leading

to a de nite choice, it is the result of delib erations and do es not haveany

role in that delib eration. It would not makeany sense that without monism

we cannot determine what we should do, b ecause what we should do is not

determined by this monistic value but by some other pro cess of evaluation.

Of course there is a sense in which one can say without monism no de nite

solution, but this is only a strict logical sense similar to the way in which the

dictator in Arrow's theorem is a dictator in a sp ecial sense. He mighteven

not know that he is a dictator, or that he is in uencing the so cial ordering

in a direct way; the orderings of so ciety and the dictator happ en to b e

congruent. In a similar way the option that is maximal on the ordering

is decided to, but not b ecause it is maximal; it could have b een a di erent

one that would then b e maximal on the ordering.

So it is argued that the maximisation view could b e saved but only at

the exp ense of making the monistic value dep endent on the delib eration

itself. It do es not help in the delib eration and thus the main reason for the

See also [Bro ome, 1991, p. 17] against Philippa Foot

See chapter 3 p. 66

122 CHAPTER 4. MORAL VALUE MONISM

maximisation view is undermined. The reasoning shows that one can stick

to monism and the view that all options can b e ordered along a complete

ordering, but the value which should exhibit this ordering is not helping

in decisions; it do es not o er reasons for choice; it represents choices. One

could wonder whether it is a value at all for whichwe care. We can cho ose

up on other values as in the pluralistic variantby delib eration. This delib er-

ation can result in a complete ordering without there b eing any sup ervalue

at all, but just an ordering ab out what to do. It is not shown that this

ordering is based on merely one sup ervalue, it is just re ecting the result of

delib eration and decision. Actually, it are the other values we care ab out

and with the help of these, we try to evaluate the options in particular

circumstances.

Summarising, the reason for monism that is based on the idea that

morality is concerned exclusively with action-guiding judgements, and b e-

cause of this, de nite answers havetobegiven, namely b eing a maximum

of a complete ordering of options, whereby the ordering exhibits one value,

is not quite a valid reason. Morality is not concerned merely with action-

guiding judgements, as is shown by the p ossibility of moral con icts in an

imp erfect world. But even if it were, the maximisation view is not the

only view p ossible, the satis cing theory could b e o ered as an alternative.

The maximisation view could b e maintained but at the cost of lo osing its

relevance for delib eration and choice and the reason for monism that it is

of help in answering the question: `What to do?' would b e undermined.

A complete ordering of options do es not imply monism. There can b e a

complete ordering without there b eing one value b eing resp onsible for this

ordering. Of course one could give the ordering a name for example `to

b e p erformedness' but sucha`value' would barely have the status of value

b ecause we do not care ab out it. We care ab out the other values which

havetobeweighted in particular circumstances; the supp osed sup ervalue

`to b e p erformedness' do es not help as it supp osed to b e. Let me turn to

the other reason o ered for monism that is rather sp eci c for distributive

contexts.

Comparing and one sup ervalue

The second reason for monism is rather particular for contexts concerning

equality. In these contexts the concept of comp ensation is imp ortant. It

is held that lack of some go o d can b e comp ensated by some other go o d.

These comparisons of lacks and comp ensations call for one value that is

resp onsible for these comparisons. Without this value, such comparisons

are argued to b e imp ossible. As a consequence of this thought, it is seen

4.2. MONISM VERSUS PLURALISM 123

that the discussion ab out the equalisandum, what has to b e distributed

equally, is essentially ab out the nature of this value.

Mainly twoways of comparing are in uential, one is in terms of mone-

tary value, the other is more directly in terms of welfare. In the previous

chapter, we met these already in discussing the non-envy analysis as re-

sources and income which are held to b e easily comparable and welfare

which is held to b e non-comparable.

According to the rst, money is the go o d to which all others are com-

pared. The value of all go o ds is measured in terms of money,itisthe

price of the go o ds and that price is determined by the free and comp etitive

market. It represents howmuch the p eople care ab out a go o d. The value

is determined by the alternative uses of this go o d i.e. the monetary value.

The value is measured by what one could do by trading that go o d for some

other go o d. It shows the purchase p ossibilities, which on their turn show

the p ossibilities of choice. It will b e clear that this view is asso ciated with

a resourcist theory of distribution according to which the equalisandum are

the resources. We met this kind of theories in the previous chapter. So,

one prop osal for the central value in distribution problems is money. Ac-

cording to this prop osal the prices settled by the market determine the just

comp ensation.

The other way of comparing lack and comp ensation is not through

money but by some rather abstract value. Comp ensation and lack are

compared on some scale such that the lack is considered to b e comp en-

sated if and only if the option with lack together with the comp ensation

for this lack, is equivalent on that scale to the option without this lack and

without its comp ensation. This comparison is more directly linked to what

some go o d or lack means to someone, it is not directly inspired by the idea

of alternative uses as in the previous metho d. One could call this value

welfare.

The question to whichwehave to turn here is: is this reason for compar-

isons sucient for monism to b e accepted? Let me turn rst to the metho d

which is mediated by the free and comp etitive market and subsequently to

the more direct metho d of comparison.

The rst metho d of comparing lack and its comp ensation by the mar-

ketprices is criticised on the ground that this way of measuring is to o much

in uenced by irrelevant in uences i.e. by for example how other p ersons

evaluate some go o d. If almost nob o dy cares ab out something, for instance

See chapter 3 p. 67.

This idea of comp ensation is for example also seen in Nozick's idea of comp ensation

[Nozick, 1974, p. 63]. Kolm argued that this idea ends in egalitarianism [Kolm, 1985,

p. 162 .].

See [Kolm, 1984, p. 60 .] for some problems of the idea commonly advo cated in

124 CHAPTER 4. MORAL VALUE MONISM

philosophical ideas, the demand for it is negligible and consequently almost

worthless. Furthermore, prices are determined not only by the demand in

particular circumstances but also by the supply.Water in the desert is

much more valuable than in Holland nowadays. This means, that b ecause

supply and demand are not intrinsic features of go o ds, the prices are not

intrinsic features of these go o ds. Go o ds cannot have prices in general but

only in particular circumstances. The prices are dep endent on the particu-

lar situations, which are external to the go o d. Hence this way of measuring

the value of something is not re ecting the value of the go o d itself.

A resp onse to this ob jection of prices b eing to o much arbitrary could

b e that go o ds have to b e particularised to a particular time and situation,

meaning that water in the desert is some other go o d than water in Hol-

land on one of its rainydays. This will account for some of the external

in uences, but the in uence of how others care ab out something is not neu-

tralised. For example, lo osing a photograph having merely value to you but

to nob o dy else would mean that you lost nothing valuable. Such a loss has

not to b e comp ensated.

But apart from the problem that the value of something that has only

some value to you would b e determined by what others want tot give for

it, monism is not yet argued to b e reasonable. Supp ose one could in a rea-

sonable way compare go o ds mediated by money, whether or not via market

prices, would this mean that money was the only value? The answer can

b e clear, monism would not b e implied. Money is just a way of measuring a

certain amount and it do es not mean it is the only value. Instead of money

chickens could have b een chosen, which do es not mean that only chickens

matter in this world. Even if all values can b e represented by money, then

it is not implied that money is the only value. So, the metho d of com-

paring go o ds through prices on the market is not a reason for adhering to

monism. Let me turn now to the second metho d of comparing lacks and

comp ensations.

By the second metho d the lack of some go o d and the comp ensation

for it is more or less directly compared. If the option with lack together

with comp ensation for this lack and the option without lack and without

its comp ensation are equivalent, then the comp ensation is sucient for the

lack. A complete ordering will, in combination with some sort of continuity

prop erty for comp ensation, ensure that there exists for each lack a sucient

comp ensation. This complete ordering is, just as in the previous section,

re ecting a value. As in the previous section, we can argue that this value

is not the reason for the comp ensation to b e sucient for the lack. The

lib eral environments that the value of a go o d is determined by the market prices.

It is tempting to take this constructed `value ' such as `to b e p erformedness' as

4.2. MONISM VERSUS PLURALISM 125

comp ensation is not mediated by asking for equivalenve regarding this value

but by considering the lack and comp ensation directly. It is not asked how

much of this value is lacking and thus such and such comp ensation is right.

The delib eration is p erformed on the ground of evaluating the lack of some

go o d and the p ossible comp ensations in a particular circumstance.

Here again the idea that a complete ordering implies there to b e just

one value, is used to conclude to monism. But as stated already, in order

there to b e a complete ordering in a particular situation of options, the as-

sumption of there b eing just one sup ervalue is not a necessary one. It is not

implied that all options from all situations can b e arranged in a complete

ordering on the basis of one value as would b e held by monists. It is not

valid to conclude from the idea that for all situations there is a complete

ordering or scale of comparing based characterised by a particular value or

go o d to the idea there is one ordering or scale for all situations characterised

by one value. It is quite intelligible that in di erent situations, go o ds are

di erently evaluated vis- a-vis each other. Actually, in the previous chapter

this was illustrated. It was shown that the evaluation of dioptrics against

ampli cation of acoustic signals is dep endent on the particular situation,

i.e. whether the visit to the theatre concerns a pantomime, a concert or a

play.

