Sunteți pe pagina 1din 25
Habermas and Foucault: Thinkers for Civil Society? Author(s): Bent Flyvbjerg Source: The British Journal of
Habermas and Foucault: Thinkers for Civil Society? Author(s): Bent Flyvbjerg Source: The British Journal of

Habermas and Foucault: Thinkers for Civil Society? Author(s): Bent Flyvbjerg Source: The British Journal of Sociology, Vol. 49, No. 2 (Jun., 1998), pp. 210-233 Published by: Blackwell Publishing on behalf of The London School of Economics and Political Science

Accessed: 18/04/2011 02:03

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact

information about JSTOR, please contact Blackwell Publishing and The London School of Economics

Blackwell Publishing and The London School of Economics and Political Science are collaborating with JSTOR to digitize, preserve and extend access to The British Journal of Sociology.

Bent Flyvbjerg

Habermasand Foucault:thinkersfor civil



Takentogether,theworksofJurgenHabermasandMichelFoucaulthighlightan essentialtensionin modernity.Thisis the tensionbetweenthe normativeandthe real,betweenwhatshouldbe done andwhatis actuallydone. Understandingthis tension is crucialto understandingmodern democracy,what it is and what it could be. It hasbeen arguedthatan effectivewayof makingdemocracystronger is to strengthencivilsociety.This articlecontainsa comparativeanalysisof the centralideasof HabermasandFoucaultastheypertainto the questionof democ- racyand civilsociety.Morespecifically,the discourseethicsof Habermasis con- trastedwiththe poweranalyticsandethicsof Foucaultevaluatingtheirusefulness for those interestedin understanding,and bringing about, democraticsocial change.




Havel(1993:3)hasobservedthata strongcivilsocietyis a crucialcondition of strongdemocracy.Empoweringcivilsocietyis a centralconcernfor the projectof democracy,just as the questionof how best to thinkaboutsuch

empowermentis importantto socialand politicaltheory.Butwhatis 'civil society'?A searchfor clear definitionsin the relevantliteratureis in vain. Not because the concept lacksdefinitions;ratherthe definitionsare too

multiple and varied to bring clarity.Most writerson

however,thatcivilsocietyhasan institutionalcore constitutedbyvoluntary associationsoutsidethe sphere of the stateand the economy.Such associ- ationsrangefrom,for example,churches,culturalassociations,sportclubs and debatingsocieties to independent media, academies,groupsof con- cernedcitizens,grass-rootsinitiativesandorganizationsof gender,raceand sexuality,all the way to occupationalassociations,political parties and labourunions (Habermas1992a:453).

civil society agree,

Brit.Jrzl.of SociolonVolume no. 49 Issue no. 2

June 1998

ISSN 0007-1315

C)London School of Economics 1998

Habermasand Foucault


The fundamentalact of citizenshipin a pluralistdemocracyis that of forming an associationof this kind. Keane (1988a:14)ascribesto these associationsthe taskof maintainingandredefiningtheboundariesbetween civilsocietyand state throughtwo interdependentand simultaneouspro- cesses:the expansionof social equalityand liberty,and the restructuring anddemocratizingof stateinstitutions.Thisexplainsthe importanceof civil societyto democracy.Thatimportanceis supportednot onlybysocialand politicaltheorybut byhistorical-empiricalevidenceaswell (Putnam1993). The worksofJurgen Habermasand MichelFoucaulthighlightan essen- tial tension in modernity.This is the tension betweenconsensusand con- flict. With a point of departurein Kant,Habermasis the philosopherof Moralitatbasedon consensus.Foucault,followingNietzsche,is the philoso- pher of wirklicheHistorie(realhistory)told in termsof conflict and power. This articlepresentsa comparativeanalysisof the centralideas of Haber- mas and Foucaultas they pertainto the issue of empoweringcivilsociety anddemocracy.Wewillaskwhethersuchempowermentisbestunderstood, and acted, in terms of consensus,or whether conflict is a more suitable frameof reference.Keane(1988b:21)explicitlywarnsus thatinequalityand

'civilsociety'from the start.

dominationhasbeen builtinto the concept of

Historically,'civilsocietyisestablishedafterthe imageof the civilised[Euro- pean] male individual,'Keanesays,'it restson a foundationof excluded women, who are expected to live under conditionsof household despot-

ism'.Today,the problemof exclusionis raisednot only bygender groups but also by groups defining themselveson the basisof, for instance,eth- nicity and sexuality.Clearly,if presentlywe are to build anything- and somethingas importantas democracy- on the concept of civilsociety,we need to deal with the problemsof exclusion,difference,diversityand the politicsof identity.Therefore,asa sub-themeto thisarticlewewillask,what do Habermasand Foucaulthaveto contributeto this task?


