University of Richmond

UR Scholarship Repository
Political Science Faculty Publications Political Science

9-1981

Understanding the General Will

Richard Dagger
University of Richmond, rdagger@richmond.edu

Follow this and additional works at: http://scholarship.richmond.edu/polisci-faculty-publications

Part of the Ethics and Political Philosophy Commons

Recommended Citation
Dagger, Richard. "Understanding the General Will." Western Political Quarterly 34, no. 3 (September 1981): 359-71. doi: 10.2307/
447216.

This Article is brought to you for free and open access by the Political Science at UR Scholarship Repository. It has been accepted for inclusion in
Political Science Faculty Publications by an authorized administrator of UR Scholarship Repository. For more information, please contact
scholarshiprepository@richmond.edu.
 UNDERSTANDING THE GENERAL WILL
RICHARD DAGGER
Arizona State University

EW CONCEPTS in the historyof politicalthoughthave proved so

troublesome as Rousseau's notion of the general will. Rousseau must
bear much of the blame for this,of course, for the discussion of the
general will in his Social Contractis uncharacteristicallyterse and abstract.
Troublesome as it has been, though, there is reason to believe that we are
now approaching an adequate understandingof the general will. I say this
because there seems to be growing agreement among Rousseau's commen-
tatorsthatthe general will not onlycan be understood,but thatit can best be
understood in rationalisticterms.' Indeed, where explications once were
couched in termsof "real" and "higher" wills,one is now more likelyto find
the general will explained in terms of the prisoners' dilemma and Pareto-
optimality.2
While I do not accept all of these rationalisticreadings of the general
will, I do share the general convictionthat we can make sense of Rousseau's
concept, and his argument,withoutresortingto metaphysicsor psychology.
What I shall offerhere, accordingly,is in some respectsonly a variationon a
theme now well known to studentsof Rousseau's politicalphilosophy.It is an
importantvariationnonetheless,for it enables us to reconcile passages in the
Social Contractwhich otherwiseappear to be contradictory.That, at least, is
what I shall argue in this essay.
I proceed in the followingmanner. First I set out a general account of
what Rousseau means by "the general will"- an account whichresemblesin
its main lines, if not all its details, Brian Barry'sanalysisof the general will.3
This account is defended in the second part of the essay, where I show how
it helps us to understand two of the more controversialaspects of Rousseau's
argument in the Social Contract.In part three I extend this account (and
provide the variationmentioned in the last paragraph) by drawing a distinc-
tion, implicitand almost unmarked in Rousseau's writings,between thegen-
eral will and a general will. With the aid of this distinction,I argue, we can
make sense of Rousseau's bafflingand apparently contradictoryremarks
about voting. Once this is demonstrated,I conclude by raising some ques-
tions about the utilityof the concept of the general will.

NOTE: I wish to thank Terence Ball, Peter Fuss, C. B. Macpherson, and thisjournal's anony-
mous refereesfor theirvaluable commentson earlier draftsof this paper.
'I include the following,interalia, in thiscategory:Glen Allen, "La Volontede tousand La Volonte
generale:A Distinctionand Its Significance,"Ethics71 Uuly 1961): 263-75; George Kateb,
"Aspects of Rousseau's Political Thought," PoliticalScienceQuarterly76 (December 1961):
519-43; James McAdam, "What Rousseau Meant by the General Will,"Dialogue 5 (1967):
498-515; Roger Masters,The PoliticalPhilosophy ofRousseau (Princeton: Princeton Univer-
sityPress, 1968), esp. pp. 323-35; and JudithShklar,Men and Citizens:A StudyofRousseau's
Social Theory(Cambridge: Cambridge UniversityPress, 1969), esp. pp. 184-97. Further
examples of rationalisticreadings of the general will are cited in the next two footnotes.
2 For an
explanation of the general will in termsof the prisoners'dilemma, see W. G. Runciman
and A. K. Sen, "Games, Justiceand the General Will," Mind 74 (1965): 554-62; for an
explanation in terms of Pareto-optimality, see Virginia Held, The Public Interestand Indi-
vidual Interests
(New York: Basic Books, 1970), pp. 99-107.
Brian Barry, "The Public Interest," in Political Philosophy,ed. Anthony Quinton (Oxford:
Oxford UniversityPress, 1967), esp. pp. 119-26. For similarinterpretations,see Andrew
Levine, The Politicsof Autonomy: A Kantian Reading of Rousseau'sSocial Contract(Amherst,
Mass.: Universityof MassachusettsPress, 1976), pp. 40-43, and John Charvet, The Social
Problemin the Philosophyof Rousseau (Cambridge: Cambridge UniversityPress, 1974),
chapter 4.
360 PoliticalQuarterly
Western