The reason for monism that is based on comparing and determining

comp ensation for lackofgoodsisnotaconvincing reason. It is even more

serious for monism. It is not only that monism do es not follow from the

p ossibility of comparing di erent go o ds, it makes even talking ab out lacks

and comp ensations imp ossible.

In monism in which only one value is acknowledged there is hardly

lack and comp ensation, there is just less, or more, or an equal amountof

one value. The ab ove-cited principle of comp ensation in kind recognised

by Wiggins, assumes that there is one value relevant, which implies that

there is no imp ortant distinction to b e made b etween the option with lack

together with comp ensation and the option without lack and without com-

p ensation, they just have an equal amount of the one and only value and

that is all. The whole idea of comp ensation for some particular lackis

not intelligible, there are just amounts of value, whether these are instan-

tiated by a particular combination of go o ds or some other combination, is

the equalisandum.However as is clear from the main text here and b elowitisan

empty equalisandum hiding what we care for. It do es not makeintelligible whyit

should b e divided equally b eyond the values we care for that form the basis of the

constructed arti cial `value'. It do es not add anything and it hides the `desiderative

structure discussed on p. 129 . in this chapter

See chapter 3 p. 104.

10 See the principle of comp ensation in kind cited at p. 117.

126 CHAPTER 4. MORAL VALUE MONISM

not relevant at all. All go o ds are only to b e seen as instantiations of this

singular value. And in case of comp ensation there is actually no reason to

sp eak of comp ensation of some lack. By these comp ensations in kind there

will remain no trace of the lackaswe normally recognise there to b e. For

instance, b eing comp ensated with money for having no longer p ersonal pho-

tographs, do es not exclude regretting having no photographs. This regret

is not intelligible if one is a monist. A monist can only regret having less of

the one and only value than one could have, but not that it consists of this

comp osition rather than that; only the amount of the value is relevant; its

way of instantiating is not relevant at all. This means that, although until

now I undermined the reasons for monism, I touch here up on an argument

against monism namely that monism cannot allow for our way of talking

in a sensible way ab out lack and comp ensation. This argument is turned

to in the next section.

4.2.2 Reasons for moral pluralism

In the previous sections, I have shown that the reasons for monism were

not convincing. Pluralism is still an option. It was not yet shown that

monism is less satisfactory than pluralism. In the last section, one argument

against monism was touched up on, namely the fact that monism cannot

account for lack and comp ensation. In this section I discuss this kind of

argument against monism more extensively. The intro duction to this eld

of arguments against monism is the phenomenon of akrasia, weakness of

will, which consists of doing something of which it is clear, also to the actor,

that it is not rational to do. For example staying in b ed while it is clear

that getting up is much b etter. This phenomenon is cited in order to show

that monism is wrong, for example by Wiggins [Wiggins, 1982b, p. 262 .].

Davidson stated a view on akrasia, which also accounts for moral con icts

in general [Davidson, 1970]. But as is shown shortly this view of Davidson

elab orated by Jackson lures us to moral universalism [Jackson, 1985]. So,

with akrasia we meet two theoretical problems: monism versus pluralism

and moral universalism versus particularism.

The main problem now is to articulate a view that can account for moral

particularism, moral pluralism and moral realism, of whichitwas argued

that it would b e convenient for the development of the ideal of equality.

Let me turn to the phenomenon of akrasia for the discussion monism versus pluralism.

4.2. MONISM VERSUS PLURALISM 127

Akrasia and moral pluralism

In akratic actions showing weakness of will, for example remaining in b ed

while it is known to b e b etter to get up, the actor is acting irrationally.

On the other hand, there is something in the akratic action which makes it

attractive; it is not just irrational. Remaining in b ed gives some pleasing

b o dily sensation of warmth and rest that is disturb ed by getting up. But if

there was merely one value then it is not explicable why the actor remained

in b ed; there would b e nothing attractive to that action. The akratic action

has less value than getting up, so there would not even b e a single reason for

the actor to cho ose for the akratic action; it would b e highly unintelligible.

But akratic actions are known to b e p ossible and also tempting and not

merely irrational. It seems that only intervention of some other value can

account for akratic actions. Getting up is b etter, but the b o dily sensations

interfere. Brie y, this is the reasoning against monism based on akrasia.

This argument based on akrasia is not valid in this form. Some com-

ments have to b e made in order to get the problem of akrasia clear. It

can b e admitted that something has to intervene b etween seeing what the

b est option is and cho osing the akratic action. But it is not necessarily so

that some other value has to intervene. Akratic actions although under-

standable, are nevertheless irrational actions, one could for example, p oint

to some rather common psychic defect. It is not without reason that it

is weakness of will, which is in a particular way irrational and not to b e

admired. Hence it is not clear at all what akrasia has to do with monism.

The real problem will b ecome clear if it is recognised that within monism

there is never genuine choice at all. With genuine choice is meanta choice

between alternatives which are b oth attractive in their own way.Was there

merely one value, a maximiser could not describ e choices as choices at all.

Only the b est would b e attractive, there would not b e any genuine choice.

The alternatives b earing less than the maximum of value are not really

alternative options, and if the alternative options are equivalent the di er-

ences don't matter and there is no genuine choice either. So the problem

p ointed to by akrasia is not akrasia and its irrationality, but concerns dif-

ferences in the attractiveness of actions. It is even more clearly seen in far

more common situations of choice and delib eration which do not concern

irrational actions. The reason why akratic actions call for pluralism also

showwhy the more common cases of genuine choice, even if there is no

con ict of reason, call for pluralism.

So far, it is still not yet shown that akratic actions require pluralism.

That is not bychance, b ecause it cannot b e shown. If some other way

than maximisation of the monistic value is agreed to b e p ossible, for exam-

ple a satis cing theory, then akratic actions can b e accounted for within

128 CHAPTER 4. MORAL VALUE MONISM

monism. The akratic action could b e considered to b e just satisfying while

an alternativewould b e judged to b e b etter or even b est. Deviating from the

maximisation view will not require pluralism for explaining akrasia. And as

made clear by Sto cker, sticking to the maximisation view is not of any help

in making clear how within pluralism akrasia is explained [Sto cker, 1990,

p. 242 .].

Supp ose one is a pluralist and a maximiser, then if the values in the

akratic action and the alternative action that is b etter are not comparable,

then the statement that one option is b etter than the akratic cannot b e

made seriously. On the other hand were the values comparable than the

akratic action would lo ose its attractiveness vis- a-vis its alternative in the

maximisation view. So, akrasia is not explained by pluralism either. The

call for pluralism is not dep endent on the phenomenon of akrasia.

Moral con ict, genuine choice and moral pluralism

Instead of arguing in favour of pluralism b ecause of akrasia one can b etter

argue for pluralism b ecause we recognise that there are moral con icts and

genuine choice. Although as stated already, it is not true that all forms of

moral con ict are imp ossible within monism, a limited sort of con icts is

p ossible. The con icts which can b e accounted for in monism, as was stated

ab ove, are those for example in which it is regretted that there were no other

alternatives available than there actually were b ecause of the imp erfection

of the p eople inhabiting this world. But we recognise there to b e other

forms of con ict that can b e detected, even in normal reasoning in which

there is genuine choice. Precisely this was seen to b e imp ossible in monism.

Monists could argue that some more con icts than those b ecause of

imp erfect p eople, are p ossible. If there is just one value then it is not

excluded that there can b e a con ict ab out who is getting that value, or

to whom it b elongs, or applies. As for example in the argument to answer

Williams' argument against the p erception view of ethics in which doing

philosophywas the only value, there was still a con ict p ossible, b ecause

doing philosophy is not something merely abstractly in the air, but is linked

with p ersons doing philosophy and it can b e the case that one p erson doing

philosophy excludes another doing the same. Con icts can arise b ecause

the values can b e linked with di erent p eople, thus a monist.

Although it seems to b e a nice defence for monists, it is not a go o d

one. It is even showing monism to b e wrong. The indexicality of some

value with resp ect to p ersons p oints immediately to the issue of a fair dis-

tribution. The question of the distribution has to b e solved byinvoking

See p. 118 in this chapter.

See chapter 2 p. 49.

4.2. MONISM VERSUS PLURALISM 129

some other value than the indexed one. The indexed value cannot func-

tion itself as a canon for its own distribution, just b ecause it is indexed.

Some other value has to b e invoked, for example equality or some other

distribution relevant notion. It can even b e stated that the whole idea of

equality presupp oses value pluralism, namely b eyond the value `equality',

some value that has to distributed equally. In this lightitiseven strange

that the whole discussion ab out the equalisandum seems to assume there to

b e just one equalisandum, while monism has already to b e abandoned for

acknowledging the ideal of equality. Returning to the p ossibility of con icts

for monists, it is concluded here that monists can not account for con icts

which came into existence through indexing the value. Similarly, indexing

some value with resp ect to time and place will give the same problems as

indexing with resp ect to p ersons. Summarising monism can allow for

some con icts but not for genuine choice and moral con icts resulting from

con icts of reasons. But what do these con icts and genuine choices show?

How should they b e describ ed?

Re ecting on what monism is missing, namely the capacity of making

moral con icts and genuine choice intelligible, leads to realising there is lack

of so-called desiderative structure, a term used byPettit. Pettit uses this

term in order to indicate some defects in decision theory as a full theory of

explanation of actions. Like decision theory in which that action is to b e

chosen which maximises exp ected utility, monism explains that an action

has to b e chosen by citing: `It is the b est'. But this explanation is hardly an

explanation b ecause an option is necessarily taken to b e b est if it is chosen,

just as in decision theory it is necessarily the option with the maximum of

exp ected utility. The reason why remains obscure.