'WithKant,the modern age is inaugurated,'saysHabermas(1987:260), who cites the importanceof Kant'sattemptto developa universalrational foundationfor democraticinstitutions.2Habermasagreeswith Kantas to the need to developsuch a foundationfor democracyand its institutions, but he pointsout thatKantfailed to achievehis goal.Accordingto Haber- mas (1987:18-21, 302), thiswasbecauseKant'sthinkingwasbasedupon a subjectwenteredrationality.Moreover,Habermaspoints out thatthe later philosophers,from Hegel and Marxto contemporarythinkers,have also been unableto developthe much sought-afterrationaland universalfoun- dationfor suchsocialinstitutions.Accordingto Habermas(1987:294), this is becausetheyhaveall workedwithina traditionhe calls,'the philosophy of the subject'.



Mostcontemporaryphilosophersand socialscientistshaveacceptedthe consequencesof more than twomillenniaof failedattemptsto establisha universalconstitutionof philosophy,socialscienceandsocialorganization, having concluded that such a foundation does not seem feasible. Not Habermas,however,who thinksthathis ownworkcan providethis consti- tution, and that the consequences of abandoning it are unacceptable. Withouta universallyconstitutedphilosophy,science and democracy,says Habermas,the resultwould be contextualism,relativismand nihilism;all of whichHabermassees as dangerous. According to Habermas,the problem with Kantand with subsequent thinkerson modernityis not that theywere mistakenin their goal of con- stitutingsociety rationally,but that they had the wrong ideas of how to achievethe goal.ForHabermas,the pathtowarda rationalconstitutionand the establishmentof a bulwarkagainstrelativismis a reorientationfrom earlierphilosophers'focuson subjectivity,withinwhichHabermasclassifies both Hegel's 'worldspirit'and Marx's'workingclass,'to a focus on inter- subjectivity.And Habermas'sown work, particularlyhis 'theory of com- municativeaction' and 'discourseethics', is located in the intersubjective approachto the problematicof modernity(Habermas1983, 1987, 1990,


The goal of Habermas'stheoryof communicativeactionis thatof 'clari- fyingthe presuppositionsof the rationalityof processesof reachingunder- standing, which may be presumed to be universal because they are unavoidable'(Haberrnas1985:196).In his PhilosophicalDiscourseof Mod- ernity,Habermasdevelopshis intersubjectiveapproachto modernityusing the concept of 'communicativerationality'.

The communicativerationalityrecallsolder ideas of logos, inasmuchas it bringsalongwithit the connotationsof a noncoercivelyunifying,con- sensus-buildingforce of a discoursein which the participantsovercome their at firstsubjectivelybased viewsin favorof a rationallymotivated agreement (Habermas1987:294, 315).

AlthoughHabermassees communicativerationalityasbeing threatenedby actualmodern society,he neverthelessarguesthat the core of communi- cativerationality,'theunconstrained,unifying,consensus-bringingforceof argumentativespeech',isa 'centralexperience'in the life of a humanbeing (Habermas1983: 10). According to Habermas(1983: 316), this central experience is inherent in human social life: 'Communicativereason is directlyimplicatedin sociallife processesinsofaras actsof mutualunder- standingtakeon the role of a mechanismfor coordinatingaction'.Haber- masleavesno doubt thatby 'inherent'he means universallyinherent.The universalityderivesfrom the fact that for Habermashuman social life is basedupon processesforestablishingreciprocalunderstanding.Thesepro- cessesare assumedto be 'universalbecausetheyare unavoidable'(Haber- mas 1985:196). In an earlierformulation,Habermas(1979:97) statesthis vieweven more clearly

Habermasand Foucault

In actionoriented to reachingunderstanding,validityclaimsare 'always already'implicitlyraised.Theseuniversalclaims. s . areset in the general structuresof possiblecommunication.In thesevalidityclaimscommuni- cation theorycan locate a gentle, but obstinate,a neversilent although seldom redeemed claim to reason,a claim that must be recognisedde factowheneverand whereverthere is to be consensualaction.

The consequence, for Habermas,is that human beings are defined as democraticbeings,as homodemocraticus. As for the validityclaims,Habermas(1990:93) explainsthatvalidityis

defined as consensuswithoutforce: 'a contested norm cannot meet with

the consentof the participantsin a

canfreely[zwanglos]accept the consequencesand the side effectsthat the generalobservanceof a controversialnormcan be expected to havefor the satisfactionof the interestsof eachindividual'(italicsin original).Thisprin-

