Michael Walzer

ipremont
I Basis of
'tudies 44
' and the
Eeonomic
Shts and
r Social
(re77);
nin and
ization,'
K.W.S.
with
Levels,"
o); PJ.
rometica
MichaelWalzer
Grading
'ith conbetween
Pluralism
and a host of goods more narrowlv and materially
conceived -
food, shelter, clothing, transporta-
"second
nrerper'ith perst off in
ranslate
Distributive iustice is a large idea. It draws the
entireworld of goods within the reach of philo-
tion, medical care, commodities of every sort, and
all the odd things (paintings, rare books, postage
sophicalreflection. Nothing can be omitted; no
f e a t u r eo f o u r co m m o n life c:r n e scxp e scr u tin \.
stamps) that human bei ngs col l ect. And thi s mul tiplicity of goods is matched by a multiplicity of,
d (1979)
H u m a n s o c i e tv is a d istr ib u tive co m m u n it.v.
T h a t ' sn o t a l l i t is, b u t it is im p o r ta n tly th a t: we
distributive procedures, agents, and criteria.
There are such things as simple distributive svs-
cometogether to share, divide, and exchange. We
alsocome together to make the things that are
shared,divided, and exchanged; but that very
tems - slave galleys, monasteries, insane asylums,
kindergartens (though each of these, looked at
making- work itself - is distributed among us in
a division of labor. NIv place in the economy, m1''
but
r979).
'iority,"
(re73);
Rapls
uality,"
standing in the political order, mv reputation
amongm-y fellows, m1' materill holdings: all these
cometo me from other men and \\'omen- It can be
saidthat I have what I have rightlv or wrongly,
j u s t l yo r u n j u s t ly; b u t g ive n th e r a n g e o fd istr ib u t i o n sa n d t h e n u m b e r o f p a r ticip a n ts, su ch ju d g -
closely, might
show unexpected complexities);
no full-fledged
human
society has ever
avoi ded the mul ti pl i ci ty. We must studl - i t al l , the
goods and the distributions, in many dilferent
times and places.
There is, however, no single point of accessto
this world of distributive arrangements and ideologies. There has never been a universal medium
mentsare never easv.
of exchange. Since the decline of the barter economv, money has been the most common
The idea of distributive justice has as much to
d o w i t h b e i n g an d d o in g a s with h a vin g , a s m u ch
medi um. But the ol d maxi m accordi ng to w hi ch
there are some thi ngs that money can't buy i s not
to do with production as with consumption, xs
m u c ht o d o w i t h id e n titv a n d sta tu s a s with la n d ,
onlv normatively but also factually true. What
shoul d and shoul d not be up fbr sal ei s somethi ng
capital,or personal possessions.Different politicalarrangementsenforce, and dilferent ideologies
justify', different distributions of membership,
men and women always have to decide and have
power,honor, ritual eminence, divine grace, kin-
tant mechanisms for the distribution of social
ship and love, knorvledge, wealth, physical secur i t y , w o r k a n d l e isu r e , r e w' a r d sa n d p u n ish m e n ts,
a compl ete di stri buti ve system.
ol'Justict: .1 De.linseo/
Originalll'publishedin Spheres
PlurulivnurulEt1ulit.1,(\lxrin Robcrtson,lgtt:i), 3 30.
Gl 1983 br Basic Books, Inc., a division ol'
Cop,'-right
HarperCollins, Inc. Reprinted b1 permission of
ell Publishers.
Blacku
decided in many different ways. Throughout history, the market has been one of the most imporgoods; but i t has never been, i t now here i s today,
Similarll'', there has never been either a single
decision point from which all distributions are
control l ed or a si ngl e set of agents maki ng deci sions. No state power has ever been so pervasive
as to regul ate al l the patterns ofshari ng, di vi di ng,
and exchanging out of which a societl' takes
@
I
M i c h a e l Wa l z e r
shape. Things slip awav from the state's grasp;
new patterns are worked out - familial networks,
black markets, bureaucratic alliances, clandestine
political and religious organizations. State officials
can tax, conscript, allocate, regulate,
appoint, reward, punish, but they cannot capture
the full range of goods or substitute themselves
for every other agent of distribution.
Nor can
anyone else do that: there are market coups and
cornerings, but there has never been a fully succ e s s f u ld i s t r i b u t i v c co n sp ir a cl .
And finally, there has never been a single criterion, or a single set of interconnected criteria, for
a l l d i s t r i b u t i o n s . D e se r t, q u a lifica tio n , b ir th a n d
blood, friendship, need, free exchange, political
loyaltv, democratic decision: each has had its
p l a c e , a l o n g w i t h m a n v o th e r s, u n e a sill' co e xisting, invoked by competing groups, confused with
one another.
I n t h e m a t t e r o f d istr ib u tive ju stice , h isto r v
d i s p l a y s a g r e a t v ar ie ty o f a r r a n g e m e n ts a n d id e o l o g i e s . B u t t h e f i r st im p u lse o f th e p h ilo so p h e r
i s t o r e s i s t t h e d i sp la vs o f h isto r y, th e r vo r ld o f
a p p e a r a n c e s ,a n d to se a r ch fo r so m e u n d e r lyin g
u n i t y ' : a s h o r t l ist o f b a sic g o o d s, q u ickly
a b s t r a c t e d t o a s i ng le g o o d ; a sin g le d istr ib u tive
c r i t e r i o n o r a n in te r co n n e cte d se t; a n d th e
p h i l o s o p h e r h i m s e lf sr a n d in g , sym b o lica llv a t
l e a s t , a t a s i n g l e d ecisio n p o in t. I sh a ll a r g u e th a t
t o s e a r c h f b r u n i t y is to m isu n d e r sta n d th e su b j e c t m a t t e r o f d i s tr ib u tive iu stice . Ne ve r th e le ss,
in
s o m e s e n s e th e p h ilo so p h ica l im p u lse is
u n a v o i d a b l e . E v e n if r ve ch o o se p lu r a lism , a s I
s h a l l d o , t h a t c h o ice still r e q u ir e s a co h e r e n t
d e f e n s e . T h e r e m u st b e p r in cip le s th a t ju stily
t h e c h o i c e a n d s e t lim its to it, fcr r p lu r a lism d o e s
n o t r e q u i r e u s t o e n d o r se e ve r t' p r o p o se d d ist r i b u t i v e c r i t e r i a o r to a o ce p t e ve r v lvo u ld - b e
a g e n t . C o n c e i v a b lv, th e r e is a sin g le p r in cip le
a n d a s i n g l e l e g i tim a te kin d o f p lu r a lism . Bu t
t h i s r v o u l d s t i l l b e a p lu r a lism th a r e n co m p a sse d
a w i d e r a n g e o f d istr ib u tio n s. Bv co n tr a st, th e
d e e p e s t a s s u m p t i o n o f m o st o f th e p h ilo so p h e r s
r v h o h r v e r v r i t t e n a b o u t iu stice . fr o m Pla to
o n w a r d , i s t h a t t h e r e is o n e , ir n d o n ly o n e , d istr i b u t i v e s y s t e m th xr p h ilo so p h l ca n r ig h tly
encompass.
' l ' o d a v t h i s s v s te m is co m m o n lv d e scr ib e cl
as
th e o n e t h a t i d e alh ' r a tio n a l m e n a n d u o m e n
rvoulcl choose if thel- lvere forced to choose
im p a r t i a l h , k n o r v i n g n o r h in g o f th e ir o r vn sir u ,
a t i o n , b a r r e d f r o m m lkin g p a r r icu lxr ist cla im s,
c o n t i o n t i n g a n a b str a ct se t o f g o o d s.r lf th e se
@
constraintson knowing and claiming are suitably
shaped,and ifthe goodsaresuitablyde6ned,it is
probably true that a singular conclusioncan be
produced.Rationalmen and women,constrained
this way or that, will chooseone, and only one,
distributivesystem.But the force of that singular
conclusionis not easy to measure.It is surely
doubtful that those same men and women,if
they weretransformedinto ordinary people,with
a firm senseof their own identity, with their own
goods in their hands, caught up in everyday
troubles, would reiterate their hypothetical
choice or even recognizeit as their own. The
problem is not, most importantlv, with the particularism of interest, which philosophershave
alwavs assumed rhey could safely - that is,
uncontroversiallv- set aside. Ordinary people
can do that too, for the sake,say, of the public
interest.The greaterproblemis with the particularism of history,culture, and membership.Even
if thev are committed to impartiality, the question most likely to arisein the minds of the membersof a political community is not, What would
rational individuals chooseunder universalizing
conditionsof such-and-sucha sortl But rather,
What would individuals like us choose,who are
situatedas we are, who share a culture and are
determined to go on sharing it? And this is a
question that is readily transformedinto, What
choiceshave we already made in the courseof
our common life? What understandingsdo we
(really)share?
Justice is a human construuion, and it is
doubtfulthat it can be madein only one way.At
lny rate, I shall begin by doubting, and more
than doubting, this standard philosophical
assumption.The questionsposed b1.the theory
of distributive f ustice admit of a rangeof
answers,and there is room within the rangefor
cultural diversity and political choice.It's not
onlv a matter of implementingsome singular
principleor setof principlesin dilferenthistorical settings.No one would deny that thereis a
rangeof morallv permissibleimplementations.
I
want to arguefor more than this: that the principlesof justiceare rhemselves
pluralisticin form;
that different social goods ought to be distributed for different reasons,in accordance
with
diff-erent procedures,by different agents;and
that all thesedifferencesderive from different
understandings
of the socialgoodsthemselves
the inevitablcproduct of historicaland cultural
particularism.
A Theory ofGoods
Theories of distributive justicefocu
process commonly describedas if
form:
Peopledistributegoodsto (other)peo
Here, "distribute" meansgive,alloca
and so on, and the focusis on theind
stand at either end of theserctions
ducersand consumers,but on distri
and recipients of goods.We are as:
ested in ourselves,but, in this case
as
and limited vercion of ourseJves,
give and take. What is our nature?W
rights? What do we need, want, des
are we entit\ed to? What would we a
ideal conditions? Answers to these q
turned into distributive principle
supposedto control the movementol
goods, defined by abstraction.are
movablein any direction.
But this is too simplean understa
actually happens,and it forcesus to
make large assertionsabout human
unlikely,e
moral agency- assertions
I want t
general
agreement.
mand
more preciseand complexdescripti
tral process:
Peopleconceiaeand creategoods,wl
distribute amongt hemselues.
Here, the conceptionand creation
control the distribution. Goodsdon
in the hands of distributiveagents
them as they like or give them out i
with some generalprinciple.2Rathe
their meanings- becauseof their m
the crucial medium of socialrelatio
into people's minds beforethey co
hands; distributions are patternedi
with shared conceptionsof what t
and what they are for. Distributi
constrainedby the goodstheY hol
almost sav that goods distribu
amongpeople.
Things arein the saddle
And ride mankind.3
But theseare alrvaysparticularthin
ular groupsof men and women.At
ComplexEquality
tuirably
".d,it is
can be
:rained
y one,
ngular
surely
ren, if
,, with
r own
ryday
retical
The
: parhave
It is,
.'ople
ublic
ticuiven
uesremruld
zing
her,
are
are
i sa
'hat
' of
we
ls
At
)re
cal
,rY
of
br
ot
ar
ia
I
i:
1
I
t
I
A Theory
o f Go o d s
Theories of distributive justice focus on a social
process commonly described as if it had this
form:
Peopledistribute goods to (other) people.
Here,"distribute" means give, allocate, exchange,
andso on, and the focus is on the individuals who
s t a n da t e i r h e r en d o f r h e se xctio n s: n o t o n p r o _
d u c e r sa n d c o n su m e r s, b u t o n d istr ib u tive a g e n ts
and recipients of goods. We are as alwavs inter_
estedin ourselves, but, in this case, in a special
and limited version of ourselves, as people who
give and take. What is our naturel What are our
rights?What do u'e need, want, deserve? What
are we entitled to? What would we accept under
ideal conditions? Answers to these questions are
turned into distributive principles, which are
supposedto control the movement ofgoods. The
goods, defined by abstraction, are taken to be
m o v a b l ei n a n v dir e ctio n .