I
The idea of the general will rests on a fundamentaldistinctionbetween
two aspects of a person. Everyone,in Rousseau's view, may be thoughtof as
both a man - an actual, identifiableperson - and a citizen. Insofar as we
are men, we are each unique; each of us, that is, has his own particular
identityand set of interests.Insofar as we are citizens,however,we are alike
in thatwe are membersof the public; and as membersof the public we share
a common interest in the welfare of the body politic. Everyone, con-
sequently,has both a particularinterestas a man and a general interestas a
citizen.
From the distinctionbetween man and citizen Rousseau moves to a
correspondingdistinctionbetween the particularor privatewill and the gen-
eral will. The private will aims at the fulfillmentof the particularinterestof
a man, an actual individual,while the general will seeks to furtherthe inter-
est of the citizen. Since the interestsof all citizensare the same, the object of
the general willmay be said to be the common good or public interest- that
is, the interestwe all share as membersof the public. Rousseau distinguishes
the privatefromthe general will in thismanner: "the privatewill tends by its
nature toward preferences,and the general will toward equality." (II,1:59.)4
As men withprivatewills,we naturallytend to grant precedence to our own
interestsand desires. The private will is partial both because it is the will of
an identifiable individual, one who is only part of the body politic, and
because it is biased: it places a higher value on the interestsof the self than
on the interestsof others. The general will, in contrast,tends to equality
because it necessarilygrants equal consideration to everyone's interests;or,
at least, to everyone withinthe body politic. Because it focuses on the com-
mon interestwe share as citizens,the general will is impartial:it considers
only the interestsof the abstractperson Rousseau calls the citizen. Since we
are all the same qua citizen,the general will is devoted equally to all.
An example may help to clarifythis distinction.Consider the case of
someone who has the abilityto be an unusually adept burglar. Such a person
may well find it in his interest,as a man, to put his skillsto use. His private
will may then be to steal fromothers in order to satisfyhis own desires. The
general will,however,directsour potentialburglar to obey the law. Since it is
the will of the citizen, the general will requires us to ignore at times our
particular interestsand personal attributes- our possessions, our position
in society,our abilities- and to thinkof ourselves onlyas members of the
public. Laws against burglary are in the interestsof all citizens, if not all
men; and if our potentialburglar thinksof himselfonly as a citizen,he will
recognize that he ought to respect the law.
As I understand it, then, the general will performs much the same
functionas John Rawls's "veil of ignorance."5Rawls argues that a hypotheti-
cal social contractwill produce generally acceptable principles of justice if
the parties to that hypotheticalcontractchoose these principlesfroma prop-
erly defined "initialsituation."The main featureof thissituationis the "veil
of ignorance,"whichdeprives the partiesto the contractof most informnation
about theirpersonal identities.More precisely,the parties do not know what
their socioeconomic positions are, what abilitiesor disabilitiesthey have, or

4All citations in parentheses refer to the book, chapter, and page number in On the Social
Contract,ed. Roger Mastersand trans.JudithMasters (New York: St. Martin'sPress, 1978).
The referencehere is to Book II, chapter one, page 59.
5See John Rawls,A TheoryofJustice(Cambridge, Mass.: Harvard UniversityPress, 1971), ?24,
"The Veil of Ignorance."
 theGeneralWill 361
Understanding

even what their particular aspirations happen to be. Thus Rawls asks the
individual to blind himselfto his private interestsand to assume an abstract
identityso that the principles he choosesfor all will be acceptable to all.
Behind the veil of ignorance, Rawls points out, "No one knows his situation
in societynor his natural assets,and thereforeno one is in a position to tailor
principles to his advantage."6 In this way the veil of ignorance forces the
parties to the contractto adopt a common viewpoint; or, to put it in Rous-
seau's terms,the veil of ignorance leads the parties to promote their com-
mon interestas citizensratherthan theirprivateinterestsas men. The terms
are different,but the point is much the same.
There is also a moral dimension to the general will,just as there is to the
veil of ignorance. Rousseau does not invoke the general will simply as a
counterweightto selfishinterests.This would be futile,for one's interestas a
man may well outweigh one's interestas a citizen. An industrialistmay find,
for instance, that when all things are considered it is not in his interestto
installpollution-inhibiting devices in his smokestacks,even though it is in his
interestas a citizento reduce air pollution. Rousseau's claim, however,is that
in some cases we ought to consider onlyour interestsas citizens.The purpose
of the general will,therefore,is to provide a principle which will lead to just
public policy. If we consider the interestsof the citizenand not those of men,
Rousseau maintains,we will reach decisions which establishlaws and policies
in accordance with the common interest.This is possible because from the
standpointof the citizen- as members of the public - we share the same
interests.Considered abstractly,as citizens,the burglar and his victimshare a
common interestin the enforcementof a law against burglary,just as the
industrialistand the person who lives near his factoryshare a common
interestin eliminatingair pollution. What Rousseau suggests,in sum, is that
only the general will of the citizen, and not the private will of the man, is
relevantto public policy decisions.
When Rousseau uses the notion of the general will,then, he singles out
a certainperspective- a moral perspective- and claims thatthisis the only
perspectivewe ought to consider in reaching public decisions. In this sense
the general will is a moral imperative or principle, and its purpose is to
guarantee that the claims of no particularindividuals are given preference.
Everyone receives equal consideration because public decisions take into ac-
count only the viewpoint of the citizen. Any other perspective is at best
morally irrelevant. Here we see the emphasis on equality which is such a
marked characteristicof Rousseau's political philosophy. Thus Rousseau
writesthat
thesocialcompactestablished an equalitybetweenthecitizenssuchthatthey
all engagethemselves underthesameconditions and shouldall benefitfrom
the same rights.Thus by the verynatureof the compact,everyact of
sovereignty, whichis to say everyauthentic or
act ofthegeneralwill,obligates
favorsall citizens so thatthesovereignknowsonlythenationas a body
equally,
and makesno distinctions betweenany of thosewho composeit. (II, 4:63;
emphasisadded.)
A furtherimplication is that laws established in accordance with the
general will do more than grant equal consideration to everyone: theyactu-
ally are in everyone's interest.For a law sanctioned by the general will pro-
motes everyone's interestqua citizen. Such a law may not be the firstchoice
of many individuals, but it is acceptable to all because it makes some con-