ciple of validity,Habermas(1990:120-1) calls '(U)', the 'universalisation principle'of discourseethics. Similarly,in a keypassageon truth,Haber- mas (1990:198) states:'Argumentationinsuresthatall concernedin prin- ciple takepart,freelyand equally,in a cooperativesearchfor truth,where nothingcoercesanyoneexcept the force of the betterargument'.The only 'force'which is activein the ideal speech situationand in communicative rationalityis thus this 'force of the better argument',which consequently obtainsa criticalplace in Habermas'swork. Validityandtruthareensuredwherethe participantsin a givendiscourse respectfive key processualrequirementsof discourseethics: (1) no party affectedbywhatis being discussedshouldbe excludedfrom the discourse (the requirementof generality);(2) all participantsshould have equal possibilityto presentand criticizevalidityclaimsin the processof discourse (autonomy);(3) participantsmust be willingand able to empathizewith each other'svalidityclaims (ideal role taking); (4) existing power differ- ences betweenparticipantsmustbe neutralizedsuch thatthese differences haveno effecton the creationof consensus(powerneutrality);and (5) par- ticipantsmust openly explain their goals and intentions and in this con- nection desist from strategicaction (transparence)(Habermas1993:31, 1990:65-6, Kettner1993). Finally,given the implicationsof the firstfive requirements,we could add a sixth:unlimitedtime. In a societyfollowingthis model, citizenshipwouldbe defined in terms of takingpartin publicdebate.Participationis discursiveparticipation.And participationis detachedparticipation,in as much as communicativeration- alityrequiresideal role taking,power neutrality,etc. Habermas'smodel, i.e., discourseethics,should not be confusedwithcontingenttypesof bar- gainingorwithmodelsof strategicallynegotiatedcompromisesamongcon- flicting particularinterests. What is missing in strategic pursuits and rational-choicemodels is the recourseto ultimatenormativejustification thatHabermasclaimsto giveus (Dallmayr1990:5). Empirically,Habermas



Bent F7yvbjerg


social movements as agents of communicative rationality and the

the public sphere.


communicative ration-

clear that

of discourse ethics and

above, make it

procedural requirements mentioned

procedural as



does not

set up

substantive rationality: 'Dis-

Habermas is a uni-

the rules for correct


substantive orientations. Instead it establishes

to guarantee the impar-

on presuppositions and designed

(Habermas 1990:

sees new




ality, andthe

wetalkingare about

course ethics



tiality ofthe

process ofjudging'

moralist as concerns process:

advance, in the

form of the requirements

Habermas is

given communicative



establishing consensus and

at the centre of

normatively given in







the study







claims on

Habermas's work.

linked tojudicial





Conversely, as regards content,

andtrue in a


participants in

of processes for

situationalist: whatis right

solely by

are built stands

which the processes

process is directly

conceive of the democratic

Habermas's view of

the democratic



institutionalization: 'Iwish to


institutionalization of




forms of communication

forrational political willformation' Habermas (undated: 15)says.

in this process,




(undated:8) states





'authorisation ofpower bylaw and

thesanctioning of law

Habermas thus

(emphasis in original).

perspective of law

occur uno acto'

and sovereignty.

contrasts with Foucault 'by no means adequate'.

that it 'can

representation of



he operates within a

this is a

finds this conception

90) says about

if it frees itself



Aswewill see below,

(1980a: 87-8)

Foucault (1980a: 82,

beconstituted only that I would


power-law, of

(1980a: 89)


clear that


perspective which

of power

'analytics of power'


. a certain



off the



his own


completely from



that Foucault of the king' in

understanding of power. For

power-sovereignty'. It is


argument to 'cut

made his

replace it bya decentred

is still very

analysis and

much on, in

sense that sov-


Habermas the

head of the king

the regulation of power



ereignty is

a prerequisite for

uncritical about mod-

such as

'methods of

forinstance forstrengthening civil society, arethe writing of con-

substantially more optimistic

Max Weber and members of




Habermas is

ernity than


the Frankfurt School,

Habermas's main



Max Horkheimer


stitutions and

elements in

size the

constitutions as

become central

institutional development,

Habermas's project. Itishard to over-empha-


(1994: 514)

citizens in a


simply sees

pluralist society

and endpoints for

importance of


this point.

main device

for uniting

bysocial, cultural, and philo-

abstract principles

citizens of a society shaped

What unites


of an


first of all the

[weltanschaulich] pluralism are

the medium of law.

artificial republican order,

created through

Habermasand Foucault


If Habermasis rightaboutthe importanceof constitution-writingand insti-

tutionalreforms,the prospectslook good indeed for changinggovernment

in a moredemocraticdirectionbymeansof discourseethicsand the theory

of communicativerationality.The problem, however,as pointed out by Putnam(1993:17-8), is that'[t]wocenturiesof constitution-writingaround

the worldwarnus

Thatinstitutionalreformsalterbehavioris an hypothesis,not an axiom.' The problem with Habermasis that he has the axiom and the hypothesisreversed:he takesfor grantedthatwhichshouldbe subjectedto empiricaland historicaltest. The basic weakness of Habermas'sproject is its lack of agreement between ideal and reality,between intentions and their implementation. This incongruitypervadesboth the most general as well as the most con- crete phenomena of modernityand it is rooted in an insufficientconcep- tion of power.Habermashimself observesthat discoursecannot by itself insurethatthe conditionsfor discourseethicsand democracyaremet. But discourseabout discourseethics is all Habermashas to offer. This is the fundamentalpoliticaldilemmain Habermas'sthinking:he describesto us the utopia of communicativerationalitybut not how to get there. Haber- mas (1990: 209) himself mentions lack of 'crucialinstitutions',lack of 'crucialsocialization'and 'poverty,abuse,and degradation'as barriersto discursivedecision making.But he has little to sayabout the relationsof powerthat createthese barriersand how powermaybe changedin order to begin the kindsof institutionaland educationalchange, improvements in welfare,and enforcementof basichuman rightsthat could help lower the barriers.In short,Habermaslacksthe kind of concreteunderstanding of relationsof powerthatis needed for politicalchange. With his characteristicallycomprehensiveapproach,Habermas(1987:

322) lets us know that his theory of communicativeaction opens him to criticismasan idealist:'Itis not so simpleto counterthe suspicionthatwith the conceptof actionorientedtovalidityclaims,the idealismof a pure,non- situatedreasonslipsin again'.I will arguehere that not only is it difficult to counterthissuspicion,it is impossible.And thisimpossibilityconstitutes

a fundamentalproblemin Habermas'swork. 'There is a point in everyphilosophy,'writesNietzsche (1966:15[§8]), 'whenthe philosopher's"conviction"appearson the stage'.3ForHabermas thatpointis the foundationof hisidealspeechsituationanduniversalvalid- ity claimsupon a Kirkegaardian'leap of faith'.4Habermas,as mentioned, statesthatconsensus-seekingandfreedomfromdominationareuniversally inherent as forces in human conversation,and he emphasizesthese par- ticular aspects. Other important philosophers and social thinkers have tended to emphasize the exact opposite. Machiavelli(1984: 96), whom Crick(1983:1S, 17) and othershave called 'a mostworthyhumanist'and 'distinctlymodern',andwhom,like Habermas,is concernedwith'the busi- ness of good government,'states:'One can makethisgeneralisationabout men:theyareungrateful,fickle,liars,and deceivers'.Lessradically,butstill

thatdesignersof newinstitutionsareoftenwritingon



Habermas, are statements by Nietzsche,

thatcommunication is at all times

always present',


thinkers, to

in which power

is absent.



incontrast to


power: 'power is

meaningless, according to these

already penetrated by

Itis therefore

concept of com-

(1988: 11,18).

operate witha


Forstudents of power,

non-rational rhetoric

domination and

themode of communication - e.g.,

ation, charisma, using dependency


thisperspective Habermas

contrasts 'successful' with

when he

sation, because

Whether the

important here. Whatis decisive, rather,

communication is more typically characterized by

freedom from

established via

and maintenance of interests than by

rhetoric, 'validity' is


consensus-seeking. In

control, rationaliz-


relations between participants - rather


seems overlynaive andidealistic

rational arguments concerning the matter athand.

(1987: 297-8)

'distorted' associated utterance precisely withdistortion. in human conver-

is 'correct' is not

non-idealistic point of

is that a

that both positions are possible,

In an empirical-scientific context, some-

define himself, the


must therefore

success in rhetoric is

communicative or the rhetorical position

must take account of the

simultaneously possible.




Habermas otherwise takes great painsto


thing to

question of communicative

remain open. The

rationality versus rhetoric

mustbe settled byconcrete

examination of the

researcher mustaskhowcommunication takes place, and

democracy operate. Is

communication characterized by

caseat hand. The



Or is communication the exer-

consensus-seeking and rhetoric,

power, eventually come together

one can meaningfully

consensus-seeking and

ciseof power

freedom from

inindividual actsof communication?5

absence of power?


rhetoric? How do

domination andexercise of


The basic question being

here is whether

in communication and

distinguish rationality and

whether rationality can be viewed in isolation

mas.To assume an

suming that one can ultimately

humans are

and build a theory

makes for problematic philosophy


rationality to

onereason wehaveto

power from each other

from power, as does Haber-

przorzisjust as invalid as pre-

answer to this question


answer the biblical question of whether

either position

as Habermas does,

and speculative social science. This is

basically good

or basically evil.And to assume

upon it,

cautious whenusingthe

to universalize it


theory of communicative


leap of faith is hardly sus-

precisely in

Richard Rorty does not

understand and act in relation to

Constituting rationality and democracy on a

seems to

to be subject to

68) to

but it

his ownaxiom thatphilosophical


tainable. Habermas here

questions ought

thissense that Habermas mustbe seen as utopian. same issues which impel

use these exact words,


status in

which will do the workonce

empirical verification. Andit is

is nevertheless the

criticize communicative rationality for having religious

andunitring power

Habermas's thinking, andforbeing 'a healing

done by God'. As

Rortysays,'Weno longer

need [that]'.