B u t t h i s i s t o o sim p le a n u n d e r sta n d in g o f wh a t
actually happens, and it forces us too quicklv to
make large assertions about human nature and
moral agency - assertions unlikely, ever, to com_
mand general agreement. I want to propose a
more precise and complex description of the cen
tral process:
People conceixe and create goods, mhich the.),then
distribu t e a mong t hentselt: es.
Here, the conception and creation precede and
control the distribution. Goods don't just appear
i n t h e h a n d s o f d istr ib u tive a g e n ts r vh o d o with
them as thev like or give them out in accordance
we make the things - even the saddle. I don't
want to deny the importance of human agency,
only to shift our atention from distribution itself
to conception and creation: the naming of the
goods, and the giving ofmeaning, and the collec_
tive making. What we need to explain and limit
the pluralism of distributive possibilities is a theory of goods. For our immediate purposes, that
theory can be summed up in six propositions.
l. All the goods with which distributive iustice
is concerned are social goocls. Thev are not and
they cannot be idiosyncratically valued. I am not
sure that there are any other kinds of goods; I
mean to leave the question open. Some domestic
objects are cherished for private and sentimental
reasons, but only in cultures where sentiment
regularly attaches to such obfects. A beautiful
sunset, the smell of new-mown hay, the excite_
ment of an urban vista: these perhaps are pri_
vately valued goods, though thev are also, and
more obviously, the obiects of cultural assess_
ment. Even new inventions are not valued in
accordance with the ideas of their inventors: thev
are suhj ect ro a w i der process of concepti on and
creation. God's goods, to be sure, xre exemDt
from thi s rul e - as i n the fi rst chapter of Generi s:
"and God sarl' every thing thar He had made.
and, behol d, i t w as very good" (1:31). That eval u_
ation doesn't require the agreement of mankind
(who might be doubtful), or of a maioritv of men
and women, or of any group of men and ll'omen
meeting under ideal conditions (though Adam
and E ve i n Eden w oul d probabl y endorse i t). But
I can't think of anv other exemptions. Goods in
the world have shared meanings because conception and creation are social processes. For the
same reason, goods have different meanings in
different societies. The same "thing" is valued for
with some general principle.: Rather, goods with
their mernings - because of their meanings - are
the crucial medium of social relations; thev come
into people's minds before thev come into their
hands; distributions are patterned in accordance
dilferent reasons, or it is valued here and disvalued there. John Stuart Mill once complained that
"people like in crowds," but I know of no other
with shared conceprions of rvhat the goods are
and rvhat thev are for. Distributive agenrs are
c o n s t r a i n e d b y th e g o o d s th e v h o ld ; o n e m ig h t
a l m o s t s a y r h a t g o o d s d istr ib u te th e m se lve s
the goods or figure out the reasons for taking
them as l i kabl e or di sl i kabl e. Once peopl e l i ke i n
among people.
T h i n g s a r e i n th e sa d d le
A n d r i d e m a n kin d .r
But these are alrvavs pxrticular things and particu l a r g r o u p s o f me n e n d wo m e n . An d . o f co u r se ,
wav to like or to dislike social goods.a A solitar_v
person could hardly understand the meaning of
crowds, it becomes possible for individuals to
to latent or subversive
meani ngs,ai mi ng at al ternati veval ues- i ncl udi ng
break alvav, pointing
the values, for example, of notoriety' and eccentricitl'. An easv eccentricity has sometimes been
one of the privileges of the aristocracy: it is a
social good like any other.
2. NIen and women txke on concretc identities
@
ili
'lr
i
M ic h a e l Wa l z e r
bec a u s eo f t h e w a y t h ey co n ce ive a n d cr e a te , a n d
then possessand emplol' social goods. ,,The line
bet w e e n w h a t i s m e an d m in e ," wr o te Willia m
J lme s , " i s v e n h a r d r o d r a u ."i Distr ib u tio n s ca n
not be understood as the acts of men and women
who do not vet have particular goods in their
mind s o r i n t h e i r h a nd s. In fh ct, p e o p le a lr e a d v
stand in a relation to a set of goods; they have a
his t o r v o f t r a n s a c t i o n s , n o t o n ll' with o n e a n o th e r
but also u'ith the moral and material world in
which t h e y l i v e . Wi t h ou t su ch a h isr o r v, r vh ich
begin s a t b i r t h , t h e y w o u ld n ' t b e m e n a n d wo m e n
in an v r e c o g n i z a b l es e n se ,a n d th e v wo u ld n ,t h a ve
t he f i r s t n o t i o n o f h o w to g o lb o u t th e b u sin e sso f
giv ing , a l l o c a t i n g ,a n d e xch a n g in gg o o d s.
3. T h e r e i s n o s i n gle se t o f p r im a r v o r b a sic
goods conceivable across all moral and material
worlds - or, anv such set lvould have to be con
c eiv e d i n t e r m s s o a b str a ct th a t th e v wo u lcl b e o f
lit t le u s e i n t h i n k i n g a b o u t p a r ticu la r d istr ib u _
t ions . E v e n t h e r a n g e o f n e ce ssitie s,if we ta ke
lnt o a c c o u n t m o r a l a s w e ll a s p h vsica l n e ce ssitie s,
is v er v r v i d e , a n d t h e r a n k o r d e r in g s a r e r .e r v d ifferent. A single neccssarv good, and one that is
always necessarv
food, for example - carries
different meanings in dilferent places. Ilread is
the staff of life, the body of Christ, the sl.mbol of
t he S a b b a t h , r h e m e a n s o f h o sp ita lin ., a n d so o n .
Conce i v a b l v , t h e r e i s a lim ite d se n sein r vh ich th e
Iirst of these is primarv, so that if rhere were
t rv en$ ' p e o p l e i n t h e u , o r ld a n d iu st e n o u g h b r e a d
to feed the twent\', the primacv of bread ls-stalfof lif e r v o u l d v i e l d a s u fficie n t d istr ib u tive p r in ciple. B u t t h a t i s t h e o n l v cir cu m sta n ce in wh ich it
rv ould d o s o ; a n d e v e n th e r e , wc ca n ' t b e su r e . If
t he re l i g i o u s u s e s o f b r e a d we r e to co n flict with
it s nu t r i t i o n a l u s e s i f th e g o d s d e m a n d e d th a t
bread b e b a k e d a n d b u r n e d r a th e r th a n e a te n - it
is bv n o m e a n s c l e a r w h ich u se r vo u ld b e p r im a r v.
IJorv, then, is bread to be incorporatecl into the
univ er s a l l i s t ? ' I ' h e q u e stio n is e ve n h a r d e r to
ans we r , t h e c o n v e n t i o na l a n su e r s le ss p la u sib le ,
x s \ \ 'e p a s sf r o m n e c c s : i ticsto o p n o r lu n ilie s, p o *ers , re p u t a t i o n s ,a n c l s o on . T h e se ca n b e in co r p o _
rated onlv if thel, are abstractcd lrom everv
part icu l a r m e a n i n g - h en ce , tb r a ll p r a ctica l p u r pos es ,r e n d e r e d m e a n i n glcss.
. 1. t s u t i t i s t h e m e a n in g o f
so o d s th a t d e te r m ines t h e i r m o v e m e n t . Disr r ib u tir .e cr ite r ia a n d
arrt ng e m e n t s r r e i n t r i n sic n o t to th e g o o d _ in
itself but to the social good. Ii' rve understand
lvhat it is, lvhat it means to those for rvhom it is a
good, u e u n d e r s t a n d h o iv, b 1 - u h o m , a n d fb r
@
whar reasons it ought ro be distributed. All distri_
butions are just or unjust relative to the social
meanings of the goods at stake. This is in obvious
rvays a principle of legitimation, but it is also a
critical principle.6 When medieval Christians, for
example, condemned the sin of simony, they were
claiming that the meaning of a particular social
good, ecclesiastical office, excluded its sale and
purchase. Given the Christian understanding of
office, it follorved - I am inclined to say, it
necessarilv follorved - that olfice holders should
be chosen for their knowledge and piety and not
lbr their wealth. There are presumably. things
th a t money can buy, but not thi s thi ng. Si mi l arl y,
the words prustitutizn and hribery, llke simony,
describe the sale and purchase of goods that,
given certain understandings of their meaning,
o u g ht never to be sol d or purchased.
5. Social meanings are historical in character;
a n d so di stri buti ons, and j ust and unj ust di stri butions, change over time. 1'o be sure, certain key
goods have rvhat we might think ofas characteristic normative structures, reiterated across the
lin e s (but not al l the l i nes) ofti me and space.It i s
bccause of this reirerarion that the British
philosopher Bernard Williams is able to argue
that goods should ahvays be distributed for ,,relevant reasons" - where relevance seems to connect
to essential rather than to social meanings.TThe
idea that offices, lor example, should go to qualiIie d candi dates- though nor rhe onl r- i ci eathat
has been held about offices - is plainly visible in
r,erv diff-erent societies rvhere simonv and nepotism, under dilferent names, have similarly been
th o u ght si nl ul or unj ust. (But there has been a
wide divergence of viervs about what sorts of
p o sit i on and pl ace are properl y.cal l ed ,,offi ces.,')
Again, punishment has been rvideh. understood
as a negarive good that ought to go to people who
a r e judged to deserve i t on the basi s ofa verdi ct,
n o t o fa pol i ti cal deci si on. (B ut l vhat consti tutesa
verdict? Who is to delir,er it? How, in short, is
ju stice to be done to accused men and w omen?
Ab o u t these questi ons there has been si gni fi cant
disagreement.) T'hese examples invite empirical
in vcsti gati on. 'fhere i s no merel v i ntui ti ve or
procedurelor seizingupon relevant
:ffi*::t.,.
6 . When meani ngs are di sti nct, di stri buti ons
m u st be autonomous. l tvery soci al good or set of
goods constitutcs, as it lvere, a distributive sphere
r vith in rvhi ch onl v certai n cri teri a and arrangem e n ts are appropri ate. trIonev i s i nappropri atei n
the sphereof
from another
no advantagein tht
place has
can rightly be sold
and women and
sinful men and
much business).
ers;the churchis
social meanings
in one distributive
the othersi lve un
autonom.v. But
meaning,is a cril
be argutng t
ple. It is radical
single standard agai
to be measured.
there are standards
they are also
and everv di
society; and these
the goods usurped,
erful men and
Dorninance
and
In fact, the violat
a matter of social
it is more likelvto
tion and rebellion
For all the
arrangements,most
what we might thinlt
gold stenderd: one
dominant and
spheresof di
goodsis commonly
bv the strengthand
good dominantif
because they haveit,
of other goods.It is
gle man or woman'
vxlue - or a groupof
holdit
successfull-v
describesa wav of
limited b1'their intri
those meanings in
describes a wav of
goods i n order to
goods are scarceand
the desert, monopoly
ComplexEquality
rted.All distrie to the social
is is in obvious
)ut it is also a
Christians,for
ony, they were
rrticular social
d its sale and
Jerstandingof
red to say, it
roldersshould
piety and not
rmably things
ing. Similarly,
, like simony,
f goods that,
reir meaning,
in character;
rjust distribue, certain key
rscharacterisd across the
rd space.It is
the British
Lbleto argue
ted for "relens to connect
,anings.7
The
I go to qualinly idea that
nly visiblein
ry and nepomilarly been
: has been a
hat sorts of
:d "offices.")
understood
r peoplewho
of a verdict,
constitutesa
in short, is
Lndwomen?
n significant
te empirical
intuitive or
on relevant
Listributions
od or set of
rtivesphere
td arrangerropriatein
thesphereofecclesiastical office; it is an intrusion
from another sphere. And pietl' should make for
no advantagein the marketplace, as the marketplacehas commonlv been understood. Whatever
canrightly be sold ought to be sold to pious men
and women and rlso to profane, heretical, and
sinful men and rvomen (else no one would do
much business). The market is open to all comers;the church is not. In no society, ofcourse, are
nant. Nlostll*, however, dominance is a more
elaborate social creation, the work ofmany hands,
mixing reality and svmbol. Physical strength,
familial reputation, religious or political office,
landed wealth, capital, technical knowledge: each
of these, in dilferent historical periods, has been
dominant; and each of them has been monopolized by some group of men and women. And
then all good things come to those who have the
socialmeanings entirely distinct. What happens
in one distributive sphere affects what happens in
one best thing. Possessthat one, and the others
come in train. Or, to change the metaphor, a
the others; we can look, at most, for relative
autonomy. But relative autonomv, like social
dominant good is converted into another good,
into many others, in accordance with lvhat often
meaning,is a critical principle - indeed, as I shall
be arguing throughout this book, a radical princi-
appears to be a natural process but is in fact magi-
p l e .I t is r a d ica l e ve n th o u g h it d o e sn ' t p o int to a
singlestandard against which all distributions are
to be measured. 'l'here is no single standard. But
thereare standards (roughl-v knowable even when
they are also controversial) for ever,v social good
and everv distributive sphere in ever]' particular
society';and these standards are often violated,
the goods usurped, the spheres invaded, by pow
erful men and rvomcn.
cal , a ki nd ofsoci al al chemy.
