Foucault & Habermas on Discourse & Democracy

Author(s): Nancy S. Love

Source: Polity, Vol. 22, No. 2 (Winter, 1989), pp. 269-293
Published by: Palgrave Macmillan Journals
Stable URL: http://www.jstor.org/stable/3234835
Accessed: 21/03/2009 03:00

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. 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
http://www.jstor.org/action/showPublisher?publisherCode=pal.

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 organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access to Polity.

http://www.jstor.org
Foucault & Habermason
Discourse & Democracy*
Nancy S. Love
The PennsylvaniaState University

Theproblem of modernityis the subject of a continuingdebate that

revolvesaroundthree issues:rationality,subjectivity,and democracy.
JurgenHabermasand MichelFoucaultare majorfigures in this
debate. The authorarguesthat theirpositions on democraticdiscourse
are one-sided.Still, theirdebatedefines the tasks of democratictheory
by raisingthe questionsa balancedtheoryof democraticdiscourse
must consider:Whendoes subjectivitybecomesubjection?Whendoes
communicationbecomeconfession? Whendoes democracybecome
domination?

Nancy S. Love is Associate Professor of Political Scienceat The

PennsylvaniaState University.She is authorof Marx, Nietzsche,and
Modernity(ColumbiaUniversityPress) and editor of Ideologyand
Revolution(ChathamHouse).

The problem of modernityis the subject of continuing debate. Two

themes emergeclearlyin the various formulationsof that problem:ra-
tionality and subjectivity.What foundations does reason retain in a
pluralizedworld? What autonomy remains for individualsin a mass
society? Democracyis a less frequentbut equally importanttheme. Is
participatorypolitics all but obsolete in the modern state? And, since
democracydependsupon rationalsubjects,is it implicatedin the prob-
lem of modernity?Ultimately, the question becomes what should be
saved from the modernproject.
Criticaltheorists and post-structuralists,the primaryparticipantsin
this debate, have taken opposing positions. Jiirgen Habermas and
MichelFoucaulthave been majorvoices for their respectivesides.1Un-

*Theauthorwouldlike to thankWilliamConnolly,FredDallmayr,Diane Rubenstein,

andStephenWhitefor helpfulcommentsand suggestions.A fellowshipfromthe American
Councilof LearnedSocietiessupportedresearchfor this article.
1. MartinJay has arguedthat the futureof westernMarxismlies in the confrontation
betweenHabermasand Foucault,in Marxismand Totality:TheAdventuresof a Concept
from Lukdcsto Habermas(Berkeley:Universityof CaliforniaPress, 1984),p. 509.

Winter 1989
Polity
Polity VolumeXXII, Number 22 Winter1989
270 Foucault& Habermason Discourse& Democracy

fortunately,Foucault'suntimelydeathcut theirconfrontationshort,but
not beforethey had identifiedtheircommonconcerns.Foucaultspokeof
"a strangecase of non-penetrationbetweentwo very similartypes of
thinking," similar in their basic question: "What is the history of
reason?"2Even as he labeledpost-structuralists"young conservatives,"
Habermasacknowledgedsimilaritiesbetween "the critique of instru-
mentalreasonand the analysisof formationsof discourseand power."3
Yet Foucault and Habermashad also begun to disagree.Habermas
portrayedFoucault as "anti-modern,"arguingthat he rejectedmuch
which should be saved from the still uncompletedmodern project.4
Foucaultdid say that "man would be erased, like a face drawnin the
sand at the edge of the sea," but he regardedthe choice "for" or
"against"modernityas "intellectualblackmail."5As theseremarkssug-
gest, Foucaultand Habermasfrequentlytalkedpast one another.In their
debate, they often missed opportunitiesfor dialogue.6
HereI attemptto constructthat misseddialogue.My focus is the prob-
lem of democracy,specificallythe relationshipbetweendemocracyand
discourse,though I also discuss rationalityand subjectivityas related
themes. I arguethat the positionsof Habermasand Foucaulton demo-
craticdiscourseare one-sided.As Nancy Fraserhas suggested,they are
blindedby their fears.7Foucault, fearinga disciplinarysociety, focuses
on the constrainingpowersof democraticdiscourse.Habermas,fearing
the end of individuality,emphasizesits enablingpowers. Neither con-
siders how rational communicationboth enables and constrains,and
neither adequatelybalances the two in his approachto discourseand
democracy.
Before beginning,however,I should addressa possibleobjection.To
some, my approach-debate and balance-may seem to favor Haber-
mas. I do assume that Habermasand Foucault are arguingwith one
another.By presentingtheirpositionson modernity,I placethemwithin
Habermas'ssphere of rational argumentation.This bias, if it is one,