6Ibid.,?24, p. 139.
362 Western
PoliticalQuarterly

tributionto everyone's well-being.It is, as Barry has said, "a sort of highest
common factorof agreement."7
There is also a close connectionbetween the general will and Rousseau's
conception of moral freedom. If the people are furnished with adequate
information,Rousseau says, and are allowed no communicationwhen they
deliberate (to preventlogrolling),the outcome of their vote will conform to
the general will. Everyone can vote for and obey laws which are in his
interestas a citizen (although everyone might prefer to vote for and obey
laws which are in his interestas a man), and this satisfiesRousseau's defini-
tion of moral freedom as "obedience to the law one has prescribed for
oneself.. ." (1,8:56.)
Two other points should be noted before I take up the defense of this
interpretation.This way of understanding the general will is consistent,in
the firstplace, withwhat Rousseau has to say about groups withinthe body
politic. There is a sense in which each of these groups, or "partial associa-
tions,"has its own general will,for the individuals who compose each group
share a common interestin their capacities as members of that association.
These groups, in other words, may be considered publics. When we regard
them from the perspectiveof the body politic,however, it is clear that the
will of each association is a privatewill- a will which may be antagonisticto
the general will of the state. Rousseau disapproves of these associations,
consequently,because he fears that theywill divide and divertthe loyaltyof
the people; men will come to thinkof themselvesas merchantsand farmers,
Catholics and Protestants,ratherthan citizens.When thishappens, Rousseau
warns, the private wills of these groups will prevent the general will from
prevailingwhen the people vote.
Understanding the general will as a principle meant to guide public
decisions is also consistent,secondly, with Rousseau's remarks on the limits
of the general will. According to Rousseau, the general will applies to laws,
but not to decrees. A law is a general policy, a rule which governs the
conduct of every member of the body politic,while a decree is an act which
refers to particular, identifiable individuals. The general will "loses its
natural rectitude when it is directed toward any individual, determinate
object. Because then,judging what is foreignto us, we have no true principle
of equityto guide us." (II,4:62.)8 There is no true principleof equityin these
cases because men are involved, not citizens. Instead of appealing to the
general will,then,we mustdecide on the meritsof the particularcase - and
these decisions are best left to executives and judges, not to the people as a
whole.

II
The foregoingaccount of the general will is littlemore than a sketchof
what I take to be the most plausible interpretation.What I want to do in this
section is to elaborate and defend this account by presenting two kinds of
evidence for it.
There is, to begin with,the evidence to be found in the way Rousseau
talksabout the general will. The factthat Rousseau distinguishesthe general
will fromthe will of all is evidence enough that the general will is not merely
whateverthe people, or most of them,may want at any particulartime. This
does not lead immediatelyto the conclusion that the general will should be
understood as a principle,of course, since it leaves open the possibilitythat

7Barry, "The Public Interest,"p. 120.

8See also the chapter on "Law" (11,6).
 theGeneralWill 363
Understanding

the general will is some sort of metaphysicalentity,such as a "higher will."

But it is difficult,at best, to square this metaphysicalview with Rousseau's
reference in PoliticalEconomyto "the maximsof the general will ...." For
while maxims may be derived fromprinciples,theycan not be derived from
metaphysicalentities.
Other evidence of this sort provides more direct support for the claim
that the general will is best understood as the principle to be followed in
mattersof public policy. In another passage in PoliticalEconomy,Rousseau
describes the general will as "the firstprinciple of public economyand the
fundamentalrule of government.. .."10 And in the Social Contract,Book II,
Rousseau proclaims that
whatgeneralizesthewillis notso muchthenumberof votesas thecommon
interestthatunitesthem,because in thisinstitution everyonenecessarily
subjectshimselfto the conditionshe imposes on others,an admirable
agreementbetweeninterest andjusticewhichconferson commondelibera-
tionsa qualityof equitythatvanishesin thediscussionof privatematters,for
wantof a commoninterestthatunitesand identifies the rule of thejudge
withthatof theparty.
However onetracestheprinciple,
one alwaysreachesthesame conclusion,
namelythatthe socialcompactestablishedan equalitybetweenthecitizens
suchthattheyall engagethemselves underthesame conditionsand should
all benefitfromthesamerights.(11,4:63;emphasisadded.)1'
In reaching public decisions, thatis, the sovereignshould consider the inter-
ests of the citizens,and not the men, who compose it; for it is this concern
for "the common interest that unites them" that generalizes the will. We
generalize the will,therefore,when we act on Rousseau's principle: vote as a
citizen,not as a man.
If these passages do not absolutely prove that the general will is best
understood as a principle or imperative,they certainlyprovide strong sup-
port for that interpretation.Further support can be found, moreover, in
evidence of a second kind. This consistsin showing that the interpretationI
have set out enables us to make sense of some of the more controversialand
perplexing aspects of the Social Contract.Here I shall consider the following
questions: (1) What does Rousseau mean when he says that one may be
"forced to be free"? and (2) What is the differencebetween the general will
and the will of all?
(1) The claim that someone may be "forced to be free," found in Book
I, chapter seven of the Social Contract, strikes many readers as self-
contradictory.How can one be forced to be free,it is asked, when freedom is
the very absence of coercion? Much depends on what one means by "free-
dom," of course, and this objection holds only if the word is understood in
this narrow,"negative" sense. Rousseau himselfconceives of three orders of
freedom- natural,civil,and moral - and he defines moral freedom,as we
have seen, as "obedience to the law one has prescribedfor oneself." When he
says that one may be forced to be free, then, he means that one may be
compelled to obey a law to which he has given his consent.
This does not meetall objections,however.For whyshould one have to be
forcedto obey a law to whichone has consented?Isn't thiscontradictory? Given

9PoliticalEconomy,in On theSocial Contract,ed. Masters,p. 223; emphasis added.