Within momism actions are seen as b earing one value and all other prop-

erties are not relevant for delib eration and choice. Only b eing a b earer of a

particular amount of that value is interesting. So, any explanation in virtue

of what an action is seen as go o d is missing. But this reasoning is defective

b ecause in our daily life we do delib erate on actions and evaluate them

b ecause of their particular prop erties. They are right or wrong in virtue of

This recognised byPar t in [Par t, 1989].

Sto cker argues that di erences in time are mostly di erences in the comp osition of

some go o d and so contradictingvalue monism according to which comp osition do es not

matter at all. It matters much if someone has a desert, say ice, after dinner or b etween

the ap eritif and the soup. The whole meal changes, it is not merely a matter of time. In

this way time do es matter and a change in the ordering of time changes the whole go o d

and the whole value of some complex go o d. Here to o indexicality will involve pluralism,

b ecause of the change of the whole meal by p ermutating the parts of a meal in a di erent

order, the value is changed. There is not one go o d indexed to times but there is some

value consisting of the whole complex, just as was seen to b e the case in the issue on

distribution resulting from indexing the value to p ersons [Sto cker, 1990, p. 250-260].

130 CHAPTER 4. MORAL VALUE MONISM

something. And this something is the reason for considering the action as

desirable or not. In virtue of these prop erties actions are further delib er-

ated ab out or not. Those prop erties matter and are not merely the b earer

of the single value re ecting for example the value `to b e p erformedness'.

Apart from repairing the lack of reasons for choice by this delib erative

structure Pettit argues also in favour of this structure b ecause it can explain

why in attitudinal prop ositional contexts di erent descriptions of the same

action cannot simply b e substituted without altering its truthvalue. For

example if telling a certain lie in a particular situation happ ens to b e the

same action as saving someone's life, then the description of saving a life

cannot b e substituted simply by the description telling a lie. `I am happy

to havesaved a life' is not similar to `I am happytohave told a lie'. These

cases of non-substitutability show prop erties of actions to b e relevant.

Another argumentofPettit for this desiderative structure is that choices

which seem to b e irrational are by considering them more precisely and

acknowledging moral relevant prop erties, app ear to b e p erfectly rational.

For example one of the axioms for rationality is the sure thing principle

meaning that if some alternative is preferred just b ecause of the discerning

prop erties of the alternative it is also in another situation in which it is also

the discerning prop erty.For example if you prefer a b et 1 to 2:

20 chance: 80 chance:

1: an ice cream of strawb erry liquorice

2: an ice cream of vanilla liquorice

then it would b e strange if you did not prefer b et 3 to 4:

20 chance: 80 chance:

3: an ice cream of strawb erry a bar of cho colate

4: an ice cream of vanilla a bar of cho colate

Of course indep endent of what is won in the 80 chance case, you prefer

strawb erry ice to vanilla. So the sure thing principle is formulated as a rule

of rationality.

Consider now the following example from Diamond, in whichyou prefer

distributing some go o d by lottery to giving it directly to one of to p ersons

Mary or John. This would b e irrational. Because if you prefer b et 1 to 2:

head: tail:

1: Mary wins John wins

2: John wins John wins John gets it whatso ever

then you should, following the sure thing principle, prefer 3 to 4:

4.2. MONISM VERSUS PLURALISM 131

head: tail:

3: Mary wins Mary wins Mary gets it whatso ever

4: John wins Mary wins

But b ecause of fairness, you intro duced the lottery and you just prefer 1

to 2 and 4 to 3. The prop erties of the options are relevant, not only the

bare options characterised by their abstract formal structure. Desiderative

structure in which prop erties of options play an imp ortant role, is prop osed

byPettit to complement decision theory.

In the desiderative structure, actions are judged in virtue of their prop-

erties, which means generally b eing go o d qua action b eing of a particular

typ e say instead of go o d simpliciter. This makes much more con icts

p ossible than allowed for by monism. It also makes genuine choice p ossible.

In this view, an action can b e called prima facie desirable b ecause it has

some features. Con icts can arise if an action is simultaneously prima facie

desirable and prima facie undesirable, i.e. if there is a reason whyitwould

b e go o d to p erform the action and a reason whyitwould b e go o d to forego

the action. But this should b e carefully stated.

What do es prima facie go o d mean? An action of typ e b eing prima

facie right could mean that if ab out some action it is only known to b e of

typ e , it should b e chosen. Similarly, prima facie wrong b ecause of b eing

of typ e means that if the action is only known to b e of typ e , it should

not b e chosen. This is a hint to the right meaning of prima facie but it is

not capturing the idea of prima facie correctly. Stated in this wayitisan

epistemological concept.

Against this epistemological interpretation Jackson states that if one

knows that a certain action is wrong in general, and cho oses that action

and in the particular circumstances foreseen by the agent with some degree

of reliability, it turned out to b e right then there is no moral con ict.

There are no traces whichwe are familiar with in con icts, although this

action was considered as prima facie wrong, b ecause the knowledge of the

badness in general and ignorance of the particular circumstances should

have lead to abstaining from that action. The situation is more like b eing

told that some action is wrong and someone else telling you it is right. But

this epistemological problem is di erent from the genuine moral problem,

which is not merely epistemological, b ecause in these some traces remain

if an action has to b e done which is bad in some way. It is not like the

reasoning ab out Tweety that in learning that Tweety is a bird and a p enguin

something of the capacity of ying remains, the whole idea of the capacityof

ying is cancelled by learning Tweety is a p enguin. It is di erent with moral

con icts, there traces do remain. If you save a life by lying, it still remains

true although the actions was right, that it had something regrettable,

132 CHAPTER 4. MORAL VALUE MONISM

namely that it was a lie. Thus comp ensation of the lack of some go o d in

some option is comp ensation and not just comp ensation in kind as in the

principle of comp ensation in kind. But if this epistemological meaning of

prima facie is not right, how should prima facie b e explained?

One could argue that some action is called prima facie right or wrong,

if it is right wrong if all other things considered the same, the action

would b e b etter, worse. The action would b e considered prima facie

b etter in virtue of some prop erty if in comparing the state of the world

in which all other valuable asp ects remain indep endently the same, the

option with would make it b etter. In itself this cannot b e of help to

explain moral con icts, b ecause there is nothing said ab out the relation

between the judgements based on a prop erty in situations in which the

other asp ects are kept the same and judgements of situations in which

these others asp ects are varied to o. Moral universalism as was seen in the

intro duction, would b e a natural answer. b eing b etter than : means

that under all conditions C in which all other asp ects were kept constant

C; is b etter than C; : . Thus in every context an action which lacks

lacks something which an action with do es not lack. This reasoning

using separabilityofvalues as in moral universalism, can help to account for

moral con ict and genuine choice. But nowwehave a problem b ecause this

account contradicts moral particularism. Moral con icts are now accounted

for, but in a way not consistent with moral particularism.

So far, it has b een argued that moral con ict and genuine choice demand

moral pluralism. Without pluralism there would b e a lack of what Pettit

called desiderative structure. In this structure actions are judged by their

prop erties that refer to the idea of b eing prima facie right or wrong. But

the idea of b eing prima facie right or wrong seems to b e based on moral

universalism. So, the question is now: how to account for moral con icts

and genuine choice accepting moral pluralism in a way that is consistent

with moral particularism?

Moral con ict, genuine choice and moral particularism

At a rst glance it seemed that particularism would b e consistent with plu-

ralism which can account for moral con ict. By indexing the values to the

particular situations pluralism seemed to b e the result; cho osing not in the

This di erence b etween epistemological con ict and moral con ict is describ ed by

Hurley in [Hurley, 1989, p. 130 .]. Pro tanto reasons do not cancel each other completely

b ecause they grew from di erent reasoning systems each with its own central value.

Also Dancy argues that there is a di erence b etween moral con icts and epistemological

con icts in his argument against induction as a way of arriving at moral knowledge

[Dancy, 1983].

16 See p. 117.

4.2. MONISM VERSUS PLURALISM 133

light of more of some abstract value called `friendship', but b etween this

or that friendship. Because of this indexing to particular situations the

desiderative structure dissolves and only the value seen in the relation `b et-

ter to b e p erformedness' remains, which cannot allow for con icts. Actions

were merely bare b earers of this value `to b e p erformedness'. It app eared

subsequently that other reasons for actions invoke moral universalism by

taking the judgements in situations in which all other asp ects were kept

constant but the prop erty , to b e relevant for situations in which these

other asp ects C are di erent. But moral particularism denied this reason-

ing to b e valid. It is denied that judgements in situations in which other

asp ects were held constant can b e applied directly to situations in which

they are di erent. There seems to b e a dilemma: either b eing a pluralist

and b eing capable of accounting for moral con icts but accepting moral

universalism, or b eing a particularist but not b eing capable to account for

con icts. How to solve this?

Moral particularism denies that if something is a reason, or is valuable

in one situation it is necessarily a reason or valuable in any other situation.

But moral particularism cannot of course, on the cost of b ecoming senseless,

deny that what is imp ortant in one situation can b e imp ortant in an other.