2. Michel Foucault, "Structuralismand Post-structuralism."Interviewwith Gerard

Raulet, Telos55 (Spring1983):200.
Profile," New Left Review151(May/-
3. JiirgenHabermas,"A Philosophico-Political
June 1985): 81.
4. Ibid., p. 83.
5. Michel Foucault, The Order of Things: An Archaeology of the Human Sciences (New
York:VintageBooks, 1973),p. 387.
and criticaltheorists
6. Unfortunately,a previousconferencebetweenpoststructuralists
was no exception.See RainerRochlitz,"The MissedMeeting-A ConferenceReporton
Frenchand GermanPhilosophy," Telos66 (Winter1985-86):124-28.
7. Nancy Fraser, "Michel Foucault:A 'Young Conservative'?",Ethics 96 (October
1981): 165-84.
 Nancy S. Love 271

seemsunavoidable.As Habermasputs it, "if we are not free ... to reject

or to acceptthe validityclaimsbound up with the cognitivepotentialof
the human species, it is senseless to want to 'decide' for or against
reason, for or against the expansion of the potential for reasoned
action."8Rationalargumentis our tool, even for assessingreason. Late
in life, Foucaultexpressedsome sympathyfor this approach,saying, "in
the serious play of questions and answers, in the work of reciprocal
elucidationthe rightsof each personare in some sense immanentin the
discussion."9
With this referenceto rights, Foucault also grants another of my
assumptions.My searchfor balancepresumesthat thereis emancipatory
potentialin the modernproject.SinceFoucaultoften gesturestowardthe
achievementsof modernity,this is not necessarilya Habermasianbias. I
will argue, however, that it is Habermas who provides criteria for
distinguishingthose achievementsfrom liabilities. For this reason,
HabermascharacterizesFoucault'sconceptof rationalityas one-sidedly
instrumentaland his own, which adds communicativeforms, as more
complete.'?This impliesthat HabermasincorporatesFoucault'sinsights
within his own balanced philosophical framework. Others have fur-
theredthis interpretationby drawingupon Habermasto analyzegaps in
Foucault's philosophy." But I do not think that by providing such
criteria,HabermasencompassesFoucault or achievesa balance. Their
debate also allows Foucaultto raise serious reservationsabout Haber-
mas's debts to modernity. By using each philosophy to illustratethe
other's limitations,I intendto pose the questionsthat a more balanced

8. JiirgenHabermas,"Towarda Reconstructionof HistoricalMaterialism"in Com-

municationand the Evolutionof Society,ThomasMcCarthy,tr. (Boston:BeaconPress,
1979),p. 177.
9. Michel Foucault, "Polemics, Politics, and Problematizations:An Interviewwith
Michel Foucault," in The FoucaultReader, ed. Paul Rabinow(New York: Pantheon,
1984),p. 379. Citingthis passage,David Ingramspeaksof a "remarkableconvergence"
betweenHabermasand Foucaultin "Foucaultand the FrankfurtSchool:A Discourseon
Nietzsche,Power, and Knowledge,"PraxisInternational6 (October1986):311-27.
10. Jiirgen Habermas, Interviewby Angelo Bolaffi, Telos 39 (Spring 1979): 170;
"ModernityversusPostmodernity,"New GermanCritique22 (Winter1981):13.
11. See PeterDews, Logicsof Disintegration:PoststructuralistThoughtand the Claims
Of CriticalTheory(London:Verso, 1987),pp.220-42 and his earlier"Powerand Subjec-
tivity in Foucault," New Left Review 144 (1984): 73-95; Nancy Fraser, "Foucaulton
ModernPower: EmpiricalInsightsand NormativeConfusions,"Praxis International1
(October 1981):272-87 and "Michel Foucault:A 'Young Conservative'?",p. 180. In
"Foucault'sChallengeto CriticalTheory,"AmericanPoliticalScienceReview,80 (June
1986):419-32, StephenK. Whitearguesthat a Foucauldianexaminationof Habermas's
categoriesis also needed.
272 Foucault& Habermason Discourse& Democracy

view of democraticdiscourseand a more balancedview of modernity

must address.