No social good ever entirely dominates the
range of goods; no monopoly is ever perfect. I
mean to describe tendencies only, but crucial tendencies. For we can characterize whole societies
in terms of the patterns of conversion that are
established within them. Some chlracterizations
are simple: in r capitalist society, capital is dominant and readily converted into prestige and
power; in a technocracy, technical knowledge
play's the same part. But it isn't difficult to imagine, or to find, more complex social arrange-
D o m i na n ce
a n d M o n o p o ly
ments. Indeed, capi tal i sm and technocracy are
more complex than their names imply, even if the
In fact, the violations are systematic. Autonomf is
a m a t te r o f so cia l m e a n in g a n d sh a r e d va lues,but
names do convey real information about the most
important forms of sharing, dividing, and
it is more likeh' to make for occasional reformation and rebellion than lbr evervdav enforcement.
exchanging. Nlonopolistic control of a dominant
good makes a ruling class, whose members stand
complexitl' of their distributive
arrangements, most societies are organized on
what we might think of as a social version of the
gold standard: one g;ood or one set of goods is
atop the distributive system - much as philosophers, claiming to have the wisdom they love,
d o m i na n t a n d d e te r m in a tive o f va lu e in al l the
spheresof distribution. Ancl that good or set of
g o o d sis co m m o n lv m o n o p o lize d , its va lu e uphel d
every ruling class is unstable. It is continuall,vchallenged by other groups in the name of alter-
bv the strength and cohesion of its orvners. I call a
g o o d do m in a n t if th e in d ivid u a ls ivh o have i t,
Distribution is what social conflict is all about.
N[arx's heavv emphasis on productive processes
should not conceal from us the simple truth that
For all the
becausethev hirve it, can commancl r wide range
o f o t h e r g o o d s. It is m o n o p o lize d r vh e n e vera si ngle man or \r'oman) a monarch in the lvorld of
value
or a group of men and women, oligarchs -
s u c c e sslu llvh o ld it a g a in st a ll r iva ls. Do mi nance
describes a wav of using social goods that isn't
l i m i t e d b 1 ' th e ir in tr in sic m e a n in g s o r th a t shapes
those meanings in its olvn image. X'Ionopoll'
d e s c r ib e sa u :r v o f o lvn in g o r co n tr o lling soci al
g o o d s in o r d e r to e xp lo it th e ir d o m in a n ce. When
goods are scarce and rvicleli' needed, like water in
thc desert, monopoli' itself rvill make them domi-
mi ght l i ke to do. B ut si nce domi nance i s al w ays
i ncompl ete and monopol y i mperfect, the rul e of
nati \ e patterns uf convcrsi on.
the struggle for control of the means of production
is a distributive struggle. Land and capital are at
stake, and these are goods that can be shared,
divided, exchanged, and endlessly converted. But
land and capital are not the only dominant goods;
it is possible (it has historicall.v been possible) to
come to them by wav of other goods - military or
political power, religious office and charisma, and
so on. Histor-v reveals no single dominant good and
no naturallJ' dominant good, but only different
kinds of magic and competing bands of magicians.
@
M ic haelW alz er
The claim to monopolize a dominant good when worked up for public purposes - constitutes an ideology. Its standard form is to connect
legitimate possession with some set of personal
qualities through the medium of a philosophical
principle. So aristocracv, or the rule ofthe best, is
the principle of those who lay claim to breeding
and intelligence: they are commonly the monopolists of landed wealth and familial repurxtion.
Divine supremacy is the principle of those who
c l a i m t o k n o w t h e wo r d o f Go d : th e v a r e th e
monopolists of grace and office. N{eritocracv, or
t h e c a r e e r o p e n t o t a le n ts, is th e p r in cip le o fth o se
who claim to be talented: they are most often the
ward. Though these are of manv different sorts,
three generalsorts are especiallvimportant:
I The claim that the dominant good, whateverit
is, should be redistributed so that it can be
equally or at leasr more widely shared: this
amounts to saying that monopoly is unjust.
2 The claim that the way should be opened for
the autonomous distribution ofall social goods:
this amounts to saying that dominance is
unj ust.
3 The claim that some new good, monopolized
by some new group, should replace the currenrly dominant good: this amounts to saying
monopolists of education. Free exchange is the
pr i n c i p l e o f t h o s e w h o a r e r e a d v, o r wh o te ll u s
that the existing pattern of dominance and
monopol y i s uni ust.
thev are readv, to put their monev at risk: thev
are the monopolists of movable wealth. These
groups - and others, too, similarlv marked off by
their principles and possessions- compete with
The third claim is, in Marx's view, the modelof
every revolutionary ideology - except, perhaps,the
proletarian or last ideology. Thus, the French
one another, struggling for supremacy. One
group wins, and then a different onel or coalitions
ar e w o r k e d o u t , an d su p r e m a cy is u n e a sily
shared. There is no final victory, nor should there
be . B u t t h a t i s n o t to sa y th a t th e cla im s o f th e
different groups are necessarilv wrong, or that thc
pr i n c i p l e s t h e y i n v o ke a r e o fn o va lu e a s d istr ib u tive criteria; the principles are often exactly right
within the limits of a particular sphere. Ideologies
are r e a d i l v c o r r u p t e d , b u t th e ir co r r u p tio n is n o t
t he m o s t i n t e r e s t i n g th in g a b o u r th e m .
It is in the studv of these struggles that I have
sought the guiding thread of mv own argumenr.
The struggles have, I think, a paradigmatic form.
Some group of men and women - class, caste,
strata, estate, alliance, or social formation - comes
to enjov a monopolv or a near monopolv of some
dominant good; or, a coalition of groups comes ro
enjov, and so on. This dominant good is more or
less systematicallv converted into all sorts of other
things - opportunities, porvers,and reputations. So
wealth is seized by the strong, honor b-v''the wellborn, officc b-v the well educated. Perhaps the ideologv that justifies rhe seizure is u.idely believed to
be true. But resentment and resistancearc (almost)
as pervasive as belief. There are alwavs some
people, and aftcr a time there are a great manv, who
think the seizure is not ,usrice but usurpation. The
ruling group does not possess,or does not uniquely.
possess, the qualities it claims; the conversion
process violates the common understanding of the
goods at stake. Social conflict is intermittent, or it is
endemic; at some point, counterclaims are put for
@
Revolution in Marxist theorv: the dominance of
noble birth and blood and of feudal landholdingis
ended, and bourgeois wealth is establishedin its
stead. The original situation is reproduced with
different subiects and obiects (this is never unimportant), and then the class war is immediately
rcnewed. It is not m]' purpose here to endorseor to
criticize N{arx's view. I suspect, in fact, that there
is something of all three claims in every revolutionary ideology, but that, too, is not a position
that I shall try to defend here. Whatever its socio.
logical significance, the third claim is not philosophicallv interesting - unless one believes that
there is a naturallv dominant good, such that its
possessorscould legitimately claim to rule the rest
of us. In a sense,Marx believed exactly that. The
means of production is the dominant good
that one seemsto me to
of social meaningsand the
tributive systems.But the
mon among PhilosoPhers;
searchfor unitY and si
to explain its difficultiesat
Men and women who
challengethe monoPolY
a particular socialgood.Thi
monopolyin general;forif
dominant and widelY
possiblY be monoPolized'
which everYthing is uP for
hasas much moneyas
the "regime of simPle
tiplied through the
extendsacrossthe full
regime of simPle
becausethe further
exchange in
the
inequalitiesin its train.If
simple equality over t
"monetar-v law" like the
times or the Hebrew
periodic return to the orl
centralized and activist
enough to force sucha
that state officials would
ing to do that, if moneY
In any case,the original
another waY. It's not
reappear, but also that
In practice, breaking
neutralizesits dominancc'
play, and inequalitYtakes
throughout history, and Marxism is a historicist
doctrine insofar as it suggests that whoever con-
again the regime of si
up for sale, and everYone
money. So everYonehag
trols the prevailing means legitimately rules.8After
the communist revolution, we shall all control the
buy an education for his
and others don't' It
means of production: at that point, the third claim
collapses into the first. N{eanwhile. Marx's model
ment: other socialgoods
for sale onlY to PeoPle
is a program for ongoing distributive struggle.It
will matter, of course, who wins at this or that
cates. Soon everyone
moment, but we won't know why or how it matters if we attend only to the successiveassertions0f
dominance and monopoly.
Simple Equality
It is with the first two claims that I shall be concerned, and ultimatelv with the second alone,for
more likely, the Purchase
the tax sYstem.But then
a competitive world wi
longer dominant
upbringing or skill in
dominant instead,and
certificationare
Let's call them (what
"group of the talented."
of this grouP claim t
C ompl ex Equal i ty
rent sorts,
rt:
thatone seems to me to capture best the plurality
of socialmeanings and the real complexity of dis-
should be dominant ourside the school: of6ces,
titles, prerogatives, wealth too, should all be pos-
hateverit
it can be
rred: this
rjust.
rened for
al goods:
nance is
tributive systems. But the first is the more common among philosophers; it matches their own
searchfor unity and singularity; and I shall need
sessed by themselves. This is the career open to
talents, equal opportunitv, and so on. This is
what fairness requires; talent will out; and in any
to explain its difficulties at some length.
case, talented men and women will enlarge the
resources available to everyone else. So Michael
Young's meritocracy is born, with all its atten-
opolized
the curo saying
rce and
nodelof
aps,the
French
ance of
lding is
I in its
:d with
unim:diately
ie or to
t there
'evoluosition
sociophilos that
5at its
le rest
. The
god
rricist
conAfter
rl the
claim
rodel
le. It
that
matns of
)onfor
Men and women who make the first claim
challengethe monopoly but not the dominance of
a particular social good. This is also a challenge to
monopolyin general; lor if wealth, for example, is
dominant and widely shared, no other good can
possibly be monopolized. Imagine a society in
which everything is up for sale and every citizen
hasas much money as every other. I shall call this
t h e " r e g i m e o f sim p le e q u a lity." Eq u a litv is m u ltiplied through the conversion process, until it
extendsacross the full range of social goods. The
regime of simple equality rvon't last for long,
becausethe further progress of conversion, free
exchange in the market, is certain to bring
inequalitiesin its train. If one wanted ro sustain
s i m p l e e q u a l i tv o ve r tim e , o n e wo u ld r e q u ir e a
"monetarv law" like the agrarian laws of ancient
times or the Hebrew sabbatical, providing for a
periodic return to the original condition. Only a
centralized and activist state rvould be strong
enough to force such a return; and it isn't clear
that state officials would actually be able or willing to do that, if money were the dominant good.