Judgements in di erent situations are of course related to each other. If

this was denied, then a disastrous fragmentation of value and meaning was

implied. Nothing intelligible was p ossible any longer. Even a simple de-

scription of situations would b ecome imp ossible. By describing situations,

situations are compared to each other. If the terms used in descriptions

would have di erent meanings in di erent situations, wewould lo ose sense

of meaning, whether it concerns moral or non-moral terms.

Terms, whether they are moral or non-moral, havetobeinterpreted as

is stated byDavidson's radical interpretation view on language. They get

their meaning through interpretation for which the principle of charitywas

indisp ensable. We assume we live in the same world. Similarly,we can

see this principle to b e imp ortant for interpreting what is said in di erent

situations by ourselves or by others. We can consider each sp eaker to b e

particularised to situations. Once this is realised, the principle of charity

will care for the relation b etween moral judgements in di erent situations.

The principle of charity means that in interpreting we should take the other

or the indexed other as similar as ourselves in our situations, assuming we

see the same things, we care ab out similar things etc. Otherwise, we could

not understand what was said. This was not a choice up to us, but we do,

otherwise wewould b e solipsists not capable of understanding others, not

even ourselves. Do es this mean that radical interpretation and the principle

of charity do imply moral universalism? No, it do es not.

134 CHAPTER 4. MORAL VALUE MONISM

Moral universalism holds that a reason in one situation is necessarily

a reason in any other to o. Moral particularism do es not deny that a

reason can b e a reason in some other situation, but it is denied to b e

necessarily so. It could have b een di erent, it all dep ends on the situation.

In one it is a reason, in another it is not. The principle of charitydoes

not imply that in all interpretations there should b e the same similarities

as basis for interpretation. They can b e di erent. In for example contexts

of p olitics, so cial inequality is seen to b e an imp ortantvalue in another

for example in dangerous situations, safety will b e a most imp ortantvalue.

Some situations are seen as mainly determined by the context of p olitics

and some by the context of danger, rescue and survival. Here the idea of

moral vision can b e of help. Moral particularism will not deny there is

some relation b etween judgements in di erent situations, but it denies the

necessary character of it. It is not said that one can do without common

reasons. It can b e held that in every comparison or description there are

some common values or reasons, but it can b e denied that there is some set

of reasons or values necessarily the same in all. So, it is admitted that for

all limited sets of situations there is a set of common reasons and values,

but not that there is one set of common rules for all sets of situations.

The recourse to moral universalism in order to account for moral con-

icts was premature. Judgements in one situation can b e relevant for other

situations, and this is sucient. Moral universalism was not necessary.It

is sucient that it is acknowledged that there are relations b etween judge-

ments in di erent situations. In contrast to the moral universalist, who do es

not even recognise the problem, the particularist has of course to answer

the question howwe know in which situations some reasons are relevant and

in which not. The answer has to b e found in epistemology. In the second

chapter a coherentist view was argued for, according to which something

is knowledge if it ts in more prop erly with other things already known

than its alternative, whether it concerns moral or non-moral knowledge.

Moral p erception will play a role. By the question `Howdowe know when

A rather p opular idea that the adjustment that the imp ortance of the reasons can

vary is inconsistent. If reasons are imp ortant in all situations than that is also valid for

the reasons for a particular weighing of these reasons. Consequently, di erentweights

are not p ossible, b ecause that would require di erent reasons applying to the situations.

Non-monotone reasoning could b e a way out but it do es not save moral universalism as

was argued in chapter 2 p.33.

In order to preclude some misunderstandings namely that all situations are to b e

considered as similar, we should b ear in mind that this is not implied, just as the principle

of charity do es not imply there are no di erences b etween p eople. It is acknowledged

that there are di erences and that p eople have di erentwants, but in order to discern

these di erences, one has to assume a lot of similarity. Similarly, situations di er but in

order to discern the di erences it has to b e assumed that they have a lot in common.

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 135

reasons are relevant and not?', it should not exp ected to nd a general

rule which can b e applied as a kind of algorithm. As was explained in the

previous chapters such an algorithm is not likely to b e found as explained

by coherentism and particularism.

Summarising, moral particularism can account for con icts by accepting

that judgements in di erent situations can b e related to each other, i.e.

one reason in one situation is relevant for another. Although it is admitted

that it is necessarily so that some judgements are related to each other in

di erent situations, without these relationships there would even no sensible

descriptions p ossible, it is denied that these judgements are necessarily

related in this way. The dilemma, either accounting for con icts but losing

moral particularism or keeping particularism but not b eing able to account

for moral con ict, can b e solved by realising it is not a genuine dilemma.

We can accept moral pluralism within a particularistic framework.

4.3 Equalisanda instead of one equalisandum

So far, I argued in favour of moral pluralism meaning that there are more

values than one. I have shown that the arguments for monism based on

the narrow view on morality, that it has to b e action-guiding, and the idea

that comparability implies one value, are not valid. Furthermore, I argued

that monism is defective b ecause it cannot account for moral con icts and

even not for genuine choice. It was shown that monism lacks a desiderative

structure. So moral value pluralism is a more promising starting p ointin

developing ideas ab out the equalisandum or even b etter, equalisanda. Value

pluralism suggests already that there are more equalisanda than just one,

contrary to what is assumed in the discussion on the prop er equalisandum.

In the discussion on the prop er equalisandum, ingenious examples are

constructed to show that some particular equalisandum is not the prop er

one. For example, in arguing that well-b eing cannot b e the prop er equal-

isandum, it is argued that it would lead to a strange distribution if some

p erson was only happyby eating caviar and living in a palace, and someone

else was as well-o as the former in his b est days, by just sitting in front

of his little house, some piece of strawbetween his lips, lo oking at the sea

extending b efore him. The former would get to o much of the resources than

would b e fair, and it is argued consequently that well-b eing cannot b e the

prop er equalisandum but that resources are. Against the latter view that

b eing a resource is the de ning characteristic of the equalisandum some

other examples can b e given. As will b ecome clear, pluralism is of help in

solving this discussion ab out the equalisandum.

See chapter 2 p. 46 and chapter 3 p. 104.

136 CHAPTER 4. MORAL VALUE MONISM

This discussion on the prop er equalisandum concerns for its most im-

p ortant part the incorp oration of resp onsibility in the equalisandum. As

was argued in the previous chapter, incorp orating resp onsibility is a serious

complication for the pro ject of this thesis, i.e. an evaluation of distributions

without taking into account the way they came ab out. Because a pattern of

distribution could b e the result of di erentchoices of individuals and even

by gambling, as was illustrated by the Babylonian lottery, resp onsibility

cannot play a role in pattern principles. Resp onsibility p oses a problem for

the ideal of equality as treated in this study.

In the subsequent section, I discuss the prop osals of Rawls, Dworkin

and Cohen. They suggest an equalisandum in which resp onsibilityisin-

corp orated b ecause as they argue, neglecting resp onsibility will lead to an

unequal distribution. The situation in which someb o dy pays continually for

the pleasures of his lazy incontinent neighb our represents such an unequal

distribution. Thus it is argued that egalitarian ideas themselves are the

basis for taking resp onsibilityinto account. These ideas can also b e discov-

ered in the debate in Holland on the health insurance, which is aimed at

equal care for all. It is suggested for example by the Minister of V.W.S.,

Borst, that p ersons who are resp onsible for their injuries by doing sp orts

should pay for their own treatmentbyphysiotherapists, in order to make

as muchavailable for all as p ossible. I argue against these prop osals in

which resp onsibility is densely interwoven in the equalisandum. My argu-

ment is based on Scanlon's ideas on what should b e distributed equally

and his ideas on resp onsibility. I arrive at equalisanda. This is p ossible

b ecause of moral value pluralism. The consequence is that the meaning of

the ideal of equality has to b e articulated indep endent from a particular

equalisandum.

4.3.1 Sub jectivewelfare as equalisandum

Awell-known idea of equality is equalityofwelfare in which a distribution

is considered to b e rightifeveryone has reached an equal level of welfare

whereby the level of welfare is determined from the p oint of view of each

p erson's own tastes and preferences. This principle assumes a sub jective

criterion for welfare levels. The level of welfare of a p erson is determined

by that p erson himself. This idea of equality has some attractive features.

In the rst place, it takes seriously the idea we met in the previous chap-

ter, that go o ds are not go o ds simpliciter but that go o ds do mean something

to p eople. And b ecause go o ds do mean something to p eople, dealing with

Algemeen Dagblad, 7-3-1995

See chapter 3 p. 67.

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 137

go o ds on their own is not sucient for determining a fair or equal distri-

bution.

In the second place, it takes into accountavariety of tastes, interests

and preferences. This feature is particularly imp ortant in a so ciety in which

there is a great diversityofvalues to which p eople are committed, as is taken

to b e the case in our so ciety. It is commonly held that by turning to equality

of sub jectivewelfare, totalitarian intervention in p ersonal developmentin

order to reach consensual preferences can b e avoided. It is held that freedom

and autonomy are not necessarily overthrown by the ideal of equalityof

sub jectivewelfare.