Habermasand Foucault


There may be a substantial element of truth in the benefits of consti- tution-writing a la Habermas. And Habermas's home country, Germany, clearly needed new constitutional principles after World War II, a fact that seems to have been formative for Habermas's thinking. But Habermas (1994: 513-4) relies on something as weak as Verfassungspatriotismus(consti- tutional patriotism) as the main means to have constitutional principles take root and gain practical importance in a society

[C]onstitutional principles can only take root in the hearts of citizens once they have had good experiences with democratic institutions and have accustomed themselves to conditions of political freedom. In so doing, they also learn, within the prevailing national context, to com- prehend the republic and its Constitution as an attainment. Without a historical, consciously formed vision of this kind, patriotic ties deriving from and relating to the Constitution cannot come about. For such ties are connected, for example, with pride in a successful civil rights move- ment.

Studies of struggles over the actual writing, implementation and modifi- cation of real constitutions in real societies prove this account - with its emphasis on conflict-free phenomena like 'good experiences', 'vision' and 'pride' - to be far from sufficient (Putnam 1993). Something infinitely more complex than these phenomena are at work in real life situations, perhaps because humans are infinitely more complex than Habermas's homodemocraticus.People know how to be, at the same time, tribal and democratic, dissidents and patriots, experts atjudging how far a democratic constitution can be bent and used in non-democratic waysfor personal and group advantage (Flyvbjerg1998). Machiavelli is a more enlightened guide to social and political change

than Habermas

Machiavelli (1983: 111-2[§I.3]) recapitulates that '[a]ll writers on politics

have pointed out

wealth it must be taken for granted that all men are wicked and that they will alwaysgive vent to the malignity that is in their minds when opportunity offers'.7 If Machiavelli and other writers are right in this 'worst-case'think- ing, then we might clearly end up in trouble if we use Habermas's discourse ethics as a basis for organizing our society, as Habermas advocates we do, since discourse ethics contains no checks and balances - other than an abstract appeal to reason - to control the wickedness which Machiavelli talks about. Such wickedness is assumed awayby Habermas's leap of faith for the good. History teaches us, however, that assuming evil awaymay give free reign to evil. Thus, the lesson to be learnt from Machiavelli is not so much that moralism is hypocrisy. The lesson is that the first step to becom- ing moral is realizing we are not. Furthermore, by determining validity, truth,justice, etc., as an outcome of 'the better argument', Habermas simply moves the problems of determi- nation from the former concepts to the latter. As Bernstein (1992: 220)

when it comes to constitution-writing. In The Discourses

that in constituting and legislating for a common-



correctly points out, 'the better argument', and with it communicative

there is something

empty concept: 'Abstractly,

to the


enormously attractive

argument" until we

problem here


different arguments

mean that

Yetas Bernstein (1992: 221) states, 'Any

fordealing with conflicts that cannot to rational argumentation.'

when all


ofconflicts which

Agnes Heller, Albrecht Wellmer, Herman

have expressed similarly Habermas's


Heller (1984-5: 7) simply rejects

we look to moral

bluntly, if

now, we cannot

of the categorical

is an empirically

"force of the better The

about Habermas's appeal

ask ourselves what this

means and presupposes'.

are few clear criteria for

how good it is, and how

have some procedures - even

real civil

precisely these kinds


that in non-trivial situations there

what is considered an argument,

are to be evaluated

against each other. This does not

and evaluate them.

should not attempt to identify arguments

society must


be resolved by argumentation


parties are committed

as opposed

to Habermas's ideal types - it is

are of interest, both empirically


criticisms of

and normatively.

Lubbe and Niklas Luhmann

discourse ethics. In comment-

(U) mentioned


universalization principle

the value of

philosophy for guidance


Habermas's approach: 'Put

in our actions here and from the Habermasian

(1986: 63) is equally

universalization principle in

obtain any positive


imperative'. Wellmer

reformulation harsh when he writes that

moral judgment [einem Ding der

(1990) and Luhmann


point of a breakdown

adhering to the

make justified moral

an impossibility

institutional analysis, Lubbe concrete institutions

institutional life to the


Unmoglichkeit]'. At the level of

comments that upholding any

discourse ethics would paralyse

(Benhabib 1990).

most sympathetic interpreters,

to criticize

demands of

such as Seyla Benhabib

Habermas for his formal-

to provide a cor-

weak points and to

(Ferrara 1989).8 I also need to bring in


Even Habermas's

and Alessandro Ferrara, have begun


rective to

introduce an element

would argue that critical theory

theelement of power. Habermas (1996a, b;

atthe same time, developed

and Rene 1996). remains strongly



face of

regard the

well as the

massive non-communicative

and insensitivity to

context. They are trying

on precisely these

Habermas's thinking


phronesis into critical theory

and Habermas's

Norms and other recent work,

with power, and he has,

of civil society (Carleheden

however, Habermas's approach

scant attention to the

ethical values and to the

a foothold in society in the

In his Between Facts and

1995) has attempted to deal

a deeper analysis

these efforts,


normative and procedural, paying

of actual discourse, to substantive

relating to identity


Habermas also continues to dis-

and cultural divisions as

reason that are being devel-



how communicative rationality forces.



non-discursive waysof safeguarding

and new social movements.



oped by so-called minority groups of



Habermas's universalization also be unnecessary.