I. Enlightenment
The debatebetweenHabermasand Foucaultbeginswith a discussionof
the Enlightenment,as that is when the rational subject of modern
democracies originates. Habermas frequently distinguishes critical
theorists from post-structuralistsby arguingthat the former radicalize
and the latter reject Enlightenment.'2Habermas includes Foucault
among the post-structuralistswhose "anti-modernism"abandons the
still uncompletedmodern project. Against the post-structuralistposi-
tion, Habermasmaintainsthat "modernlife-worldsare differentiated
and should remainso, in orderthat the reflexivityof traditions,the in-
dividuationof the social subject, and the universalisticfoundationsof
justice and moralitydo not go all to hell."'3 If these are the characteris-
tics Habermasassociateswith modernity,then his critiqueof Foucaultis
far too blunt. A more subtle analysis suggests that Foucault rejects
universalism,retains differentiationand reflexivity, and relies upon
individuality.
Nancy Fraserprovidesthe basis for such an analysisby arguingthat
Foucaultdoes not criticizeEnlightenmentper se but ratherone aspectof
it: humanism."4Even humanismhe criticizesless for its concepts and
values than for its universalisticinterpretationsof them. Accordingto
Foucault, humanismhas a distinctiveobject-Man-who emergedwith
a new power/knowledgeregimein the late eighteenthcentury.Man is
simultaneouslyconstitutedas the epistemologicalobjectand the political
subjectof that regime.As this subject/object,Manis a strangetwo-sided
entityor "doublet." Foucaultidentifiesthreemanifestationsof this Man
"doublet." First, in the transcendental/empirical double, Man consti-
tutes the world of empiricalobjects and is constitutedby them. Second,
in the cogito/unthoughtdouble, Man knows himself to be determined
by unknownforcesand is determinedby them. Third,in the retreat-and-
return-of-the-origindouble, Man createshistory and is createdby past
events. In each case, the subjectpole suggestsman's autonomyand ra-

12. Jiirgen Habermas, "The Entwinement of Myth and Enlightenment: Re-Reading

Dialectic of Enlightenment, " New German Critique 26 (Spring/Summer 1982): 13, 28-30;
and "A Philosophico-Political Profile," p. 82.
13. Jiirgen Habermas, "The Dialectics of Rationalization," Interview with Axel Hon-
neth, Eberhard Knodler-Bunte, and Arno Widmann, Telos 49 (Fall 1981): 15.
14. Fraser, "Michel Foucault: A 'Young Conservative'?," p. 166. I am indebted to
Fraser for this analysis of Foucault's critique of humanism.
 Nancy S. Love 273

tionality, but the object pole denies them. The humanistproject is to

make the subjectpole masterthe object pole, to overcomethe other in
nature, self, and society.'5
This humanistprojectclearlyhas universalistictendencies.These ap-
pear in anthropologicaland teleologicalidealizationsof the Enlighten-
ment as a stage in Man's self-formation.Foucaultrejectsthese idealiza-
tions but not Enlightenmentitself. Anthropologicalconceptionsof Man
expressthe self-defeatingcharacterof the humanistproject:the subject
Man is itself an objectification.Teleologicalconceptionsof Man's self-
formationcorrelatewith disciplinarytechniquesfor Man's subjugation,
specificallywith new ways of administeringand utilizingtime.16Unlike
humanists,FoucaultregardsEnlightenmentreasonas only one of many
rationalities.He arguesthat it can be criticizedwithoutlapsinginto irra-
tionality, that there can be rationalcritiquesof Enlightenmentreason.
Indeed,he regardssuch criticismas the ethos of Enlightenment,and he
concludesthat humanismand Enlightenmentare at odds.7
This stance toward Enlightenmentclearlyconflicts with Habermas's
universalism.Although Habermasassociatesmodernitywith the differ-
entiation of value-spheres,each of which has its own rationality,he
regardsthis differentiationas part of a universaldevelopmentalprocess
and arguesthat universalstructuresof consciousnesslink these multiple
rationalities.Cultural relativists(Habermasincludes Foucault among
them) merely operate at a different, a less abstract, level of analysis:
"Whetherand, if so, how the relativismof value contents affects the
universal character of the direction of the rationalizationprocess,
depends... on the level at whichthe pluralismof 'basicpoints of view'
is set."'8 Habermassets this level betweenmultiplecontentsand univer-
sal forms: "The unity of rationalityin the multiplicityof value spheres
rationalizedaccordingto theirinnerlogics is securedpreciselyat the for-
mal level of argumentativeredemptionof validityclaims."'9
Foucault, however, would not identify the issue here as "levels of
analysis." He does not acceptHabermas'sdistinctionbetweenuniversal
forms of argumentationand particularcontentsof arguments.My com-