Although equality of sub jectivewelfare is an app ealing idea there are

several problems. Apart from the problem of interp ersonal comparisons,

with which I dealt in the previous chapter, I mention two that are particu-

larly imp ortant in the coming dispute on equalisanda:

o ensive tastes

exp ensive tastes

The rst ob jection against sub jectivewelfare as the equalisandum holds

that there is no way to exclude claims of sadists and racists for comp en-

sation if they endure for example lack of sadistic or racist pleasure. Such

claims that are based on o ensive tastes should not count for establishing

an equal distribution. However, adherents of equality of sub jectivewelfare

have no reason to exclude these claims. The mere p ossibility of these of-

fensive tastes form an ob jection to sub jectivewelfare as the equalisandum.

The second ob jection concerns exp ensive tastes. It holds that it is not

right according to the idea of equality, to transfer commo dities from rela-

tively p o or but content p eople to rich p eople who are discontent b ecause

of their re ned exp ensive unsatis ed desires. Again it is argued that in the

idea of equality of sub jectivewelfare there is no reason for excluding such

exp ensive tastes and no reason to condemn such transfers. The recent dis-

cussion on the prop er equalisandum was fo cussed on these two ob jections

to sub jectivewelfare.

4.3.2 O ensive and exp ensive tastes excluded by so cial

ideals

One reason why o ensive tastes should not b e taken into accountby estab-

lishing an equal distribution is that such tastes are contrary to the ideals

by which they give rise to claims of comp ensation. O ensive tastes are ex-

cluded by the so cial ideals of justice. They should not b e given anyweight

in distribution problems. It is even argued that utilitarianism based on

138 CHAPTER 4. MORAL VALUE MONISM

sub jectivewelfare can exclude o ensive tastes b ecause such tastes are con-

trary to utilitarianism itself. It is b etter that no one has those tastes in

a so ciety considered from an utilitarian p oint of view [Rawls, 1971,p.30-

31]. Generalising this line of reasoning, one arrives at the idea that an

ideal P using sub jectivewelfare, can exclude tastes and preferences which

are contrary to P itself. So it can restrict itself to ino ensive tastes. This

formulation has a circular app earance. How to determine whether tastes

and preferences are contrary to the ideal if it is not clear which tastes to

include? This question however, shows merely an apparent ob jection. It

is p ossible to compare the situation in which the ideal P is satis ed and

in which all tastes are included with the one in which o ensive tastes are

not present. The latter could b e b etter according to the ideal P, b ecause

satisfaction of those o ensive tastes harm others.

For some ideals this might b e the way to exclude o ensive tastes, but

for the ideal of equality of sub jectivewelfare I am sceptical. Let me use

the ab ove stated metho d to circumvent circularity in the case of the ideal

of equality of sub jectivewelfare.

Let me compare two situations. In one situation a sadist is satis ed

in his sadistic preferences and his level of sub jectivewelfare equals that

of all the others let us say n units. I assume that such a comparison

is p ossible just for the sake of the argument. In the other situation the

sadist's preferences are neglected and not satis ed altogether and his welfare

is m units equally for all. Is the former worse according to equalityof

welfare? Of course all mightbeworse o , n

to the victims, but that is not relevant here. The question here is whether

regarding the ideal of equality the former is worse? The answer is: `No

it is not.' Both situations can b e considered as equivalent. In the latter

the ideal of equality is just restricted to taking only ino ensive tastes into

account. It could b e worse for the sadist, but only in a way which is not

relevant, b ecause his sadistic preferences should b e neglected. Circularity

is not avoided by the comparison metho d in case of utilitarianism. However

this answer is not nal.

If it is acknowledged that the ideal of equalityofwelfare is just one

among other so cial ideals, it is p ossible to exclude o ensive tastes; not by

the ideal of equality itself but by the combination of ideals in the way the

comparison metho d indicates. Ideal P refers in this case to a combination

of ideals. Supp ose wehave a combination of utilitarianism or p erfection-

ism and equality of sub jectivewelfare. It is p ossible that o ensive tastes

should b e neglected by the ideals themselves. It is p ossible that by com-

p ensating the sadist for his lack of sadistic pleasures, the total amountof

See [Nagel, 1991] for what is meant here by p erfectionism.

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 139

welfare is lower in which case there is a reason for neglecting those tastes

b ecause it is contrary to the ideals particularly utilitarianism. Of course, in

theory it remains p ossible that it is b etter to take o ensive tastes into ac-

count in case the pleasures generated by o ensive tastes outweigh the costs

in terms of welfare of others. So o ensive tastes are not de nitely excluded

but I b elieve the sting is removed from the ob jection that sub jectivewelfare

theories lackany reason for excluding o ensive tastes. It has b een shown

that there can b e a reason for the exclusion of o ensive tastes.

O ensive tastes can b e excluded by so cial ideals. Rawls' theory of jus-

tice, particularly his theory on primary go o ds can b e referred to as an

example. The primary go o ds he prop oses as candidates for b eing the equal-

isandum are not a ected by the problem of o ensive tastes. Although the

primary go o ds can b e seen as derived from sub jectivewelfare, o ensive

tastes are excluded by his so cial ideals even b efore they can get anyweigh.

This is demanded by his metho d of constructing so cial principles which

is essentially based on reasonable morally inspired p eople [Rawls, 1971]

[Rawls, 1985].

What ab out exp ensive tastes? The reasons why they should b e excluded

could b e that such tastes fall within the domain of volition. Although it

is not said as in volitional individualism that all preferences have their

origin in a well of volitions, it is stated that moral p ersons have some part

in forming and cultivating their nal ends and preferences. It is likely

the case that those with less exp ensive tastes have adjusted their lives

and their nal ends to their income or wealth that they could reasonably

exp ect. Consequently, it is regarded as unfair that they should receive less

in order to spare others from the consequences of their lack of foresightor

self-discipline [Rawls, 1985, p. 168 -169]. It is b ecause of resp onsibility that

exp ensive tastes should b e neglected in determining equal distributions.

This idea to o is incorp orated in Rawls' theory on primary go o ds. Being a

primary go o d as de ning characteristic of the equalisandum prop osed by

Rawls do es not su er the problems of o ensive and exp ensive tastes.

4.3.3 Ob jectivewelfare as equalisandum

Are the aforementioned reasons for excluding o ensive and exp ensive tastes

b ecause of their anti-so cial content and their voluntary nature resp ectively

To prevent misunderstandings on my description of Rawls, these ob jections to equal-

ity of sub jectivewelfare are not the decisive reasons for Rawls to condemn sub jective

welfare theories; they are just consequences of his argumentation. His main argument

for primary go o ds is based on the idea that primary go o ds are necessary conditions for

realising the p owers of moral p ersons and are all-purp ose means for a suciently wide

range of nal ends of reasonable morally inspired p eople. These p eople determine the

basic ideas of so cial justices according to Rawls.

140 CHAPTER 4. MORAL VALUE MONISM

indeed the appropriate reasons for exclusion? The former can b e questioned

b ecause not all o ensive tastes are anti-so cial; morality demands more of us

then just conformity to the so cial moral rules. For example, the p erson

who do es not wanttohaveany moral feelings at all has an o ensive prefer-

ence for b eing inhuman, although it is not a preference which is necessarily

contrary to so cial moral rules. The latter can b e questioned b ecause rea-

sons for excluding o ensive tastes can b e also applicable to exp ensive tastes

and could put aside the reason based on resp onsibility for one's exp ensive

tastes.

As is said ab ove, morality demands more from us than mere conformity

to so cial moral rules. It suggests that o ensive tastes have to b e excluded

directly. They are not worth to b e taken into account. So, there is invoked

a system of ideas by which tastes and preferences are judged as valuable

or worthless. It means that the idea of equality of sub jectivewelfare has

to b e abandoned and that we should use ob jective judgements ab out the

imp ortance of what go o ds do to p eople. Ob jective here means the con-

trary of sub jective and refers to the basis of interpretation of preferences

as explained in the previous chapters in radical interpretation and realistic

individualism. Ob jective judgements on the level of welfare of a p erson

means that these judgements are indep endent of this p erson's tastes and

preferences; they mayeven con ict with those p ersonal tastes. Sub jective

judgements are not to b e seen as unimp ortant altogether, but their value is

determined by ob jective judgements on for example the value of freedom of

choice and the value of the freedom of making mistakes. Sub jective judge-

ments are not any longer directly relevant for determining equal allo cations

they are relevant only via an ob jectiveevaluation.

This view is plausibly illustrated by Scanlon in taking it for granted that

we judge a claim for help for building a religious monumentby someb o dy

who forego es a decent diet for it, not a priori as strong as a claim for aid in

obtaining enough to eat [Scanlon, 1975]. Someone's own judgementonthe

imp ortance of claims is not decisive. To b e clear, it is not held by Scanlon

that b oth claims, the one for help for the building and the one for obtaining

fo o d, should not count at the same degree of urgency, but it is not a priori

necessarily so b ecause of their b eing sub jectiveevaluations. The degree of

urgency dep ends on the ob jectivevalue of such a religious building.

According to the view that ob jectivewelfare is the prop er equalisandum

o ensive tastes are excluded directly by ob jectiveevaluations. Furthermore

exp ensive tastes can b e treated similarly. On the basis of those ob jective

judgements, they do not have a high priority in a distribution problem.

Contrary to Rawls' prop osal, they should not b e excluded b ecause p eople

See for example P.F. Strawson's article: Social morality and individual ideal.