I n a n y c a s e , t he o r ig in a l co n d itio n is u n sta b le in
another way. It's not only that monopoly will
r e a p p e a rb, u t also th a t d o m in a n ce will d isa p p e a r .
In practice, breaking the monopolv of money
neutralizesits dominance. Other goods come into
p l a y ,a n d i n e q u a lity ta ke so n n e w fo r m s. Co n sid e r
againthe regime of simple equality. Er,ervthing is
up for sale, and evervone has the same amount of
monev. So evervone has, say, an equal abilitl' to
b u y 'a n e d u c a t io n lo r h is ch ild r e n . So m e d o th a t,
dant inequalities.e
What should we do now? Ir is possible to set
limits to the new conversion patterns, to recognize but constrain the monopoly power ofthe talented. I take this to be the purpose of John
Rawls's dilference principle, according to which
inequalities are justified only if they are designed
to bring, and actually do bring, the greatesr possible benefit to the least advantaged social class.r0
More specifically, the difference principle is a
constraint imposed on talented men and women,
once the monopoly of wealth has been broken. It
works in this way: Imagine a surgeon who claims
more than his equal share of wealth on the basis
of the skills he has learned and the cerrificates he
has won in the harsh competitive struggles ofcolIege and medical school. We will grant the claim
if, and only if, granting it is beneficial in the
stipulated ways. At rhe same time, we will act to
limit and regulate the sale ofsurgery - that is, the
direct conversion ofsurgical skill into wealth.
'I'his regulation will necessarily be the work
of
the state, just as monetary laws and agrarian laws
are the work of the state. Simple equalitv would
require continual state intervention to break up or
constrain incipient monopolies and to repress
new forms of dominance. But then state power
itself will become the central object of competitive struggles. Groups of men and women will
seek to monopolize and then to use the state in
a n d o t h e r s d o n ' t. It tu r n s o u t to b e a g o o d in ve stment: other social goods are, increasingly, offered
order to consolidate their control of other social
goods. Or, the state will be monopolized by its
own agents in accordance rvith the iron law ofoli-
for sale onll' to people rvith educational certific a t e s . S o o n e ve r vo n e in ve sts in e d u ca tio n ; o r ,
more likelr', the purchase is universalized through
garchy. Politics is always the rnosr direct path to
domi nance, and pol i ti cal pow er (rather than the
means of production) is probably the most impor-
t h e t a x s y s t e m . Bu t th e n th e sch o o l is tu r n e d in to
a c o m p e t i t i v e wo r ld with in wh ich m o n e y is n o
tant, and certainly the most dangerous, good in
human hi storv.rr H ence the need to constrai n the
longer dominant.
Natural talent or firmily
u p b r i n g i n g o r skill in r vr itin g e xa m in a tio n s is
d o m i n a n t i n s t ea d , a n d e d u ca tio n a l su cce ss a n d
agents of constraint, to establish constitutional
checksand bal ances.These are l i mi ts i mposed on
political monopolv, and thev are all the more
c e r t i f i c a t i o na r e m o n o p o lize d b r .so m e n e w g r o u p .
L e t ' s c a l l t h e m ( r vh a t th e v ca ll th e m se lve s) th e
important once the various social and economic
monopol i es have been broken.
" g r o u p o f t h e ta le n te d ." Eve n tu a llr . th e m e m b e r s
of this group claim that the good the1, control
One w av of l i mi ti ng pol i ti cal pow er i s to di stribute it rvidell'. 'I-his may not work, given the
@
Michael Walzer
well-canvassed dangers of majority tyranny;
but
these dangers are probablv less acute than
thev
a r e o f t c n m a d e o u t to b e . T h e g r e a r e , d u n g .,
tr f
d e m o c r r t i c g o v e r nm e n t is th a t it r vill b e we a k
t< r
cope with re-emerging monopolies in society
at
lr r g e , u i t h t h e s o cie l sr r e n g th o f p lu to cr a r s,
b u r e a u c r a l s , t e c h n o cr a ts, m e r ito cr n ts, a n d
so o n .
In theory, political power is the dominant
sood in
a d c m o c r a c _ \a
. ,n d i r is co n ve r r ib le in a n v *a y
th e
c i t i z e n s c h o o s e . B u t in p r a ctice , a g a in , b r e a kin e
rh e m o n o p o i r . o f p owe r n e u tr a lize sir s d o m in a n ce .
Political power cannot be widelv sharetl without
be i n g s u b j e c t e d t o th e p u ll o fa ll th e o th e r
so o d s
t h a t r h e c i r i z e n s e l r e a d l. h r r c o r h o p e ,o
h r r ..
H e n c e d e m o c r a c v is. a s
- \la r x r e co g n ize d , e sse n _
tiallv a reflective svstem, mirroring the prevailing
an d e m e r g i n g d i s tr ib u tio n o f so cia l
so o ,ls.i
De m o c r l r i c d e c i s i on m a kin g sill b e sh a n e tl
bv
t h e c u l r u r r l c o n c e p tio n s th a t d e tcr m in e n r - u n d e .l
lvrite the nerv monopolies. To prevail
against
t he s c m o n o p o l i e s , po we r u .ill h e r .e to b e ce n t.a l_
iz e d , p e r h a p s i t s e l f m o n o p o lize d . On ce a g a in ,
th e
state musr be verv powerful if it is to fulfill
the
purposes assigned to it by the difference princiole
or h v a n r s i m i l a r l . r i ntr r r e n tio n ist r u le .
S t i l l , t h e r e g i m e o f sim p le e q u a lity m ig h t wo r k.
O n e c a n i m a g i n e a m o r e o r le ss sta b le te n sio n
bet r v e e n e m e r g i n g m o n o p o lie s a n d p o litica l
co n _
strarnrs, between the claim to privilege put for_
ward by the talented, say, and the enforcement
of
the clifference principle, and then between
the
agents of enfbrcement and the democratic
consti_
t utr o n . B u t I s u s p e ct th a t d ifficu ltie s will r e cu r .
and t h u t : l t m x n \ . p o in ts in tim e th e o n lr . r e m e .lr i
f or p r i v a t e p r i v i l e g e r vill b e sta tism , a n d th e
o n lv
c s c r p c f r o m s r a t i s m i r .ill h e p r ir .e te p r ivilcg e .
We
will m o b i l i z e p o w e r r o ch e ck m o n o p o lv. th e n
lo o k
f t rr s o m e $ t . r o f c he ckin g th r p o u e r $ e
h r r vc
mob i l i z e d . B u t t h e r e is n o wa v th a t d o e sn ' t
ooen
opp o r t u n i r i c s f i r r s t re r cg ica llv p la ccd
m .n
"n d
women to serze :rnd exploit important
social
goo d s .
1 - h e s e p r o b l e m s de r ive lr o m tr e a tin g m o n o _
polr . a n c l n o l d o m i n rn ce , a s th e ce n tr a l issu e
in
dis tr i b u t i v e i u s t i c e . I t is n o t d ifficu lt, o fco u r se ,
ro
undersrand u'h],' philosophers (and political
lc t r\ i s t s , t o 9 ) h " 1 ' q l b cu scd o n m o n ,,p o ir .
1 .h c
dis tr i h u r i v c s t r u g g l e s o t' th e m o tle r n .g .
b .g in
rvith a lvar against the aristocracv,s singular
hold
on la n d , o f f i c c , a n d ho n o r . T h is se e m s a n e sDe _
r'irrl l r l l r r n i c i o u s m o n o p o lr .b e clu sc it r e sts u n o n
hirt h r n d h l o o d . r r i r h r r h ich r h e in d ir .id u r l
h :r s
not h i n g t o d o , r a t h e r t h a n u p o n r ve a lth ,o r
Do r ve r .
@
or education, all ofwhich _ ar least in principle_
can be earned. And when every man and
woman
becomes, as it were, a smallholder in the
sphereof
bi rth and bl ood. an i mportant ui ctory
i s i ndeed
won. Birthright ceases to be a dominant
good;
henceforth, it purchases very little;
wealth.
power, and education come to the
fore. With
regard to these latter goods, however,
simplc
equality cannot be sustained at all, or it
can onlv
be sustained subject to the vicissitudes I
haveiust
descri bed. Wi thi n thei r or.rn spheres,as
theyare
currentl _vundcrsrood. these three tend to qenei _
ate natural monopol i es that can be repressed
onl y
if stare power is itself dominant and
if it is
monopolized by officials committed to rhe
reDres_
si on. Bur there i s. I thi nk, another parh
ro an;ther
ki nd ofequal i tv.
T yranny
and C ompl ex
E qual i ty
I want to argue that rve should focus
on the
reduction of dominance - not, or not primarily,
on the break-up or the constraint ofrnonopolv.
W e shoul d consi der l r.het i t mi ght *.rn
,o'n.r_
row the range within which particular goods
are
convertible and to vindicate the autonomv
ofdis_
tr i huti ve sphercs. Bur thi s Ii ne of ,rl urn.n,,
though it is nor uncommon historicallv.
has
n ever ful l r emerged i n phi l osophi cal w ri ti nc.
Phi l osophcrs have tended ro cri ri ci zc
1or to i us_
tif v) eri sri ng or emergi ng monopol i es of w eal th,
polver, and education. Or, they have criticized (or
.iustified) parricular conversions - of wealth into
education or of office into wealth. And
all this,
most often, in the name of some radically
simnli_
fied distributive s].stem. The critique of
domi_
nance will suggesr instead a way of reshaping
and
then living with the actual complexity of
distribu_
tio nsImagine now a society in which clifferent
social
goods are monopolistically held _ as rhev
are in
fa c t l nd al w l r.s ui l l he. barri ng conti nual
state
intervention - but in which no particular good
is
g e ncral l r convcrri hl e. As I go al ong, I shal i
try to
d e fi ne thc preci se l i mi rs on converri bi l i tr. hui
fo,
n o rv the gcncral dcscri pti on u i l l suffi ce.-Thi s
is a
co mpl ex egal i tari ansoci erv.'fhough there
w i l l be
m a nr, smal l i nequal i ti cs, i ncqual i tv w i l l
not be
m u l ri pl i ed through rhc conrcrsi on process.
N or
u'ill it be summed across diff'erent goods,
because
th e autonomv of di stri buti ons w i l l tend to pro_
d u c e a rrri crr ot'l ot.rl monopol i cs, hcl d bt.
di i fer_
Cnt groups of men and
claim that comPlexeq
more stable than si
inclined to think thatit
more diffused and Parti
conflict. And the resi
in
would be maintained,
men and women within
competencexnd controlt
action.
This is, I think, an
have not yet exPlain€d
i
The argument for
our understanding- I
positive. and Particular
varioussocialgoods.And
account of the way we
through those goods.
distributive condition,m
hats and you have
And it is all to the good
then our equality is
spheresof sociallife.
here, however, we
of hats, and it is unlikelY
i
nant for long. EqualitY
persons,mediatedbY
and divide among
It requi
of possessions.
tributive criteria that
social goods.
The argument for
beautifullyput bYPascal
the whole world and
There are different
the hand s ome,thei
each man reigns in hi
sometimes they meet,
handsome hght for
their masterYis of
of each aiming at uni
can win this, not
lessin the kingdom
Tyrannyt. The
fore, are false and
handsome,so I
am strong, therefore
" I a m . ..e tce te r a ."
Tyrann-v ts the ui
what can onlv be
C ompl ex Equal i ty
rn principle _
1 and woman
thesphereof
,ry is indeed
tinant good;
tle; wea.lth,
fore. With
ver, simple
it can only
I havejust
as they are
I to gener_
'essedonly
di f i t i s
he repres:o another
on the
'imarily,
rnopoly.
to nar_
ods are
of disument,
)r, has
'riting.
o jus,ea.lth,
ed (or
r into
this,
npliomi: and
'ibu)cial
.. in
tate
Jis
' to
for
sa
be
be
DT
:
cnt groups of men and women. I don't want
to
claimthat complex equalitv would necessarilv
be
m o r e s t a b l c r ha n sim p le e q u a lir l, h u t
I am
inclinedto think that it would open the
wav lor
m o r ed i f f u s e d en d p a r ticu la r ize d fo r m s
o f so cia l
conflict. And the resistance to convertibilirv
w o u l db e m a i n r ain e d .in la r g c d e g r e e ,
b v o r d in a r r .
m e na n d w o m e n u ith in th e ir o u n sp h e r e s
of
competence
and control, without large_scalestate
actron.