"'Ibid., p. 213; emphasis in original.
Note also the followingpassage fromtheGenevaManuscript:"If thecommon interestis the object
of theassociation,itis clear thatthe general willshould be therule of thesocial body'sactions.
This is the fundamentalprincipleI have triedto establish."On theSocial Contract, ed. Masters,
p. 174.
364 PoliticalQuarterly
Western

Rousseau's distinctionbetweenprivatewillsand thegeneralwill,itclearlyis not.

The privatewillof one who cheats on his taxes, forinstance,is in conflictwith
his general willas a citizen.The cheater may "view what he owes the common
cause as a free contribution,the loss of which will harm others less than its
paymentburdens him"and consequently"wishto enjoy therightsof the citizen
withoutwantingto fulfillthe duties of a subject .. ." (1,7:55.) These two wills
are in conflictbecause thecheaterwantsto enjoy thebenefitsof thesocial order
- "the rightsof the citizen"- withoutcontributinghis share to the mainte-
nance of thatorder - "the duties of a subject."If he followshis privatewill,he
acts against "the general will he has as a citizen." (1,7:55.) The spread of this
kind of injustice,Rousseau observes,
wouldcause theruinof thebodypolitic.
Therefore,in order for the social compactnot to be an ineffectual
formula,it tacitlyincludesthefollowing engagement, whichalone can give
forceto the others:thatwhoeverrefusesto obeythe generalwillshallbe
constrainedto do so by the entirebody;whichmeansonlythathe willbe
forcedto be free.For thisis theconditionthat,bygivingeach citizento the
homeland,guaranteeshim againstall personaldependence... and alone
gives legitimacyto civilengagementswhichwithoutit would be absurd,
tyrannical,and subjectto themostenormousabuses.(1,7:55.)
The state,according to the view set out here, is a cooperative enterprise,
and the citizensare all pledged to share its burdens. Only those who tryto
evade this contractual duty - those who "wish to enjoy the rights of the
citizen withoutwantingto fulfillthe duties of a subject" - must be "forced
to be free." These individuals are nowadays called "free riders." The tax
cheater mentioned earlier is a free rider, for example, because he wishes to
receive the benefitsof social cooperation withoutcontributinghis fair share,
whateverit may be, toward the provision of those benefits.If the number of
free riders is relativelysmall, then the benefitsof social cooperation will still
be provided; but when their numbers go beyond a threshold,they may well
render social cooperation impossible. In large groups the temptationto ride
free is especially strong,and Rousseau recognizes that some means must be
found to encourage men to honor their commitments.Otherwise, he says,
the sovereign "would have no guarantee of the subjects' engagements if it
did not find ways to be assured of their fidelity."(I,7:55.)12
One may still wonder what all this has to do with freedom. Why not
simplysay that membersof the body politicwill not be allowed to take unfair
advantage of one another? The answer is that Rousseau speaks of freedom
here because he considers the state- or at least the dejure state he sketches
in the Social Contract- to be the realm of freedom. The just state grants
equal rightsand protectionto all, therebyhelping to secure everyone from
personal dependence. Anyone who accepts and followssocial rules whenever
they favor him, yet breaks them when it serves his purposes, is contributing
to the destructionof that realm. Such a person is a parasite whose actions
threatenhis own freedom as well as that of others. And thisis why Rousseau
says that the free rider must be "forced to be free": forced, that is, to act in
accordance with his own will as a citizen.13
12For a discussion of the
relationshipbetween group size and the conduct of members of the
group, see Mancur Olson, The LogicofCollective Action(New York: Schocken Books, 1971),
esp. chap. 2. Note also Rousseau's preferencefor small communities:e.g., in Social Contract
(111,13).
13Cf. Brian (London: Routledge & Kegan Paul, 1965),
Barry'sexplanation in PoliticalArgument
p. 198: "Rousseau does not deny that it may be in your interest to breaka law which
benefits you qua member of the community; all he says is that it is certainlyin your
 theGeneralWill 365
Understanding