unsustainable, may






For instance,


the groups in



Habermasand Foucault


property-owning men to include all adult men did not necessarily have any ultimate democratic vision that voting rights should also include women. Nevertheless, their efforts unwittingly laid the groundwork for the subse- quent enfranchisement of women. Similarly, those civil rights groups who worked for the right to vote for adult women did not necessarily envisage a situation where suffrage would also include 18-year-olds,even though this

later came to pass in many countries. The struggle was carried out from case to case and utilized the arguments and means which worked in the specific

to today's new

socio-historical context. This mode of action is also pertinent

social movements, where we still do not know what will be meant by democ- racy in the future; we know only that, as democrats, we would like to have more of it. Rorty (1991: 190) is correct in noting that the 'cash value' of Habermas's notions of discourse ethics and communicative rationality consists of the familiar political freedoms of modern liberal democracies, freedoms that are essential to the functioning of civil society. But such notions are not 'foundations' or 'defences' of free institutions; they are those institutions, saysRorty:'We did not learn about the importance of these institutions by thinking through the nature of Reason or Man or Society; we learned about this the hard way,bywatching what happened when those institutions were set aside'. The vocabulary of Enlightenment rationalism, although it was essential to the beginning of liberal democracy, has become an impediment to the

preservation and progress of democratic societies (Rorty 1989: 44). One reason for this is that Enlightenment rationalism has little to offer in under- standing power and in understanding the related discrepancy between formal rationality and RBalrationalitat (real rationality) in modern democ- racies. In staying close to the Enlightenment vocabulary Habermas has developed little understanding of power and thus tends to become part of the problem he wishes to solve. Habermas's efforts to achieve more ration-

democracy, however laudable, draw attention away from critical

relations of power. The neglect of power is unfortunate, because it is pre-

ality and

cisely by paying attention to power relations that we may achieve more democracy. If our goal is to move toward Habermas's ideal - freedom from domination, more democracy, a strong civil society- then our first task is not to understand the utopia of communicative rationality, but to under- stand the realities of power. Here we turn to the work of Michel Foucault, who has tried to develop such an understanding.


Both Foucault and Habermas are political thinkers. Habermas's thinking is well developed as concerns political ideals, but weak in its understanding of actual political processes. Foucault's thinking, conversely, is weak with reference to generalized ideals - Foucault is a declared opponent of ideals,



'What ought I to do?' reflects a sophisticated

Habermas agree that in

answers to Kant's question,

be done?' - but his work


understood as definitive

orLenin's 'What is to

understanding of

politics one

Realpolitik. Both Foucault

Referring to Habermas and similar

rationalism as analysis of the

In the following com-

must 'side with reason'.

however, Foucault (1980b)

Foucault and Habermas,

democracy as a promising

warns that 'to respect


anideal should never constitute

rationalities really at work'

parison of

Descombes (1987) has called the


himself as a citizen in a democratic


Foucault was

School, just


significance for someone who rarely

phers. In an

agreement' with

dons the

into irrationality'. And, like Habermas,

evaluation of the significance

suggests, however,

interpreted by

that of

Foucault (1984b: 45),

beturned back


critique that takes the obvious consequence,

criticism is no longer

tures with universal

a blackmail to prevent the

'American Foucault',

social experiment

society working on

(Rajchman 1988: 170).


will be placed on what

the Foucault who and who regarded

the project of human

familiar with the work of Habermas and the Frankfurt



work of Foucault. Foucault

Habermas which is a fact of some


'completely in

built upon contemporary

(1984a: 248) said he was

of Kant. 'If one aban-

runs the risk of lapsing

Foucault was unequivocal in his

as Habermas is familiar with

even built upon the work of


Habermas regarding the importance


work of Kant', explained Foucault,

that the

of rationality as an object

work of Kant might


of study. Foucault been too narrowly

the Kantian question was

renounce transgressing', says

Habermas and his followers. '[I]f


knowing what limits knowledge has

'it seems to me that


a positive one

the critical question today has to

in brief, is to transform

limitation into a practical This entails an

namely 'that

The point,

conducted in the form of necessary

form of a possible transgression'.

according to Foucault (1984b:


in the search for formal struc-

going to be practiced historical investigation'.

value, but rather

as a




about Foucault is what Habermas sees as

harshly dismisses

Habermas's main complaint Thus

Foucault's relativism. Foucault's genealogical


ifby relativistic we mean and this is what Habermas (1987:

universally grounded;

'relativistic, cryptonormative illu-

for relativism is correct,


(emphasis in original).

Such critique

that can be rationally and

294) means when

the normative foun-

unfounded in norms

Foucault for not giving an 'account

his thinking. By this standard,



he criticizes



however, Habermas's own work

so far, been able to

of his discourse ethics

is also relativistic. As we have seen,


is possible, he has only postulated

Habermas has not,

rational and universal grounding

1985: 196,

alone with this problem. Despite more than

such grounding (Habermas

1979: 97). And two thousand

been able so far to thinking must be

Habermas is not

philosophers, no one has that to avoid relativism our

years of attempts by rationalistic

live up

to Plato's injunction

rationally and universally grounded.