This is, I think, an artractive picture,
but I
havenot vet explained just whl, it is attractive.
The argument for complex equality begins
from
ourunderstanding - I mean, our actual, concrete,
positive,and particular undersranding _
of the
varioussocial goods. And then it moves on
to an
accountof the wav rve relate to one
another
throu.gh
those goods. Simple equalitv is a simple
distributivecondition, so thar if I have fburteen
hatsand you have lburteen hats, rve are
equal.
And it is all to the good if hars are dominant,
for
t h e n o u r e q u a l i tv is e xte n d e d th r o u g h
a ll th e
spheres
ofsocial life. On the vierv that I shall take
h e r e h, o w e v e r , w e sim p lv h a ve th e sa m e n u m b e r
o f h a t s ,a n d i t i s un like ll,,th a t h a ts will
b e d o m i_
nant for long. I,-quality,.is a complex relation
of
persons,mediated b-v-thc goods we make,
share,
a n dd i v i d e r r m o n g o u r se lr .e slit is n o r ir n itle n titr
o f p o s s e s s i o n sl t. r e q u ir e s th cn . a tlir e r sit,
o l.,iis_
t r i b u t i v c c r i r e r i a r h r r m ir r o r s r h c d ir .e r sir i
o l.
socialgoods.
T h e a r g u m e n t lb r co m p le x e q u a litv h a s
been
beautifullvput bv Pascal in onc of'his pensy'es.
The nature of tvrannv is to desire po\\,er
over
t h e r v h o l ei v o r l d a n d o u tsid e its o r vn sp h e r e .
T h e r e a r e d i l fe r e n t co m p a n ie s _ r h e sr r o n g ,
t h e h a n d s o m e ,t he in te llig e n t, th e d e vo u t a n cl
e a c hm a n r e i g n s in h is o r vn , n o t e lse r vh e r c.
Bu t
s o m e t i m e sr h e \ . m e e t, a n d th e str o n g a n d
th e
handsome fight for masrery _ fbolishll.,
firr
t h e i r m a s t e r v i s o l d iffir e n t kin d s. T h e v
m is_
u n d c r s t a n do n c ln o th cr . r n d m r kc th c
m ista ke
o f e a c h a i m i n g a t u n ive r sa l d o m in io n . No th in g
can lvin this, not e\,en strengrh, fbr it is pou.er_
l e s si n t h c k i n g d o m r ,l th e r r isc. . . .
Tyrann.1,.T'he fbllou,ing sratcmcnrs, thcre_
fbre, arc lalse and tr,rlnnical: ,,Because
I :rn.r
h a n d s o m e ,s o I sh o u ld co m m a n d r e sp e ct." ,,I
a m s t r o n g , t h e r e fo r e m e n sh o u ld lo ve n te .
. . ."
"Iam...etcetera."
Tvrannr. is the ivish to obtain br. one means
what can onlv be had br, another. We
olve dil._
ferent duties to different qualities:
love is the
proper response to charm, fear
to strenqth. and
bel i eIto l earni ng.rr
tr{arx made a similar argument in his
earlv manu_
scriprs; perhaps he had this pensie in
mini:
Let us assume man to be man, and
his relation
to the lvorld to be a human one. Then
love can
only be exchanged for love, trust for
trust, etc.
If vou wish to enjoy art you must be
an artisti_
cally cultivated person; ifyou wish to influence
other people, you must be a person rvho
reallv
hes a sti mul ari ng and encouragi ng efl ect
upon
orhers. . . . Ifvou l otc w i rhout evoki nc
l ove i n
rcrurn. i .e.. i fyou rre not abl e. bv thc
mani fes_
tation of vourself as a loving person,
to make
yourself a beloved person _ then your
love is
l mpotent and a mi sfortune.rr
These are not easy arguments, ancl
most of mv
hook i s si mpl ,v an exposi ri on of thei r
_."ni no.
But here I shal l attempt somerhi ng
l ,i rnpi c
and schemati c:a transl ati onof the arguments
-orc
i nto
the terms I har-eal readv heen usi ng.
The fi rsr cl ai m of past.aland \i arx i s
rhar ncr.
sonal qual i ri cs and soci al goods have
rhei r orvn
spheres of operati on, w here thev
w ork thei r
effects freelv, spontaneousl]., and legitimatelv.
There are ready'or natural conversi ons
that fol _
l ow from, and are i ntui ti vel v pl ausi bl e
becauseof.
the soci al mcani ng of p:rrti cul er goods.
The
appcal i s to our ordi narl , uncl erstandi ng
ancl . at
the same ti mc, l gai nst our common ,.;r;;r;;;;
i n i l l cgi ri mate convcrsi on parrerns.
Or, i t i s an
appeal from our acquiesence to our resentment.
There is something wrong, pascal suggests,
with
the conversion ofstrength into belief.
In rrolitical
rerms. Pl scl l mcans thxt no rul er
can ri shtl v
comm:rnd mr- opi ni ons merel r hecaus.
,i tn.
porver he wields. Nor can he, N,Iarx
adds, rishtlv
c l ai m to i nfl uence mv acti ons: i f a rul er
*"n-,, ,o
do that, he must be persuasi ve,hel pful ,
encouragrng, and so on. These arguments cl epend
l or thei r
f i l rce on some shrrretl untl erstl rntl i ng
of knorrl
cdgc, i nfl uence, and pol rer. Soci i l goods
har.e
soci l l mcani ngs. ancl rre fi ntl our u:rr
to di srri bu
ti rc j usti cc throxg[ ,n i nr.rpret.ri on
of rh6se
meani ngs. We sei rrch fbr pri nci pl es
i nternal to
each di stri buri r.esphere.
The second cl ai m i s that rhe di sregard
ofthese
pri nci pl es i s t\ri nn\. To convcrl
one eo,,.l i nto
a nothcr. \\'h(n thcrc i s no i ntrrnsi c
annn".ti ,,n
@
iil
ril
] li
ili
ril
I
i
C ompl ex E qual i ty
Threecriteria, however, appear to meet the
requirements
of the open-endedprinciple and
haveoften been defendedas the beginning and
cndof distributive justice, so I must sav some_
thing
abouteachof them. Free e"ch"nge,des.rt.
rndneed:all three have real force, but none of
themhasforce acrossthe rangeof clistributions.
Theyare pa r t o f r h e s t o r \ . , n o t th e wh o le o f it
Freeexchange
Free exchange is obviouslv open_ended; it
guarantees
no particular distributive outcome. At
nopoint in anv exchange process plausibly called
"free"will it be possible to predict rhe particular
d iv is ionof s o c i a l g o o d s t h a t will o b ta in a r so m e
laterpoint.r" (It may. be possible, however. to
Dre_
dic tt he gen e r a ls r r u c r u r c o f ' t h c tlir .isio n .; tn ih e _
oryat least, free exchange creates a market within
vhich all goods are convertible into all other
goodsthrough the neutral medium of monev.
T hereare no t l o m i n a n t g o o d s an d n o m o n o o o lie r .
Hencethe successive divisions that obtain will
directlvreflect the social meanings of the goods
t hatare div i d e r l . F o r e a c h h a r g a in , tr a tle , sa le .
rnd purchaservill have been agreed to voluntarilv
b ymen and r v o m c n n h o k n o w wh n t th a r m e a n in s
i s,who are in d c c d i r s m a k c r s . Er cr l e xch r n g e is
e
r ev elat ion
ol' s o c i a l m e : r n i n g . 81 d ctin itio n . r h e n .
nor will ev c r l r r l l i n t o t h e h a n ds o f so m co n e r vh o
possesses
.y, merell' because he possessesy and
w i thoutreglr d t o r v h a t . r a c r u l llr a .a n . ,,, ,,,r n .
othermember of societr-. 'I'he market is radicallv
p l u ralis t ic
in i r s o p e r a r i o n : r n d its o u tco m e s. in fi_
n i telys ens it i v e t o t h e m e a n i n g s th a t in d ir .id u a ls
attachto goods. What possible restraints can be
imposedon lree exchange, then, in the name of
pluralism?
B ut ev en'd a v l i f e i n t h e m a r ke t, th e a ctu a l
experienceof fiee exchange, is verv diffbrent
fr om uhrr t hc t h e o r r . s u g g e s t s.\l,,n cv, su p p o s_
edly the neutral medium, is in pracice a clomi_
nantgood, and it is monopolized bv people u.ho
possess
a special talent for bargaining and trading
- th c green t h u m h o l ' h o u r g e o is so cie tv. T h e n
ot herpeoplc t l e m a n d a r c d i s r r ib u r io n u f r n u n .r .
a nd t he rs t lh l i s h m e r r r o f t h c rcg im c ,,f sim n le
e qualit v and
,
t h e s c l r t . h b c g i n s fb r so m c u r \ r ( )
su staint hat rr g i n r e . B u t r r c n i l u e fir cu s o n th c
firstuntroubled momenr of simple equalin, _ free
exchangeon the basis of equal shares we will
st i l lnc ed t o s e t l i m i r s o n w h a t ca n b e e r ch a n se d
fo r u hrt . For f r c e c x c h ; r n g c l c e vcs d i:r r ih u tio n s
e n r ir e ly in the hands of i ndi vi dual s, and
soci al
meanings are not subject, or are not alwavs
sub_
je ct, r o th e i nrerprerrri r.e
deci si ons of i nj i ui dual
m e n a n d women.
Consider an easy example, the case of political
p o we r . We can concei veof pol i ri cal po*..
rr, ,.,
o f g o o tls of varvi ng val ue, vores, i nfl uence.
o ftice s. e n d so on. A nv of rhese can be traded
on
the market and accumulated by individuals
will_
ing to sacrifice other goods. Even if the sacrifices
are real, however, the result is a form oftyranny _
petty tyranny, given the conditions
of simole
e q u a litr ' . BecauseI am w i l l i ng ro do w i rhout
mv
h a r . I sh a ll vote tw i ce; and you w ho val uc rhe
vote
less than you value my hat, will not vote
at all. I
suspect that the result is tyrannical even
with
regard to the two of us, who have reached
a vol_
untary agreement. It is certainly tyrannical
with
regard to all the other citizens lvho must
now
su b m it to my di sproporri onate pou,er. l t
i s not
the case that votes can't be bargained for; on
one
lnterpretarion, that's what democratic politics
is
a ll a b o u t. . {ntl democrari c pol i ti ci ans have
cer_
tainll' been known to buy votes, or to trv
to buv
th e m , b y p r omi si ng publ i c expen<i i turesthat
ben_
e fit p a r ticu lar groups ofvoters. But thi s i s
done i n
p u b lic, u - ith publ i c funds, and subj ect to publ i c
a p p r o va l. Pri rrre tradi ng i s rul ed out bv vi rtue
of
u h a t p o litic s, or democrari c pol i ti cs, i s
rhar i s.
h v vir r u e o f w het w e di d uhen i \.c consri tuted
rhc
p o litica l co mmuni n. and of w hat $,e sri l l
thi nk
about rvhat we did.
Free exchange is not a general criterion. but
we
r r ill h e e b le ro spcci fv the boundari cs
ui thi n
which it operates onlv through a careful
analt,sis
o f' p a n icu la r suci el gootl s. An.l havi ng $.orketl
th r o u g h su ch an rnrrl .rsi s.w e w i l l come up
.t b.rt
u ir h a p h iio sophi cal l v aurhori tari ve set
oi huun.l .
aries and not necessarilv with the set thxt
ousht
to b c p o litical l \ aul hori rxti \e. F,,r" rnune).