(2) Understanding the general will as a principle also helps to make

sense of Rousseau's distinctionbetween the general will and the will of all.
This distinctionis set out in the followingparagraph:
There is oftena greatdifference betweenthewillof all and thegeneral
will.The latterconsidersonlythe commoninterest;the formerconsiders
and is onlya sumof privatewills.But takeawayfromthese
privateinterest,
same willsthe plusesand minusesthatcanceleach otherout, and the re-
mainingsumof thedifferences is thegeneralwill.(11,3:61.)
The object of this paragraph ostensiblyis to elucidate, but many readers
have found that it has the opposite effect.John Plamenatz, for instance,has
charged that this quasi-mathematicalaccount of the general will is, if taken
literally, "sheer nonsense." According to Plamenatz, the "pluses" and
"minuses" of privatewillsmustbe thatwhich is peculiar to each of them,and
the mathematicalexplanation is then:
Let John'sbe x+a, Richard'sx+b, and Thomas'sx+c; x beingwhatis com-
monto themall,and a, b, and c, whatis peculiarto each. If thegeneralwill
is whatremainsafterthe"pluses"and "minuses"have cancelledeach other
out, it is x; but if it is the sum of the differences,it is a+b+c. Whicheverit is,
itcannotbe both;and thesecondalternativeis too absurdto be considered.
Bewareof politicalphilosopherswho use mathematics, no matterhow sim-
theirmeanings!14
ple, to illustrate
Here the problem arises from Rousseau's equation of the general will
with the "sum of the differences." If this phrase is taken literally,then
Plamenatz's complaint is probablyjustified. But the surrounding passages
indicate that what Rousseau means by the "sum of the differences"is what
Plamenatz designatesx, not a+b-+c. In a footnoteto the passage in question
Rousseau says that "the agreement of all interestsin formedin opposition to
the interestof each. If there were no differentinterests,the common inter-
est, which would never encounter any obstacle, would scarcely be felt."
(11,3:61.) In the succeeding paragraph of the text he goes on to claim that,
"If, when an adequately informed people deliberates, the citizens were to
have no communication among themselves,the general will would always
result from the large number of small differences,and the deliberations
would always be good." (II,3:61.)15 Furthermore,one of the reasons for
Rousseau's quarrel with "partial associations" is that, "The differencesbe-
come less numerous and produce a result that is less general." (II, 3:61.)
The point of these passages may be put this way: the general will is the
"sum of the differences"because Rousseau believes that the more small
differencesthere are in society,the more likelypeople are to recognize and
pursue the common interest.When there is a clash between the interestsof a
few large groups, the members of each group will tend to see only their
interestsqua members of the group or faction.But when there are a "large
number" of differencesin the particular interestsof individuals, the indi-

intereststo voteforit,and thatif you have votedin favorof a certainpunishment fora

certaincrimeyouhaveno businessto complainifyourwishfora certaingeneralpolicyis
appliedto youin a particular
case."
4JohnPlamenatz, Man and Society,
I (London: Longmans,Green,1963),p. 393. Cf. Virginia
Held'suse of Pareto-optimality
to defendRousseau'smathematics in ThePublicInterest
and
IndividualInterests,
pp. 102-4.
"This is one of theseveralcaseswhereG. D. H. Cole's translation
in thewidelyused Everyman
editionofTheSocialContract andDiscoursesis misleading.Cole translatesRousseau's"grand
nombre"as "grandtotal."See TheSocialContract and Discourses,
ed. and trans.G. D. H.
Cole (NewYork:Dutton,1950),p. 27, and compareDu Contrat Social,ed. RonaldGrimsley
(OxfordUniversity Press,1972),p. 128.
366 PoliticalQuarterly
Western

viduals are better able to see their common interestqua citizens. Each sees
that he can seldom, if ever, get all that he wants,and he soon perceives the
need for some sort of rule for allocating the benefitsand burdens of social
cooperation. The most reasonable rule, Rousseau suggests, is that which
regards us all as abstractions- as citizens- for in this way all are treated
equally. The will of all - the "sum of private wills"- will not produce an
outcome acceptable to all because, when the "pluses" and "minuses" are
tallied, it will favor the interestsof some at the expense of others. The
general will, however, is the "sum of the differences"because it represents
the common ground which all can accept despite their differences.
With the aid of the interpretationof the general will set out in section
one of this paper, then, we are able to present coherent accounts of Rous-
seau's distinctionbetween the general will and the will of all and of his claim
that upon occasion someone may be "forced to be free." This, togetherwith
the evidence fromRousseau's statementsabout the general will,suggeststhat
the general will is indeed best understood as a principle to guide decisions
on public matters.Yet this does not resolve all the problems surrounding
Rousseau's use of "the general will." Rousseau's mysteriousremarks about
the relationship between voting and the general will - and especially his
apparently inconsistentclaims about whether the individual should look to
his personal interestor the public interestwhen voting - have still to be
explained. To provide this explanation, we shall have to extend the in-
terpretationoffered so far by introducinga distinctionbetween the general
will and a general will.

III
For the most part Rousseau uses "the general will" in the sense of an
imperativeor principle;but he sometimesalso speaks of a particulardecision
as an expression of the general will (e.g., 11,3:61). If the people reach a
decision whose outcome actually promotes the common interestthey share
as citizens,Rousseau tends to say that the people have expressed or declared
the general will. But here, plainly,we have two differentuses of "the general
will," for decisions neither express nor declare principles: they follow or
conformto them. These differentuses, whichare never clearlysorted out by
Rousseau, are convenientlymarked by the distinctionbetween the general
will- the principle that all public decisions must take into account only the
interestsof the citizen- and a general will- a specificdecision which is in
accordance withthis principle.
This distinction,strange as it may firstappear, is analogous to a more
readily apparent distinctionwhich is drawn in discussions of the public
interest.16For however one may define it, the public interestis certainly
differentfroma policywhich(one believes) is in the public interest.A tax cut
may be in the public interest,for example, but it is not itselfthe public
interest.And if a tax cut is in the public interest,it is only because it some-
how promotes the public interest.In just this way a policy which conforms to
the general will is not itselfthegeneral will: it is, instead,a general will.
Rousseau never explicitlydraws this distinction,though, and his failure
to do so obscures importantparts of his argument. This is expecially true of
his remarks about voting. If we attend to the distinctionbetween theand a
general will,however,we may shed some lighton these remarks.