Habermasand Foucault

The reason may be that Plato was wrong. Perhaps the polarity relativism-

foundationalism isjust another artificial dualism that makes it easy to think

but hard to understand. Such dualisms simplit

with little reference to actual phenomena. Perhaps the horns of the dualism can be avoided by contextualism. This is the strategy of Foucault. As we will see, it is clearly wrong to criticize Foucault for being a relativist if we by

relativistic mean 'without norms' or 'anything goes'. 'I do not conclude', saysFoucault (1984c: 374), 'that one may sayjust anything within the order of theory'. Foucault resolves the question of relativism versus foundationalism by following Nietzsche (1974: 284-5) who says about what he calls 'historians of morality' that

things conceptually but


[t]heir usual mistaken premise is that they affirm some consensus of the


from this that these principles must be unconditionally binding also for you and me; or conversely, they see the truth that among different nations moral valuations are necessarily different and then infer from this that no morality is at all binding. Both procedures are equally childish. (emphasis in original)

concerning certain principles of morals, and then they infer

Employing this line of reasoning, Foucault rejects both relativismand foun-

dationalism and replaces them by situational ethics, i.e., by context. With explicit reference to Kant and Habermas, Foucault (1984b: 46) says that unlike these two thinkers he 'is not seeking to make possible a metaphysics that has finally become a science'.

and metaphysics does not leave

Distancing himself from foundationalism

Foucault normless, however. His norms are expressed in a desire to chal- lenge 'every abuse of power, whoever the author, whoever the victims' (Miller 1993: 316) and in this way 'to give new impetus, as far and wide as

possible, to the undefined

here is the Nietzschean dernocrat, for whom any form of government - liberal or totalitarian - must be subjected to analysis and critique based on a will not to be dominated, voicing concerns in public and withholding

consent about anything that appears to be unacceptable. Foucault's norms are based on historical and personal context, and they are shared with many people around the world. The norms cannot be given a universal ground- ing independent of those people and that context, according to Foucault. Nor would such grounding be desirable, since it would entail an ethical uni- formity with the kind of utopian-totalitarian implications that Foucault would warn against in any context, be it that of Marx, Rousseau or Haber- mas: 'The search for a form of morality acceptable by everyone in the sense that everyone would have to submit to it, seems catastrophic to me' (Foucault 1984f: 37 quoted in Dreyfus and Rabinow 1986: 119). In a Fou- cauldian interpretation, such a morality would endanger civil society, not empower it. Instead, Foucault focuses on the analysis of evils and shows restraint in matters of commitment to ideas and systems of thought about

work of freedom' (Foucault 1984b: 46) . Foucault



what is good for man, given the historical experience that few things have produced more suffering among humans than strong commitments to implementing utopian visions of the good. Foucault's view of the value of universals in philosophy and social science stands in diametrical opposition to that of Habermas. 'Nothing is funda- mental', saysFoucault (1984a: 247), 'That is what is interesting in the analy- sis of society'. Compare this with Foucault's (1984d: 87-8) remark that 'nothing in man - not even his body- is sufficiently stable to serve as the basis for self-recognition or for understanding other men'. Therefore, Foucault's analysis of 'the rationalities really at work' begins with the assumption that because no one has yet demonstrated the existence of uni- versals in philosophy and social science, we must operate as if the univer- sals do not exist. That is, we should not waste our time searching in vain for universals.Where universals are said to exist, or where people tacitlyassume they exist, universals must be questioned, according to Foucault. For Foucault, our history endows us with the possibility to become aware of those social arrangements which create problems, for instance a weak civil society, and those which create satisfaction, for instance empowering civil society. It follows that we have the possibility to either oppose or promote these arrangements. This is Foucault's point of departure for social and political change, not global moral norms.9 The basis for understanding and acting is the attitude among those who understand and act, and this attitude is not based on idiosyncratic moral or personal preferences, but on a context-dependent common world view and interests among a reference group, well aware that different groups typi-

cally have different world views and different interests, and that there exists no general principle - including the 'force of the better argument' - by which all differences can be resolved. For Foucault the socially and histori- cally conditioned context, and not fictive universals, constitutes the most effective bulwark against relativism and nihilism, and the best basis for action. Our sociality and history, according to Foucault, is the only foun- dation we have, the only solid ground under our feet. And this socio- historical foundation is fully adequate. According to Foucault, Habermas's (undated: 8) 'authorisation of power by law' is inadequate (emphasis deleted). '[The juridical system] is utterly incongruous with the new methods of power', says Foucault (1980a: 89), 'methods that are employed on all levels and in forms that go beyond the

state and its apparatus

further away from a reign of law.' The law, institutions - or policies and plans - provide no guarantee of freedom, equality or democracy. Not even entire institutional systems, according to Foucault, can ensure freedom, even though they are established with that purpose. Nor is freedom likely to be achieved by imposing abstract theoretical systems or 'correct' think- ing. On the contrary, history has demonstrated - saysFoucault- horrifying examples that it is precisely those social systemswhich have turned freedom into theoretical formulas and treated practice as social engineering, i.e., as

Our historical gradient carries us further and

Habermasand Foucault


an epistemicallyderived techne,that become most repressive.'[People] reproachme for not presentingan overalltheory',saysFoucault (1984c:

375-6), 'Iamattempting,to the contrary,apartfromanytotalisation- which wouldbe at once abstractand limiting-to openupproblemsthatare as con- creteand generalas possible' (emphasisin original).