,..rx
a cr o ssr ll b o undari cs thi s j s thc pri marv
i urnl of.
ille g a l im m igrati on; and j ust w here one
oushr to
tr \ to sto p it i s l questi on ofcxpecl i cncv as w .l l
as
o f llr in cip le . Fl i l ure to stop i t at some rcasonabl e
p o in t h a s co nsequencesthroughout the
range of
tlistr ib u r io n s , bur consi derari on of thcse b"l nncs
in a la tcr ch a ptcr.
Deserl
Like
free exchange, desert seems both ooen_
e n d cd ;r n d p lural i sri e. One mi ght i magi nc
r si ngl .
n cu tr r l a g e n cv di spensi ng reul rtl s anJ puni sh_
@
i
I
MichaelWalzer
ments, infinitely sensitive to all the forms of indiv id u a l d e s e r t . T h e n th e d istr ib u tive p r o ce ss
rv o u l d i n d e e d b e c e n tr llize d , b u t th e r e su lts
rvould still be unpredictablc and various. There
would be no dominant good. No .r. would ever be
distributed without regard to its social meaning;
f or , w i t h o u t x t t e n t i o n to wh a t r - is, it is co n ce p tu allv i m p o s s i b l e t o s a y th a t r is d e se r ve d . All th e
clilferent companies of men and rvomen rvould
rec e i v e t h e i r a p p r o p ria te r e wa r d . Ho w th is wo u ld
lvork in practice, however, is not easy to figure
out . I t m i g h t m a k e se n se to sa v o f th is ch a r m in g
ma n , f o r e x a m p l e , t h a t h e d e se n e s to b e lo ve d . It
makes no sense to sav that he deservesto be lor.ed
bv t h i s ( o r a n y ) p a r t i c u la r \r ,o m a n .If h e lo ve s h e r
u'hi l e s h e r e m a i n s im p e r vio u s to h is ( r e a l)
c ha r m s , t h a t i s h i s misfb r tu n e . I d o u b t r h a t $ .e
rvould lvant the situation corrected bv some outs ide a g e n c r ' . ' f h e l ove o f p a r ticu la r m e n a n d
\ \ om e n , o n o u r u n d e r stln d in g o f it, ca n o n lv b e
dis t r i b u t e d b v t h e m se h e s, a n d th e v a r e r a r e lv
guided in these mamers bJ. considerations <if
des e r t .
T h e c a s e i s e x a c tlv th e sa m e r vith in flu e n ce .
Here, let's sav, is a rvoman rvidelv thought to bc
s t im u l a t i n g a n d e n c o u r a g in g r o o th e r s. Pe r h a p s
s he d e s e r v e st o b e a n in flu e n tia l m e m b e r o f o u r
c om m u n i n - . B u t s h e d o e sn ' t d e se r ve th a t I b e
inf lu e n c e c lb r - h e r o r th a t I fir llo g , h e r le a d . No r
and so accumulate a large number of beautiful
pictures. If we assume, as painters mostly do, that
prctures are appropriatelv traded in the markeq
then there is nothing wrong with my having thc
pictures. N{y title is legitimate. But it would bc
odd to sav that I deserve ro have them simply
because I am good at bargaining and trading.
Desert seems to require an especiallv closeconnection berween particular goods and particular
persons, rvhereas justice onlv sometimes requires
a conncction of that sort. Still, rve might insist
that onh' arrisricallv cultivated people, who
deserve to have pictures, should actually havc
them. It's nor difficult to imagine a distributivc
mechanism. The state could buv all the pictures
that rvere olfered for sale (but artists would have
to be licensed, so that there wouldn't be an endless number of pictures), evaluate them, and then
distribute them ro artisticallv cultivated men and
women, the better pictures ro the more cultivated. 'I'he state does something like this, sometimes, with rcgard ro things that people needmedical care, lor example - but not with regard
to things that people deserve. There are practical
difficulties here, but I suspect a deeper reasonfor
this difference. Desert does not have the urgency
o 1 -need,and i t does not i nvol ve havi ng (ow ni ng
and consuming) in the same way. Hence, we are
lvilling to tolerate the separation of owners of
paintings and artisticallv cultivated people, or we
they would call the
tion of what is best and
insensitive to the diverse
low citizens.And then
pluralistcriterlon;we
facewith a new set(otan
do, of course'choose
to serveon juries,for
it will be worth consideri
ativesof a juror are.But
here that he operates
Desert is a strong clalm'
iudgments;and only
tionsdoesit YieldsPeciflc
Need
Finally, the criterionof
to his needs"is generallY
half of Marx's famous
the wealth of the
of its
necessities
but a radicallY
half of the maxim is also
and it doesn't fit the
"From each accordingto
iobs should be
women should be
basisof individual
u'ould rve \\,ant mv fbllorvership, as it u.ere,
as s i g n e dt o h e r b y a nl a g cn c) ca p a b le o fm ir kin g
s uc h a s s i g n m e n t s .S h e m a v g o to g r e a t le n g th s to
are unrvilling to require the kinds of interference
in the market that would be necessaryto end the
s t im u l a t e l n d e n c o u r ag em e , a n clclo a ll th e th in g s
t hat a r e c o m m o n l l c alle d stim u la tin g o r e n co u r
agin g . B u t i f I ( p e r v e rse lv)r e fu se to b e stim u la te d
separation. Of course, public provision is always
possible alongside the market, and so we might
argue that arristicxllv cultivated people deserve
candidates: rvhich
or e n c o u r a g e d , I r m n o t d cn vin g h e r a n r - th in g
t hat s h e d e s e r v e s .T h e sa m e a r g u m e n r h o ld s b 1 '
ex t e n s i o n l b r p o l i t i c i a n s a n d o r d in a r v citize n s.
n o t pi ctures bur museums. P crhaps they do, but
thev don't deserve thilt the rcst of us contribute
some non-meterial
then that need lvon't
monev or appropriate public funds for the purchase of pictures and the construction of buildin g s. Thev l vi l l have to persuade us that art i s
wo rth the money; thev w i l l have to sti mul ateand
least not to the naked
odd to ask a search
hospital director to
Cit iz e n s c a n ' t t r a d c t h e ir \,o te sfir r h a ts; th cr .ca n ' t
indiv i d u a l l v d e c i d e t o cr o ss th e b o u n d a r r . th a t
s epa r a t e st h e s p h e r c of p o litics fr o m th e m a r ke tplac e . B u t r v i t h i n t h c sp h e r e o 1 ' p o litics, th e v d o
m ak e i n d i r , i d u a l d e c i sio n s; e n cl th e v a r e r a r e lv
guid e d , a g a i n , b r . c o n sid e r u tio n so fcle se r t. It' s n o t
clear that offices can be desen'ecl - another issue
t hat I m u s t p o s t p o n e ; b u t e vcn if' th o . ca n b e , it
iv ou l d v i o l a t e o u r u n d e r sta n d in g o f cle m o cr :r tic
polit i c s r l e r c t h e i ' s i m p ll clistr ib u te clto cle se r r .in g
m en a n d l v o m e n b v s om e ccn tr ir l a g e n c\' .
encourxge our o\\'n artistic cultivation. And if
thel' fiil to do that, their ou'n lor.e of art may well
tu r n out to be "i mpotent and a mi sfortune."
Even if \\'e \\ere to assign the distribution of
lor,e, influence, offices, rvorks ofart, and so on. to
some omnipotent irrbiters of desert. how would
n e sel ect them? H orv coul d ant.one deservesuch
S i m i l a r h ' , h o r r e v e r u e d r a u th e b o u n cla r ie so f
t he sp h e r e r v i t h i n r v hich lr e e e xch a n g c o p e r a r e s,
des e r t u ' i l l p l a v n o r o l e r vith in th o se b o u n cla r ie s.I
a position? Onll' God, rvho knolvs what secrets
lu r k i n the hearts of men, rvoul d be abl e to make
th e necessarvdi stri buti ons. If human bei ngshad
to cl o the u'ork, the di stri buti ve mechani smw oul d
am s k i l l f u l a t b a r g a i nin g a n d tr a clin g , le t' s sa r .,
b e sci zed carl v on by some band ofari stocrats(so
@
don't in anv obvious
which thev are
scarce,and there area
their material needs
perhapsthey don't need
the needs of the
of the staff and the Pati
the latter set of needs,
of political di
tributive decision.
Nor will need work
Marx's maxim doesn't
the distribution of
fame, sailboats,rare
every sort. Theseare
strictly sPeaking,
view and define the
do, as the strongest
ComplexEquality
they would call themselves) with a fixed concep_
tion of what is best and mosr deserving, and
insensitiveto the diverse excellences of their fellow citizens. And then desert would ceaseto be a
pluralist criterion; we rvould find ourselves face to
facewith a new set (of an old sort) of ryrants. We
do, ofcourse, choose people as arbiters ofdesert to serveon juries, for example, or to award prizes;
it will be worth considering lirter what the prerog_
ativesof a juror are. But it is important to stress
here that he operates lvithin a narrow range.
Desert is a strong claim, but it calls for difficult
iud g m e n t s ; a n d o n l v u n d e r ve r v sp e cia l co n d it io n sd o e si t y i e l d s p e cificd istr ib u tio n s.
Need
Fin a l l y , t h e c r i t e r i o n o fn e cd . " ' I' o e a ch lcco r d in g
t o h i s n e e d s " i s g e n er a llv r a ke n a s th e d istr ib u tive
half of N'Iarx's famous maxim: we are to distribute
t he w e a l t h o f t h e c om m u n itv so a s to m e e t th e
ne c e s s i t i eosf i t s m e m b e r s.r t A p la u sib le p r o p o sa l,
but a radicallv incomplete one. In l'act, the first
hal f o f t h e m a x i m i s a lso a d istr ib u tive p r o p o sa l,
and it doesn't fit the rule of the second half.
"From each according to his abilit1." suggeststhat
job s s h o u l d b e d i s tr ib u te d ( o r th a t m e n a n d
women should be conscripted to rvork) on the
bas i so f i n d i r , i d u a l q u a lifica tio n s.Bu t in d ivid u a ls
don ' t i n a n v o b v i o us se n se n e e d th e io b s lo r
whi c h t h e v a r e q u a l ilie d . Pe r h a p s su ch jo b s a r e
s c a r c e ,a n d t h e r e a r e a la r g e n u m b e r o f q u a lifie d
c an d i d a t e s :r v h i c h c a nd id a te sn e cd th e m m o st? If
their material needs are alreadv taken care o1-,
per h a p s t h e v d o n ' t n e e d to r vo r k a t a ll. Or il, in
s om e n o n - m a t e r i a l s en se , th e v a ll n e e d to r vo r k,
t he n t h a t n e e c lr v o n ' r d istin g u ish a m o n g th e m , a t
still won't have an adequate distributive criterion.
The sorts of things that I have listed cannot be
distributed equallv ro those with equal wants
because some of them are generally, and some of
them are necessarily, scarce, and some of them
can't be possessedat all unless other people, for
reasons of their own, agree on who is to possess
them.
Need generatesa particular distributive sphere,
within which it is itself the appropriate distribu_
tive principle. In a poor society, a high proportion
o f soci al w eal th w i l l be draw n i nto thi s sphere.
But given the great variety of goods that arises
out of any common life, even when it is lived at a
very l ow materi al Ievcl , othcr di stri brrti ve cri tcri ;r
will alwavs be operating alongside of need, and it
rvill alwavs be necessary to worry about the
boundaries that mark them off from one another.
Wi thi n i ts sphere, certai nl v, need meets the general distributive rule about r and ..yr.Needed
goods distributed to needy people in proportion
to thei r needi nessare obvi ousl y not domi nated by
a n y other goods. It's not havi ngJ/, but onl v l ackin g r that i s rel evant.B ut rve can now see.I thi nk.
that every criterion that has anv force at all meets
th e general rul e w i thi n i ts ow n sphere, and not
elsewhere. This is the eff'ect of the rule: dilferent
goods to different companies of men and rvomen
lor different reasons and in accordance with different procedures. And to get all this right, or to
g e t i t roughh'ri ght, i s ro map out the enti re soci al
worl d.