'6Clarke Cochran develops a similardistinctionbetween "x is a means to the common good," "x
is a common good," and "x is the common good" in "Yves R. Simon and 'The Common
Good': A Note on the Concept," Ethics88 (April 1978): 237.
 theGeneralWill 367
Understanding

Rousseau's comments on voting actually present two related problems.

The firstof these concerns the connection between the general will and
majority-rule;the second involves the apparent contradictionin Rousseau's
statementsabout how the individual should vote. In the firstcase the dis-
tinctionbetween the and a general will proves helpful; in the second it is
essential.
Rousseau's brief discussion of majority-rulebegins with the acknowl-
edgement that the social contractrequires unanimous consent: "Since every
man is born free and master of himself,no one . . . can subject him without
his consent." (IV,2:110.) But he soon goes on to say that, "Except for this
primitivecontract,the vote of the majorityalways obligates all the others."
(IV,2:110.) This poses an obvious problem - "How can the opponents be
free yet subject to laws to which theyhave not consented?" - and Rousseau
offersthis resolution:
Whena law is proposedin theassemblyof the people,whattheyare being
asked is not preciselywhethertheyapprove or rejectthe proposal,but
whetherit does or does notconformto thegeneralwillthatis theirs.Each
one expresseshis opinion on thisby voting,and the declarationof the
generalwillis drawnfromthecountingof thevotes.
In these two sentences we have a good example of the unmarked dis-
tinctionbetween theand a general will. The proposal is to be approved, on
the one hand, if it conformsto thegeneral will (principle); but a general will
(specific policy which satisfies this principle) is discovered, on the other
hand, only by counting votes. Votes are irrelevantto the general will, for
principles are not discovered by voting,but a general will often cannot be
ascertained withoutthem.17Were it not for the differencebetween the gen-
eral willqua principleand policies whichconformto that principle,Rousseau
would not even need to discuss voting.
The point of immediate importance, however, is brought out in the
remainder of the paragraph.
Thereforewhentheopinioncontrary to mineprevails,thatprovesnothing
exceptthatI was mistaken, and whatI thoughtto be the generalwillwas
not. If myprivatewillhad prevailed,I would have done somethingother
thanwhatI wanted.It is thenthatI wouldnothave been free.(IV,2:110-
111.)
While this explanation is complicated and paradoxical, it is not simply
metaphysical nonsense. Since Rousseau admits that even the people as a
whole may fail to discern the policy which accords with the general will
(II,6:67),18 the question here is, why is the majorityopinion less likelyto be
mistakenthan that of the minority,or of a solitaryindividual? It is certainly
possible for one person to be rightwhile everyone else is wrong. But if we
make certain assumptions, as Barry has shown, then the opinion of the
majorityis more likelyto be correctthan the minority's.19 This occurs when
we assume, as Rousseau seems to do, that: (a) there is a uniquely right

171
say "often" because Rousseau allows that "the commands of leaders" can "pass for expres-
sions of the general will,as long as the sovereign,being free to oppose them, does not do
so." (11,1:59). See also PoliticalEconomy,p. 216. And see note 26, infra,for a comment on
the translationof this passage fromthe Social Contract.
18Cf.PoliticalEconomy,p. 216: "Must the whole nation be assembled at each unforeseen event?
Such an assembly is all the less necessary because it is not sure its decision would be the
expression of the general will . ."
9For a more detailed explanation of this point, see Barry,"The Public Interest,"p. 122, and
Levine, The PoliticsofAutonomy, pp. 63-72.
368 Western
PoliticalQuarterly

answer, a specificpolicyin conformitywiththe general will,to be found; (b)

everyone has an equal, better-than-evenchance of discerning the rightan-
swer; and (c) everyone wants the rightanswer to prevail. It may be difficult
to accept all, or even any, of these assumptions,but this does indicate how
one mightbe glad that his own point of view was not victorious.Rousseau's
point is not that whateverthe majoritywants is eo ipso the general will, but
that the majorityis simplymore likelythan the minorityto have discovered
the policy which conformsto the general will.
One may wonder again what this has to do withfreedom. How could a
person possiblybe less free if his opinion had carried the day? The answer
here, as in the case of the "forced to be free" passage, follows from Rous-
seau's conception of the just state as the realm of freedom. Rousseau as-
sumes that in thisstateall the voters want the policy which best satisfiesthe
general will to prevail. If the majorityis more likelyto perceive that policy,
then a person whose opinion is in the minoritymay be said to have voted
against his willqua citizen: his will was to promote the common interest,but
his opinion as to what policywould do so was mistaken."The constantwill of
all the members of the state is the general will, which makes them citizens
and free." (IV,2:110.) If one sees, therefore, that the policy he favored
actually(or ostensibly)is not in the common interest,he then recognizes that
his particularopinion was divergentfrom his constantwill. Since the realm
of freedom is preserved only through observance of the general will, free-
dom is diminished whenevera policy contraryto the general willis pursued.
Anyone who had supported such a policy - along with everyone else-
would then be less free than he would be if his opinion had not won.
The second problem with Rousseau's remarks on voting concerns the
aims of the individual voter. In the firstchapters of Book IV, Rousseau
clearlyrequires individualsto vote according to theirperceptionsof.what the
general will demands; private interestsare not to be considered. He also
provides a vivid descriptionof what happens when the people vote as men
ratherthan citizens."Finally,"he says,". . . when the social bond is broken in
all hearts; when the basest interestbrazenly adopts the sacred name of the
public good, then the general will becomes mute; all - guided by secret
motives- are no more citizens in offeringtheir opinions than if the State
had never existed, and iniquitous decrees whose only goal is the private
interestare falsely passed under the name of laws." (IV, 1: 108-109.) Yet
earlier in the Social ContractRousseau seems to take precisely the opposite
position when he asks,
Whyis the generalwill alwaysrightand whydo all constantly want the
happinessof each, if not because thereis no one who does not applythis
wordeach to himself,and does not thinkof himselfas he votesfor all?
Whichprovesthatthe equalityof right,and the conceptofjusticeit pro-
duces,are derivedfromeach man'spreference forhimselfand consequently
fromthenatureof man .... (11,4:62;emphasisin theoriginal.)