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 141

can b e held resp onsible for them, but b ecause they are less urgent. That

p eople can develop some preferences and can b e held resp onsible for them,

can b e taken to mean that p eople can do without them; they are not ne-

cessities and so lo ose some of their urgency. Resp onsibility for preferences

in itself is not a reason for rejecting claims, it is at most a sign of b eing not

very urgent if one considers ob jectivewelfare as the equalisandum.

In short, Scanlon holds contrary to Rawls that resp onsibilityisnot

densely interwoven in the tissue of egalitarian ideas, he keeps them discon-

nected. Scanlon's ob jective judgement view is simpler. It uses only one

reason for excluding o ensive and exp ensive tastes and explains the intu-

ition on which the former prop osal for exclusion was based. Of course, this

prop osal has to b e supplied with a theory that accounts for ob jective judge-

ments. Scanlon mentions two sources for such ob jective judgements, ethical

naturalism and conventionalism. The latter he thinks the most plausible

one b ecause in that one consensus has a morally prop er place, whichis

denied in the former. In the preceding chapters an alternative view was

presented, namely moral realism develop ed within a radical interpretation

framework. This view can account for ob jective judgements, i.e. moral

judgements b eing true just b ecause they are true and not b ecause someone

is convinced that they are true. What ab out this idea of ob jectivewelfare

b eing the prop er equalisandum?

4.3.4 Resources as equalisandum

The idea presented by Scanlon immediately suggests the question: `How

can we determine on the basis of these ob jective judgements whether a

distribution is equal?' This question is not easily answered. How should

one compare the welfare of the optimist with the welfare of the p essimist?

What ab out the comparison of the loss of welfare or regret b ecause of

not leading the ideal life of an ambitious p erson who judges his actual life

hardly valuable with the regret of a less ambitious p erson who judges his

actual life as quite reasonable. Should the ambitious man b e comp ensated

for regretting his actual life more than the less ambitious p erson? It is

reasonable that comp ensation should b e given, prop ortional to the degree

of di erence b etween p eople's regret leading the actual life instead of a life

which could b e reasonably exp ected. The determination however of this

reasonable exp ectation is imp ossible without assuming some distribution

of resources to b e reasonable. Because what kind of life is reasonable to

b e exp ected is dep endent on the resources one can reasonably exp ect to

have. So, an idea of entitlements to resources is assumed. Without suchan

idea no meaningful judgement on the equalityofwelfare is p ossible. This

do es not only hold for sub jectivewelfare but also for ob jectivewelfare as

142 CHAPTER 4. MORAL VALUE MONISM

it app eared in Scanlon's prop osal. Dworkin states:

Any p ertinent test of what someone should regret ab out the life

he is in fact leading, even on the b est rather than his own theory

ab out what gives value to life, must rely on assumptions ab out

what resources an individual is entitled to have at his disp osal

in leading any life at all [Dworkin, 1981a, p. 225].

On account of these arguments, equalityofwelfare, whether it concerns

sub jective or ob jectivewelfare, is dismissed as an adequate theory of dis-

tribution byDworkin. The only acceptable theory of ob jectivewelfare for

Dworkin is not a theory ab out what is valuable but consists of a list of

go o ds including for example physical and mental comp etence, education

and opp ortunities as well as material resources etc. Such a theory is a

resourcist theory in welfare language. The ob jective judgements which sat-

isfy the requirement that on the basis of them we can determine whether a

distribution is equal, are judgements on resources, thus Dworkin.

In a resourcist view, o ensive tastes and exp ensive tastes are not ex-

cluded b ecause they are not judged as valuable, but b ecause tastes and

preferences should not b e taken into account at all, only resources should.

The example of Dworkin concerning exp ensive tastes is illustrative. Sup-

p ose Louis lives in a so ciety in which there is realised a state of equality

of welfare. Now Louis is developing and growing older and happ ens to get

the b elief that it would b e b etter for him if he adjusted his tastes and he

develop ed some other more re ned tastes which happ en to b e more exp en-

sive to b e satis ed and he is aware of that. If they were satis ed his welfare

should raise considerable but if they were not satis ed his welfare would

b ecome lower. For these new insights he can hardly b e held resp onsible. To

hold that one is resp onsible for what one considers to b e true or valuable

is at least highly questionable. Although Louis cannot b e held resp onsible

for his new insights, he is resp onsible for acting up on these b elieves i.c.

developing these exp ensive tastes. Louis has a choice b etween two options:

stay in the p osition he is now in and refrain from developing the

exp ensive tastes

develop the exp ensive tastes but get less enjoyment than in the status

quo; so accept a lower level of welfare.

It is wrong according to Dworkin to give Louis a third option in whichhe

develops his tastes at the costs of other p ersons, b ecause this would dimin-

ish other p erson's shares of resources [Dworkin, 1981a, p. 237]. Exp ensive

tastes are excluded with an app eal to resp onsibility in a resourcist theory

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 143

of distribution. It suggests a tight connection b etween resp onsibility and

the idea of equality of resources, b ecause if Louis had no choice it would

b e appropriate to judge that the distribution was not egalitarian at all and

therefore he should receive some more of the resources.

In this resourcist theory resp onsibility is densely interwoven with the

idea of equality of resources. It b ecomes clearer if we followDworkin in his

distinction b etween brute luck and option luck; a distinction we already met

in the previous chapter. Brute luck happ ens to someb o dy, it is not the

result of a choice. An example of brute bad luck is a storm blowing o the

ro of of someb o dy's house or lung cancer developing in the course of a normal

life. Option luck, to the contrary, is the result of a voluntary undertaken

action. In case of for example the loss of money in a lottery, or lung cancer

after a life of heavy smoking, one has chosen an option that turned out to b e

an unsuccessful gamble. Brute bad luck should b e comp ensated, optional

disadvantages should not. Go o ds desired by one p erson can also b e valuable

to another. Tokeep a distribution right from an egalitarian p oint of view

p eople should pay the price of the go o ds they voluntary desire. They should

pay the price for the life they have decided to lead. This price is measured

in terms of what others give up in order the former can do as they want.

This idea is the basic element of equality of resources, which explains why

optional disadvantages should not b e comp ensated, but brute bad luck

should [Dworkin, 1981b]. Resp onsibility is considered to b e essential in

the ideal of equality of resources. By neglecting resp onsibility unequal

distributions will result, thus is argued byDworkin. In this resourcist

theory resp onsibility is densely interwoven in the idea of equality.

4.3.5 Access to advantages as equalisandum

We could wonder whether this interpretation of howwell-o p eople are

in resourcist terms, is appropriate; there are go o d reasons for doubt. In a

recent article Cohen shows that neither welfare nor resources are the prop er

equalisandum but that the judgements on which the distribution should b e

based concern access to advantages [Cohen, 1989].

That welfare is not a prop er equalisandum is shown by an example of

avery well to do p erson who has many opp ortunities to welfare but su ers

from a pair of paralysed legs. According to equalityofwelfare theories his

high level of welfare is a reason for rejecting his claim for help in the form

of a wheelchair. But as Cohen holds, his level of welfare has nothing to do

In the previous chapter I p ointed already to the problems with this view, here

Dworkin's theory is cited b ecause of the relation b etween resp onsibility and the

equalisandum.

144 CHAPTER 4. MORAL VALUE MONISM

with his paralysis. It is a disadvantage such that it is right to comp ensate

for it. So, equalityofwelfare is not the prop er equalisandum according to

Cohen.

In a similar wayitisshown by an example that b eing a resource is not

the de ning characteristic of the equalisandum either. The example consists

of a p erson who do es not lackany capacities and has enough resources who

can move his arms very well, but who su ers from a terrible pain after

moving his arms. According to equality of resource theories this man has

no claim to a medicine which could relief his pain, he do es not lackany

resources for which he should b e comp ensated. To describ e the p erson as

a p erson lacking the capacityofmoving his arms without pain is welfare

talk in resourcist language and will not do. But not lacking any resources

he really su ers from a disadvantage and so according to Cohen he has a

claim to the medicine. Cohen concludes that the equalisandum should not

be characterised by resources but by access to advantages.

The prop osed characterisation of the equalisandum, access to advan-

tages, do es not su er from the o ensive tastes ob jection. Unsatis ed of-

fensive preferences are not disadvantages in themselves. But contrary to

the resourcist theory such tastes could count and should count if they are

disadvantages like cravings and if a p erson would accept for example psy-

chotherapy to get rid of them. For the costs of such a therapy he should

b e comp ensated.

How ab out exp ensive tastes? Ab out these it is argued that they should

b e taken into account to the degree they are involuntary. If they are not

develop ed freely these exp ensive preferences should b e taken into account.

In Cohen's view there is a reason for helping or for giving comp ensation

such that exp ensive tastes can b e satis ed but only if these tastes are not

the result of a choice. If the preferences result from a choice no comp en-

sation should b e given. If Paul likes photography and John likes shing

and photography happ ens to b e more exp ensive then shing, Paul should

receive some comp ensation to the degree he is not resp onsible for his pref-

erences and not resp onsible for the fact that satisfying these preferences is

exp ensive. Resp onsibility cancels any claim on comp ensation. According

to Cohen it is unjust and contrary to the idea of equality that p eople should

b e exploited for the pleasure of others. This would b e the case if exp ensive

tastes for which a p erson himself is resp onsible, were comp ensated. Ex-

ploitation is contrary to the idea of equality, resp onsibility is imp ortant,

thus Cohen.