Hierarchi es
and C aste Soci eti es
least not to the naked eve. It rvould in anv case be
odd to ask a search committee looking, sav, lor a
Or, rather, it is to mxp out a particular social
worl d. For the anal vsi s that I propose i s i mmi n e nt and phenomenol ogi cali n character. It rvi l l
hos p i t a l d i r e c t o r r o m ake its ch o icc o n th e b a siso f
t he n e c d s o f t h e c a n clid a te sr a th e r th a n o n th o se
viel d not an i deal map or a master pl an but,
r a ther, a map and a pl an appropri ate to the peopl e
of th e s t a f f a n d t h e pa tie n ts o f th e h o sp ita l. Bu t
t he l a t t e r s e t o f n e e d s, e ve n if it isn ' t th e su b je ct
fbr u.hom it is drarvn, whose common lif'e it
reflects. The goal, of course, is a reflection of a
sp eci al ki nd, rvhi ch pi cks up rhose deeper under-
of p o l i t i c a l d i s a g r e e m e n r ,wo n ' t vie ld a sin g le d ist rib u t i v c c l e c i s i o n .
r v i l l n e e d u , o r k fir r m a n v o r h e r g o o d s.
I lar x ' s m a x i m d o c s n' t h clp a t a ll r vith r e g a r d to
standings of social goods ivhich are not necessarilf' mirrored
in the evervdal' practice of
d o mi nance and n-ronopol v.B ut w hat i f there are
t he d i s t r i b u t i o n o f p o litica l p o u e r , h o n o r a n d
t hm e , s a i l b o a t s , r a r e b o o ks, b e e u tifu l o b je cts o f
ev e r y s o r t . ' I ' h e s e a re n o t th in g s th a t a n vo n e ,
n o such understandi ngs?I har,ebeen assumi ng al l
a lo ng that soci al meani ngs cal l for the autonomy,
o r the rel ati \,eautonomv, of di stri buti ve spheres;
s t ric t l v s p e a k i n g . n c ed s. Eve n if r ve ta ke a lo o se
v iew a n d d e f i n e t h e ve r b to n e e dth e wa v ch ild r e n
a n d so thet, do much of the ti me. B ut i t's not
im possi bl e to i magi ne a soci ery w here domi nance
do, a s t h e s t r o n g e s t f o r m o fth e ve r b to n 4 r 7 1 ,1 1 ,g
and monopoll' are not r,iolations but enactments
\or
@
Michael Walzer
I
I
of meaning, where social goods are conceived in
hierarchical terms. In feudal Europe, for exam,
pl e , c l o t h i n g w a s n or a co m m o d itv ( a s it is to d a y)
bu t a b a d g e o f r a n k. Ra n k d o m in a te d d r e ss. T h e
m e a n i n g o f c l o t h i n g wa s sh a p e d in th e im a g e o f
the feudal order. Dressing in finerv to which one
rvasn't entitled rvas a kind of lie: it made a false
s t a t e m e n t a b o u t w h o o n e wa s. Wh e n a kin g o r a
goods - and birth and blood are dominant over
purity. Social meanings overlap and cohere.
The more perfect the coherence, the less possible it is even ro think abour complex equality. All
goods are like crowns and thrones in a hereditary
monarchy. There is no room, and there are no
cal frontiers.
almost as easilY
them. Things are
selves, back and
the
Nevertheless,
the closest we can
criteria, fbr autonomous distributions. In fact,
however, even hereditary monarchies are rarely
meanlngs
prime minister dressed as a commoner in order to
le a r n s o m e r h i n g a bo u t th e o p in io n s o f h is su b -
so simply constructed. The social understanding
of royal power commonly involves some notion of
ie c t s , t h i s w a s a k in d o f p o litic d e ce it. On th e
ot h e r h a n d , t h e d i l h cu ltie s o f e n fb r cin g th e clo th ing code (the sumptuxr-v laws) suggests that there
divine grace, or magical gifr, or human insight;
and these criteria for office holding are porentially
independent of bi rth and bl ood. So i t i s for most
wa s a l l a l o n g a n a l t er n a tivese n seo f wh a t clo th in g
m e a n t . A t s o m e p o in t, a t le a st, o n e ca n b e g in to
re c o g n i z e t h e b o u n d a r ie s o f a d istin ct sp h e r e
within which people dress in accordance u'ith
rvhat they can al}brd or rvhat thel'are willing to
spend or horv they want to look. 'fhe sumptuarl'
social goods: they are onlv imperfectly integrated
into larger systems; they are understood, at least
sometimes, in their own terms. 'I'he theory of
anywhereelse)to
ness.National
permanent mental
the sharing of
the membersof a
life. Sometimes
larvs mav still be enfbrced, but norv one can make
- a n d o r d i n a r v m e n a n d r vo m e n d o , in th ct, m a ke
- e g a l i t a r i a na r g u m en ts a g a in stth e m .
C a n w e i m a g i n e a so cie tv in wh ich a ll g o o d s a r e
hi e r a r c h i c a l l v c o n c eive d ? Pe r h a p s th e ca sr e svst em o f a n c i e n t I n d i a h a d th is fo r m ( th o u g h th a t is
a l l r - r e a c h i n g c l a i m , a n d it u o u ld b e p r u d e n t to
do u b t i t s t r u t h : f o r o n e r h in g , p o litica l p o wcr
seems allvays to have escaped the larvs of caste).
We t h i n k o f c a s t e sa s r ig id h .se g r e g a te dg r o u p s, o f
t he c a s t e s v s t e m a s a " p lu r a l so cie n ' ," a wo r ld o f
bo u n d a r i e s . t sB u t t h e s1 ' ste mis co n sr itu te d b ) ' a n
ex t r a o r d i n t r v i n t e g ra tio n o f m e a n in g s. Pr e stig e ,
lv e : r l t h , k n o w l e d g e , o fh ce , o ccu p a tio n , fb o d ,
c lo t h i n g , e v e n t h e s o cia l g o o d o f co n ve r sa ticlna: ll
are s u b j e c t t o t h e in te lle ctu a l a s lve ll a s to th e
ph v s i c a l c l i s c i p l i n c o f h ie r a r ch r ' . An d th e h ie r a r
goods explicates understandings of this sort
(where thev exist), and the theorv of complex
equality exploits them. We sav, for example, that
it is tvrannical for a man without grace or gift or
insight to sit upon the throne. And this is only
the first and most obvious kind of tvranny. We
can search for many other kinds.
Tvrannv is alwavs specific in chxracter: a particular boundarv crossing, a particular violation of
soci al meani ng. C ompl ex equal i tv requi res the
defense of boundaries; it works bv diiferentiating
goods just as hierarchy works by differentiating
people. But we can onlv talk of a regime of comp l ex equal i tv w hen there are many boundari esto
d efend; i rnd w hat the ri ght number i s cannot be
speci fi cd. There i s no ri ght number. S i mpl e
ties don't coincide,
ing number of
sensibilitiesand
the sharingtakes
perhaps,we should
distributive
those units. But
worked out
will depend uPon
the citizensabout
local autonomY,
standings that we
arguments- allof
in matters of
appeal to common
Politics,
commonality.In r
political powerisa
equalitv is easier: one dominant good rvidelv distributed makes an egalitarian society. But com-
and women, we
plexitv
gle as best theY
is hard:
horv manv
c h v i s i t s e l f d e t c r m i ne d b v th e sin g le va lu e o f r itua l p u r i t l ' . A c e r t a i n kin d o f co lle ctive m o b ilifi. is
goods nlusr be
autonomouslv conceived belore the relations thev
medi ate can become the rel ati ons of equal men
lnd llomen? -lhere is no certain answer and
possible, ftrr castes ()r subcastes can cultivate the
ou t l v a r d m a r k s o f p u r itv a n d ( r vith in se r - e r eIim
hence no ideal regime. But as soon as we start to
d i sti ngui sh meani ngs and mark out di stri buti ve
it s ) r a i s e t h e i r p o s i t io n in th e so cia l sca le .An d th e
s v s t e m a s a t v h o l c r ests u p o n a r e lig io u s d o ctr in e
spheres, w e are l aunched on an egal i tari anenterp rl sc.
t ha t p r o m i s e s e q u a l i ty'o l' o p p o r tu n in ' , n o r in r h is
lif ' e b u t a c r o s s t h e l i ve s o f th e so u l. T h e in d ivid ua l ' s s t a t u s h e r e a n d n o lv "is th e r e su lt o fh is co n i n h i s l a s t in ca r n ltio n
. and if
The Setting of the Argument
un s x t i s l a c t o r v c a n b e r e m cd ie d b r . a cq u ir in g
m e r i t i n h i s p r e s c n t l if' e wh ich u ill r a ise h is sta tu s
in t h e n c x t . " l e We s h o u ld n o t a ssu n e th a t m e n
' l' he pol i ti cal communi tv i s the appropri ate
setting for thi s enterpri se.It i s not, to be sure, a sel f-
du c t
and w o m e n a r e e \ . e r e n tir e ll co n te n t r vith r a d ica l
ine q u a l i n ' . N c r , e r t h ele ss,d istr ib u tio n s h e r e a n cl
contai ned di stri buti ve rvorl d: onl r.the rvorl d i s a
sel f-contai ned di stri buti ve u.orl d, l nd contemporarv science fiction invites us to speculate about a
now a r e p a r t o f a s i n g le svstcn l, la r g e ll u n ch a l
len g e d , i n u h i c h p u r itv is d o m in a n t o r e r o th e r
timc lvhen e\:en that lvon't be true. Social goods
a re shared, di r.i ded, and exchanged across pol i ti -
@
together (come
straints,shrpetheh
And if their desti
hands, the
ones whose
tributive criteria,
procedures,to i
this or that
set of leaders
citizensshouldbe
their own. If the
endlesslvvenal'as
some of the citi
fighting over the
The fight will be
structures of the
outcomes of previ
the product of
unav oi dehl e
tributive iustice
ComplexEquality
over
)ssiAlt
tary
Ino
bct,
rely
ling
ro f
iht;
rlly
ost
ted
ast
of
)rt
ex
al frontiers. Monopoly and dominance operate
dmost as easily bevond the frontiers as within
them.Things are moved, and people move them-
There is one last reason for adopting the view
of the pol i ti cal communi ry as setti ng, a reason
that I shall elaborate on at some length in the next
selves, back and forth
across the lines.
Nevertheless,the political community is probably
the closest we can come to a world of common
chapter. The community is itself a good - conceivably the most important good - that gets distributed. But it is a good that can only be
distributed by taking people in, where all the
meanings.Language, history, and culture come
together (come more closely together here than
anywhereelse) to produce a collective consciousness.National character, conceived as a fixed and
permanent mental set, is obviously a myth; but
the sharing of sensibilities and intuitions among
the members of a historical communitv is a fact of
life. Sometimes political and historical communitiesdon't coincide, and there mav well be a growing number of states in the rvorld today where
s e n s i b i l i t i e sa n d i n tu itio n s a r e n ' t r e a d ilv sh a r e d :
lat
t h es h a r i n gt a k e s p la ce in sm a lle r u n its. An d th e n ,
perhaps, we should look {br some way to adjust
or
ly
/e
distributive decisions to the requirements of
t h o s e u n i t s . B u t t his a d ju stm e n t m u st itse lf b e
w o r k e d o u t p o l i t i ca lly, a n d its p r e cise ch a r a cte r
w i l l d e p e n d u p o n u n d e r sta n d in g s sh a r e d a m o n g
r)f
te
th e c i t i z e n s a b o u t th e va lu e o f cu ltu r a l d ive r sity.,
local autonomv, ancl so on. It is to these understandings that ive must appeal when rve make our
b
a r g u m e n t s- a l l o f us, n o t p h ilo so p h e r sa lo n e , lo r
in m a t t e r s o f m o r a litt, a r g u m e n t sim p lv is th e
appealto common meanings.
o
Politics, moreover, establishesits orvn bonds of
co m m o n a l i t v . I n a r vo r ld o f in d e p e n d e n t sta te s,
political power is a local monopolv. These men
and women, we can sav, under rvhatever con
st r a i n t s ,s h a p e t h e ir o q ,n d e stin y. Or th e v str u g gle as best thev can to shape their orvn destinl'.