Certainlythere seems to be a contradictionbetween these passages. If

we attend to the distinctionbetween the general will and a general will,
however, we can reconcile Rousseau's, remarks. This is because Rousseau's
referencein Book II is to thegeneral will,while in Book IV he is concerned
with a general will. In the passage cited from Book II, in other words,
Rousseau's purpose is tojustifythegeneral will: the principlethatrequires all
public decisions to take only the common interestof the citizeninto account.
His argument is that this particular imperative supplies a just basis for a
politicalassociationbecause it is acceptable to everyone,foreveryonecan see
thathe is not subjectinghimselfto arbitraryrule when he agrees to the social
 theGeneralWill 369
Understanding

compact. All individuals are treated equally by the general will, and "each
man's preferencefor himself' leads to "equality of right"under the general
will. Indeed, the general will qua principle is at the heart of Rousseau's
attemptto reconcile"what rightpermitswithwhat interestprescribes,so that
justice and utilityare not at variance." (I, preface:46.)20
In Book IV, Rousseau's purpose is different.When he says that voters
must tryto decide whethera proposition is in conformitywith the general
will rather than their private wills, he is concerned with determiningwhat
specificpolicies or laws are in conformitywith the general will. Thus when
he declares that the voters are to state their opinions on the question, Is
policyx in conformitywiththe general will?,and not to followtheirparticu-
lar interestswhen they vote, he is specifyinghow a general will, or general
wills,in accordance withthegeneral will can be found. These remarksabout
the aims of the voter pertain to two differentaspects of the general will,in
sum, and theyare not contradictory.
What we find, then, is that drawing the distinctionbetween the and a
general will enables us to clarifythe firstand to solve the second of these
problems generated by Rousseau's remarkson voting.This provides power-
ful evidence for the validityof the distinction.But it also leads to a difficult
question: is this distinctionindeed implicitin the Social Contract,as I have
stated,or am I simplyreading a helpful distinctioninto Rousseau's work? I
doubt that thisquestion can be answered conclusively;yetthere are passages
in Rousseau's writingswhich indicate that he did have in mind something
verymuch like the distinctionI have drawn.
This is suggested in the discourse on Political Economy,for instance,
when Rousseau statesthat taxes ought to be levied "througha generalwill,by
majorityvote, and based on proportional rates that leave no room for an
arbitraryassessment of taxes."21Although Rousseau does not call attention
to his use of "a general will" here, it is clear that he is referringto specific
policies and decisions, not to the principle that public policy should ignore
the interestsof men. So also is the distinctionsuggested by a passage in the
GenevaManuscript,where Rousseau proclaims that,"Today's law should not
be an act of yesterday'sgeneral will, but of today's. ..."22 Insofar as the
general will is a principle,of course, it cannot varyfromday to day; but how
we can and should act on that principle may easily vary as circumstances
change. When we alter and amend our laws, then, we are not changing the
general will; we are doing what we can to keep our laws - our general wills
- in agreement withit.
Referencesto "general wills"in Emileand the Social Contractalso support
my position, for there is no reason for Rousseau to use the plural when
referringto the general will as a principle. In one passage fromEmile Rous-
seau states that,"Dependence on men . .. engenders all the vices, and by it,
masterand slave are mutuallycorrupted. If there is any means of remedying
this ill in society,it is to substitutelaw for man and to arm thegeneralwills
witha real strengthsuperior to the action of everyparticularwill."23The use
of "the general wills" here is clearly consistentwith my account of the dis-
tinctionbetween theand a general will; yet it may also be interpretedin a