Given this background theory-basedwriting of constitutions does not occupy a central place in Foucault'swork as it does for Habermas,and constitution-writingwould not be seen as an effectivewayof empowering civilsocietyin a Foucauldianinterpretation.Thisis not becausethe writing of constitutionsis without significance,but because Foucaultviewsit as more important- both for understandingand for practice- to focus on the concrete struggle over a constitutionin a specific society:how the constitutionis interpreted,how it is practicedin actualinstitutions,and especially,how interpretationsand practisesmay be changed. In other words,Foucault'sthinkingas concernslaws,constitutionsand democracy focusesmoreon howexistingconstitutionsand theirassociatedinstitutions canbe utilizedmoredemocratically,whereasHabermas'sprojectis to estab- lish more democraticconstitutionsand institutionsassuch,where'democ- racy'is defined by Habermas'sdiscourseethics. In this sense,whatFoucaultcalls'the politicaltask'is

to criticisethe workingof institutionswhich appearto be both neutral and independent;to criticisethem in such a manner that the political violencewhichhasalwaysexerciseditselfobscurelythroughthemwillbe unmasked,so thatone canfightthem. (ChomskyandFoucault1974:171)

Thisis what,in a Foucauldianinterpretation,wouldbe seen as an effective approachto institutionalchange, including change in the institutionsof civilsociety.Withdirectreferenceto Habermas,Foucault(1988:18) adds

The problemis not of tryingto dissolve[relationsof power]in the utopia

the rulesof law, .whichwouldallow

these gamesof powerto be playedwitha minimumof domination.

of a perfectlytransparentcommunication,but to the techniquesof management,andalsothe

Here Foucaultoverestimateshis differenceswithHabermas,for Habermas alsobelievesthatthe ideal speech situationcannotbe establishedas a con- ventionalrealityin actualcommunication.Boththinkerssee the regulation of actual relations of dominance as crucial, but whereas Habermas approachesregulationfrom a universalistictheory of discourse,Foucault seeksout a genealogicalunderstandingof actualpowerrelationsin specific contexts.Foucaultis thus oriented towardsphronesis,whereasHabermas's orientationis towardsepisteme.ForFoucaultpraxisandfreedomarederived not fromuniversalsor theories.Freedomisa practice,and its idealis not a utopianabsenceof power.Resistanceand struggle,in contrastto consen- sus,is for Foucaultthe mostsolidbasisfor the practiceof freedom.Itis pre- cisely on the issue of power and freedom that we find the most crucial



difference between Foucault and Habermas, a difference reflected in

Foucault's (1988: 18) labelling

(1987:253, 294) responds by

kindof 'mud-slinging' is unproductive


poweror if

for concrete social and political

remains to be discovered if everything is

of Habermas as 'utopian,' while Habermas

terming Foucault a 'cynic' and 'relativist'. This

since nothing

nothing is power, but instead ideal otopia.



with the

Whereas Habermas emphasizes

Foucault nor

macro politics, Foucault

important shared

Habermas venture to define the actual

action. This is defined by the participants. Thus, both

stressessubstantive micro politics,

featurethat neither

contentof political

Habermas and Foucault are 'bottom-up'


regardsprocedural rationality-

thinkers as concerns the content in a 'top-down' moralist fashion as sketched out the procedures to be

thinker as regards both process and

Habermas would want to tell individuals and

about their affairsas regards procedure for however, to say anything about the outcome

neither process nor outcome;

conflict and power relations as the

for the fight against domination. This

both internally, i.e., in the relationship

- groups of different gender

of civil society

business where the fight against domi-

but where Habermas thinks


followed- Foucault is a 'bottom-up'

content.In this interpretation, groupsin civil society how to go discourse. He would not want,

ofthis procedure. Foucault would prescribe

hewould only recommend a focus on

mosteffective point of departure

fight is central to civil society

between different groups within civil society

orethnicity, for instance - and externally,

to the spheres nation can be

It is because of his double 'bottom-up'

described as non-action oriented.

in the relationship

of government and

said to be constitutive of civil society.

thinking that Foucault has been

Foucault (1981) saysabout such criticism

It's true that certain people, such as those who work in the institutional

to find advice or instructions in my

'whatis to be done.' But my project is precisely to bring

know what to do,' so that the acts, gestures,

setting of the prison

books to tell them

it about that they 'no