In the resourcist view sketched ab ove as prop osed byDworkin, the

distinction b etween resources and welfare, or context and p erson, or ac-

tual circumstances of life and ideas ab out what makes life valuable, is the

morally imp ortant distinction. That what is the result from particular

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 145

ideas ab out what makes life valuable should not b e comp ensated, only lack

of resources or particular circumstances of life, or in other words the ex-

tra p ersonal elements of the context should b e comp ensated for. Tastes

do not count, resources do. But according to Cohen this distinction is b e-

sides the p oint. The more fundamental distinction is b etween involuntary

disadvantages and disadvantages which one b egot as a result of one's own

choices. Dworkin's reason for rejecting claims referring to welfare; after all

these claims are based on ideas ab out what makes life valuable and thus

on intrap ersonal characteristics; are plausible in case these claims were vol-

untary develop ed. If the p essimistic and ambitious nature were voluntary

b egotten they should not count, but if this nature is forced up on p ersons

they should count, thus Cohen.

In short, Cohen holds that we should take only involuntary disadvan-

tages into account. As a consequence it is not against equality to help a

p erson building a religious building if he is not resp onsible for having this

particular b elief. It is not against equalityeven if we do not endorse the

value of the particular religious b elieve. It is however a di erent matter if

the p erson b ecause of this particular religious b elief has to su er for build-

ing the monument and that su ering is part of his religious commission. In

that case the su ering is so intrinsically connected with his religious pro ject

that it is hardly defensible that he should b e comp ensated for his su ering.

Cohen holds that the equalisandum is access to advantages, which demands

comp ensation for those disadvantages which are not due to the sub ject's

choice and which the sub ject would not cho ose to su er. The last part of

the circumscription is b ecause of the example in which the su ering of the

p erson is an inherent element of his pro ject by which he su ers and which

should not b e comp ensated.

In Cohen's line of reasoning resp onsibility is decisive for the comp ensa-

tion demanded by equality. It is given a central role in the idea of equality,

but is it a right role? Isn't it strange that egalitarianism should lean so heav-

ily on such a complex notion as resp onsibilityby which metaphysics ab out

free will and determinism is intro duced in questions of p olitical philoso-

phy? Cohen do es not b elieve this to b e a reason for rejecting his prop osal.

If equality is complex we should not deny that complexity.Furthermore he

assures us that the question of free will or choice is a matter of degree. In

a fo otnote he refers to Scanlon's Tanner lecture: The signi cance of choice,

in which free will and choice is treated in a non-metaphysical way. He

wonders whether Scanlon's approach can b e used to improve a theory of

distribution alaDworkin. As will b e made clear b elow, it cannot b e used

as such. On the contrary, it o ers further supp ort for theories of equality

in which resp onsibility is not densely interwoven.

146 CHAPTER 4. MORAL VALUE MONISM

4.3.6 Equalisanda

The idea that resp onsibility determines what should b e comp ensated is

illustrated in the discussions by examples ab out big sp enders, lazy p ersons,

p eople with irrational or even immoral desires, in short examples in which

the actions are not very praiseworthy. Bad actions, or irrational actions, do

raise the question of resp onsibility in a particular way, but one should not

forget that if the notion of resp onsibility is applicable at all, it is certainly

applicable to rational actions, actions which are p erformed b ecause of go o d

reasons [Strawson, 1986, p. 33]. Actions that are morally right b elong to

this category and if we are resp onsible at all, we are at least resp onsible

for morally right actions. By shifting our attention from examples with

irrational and immoral actions to actions which are morally praiseworthy,it

is illuminated that resp onsibility is lo osing its imp ortant role in determining

the equalisandum. Let us lo ok at an example with a morally right action

with disadvantages for its p erformer in which, in my opinion, the claim on

comp ensation for the disadvantage is not forfeited b ecause of resp onsibility.

Supp ose you are walking alongside a canal in an exp ensive suit that you

lent from a neighb our b ecause of some imp ortantevent. A child playing

near the canal falls into the water and it app ears the child is not able to

swim. You jump into the water knowing your suit will b e sp oilt for ever

and you should pay for that. The reason you jump ed into the canal is part

of a particular morality in whichsaving children from drowning has some

place. This moralitywas not installed in you; you develop ed it voluntary.

Your jump was not a re ex. You jump ed into the water to save the child

in full awareness the suit would b e sp oilt; you realised you could not have

jump ed into it, but then the child would have b een drowned, a result which

is considered to b e awful in the moralityyou develop ed for yourself.

As the disadvantage of sp oiling the suit is clearly traceable to a choice;

the disadvantage itself was of course not chosen, it was not part of the

pro ject of saving the child; comp ensation is not demanded by equality

according to Cohen's prop osal. The same go es for Dworkin's prop osal in

which the distinction b etween option luck and brute luck is relevant here.

However I do not b elieve these judgements are reasonable. Should this

saviour's disadvantage not even b e partly comp ensated, merely b ecause of

his choice to jump?

One could argue that comp ensation is reasonable not b ecause of the

idea of equality but b ecause morally right actions should b e awarded and

not b e punished. Morally right actions should b e stimulated byawards.

By refusing any comp ensation one will probably destroy the willingness

to p erform morally right actions. And thus the example ab ove do es not

show that comp ensation is required by the idea of equality. This view,

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 147

which I call the award view, has to comp ete with another view, whichI

call the p ossibility to a normal life view, which can b e considered to b e

nally more in the line of ideas of those who argue in favour of the central

role of resp onsibility in the ideal of equality. The latter view is based on

arguments for comp ensation b ecause of equality that is emb edded within

moral realism.

The man in the example, who jumps in the water, is like the man who

is able to move his arms but su ers from pain afterwards. The man in the

latter case can freely move his arms so he could hold them relaxed in which

case he do es not su er pain. In this case to o the disadvantage is traceable

to a sub ject's choice, not in the sense that he chose the pain but in the sense

that he could avoid it. Similarly, a morally right action with consequential

disadvantages is voluntary undertaken although the disadvantages are not

chosen at all. Having pain after moving one's arm is a disadvantage just

as the bad consequences after a morally right action is a disadvantage. It

suggests that we should care ab out capabilities that p eople have. Some

of them should b e cared for in the sense that they should b e p ossible for

all equally. Some activities such as morally right ones we see as valuable

and should not b e denied to some p eople b ecause the risk of disadvantages

following such actions is to o costly for them to b ear. Some activities should

b e p ossible for all. If it is admitted for morally valuable actions why not

for many more?

There are many more actions that can b e seen as valuable that bring

with them a risk of disadvantages. These disadvantageous consequences

are not chosen. The probability distribution on the p ossible consequences

is not under control of agents. They would of course prefer doing the action

without any risk of disadvantages. But in all those cases the disadvantages

are seen as the result of voluntary undertaken actions and should according

to the ideas in which resp onsibility determines the equalisanda, not b e

comp ensated.

The example of saving the child shows that it is not righttodeny com-

p ensation for disadvantages resulting from voluntary undertaken actions

b ecause they are voluntary b egotten. Resp onsibility do es not have the role

it has b een given. This is also illuminated by the example of the p erson suf-

fering pain after moving his arms and the p erson su ering lung cancer which

develops in the course of his normal life. The lung cancer in the course of a

normal life is considered byDworkin as brute bad luckincontrast to a life

in which one smoked heavily which is considered to b e optional bad luck.

In other words, it is the result of an unsuccessful gamble. What countasa

A similar line of reasoning can b e discerned in Van Parijs' argument for real lib eral-

ism, in which the essence of lib eralism has to b e concentrated on protecting real p ossib-

ilities instead of formal opp ortunities.[Van Parijs, 1991b, p. 156,184] [Van Parijs, 1995].

148 CHAPTER 4. MORAL VALUE MONISM

normal life is of course determined by what we consider to make life valu-

able. The ideas on what is valuable are necessarily presupp osed. They are

the more fundamental ones. Without these, the distinction b etween bad

luck and option luck is imp ossible to understand and similarly the p ointof

Cohen's examples is dicult to apprehend.

The examples which are meanttoshow the imp ortance of resp onsibility

for equality are apparently convincing b ecause the actions undertaken are

not praiseworthyorvaluable or necessary for the development of a normal

life. They are irrational, bad and they do not deserve to b e cared for.

It would b e even b etter if they were not done at all. It is certainly not

according the idea of equality to care for an equal p ossibility to do these

bad actions. For many other actions it is di erent and they are comp onents

of a valuable way of life.

The actions which are a comp onentofavaluable way of life, should

b e the sub ject of our concern. These should b e available for all b ecause

they are worthwhile. That what is valuable for p eople is also worthwhile to

b e distributed equally. These consideration lead me to the suggestion that

p ossibilities to p erform actions, to e ectuate choices, or to enjoy situations

are the prop er candidates for equalisanda. One could call these lib erties.

These should b e equalised. Lib erties are those choices, actions and p ossib-

ilities that are considered to b e valuable; they are the normal elements of

a p erson's life and so in a sense necessities for a normal way of life. That

they are valuable is the reason that they should b e p ossible for all. Be-

cause of that, they are subsumed under the name of lib erties and not the

other way round. We should care that they are equalised. Something is

an equalisandum in virtue of b eing valuable and not b ecause of a sp ecial

characteristic. Because of moral value pluralism we can conclude there are

equalisanda instead of one equalisandum.