20For a
helpful account of Rousseau's attemptto unite self-interestwith the common interest,
see Nannerl 0. Keohane, "'The Masterpiece of Policyin Our Century': Rousseau on the
Moralityof the Enlightenment,"PoliticalTheory6 (November 1978): esp. 475-81.
21 PoliticalEconomy,p. 230; emphasis added.
22Included in On the Social Contract,ed. Masters, p. 181; see also p. 168 and Social Contract
(II,1:59).
23Emile,trans. Allan Bloom (New York: Basic Books, 1979), p. 85; emphasis added.
370 Western
PoliticalQuarterly

quite differentway. Consider what Roger Mastershas to say about the use of
the plural here: "when speaking of all mankind, Rousseau thereby makes
clear the inevitable division of the human species into differentsocieties,
each of which has its own general will."24Masters' reading of this passage is
not implausible. But it neitherexplains nor fitsanother passage fromEmile
where Rousseau uses the plural: "according to the social pact the sovereignis
able to act only by the common and generalwillsand that thereforeits acts
ought similarlyto have only general and common objects."25Here it seems
clear that what Rousseau is talkingabout are specificdecisions and policies
which are in accordance with the general will. This is true also of a passage
in the Social Contractwhere Rousseau says, "This does not mean that the
commands of the rulers cannot pass for general wills, so long as the
Sovereign,being free to oppose them,offersno opposition."26In at leasttwo
of these three passages, then,we have furtherevidence for myclaim that the
distinctionbetween theand a general will is implicitin Rousseau's theory.
There is, in short,sufficientreason to believe that Rousseau did indeed
perceive, though perhaps only dimly,the importanceof thisdistinction.Nor
should this be surprising. For if the general will is best understood as a
principle requiring us to act in public mattersas citizens rather than men,
then the distinctionI have drawn seems to inhere in the logic of the general
will.

IV
We can understand the general will. But this is not to say that Rous-
seau's notion is free from difficultiesor that no furtherquestions need be
asked. Even if the general will is properlyunderstood, one may stillwant to
question some of Rousseau's assumptions and conclusions. This is not the
place to attempt a full assessment of Rousseau's political philosophy, cer-
tainly,but I should like to conclude by raising some questions for further
consideration.
First,there is the question of how the general will - the principle for
deciding mattersof public policy - is to be applied. What does it mean to
say, in other words, that public policy must attend to the common interestof
citizens and ignore the private interestsof men? One problem is that the
viewpointof the citizenis not alwaysreadilyapparent. Consider the case of a
city(or any other political unit) which is dominated by one large industry.
Cities of this sort often find themselvescaught on the horns of a dilemma.
There may be reliable evidence that pollution fromthe industryis hazardous
to the health of those in the community;but eliminatingthat hazard may
involve the loss of the industrywith all the money and jobs it directlyand
indirectlyprovides. Everyone in the communitywill sufferif the hazard is
not eliminated; yet it is also conceivable that everyone will suffer if the
industryleaves. In this case is there a single viewpointwhich is distinctlythe
citizen's?What would the general will have those in this predicament do?
By altering the example slightly,we can produce yet another problem
for Rousseau. Suppose that in thiscitythere are many who will suffersevere
24Masters,The PoliticalPhilosophy Rousseau,p. 42.
of
25Emile, p. 461; emphasis added.
26Here I follow Cole's translationof the Social Contract,p. 24, with emphasis added. In the
Masters' edition the passage is translated,"This is not to say that the commands of leaders
cannot pass for expressionsof thegeneralwill,as long as the sovereign,being free to oppose
them, does not do so." (11,1:59; emphasis added.) Cf. Du ContratSocial, ed. Grimsley,p.
125: "Ce n'est point a dire que les ordres des chefs ne puissent passer pour des volontes
generales, tant que le souverain, libre de s'y opposer, ne le faitpas."
 theGeneralWill 371
Understanding

economic hardship if the industryleaves and a few who, for one reason or
another,will not sufferat all. In thiscase all the membersof the community
would share an interestin eliminatingthe hazard, but all would not share an
interest in keeping the industry in town. The general will, presumably,
would require the elimination of the hazard because that would be in the
interestof the citizens,even though it mightspell disasterfor most of the men
in the community.Given possibilitiesof this kind, is it clear that we should
always ignore the private interestsof men when deciding mattersof public
policy?
A second set of difficultiesarises in connectionwithRousseau's remarks
on the majoritariandecision-rule. If we grant the three assumptions men-
tioned earlier - that (a) political questions have uniquely rightanswers; (b)
everyone has an equal, better-than-evenchance of discovering these an-
swers; and (c) everyone wants the right answers to prevail - then the
majorityis more likelythan the minorityto hit upon the rightanswer. But
these are strong assumptions,and it is not at all obvious that we ought to
accept them. If these three conditionsare not met, moreover,then one who
voted withthe minoritymay actuallybe lessfree when and if he acquiesces to
the will of the majority.
Now it is true that Rousseau is talkingabout the de jure state when he
argues that those in the minorityare mistakenabout what the general will
requires, and perhaps in the de jure state everyone will want the right an-
swers to prevail. Even if we grant this and the other assumptions,however,
the problem of civil disobedience is stillwith us. Should one go along with
the majorityon a matterof conscience simplybecause he knows thattheyare
morelikelyto be right than he is? Are matters changed if we add, with
Rousseau, that "the more importantand serious the deliberations,the closer
the winning opinion should be to unanimity"?(IV,2:111.) The size of the
vote against him will give the conscientiousindividual pause, no doubt; but
so long as there is a possibilitythat the winningside has failed to discover a
general will whichconformsto the general will,thereis a case to be made for
civildisobedience. Rousseau failsto make thiscase. Should he have done so?
To answer these questions we shall have to go beyond Rousseau. We
may find it necessaryto refinehis analysisand concepts,or even to abandon
some of them. But the simple factthat it leads to such potentiallyrewarding
exploration may well be the most fittingtestimonyto the richnessof Rous-
seau's political philosophy - and to the importance of understanding the
general will.