A n d i f t h e i r d e s t i n v is o n lv p a r tia lh ' in th e ir o r vn
h a n d s , t h e s t r u g g l e is e n tir e ly' so . T h e - v a r e th e
ones whose decision it is to tighten or loosen dis-
sensesofthat latter phrase are relevant: they must
be physically admitted and politically received.
Hence membership cannot be handed out by
some exteinal agencv; its value depends upon an
internal decision. Were there no communities
capable of making such decisions, there would in
thi s crsc be no good w orrh di sl ri huri ng.
The only plausible alternative to the political
is humanity itself, the society of
community
nations, the entire globe. But were we to take the
globe as our setting, we would have to imagine
what does not yet exist: a community that
incl uded al l men and w omen everyw here. We
would have to invent a set of common meanings
for these peopl e, avoi di ng i f w e coul d the sti puhtion ofour own values. And we rvould have to ask
t he members of thi s hypotheti cal communi tv (or
their hypothetical representatives) to agree
among themselves on what distributive arrangements and patterns of conversron are to count as
iust. Ideal contractualism or undistorted communication, which represents one approach - not my
ow n - to j usti ce i n parti cul ar communi ti es, may
well be the only approach for the globe as a
u'hole.20 But whatever the hypothetical agreement, i t coul d not be enforced w i thout breaki ng
t he pol i ti cal monopol i es of exi sti ng states and
central i zi ng pou'er at the gl obal l evel . H ence the
agreement (or the enforcement) would make not
fbr complex but for simple equalitv - if porver
was dominant and lvidelv shared - or simplv for
t r i b u t i v e c r i t e r i a , to ce n tr a lize o r d e ce n tr a lize
p r o c e d u r e s ,t o i n t e r ve n e o r r e fu se to in te r ve n e in
t h i s o r t h a t c l i s t r i b u tive sp h e r e . Pr o b a b lr ., so m e
tyrrnnv
s e t o f l e a d e r s m a k e th e a ctu a l d e cisio n s, b u t th e
c i t i z e n s s h o u l d b e a b le to r cco g n ize r h e le a d e r sa s
to live u'ith the dilficulties I have described: the
conti nual reappearxnceofl ocal pri vi l ege, the con-
t h e i r o u n . I f t h e l e a d e r s :r r e cr u e l o r stu p id o r
e n d l e s s l v v c n a l , a s th e r .o fte n a r e , th e citize n s o r
ti nual reasserti onof gl obal stari sm. In the second
case, thev rvoulcl have to live with dilficulties that
so m e o f t h e c i t i z en s will tr i to r e p la ce th e m ,
f i g h t i n g o v e r t h e distr ib u titin o f- p o litica l p o r ve r .
'l ' h e f i g h t r v i l l b e sh a p e d b v th e in stitu tio n a l
are consi derabl yw orse. I w i l l have a l i ttl e morc to
s a1'about these di ffi cul ti es l ater. For nol v I take
st r u c t u r e s o l - t h e c o m m u n itv
th a t is, b v th e
o u t c o m e s o f p r e r , io u s fig h ts. Po litics p r e se n r is
t h e p r o d u c t o { ' p o litics p lsr - Ir e sr lb iish e s a n
u n a r - o i d a b l es e t t i n g fb r th e co n sid e r a tio n o f d ist r i b u t i r . ej u s t i c e .
if power was seized, as it probablv
l voul d be, b1'a set ofi nternati onal burcrucrats. In
the first case, the people ofthe rvorld rvould have
them to be reasons enough to l i mi t mvsel f to
c i ti es, countri es, and states that har,e,over Iong
peri ods of ti me, shaped thei r ol vn i nternal Ii fe.
lVith regard to membership, hou,ever, important questi ons ari se benreen and am<i ng such
communi ti es, and I shal l trr.to fi rcus on them and
@
MichaelWalzer
to draw into the light all those occasions when
ordinarv citizens focus on them. In a limited way,
the theory of complex equality can be extended
from particular communities to the society of
nations, and the extension has this advantage: it
will not run roughshod over local understandings
and decisions. Just for that reason, it also will not
yield a unilorm system ofdistributions across the
globe,and it will only begin to addressthe problems raisedby masspovertyin many partsof the
globe. I don't think the beginning unimportant;
in any case,I can't move beyond it. To do that
would require a different theory, which would
take as its subiectnor the common life of citizens
but the more distancedrelationsof states:a di!
ferenttheory,a differentbook,anothertime.
Notes
1
2
3
-l
5
6
7
@
Sec John Rawls, A TheoryoJ'Justirc(Cambridge,
Ilass., l97l); Jurgen Habermas, Legitimation
Cn.rro,trans. Thomas NlcCarthy (Boston, 1975),
esp. p. ll3; Bruce Ackerman,SocialJustite in the
Liberal Stare(NervHaven, 1980).
Robcrt Nozick makes a similar argument in
Anarch.y,State, unl Urapia (Nerv York, 197.1),
pp. 1,19-50,but rvith radicalll.individualisticconclusionsthat secmto me to miss the socialcharacter ofproduction.
Ralph Waldo Emcrson, "Ode," in The Complete
Essul,ran/ Other lfritings, ed. Brooks Atkinson
( N c r vY o r k , 1 9 1 0)p, .7 7 0 .
John Stuart \{ill, On Liberty, in Tht Philovph.y of
John StuurtNill, ed. \Iarshall Cohen (Nerv York,
l96l), p. 255. For an anrhropologicalaccountof
liking and not liking social goods, see Ilarv
Douglas and Baron Isherwood, The llorld r{'Coods
( N e w Y o r k , 1 9 7 9) .
William James, quotcd in C.R. Sny.der and
Horvard Fromkin, Uniqueness:The Humun Pursuit
(Nerv York, 1980),p. 108.
oJDrJlerence
Arcn't social mcanings, as }larx said, nothing
other than "the ideas ol'the ruling class," "thc
dominant materialrclationshipsgraspedas ideas"l
(Karl trIan, The (]ermun ltleology, ed. R. Pascal
(New York, 1917),p. 89), I dont think that thcv
are cver onlv that or simpll that, though the mcmbers of the ruling classand the intellectuals ther
patronizcmav lvell be in a position to cxploit and
distort social meanings in their own intcrests.
When thel do that, horveter, thcv are likclv to
encountcr resistance, rooted (intellcctuallv) in
thosesamcmeanings.A peoplc'sculturc is ahva.,,
s
a joint, er en if it isn't an cntirelv cooperative,production; and it is alrvar.sa compler production.
Thc common understandingof- ptrticular goods
incorporatesprinciples,proccdures,conccptionsof
agencv,that the rulers rvould not chooscif thelrvere choosing right non, and so provides the
terms of socialcriticism.The appealto rvhatI shall
call "intcrnal" principlesagainstthe usurpationsof
porverlulmen and rvomenis the ordinarv form of
critical discourse.
Bernard Williams, Probltms rf
the Srlf
l4
ManuscriPts,in
(London, 1963)'
note an earlier
Smith's Tieory o/
l 8l 3), v ol .I, pP .
believed that
ally conformedto
l5
Philosophicul Papers, 195G1972 (Cambridge,
England, 1973), pp. 230-'19 ("The Idea of
Equalitv"). This essavis one of the starringpoints
of mv orvn thinking about distributivejustice.See
alsothe critique of Williams'sargumenr(andof an
earlier essav of ml own) in Am1. Gutmann,
Liberal Equulity (Cambridge, England, 1980),
chap.4.
See Alan W. Wood, "The Nlarxian Critique of
Justice," Philosophy and Public Afuirs | (1972):
241-82.
9 I{ichael Young, The Rise of the Meritocracy,
1870 2033 (Harmondsworth,England, 196l) - a
brilliant pieceof socialscienceIiction.
l0 Ravls, TheoryofJustice,pp. 75ft
ll I should note here rvhatwill becomemore clearas
I go along, that political porver is a specialsort of
good.It has a twofold character.First, ir is like the
other things that men and rvomen make, value.
exchange,and share:sometimesdominant, sometimes not; sometimeswidell held, sometimesthe
posscssion
of a verv felv. And, sccond,ir is unlike
all the other things because,holever it is had and
lvhoeverhas it, political porver is the regulative
agcncv lbr social goods generalll. It is usedto
dcfcnd the boundaries of all the distributive
spheres,including its own, and to enforce the
common understandingsof what goods are and
what they'are for. (But it can also bc used,obviouslv, to invadethe different spheresand to override those understandings.)In this secr.rnd
sense,
rvemight sav,indeed,that politicalpolveris alwavs
dominant - at rhe boundaries,but not within
them. Thc central problem of polirical life is to
maintainthat crucial distinctionbetrveen"at" and
"in." But this is a problem that cannot be solved
given the imperativesof simpleequality..
t2 Sce \'larr's comment, in his "Critique of the
Gotha Program," that the dcmocraticrepublic is
the "lbrm of state" rvithin rvhichthc classstruggle
rvill be fought to a conclusion: the struggle is
immediatell' and rvithout distortion reflectedin
political lifc (Ilarx and Engels, Selrctil Worhs
l 95l l , rol . II, p.3l ).
l .\Ioscou.,
l3 Blaise Pascal, Tht Pensy'es,
trans. J.II. Cohen
(Harmondsworth,
(no. 244).
N1arx,
Karl
l6
mistake neither
See the summarY
of a Commonneak,
(Cambridge,
Cf. Nozick on
Y
I
Complex Equality
bhe
1t;
rat
,td
(Harmondsrvonh, England,
ino.244\.
l4
NS
if-
l- )
ie,
of
rts
t6
1961),
p.
96
Karl
trIarx,
Economic and
Philosophica!
Alanuscripts,in Early lfritings, ed. T.B. Bortomore
(London, 1963), pp. 193-94. k is interesting to
note an earlierecho of Pascal'sargumentin Adam
Smith's lleo4y of Morul Sentiments(Edinburgh,
l8l3), vol. l, pp.378-79; but Smith seemsro have
belieredthat distributionsin his own sociervacrually conformedto this view of appropriateness
- a
mistakeneitherPascalnor Marx evermade.
Seethe summarvaccountin JeanBodin, Sl.r Baats
o.l-a Commonntulr,ed. Kenneth Douglas NIcRac
( C a m br id g ella
, ss., 1 9 6 2 )p, p . 2 1 0 - 1 8 .
Cf. Nozick on "patterning,",lnarch1t,Stute,tnd
Utopial2l, pp. 155ff.
17 trIarx,"GothaProgram",
(above)
p. 23.
18 J.H. Hutton, Co.rrrin India: Its Nuture,Function,
undOrigins(4th ed.,Bombay,1963),pp. 127-28. I
l9
20
have also drawn on C6lestinBougl6, Essays
on the
Custe System, trans. D.F. Pocock (Cambridge,
England, 1971),esp. Pan III, chaps.3 and 4; and
Louis Dumont, Homo Hierarchus: The Caste
System and lts Impliations (revised English ed.,
Chicago,1980).
Hutton, Casrein India, (above)p. 125.
See Charles Beitz, Political Theury and
InternationalRelations(Princeton,1979),part III,
for an effort to apply Rawlsianideal conrractualism
to internationalsocietv.
tn
n,
)),
of
:):
v'
a
1S
rf
le
).e
d
e
o
e
e
d
s
n
f,
I
t
s
;
I
I
r(n\
Y.:_v/