A consequence of the view that what is valuable is also an equalisandum

is that resp onsibility has some role in equality but merely a secondary role.

It is not a decisive role, its role is dep endent on the judgement on the value

of choice and resp onsibility and on some idea ab out in what degree someone

should su er from his wrong actions as a kind of punishment in order to

discourage these actions. But why prefer this view to the alternative whichI

called the award view in which disadvantages that are morally praiseworthy

should b e comp ensated b ecause such praiseworthy actions should not b e

discouraged? There are some arguments in favour of the former.

As was stated ab ove, all the interpretations of the examples that are

meanttopoint to the imp ortance of resp onsibility given by Cohen and

See chapter 6 p. 199 for the idea in accord with moral realism: what is valuable for

me is valuable for you to o.

4.3. EQUALISANDA INSTEAD OF ONE EQUALISANDUM 149

Dworkin need some assumption on what is valuable in life and what counts

as a normal life. One element of a normal life is of course the p ossibilityof

p erforming right actions. Comp ensation should b e given not b ecause go o d

action should b e awarded but b ecause they are an element of a normal life

in this world. This was the p ossibility to a normal life view.

In the award view resp onsibility forfeits any claim on comp ensation for

disadvantages due to one's own choices but this do es not comply with the

idea that one wrong choice should not have such harsh consequences. By

applying the award interpretation, the natural jungle would b e replaced by

the so cial jungle. One lib erty that lo oks imp ortant to me is the lib ertyto

make some mistakes. A world in whicheveryb o dy is neurotically anxious

avoiding mistakes and in whichevery mistake results in a disadvantage, is

awful. There should b e some lib erty for p erforming mistakes. Howmuch

dep ends on our view on that matter. The idea that resp onsibility cancels

comp ensation for disadvantages is not at all an adequate view on equality.

The nal argumentinfavour of the p ossibility to a normal life view

against the award view is derived from Scanlon's treatment of resp onsibility

in his Tanner lecture: On the signi cance of choice. An example mainly

due to Scanlon can b e used to make the p oint clear.

Supp ose some area is seriously p olluted. The material is planned to b e

removed and some precautions havetobetaken b ecause during this work

some material will b e disp ersed into the op en air. If someb o dy inhales

the material he will get pulmonary problems. So, everyb o dy should stay

inside his home; the do ors and windows should b e closed. Those living

in the neighb ourho o d are warned by television, radio and newspap ers. Of

course it is inevitable that some p eople are a ected by a pulmonary disease

b ecause they were outside their home during the cleaning of the area.

One p erson did not receiveany message ab out the danger; he didn't

knowanything ab out the removal. On the evening of the dangerous

work he left his home as he did every evening.

Another p erson is very curious and did not b elief the seriousness of

the warnings. He left his home in order to see how this piece of work

was done.

A third p erson went outside after he weighed the reasons of pro and

con thereby taking the risk of a pulmonary disease into account and

judged his going outside to b e rational.

A fourth was used to go for a walk but had just forgotten the warnings.

The fth saw the other p ersons and ran outside to warn them and

probably prevented some more pulmonary damage to the victims.

150 CHAPTER 4. MORAL VALUE MONISM

All those describ ed ab ove did leave their home voluntary by their own

choice, no one was forced. The pulmonary problems are traceable to choice

Is this really the reason of forfeiture of comp ensation if there is one. Should

they pay for their own treatment really? Whydowe not argue that we

should acknowledge that we are small minded, strangely curious and very

often distracted as in the case of the rst, second and fourth p erson? Should

we really think these features as worthless? Should it not b e reasonable

to pay for the treatment then? But consider the rational and the heroic

typ es, should not we comp ensate for their disadvantages? Are rational and

virtuous actions not valuable? Of course they are. The conclusion must b e

that resp onsibility do es not forfeit the right to comp ensation. If that would

b e the case, all p ersons should pay for their own treatment. All should b ear

the costs of their choice alike. That is, I b elieve, contrary to the idea of

equality in which that what is valuable called lib erties, should b e equalised

and the comp ensation for disadvantages resulting from these lib erties form

claims to comp ensations of which the urgency dep ends on the lib erty that

was executed and the lib erty that as a result of that action is threatened.

To b e clear, I will not deny that the history of lib erties exhibiting the

way a particular situation came ab out is not imp ortant, it is. It shows

the other lib erties, whichifwe should not comp ensate, would b e threat-

ened otherwise. But to what extend its history and resp onsibility should

b e in uencing the claims on comp ensation dep ends on the value of all the

lib erties at hand. Lo oking once again at the lazy neighb our for whom all

the pleasures were paid by the work of others, indeed there is inequality.

However it is not simply restored by refusing to pay for the pleasure of

that lazy man. There is inequality in the relevant lib erties, which are: the

lib erty to enjoy pleasure and the lib erty to enjoy leisure. Which inequality

is more imp ortant dep ends on the urgency of these lib erties. The pleasure

of the neighb our to whichispointed, is a to o abstract term for a sensible

judgement on the urgency. All these lib erties do not make equality sim-

pler but if equality is complex in this way. let it b e so. Toacknowledge

this complexity is b etter than using simple slogans in which an app eal to

resp onsibility is used to prevent p eople to b e comp ensated and to prevent

care for their lib erties to which they have according to egalitarian ideas a

reasonable claim.

Summarising, it was argued that resp onsibility is not densely interwo-

ven with the idea of equality. The reason that it seemed to b e came from

one-sided examples in which only irrational actions were the causes of dis-

advantages. But once one turns to rational actions with disadvantageous

consequences it b ecomes clear that resp onsibility for a disadvantage do es not forfeit a claim to comp ensation for it.

4.4. SUMMARY 151

It was suggested that what is valuable should b e available for all are

the equalisanda. These are called lib erties, they are lib erties to certain

actions or activities or enjoying situations. The imp ortance of these lib erties

determines howmuch they count in distribution problems. Because of a

pluralityofvalues there is a plurality of lib erties, consistent with moral

value pluralism. There is a plurality of equalisanda. It will give rise to the

idea of several ideals of equality each for its own equalisandum or lib erty.

Equality concerning the more urgent lib erties weigh of course more than

those of the less urgent.

4.4 Summary

In this chapter, the third of the three background assumptions of the ideal

of equalitywas discussed. It was argued that monism was not tenable. First

it was seen that the reasons for monism were not sound. The assumption

of ethics to b e concerned solely with action-guiding judgements was not

right. Furthermore, a complete ordering do es not imply that there is one

sup ervalue. And even if suchavalue was p ostulated, this value was seen

to b e not of any help in deciding what to do, b ecause it is representing the

results of delib eration. It was mentioned that the whole idea of equality

actually presupp oses moral pluralism, b ecause there is a di erence b etween

the value of a distribution and the value distributed in that distribution.

Subsequently,itwas shown that monism is unsatisfactory b ecause it can-

not account for moral con ict and genuine choice. Pluralism could account

for these, but in a more complex way than one exp ect at rst sight. Par-

ticularism had to b e stated in a way such that it do es not mean that no

judgement is related to any other in other situations as would b e held in

the most extreme denial of moral universalism. Pluralism could account

for con icts and genuine choice within a particularistic framework if it is

acknowledged that there are relationships b etween judgements in di erent

situations. These judgements are not necessarily so related in all situations.

After arguing in favour of moral pluralism within a moral particularistic

framework, I turned to the problem of the equalisandum. Pluralism was

seen to b e of help in establishing equalisanda which could b e subsumed un-

der the name of lib erties. It was shown that the reasons for incorp orating

resp onsibilityinto the equalisandum were based on a myopic view, only ex-

amples in which morally wrong actions were lo oked at in arguing that some

go o d had not to b e given to someone. Once these morally wrong actions

were substituted by morally right actions the examples were lo osing their

force, indicating that resp onsibilitywas not the main reason for denying

comp ensation. Real p ossibilities for valuable actions or exp eriences with-

152 CHAPTER 4. MORAL VALUE MONISM

out b earing disadvantages that others do not su er from were intro duced

as equalisanda. They were called lib erties. Consistent with particularism

one could say that in a particular situation with a distribution problem

particular lib erties of several p ersons are imp ortant. For example in p olit-

ical contexts this will b e p olitical p ower, in situations of acute danger this

will b e chances of b eing rescued and survive, instead of p olitical p ower.

A consequence of this view on equalisanda esp ecially its denial of the

inherent link with resp onsibility is that evaluations of distributions remain

applicable without taking their history into account. It was argued in

the previous chapter, that in case resp onsibility had to b e incorp orated,

pro cedural evaluations or historically based principles were inevitable.

By taking these lib erties as equalisanda one can end the discussion on the

prop er equalisandum. The end-state evaluations as in pattern principles of

distributions remain p ossible. Howtoevaluate distributions regarding the

ideal of equality is discussed in subsequentchapters.

So far, I have only intro duced moral realism, realistic individualism

and moral value pluralism as alternatives to the traditional background

assumptions. Pro ceeding with the articulation of the ideal of equalityin

the second part of this study is p ossible after I end this part on the new

framework in the next chapter by answering the criticisms to moral realism

as basis for the ideal of equality.

See chapter 3 p. 89.