15. Michel Foucault, The Order of Things, ch. 9.

16. Michel Foucault, Discipline and Punish: The Birth of the Prison. tr. Alan Sheridan
(New York: Vintage Books, 1979), p. 60.
17. Foucault, "What is Enlightenment?" in The Foucault Reader, 42-43. For this cri-
tique of Habermas, see: Foucault, "Structuralism and Poststructuralism," p. 201 and
"Space, Knowledge, and Power," in The Foucault Reader, p. 249.
18. Jiirgen Habermas, The Theory of Communicative Action, vol. 1, Reason and the Ra-
tionalization of Society, tr. Thomas McCarthy (Boston: Beacon Press, 1984), p. 181.
19. Ibid., p. 249.
274 Foucault& Habermason Discourse& Democracy

parison of their views on discoursewill reveal that Foucault regards

Habermas'sargumentativeredemptionof validityclaims, his notion of
discursive truth, as part of a particular power/knowledge regime.
Foucault portraysthe idea of truth and the search for it as peculiarly
modern preoccupations. Habermas's universal forms are particular
contents.20
Though Foucaultrejectshumanismwith its universalistictendencies,
he retainstwo other componentsHabermasassociateswith modernity:
differentiationand reflexivity.Accordingto Habermas,modernworld
views, unlike mythicalones, differentiatebetween nature and culture.
This differentiationbeginswith the externalworld:natureis desocialized
and society is denaturalized.Later a concept of the internalworld, of
subjectivity, develops to complement external nature and culture.
Modernsocietiesnot only differentiatebetweenthese three worlds, but
also adopt differentattitudestowardeach, i.e., an instrumentalrelation
to nature,a moralrelationto society, and an expressiverelationto self.
Correspondingto these attitudes are distinct validity claims to truth,
rightness,and sincerity.
Habermasarguesthat differentiationmakes rationalcommunication
possible. Formal world concepts allow membersof society to adopt
reflexiveattitudestowardtheirnaturaland social world. Individualscan
make assertionsabout the one objectiveworld and their own intersub-
jective one. Partnersin communicationcan assess these assertions:Are
certain statementsabout objectiveconditionstrue? Are certainactions
right with regardto social norms?That is, formalworld conceptsallow
individuals collectively to assert and to examine validity claims. As
Habermasputs it, "a decenteredunderstandingof the world opens up
the possibilityof dealingwith the world of facts in a cognitivelyobjec-
tified mannerand with the world of interpersonalrelationsin a legally
and morallyobjectifiedmanner."21Decentrationalso frees subjectivity
from objectification,since individualscan also discusstheir desiresand
feelings. Habermassays that, "we call a personrationalwho interprets
the natureof his desiresand feelingsin the light of culturallyestablished
standardsof value, but especiallyif he can adopt a reflectiveattitudeto
the very value standards through which desires and feelings are
interpreted."22
Accordingto Habermas,increasinginterpretiveactivityand increas-

20. Michel Foucault, "Truth and Power," Interview with Alessandro Fontana and
Pasquale Pasquino, in Power/Knowledge, ed. Colin Gordon (New York: Pantheon, 1980),
p. 133.
21. Jiirgen Habermas, Theory of Communicative Action, vol. 1, p. 216.
22. Ibid., p. 20.
 Nancy S. Love 275

ingly rational orientations accompany decentered world views. He

describes cultural modernizationas a movement from "normatively
ascribedagreements"to "communicativelyachievedunderstandings."
Differentiation,reflexivityand the culturalrationalizationaccompany-
ing themare, he argues,necessaryconditionsfor an emancipatedsociety.
"The more culturaltraditionspredecidewhich validity claims, when,
where, for what, from whom, and to whom must be accepted,the less
the participantsthemselveshavethe possibilityof makingexplicitand ex-
amining the potential grounds on which their yes/no positions are
based."23Culturalrationalization,however,is not a sufficientcondition
for emancipation. Participantsin rational communicationprocesses
must decide what they regardas the good life.
Foucaultalso associatesEnlightenmentwith a reflexiverelationto cul-
turaltraditions.The ethos of Enlightenmentis "a critiqueof whatwe are
saying, thinking, and doing through a historical ontology of
ourselves."24He describesthis ethos as a "limit-attitude,"not only the
discovery of limits but also their transgression.Yet Foucault, unlike
Habermas,does not seek universalforms above and beyond particular
contents. Instead, he examines modern discourse, i.e., the discourse
about Man, as an historicalevent in our constitutionas subjects. His
genealogyis directedtowardidentifying,within the events which made
us what we are, the possibilityfor us to be otherwise.He asks: "In what
is given to us as universal,necessary,obligatory,what place is occupied
by whatever is singular, contingent, and the product of arbitrary
constraints?" 2
Thisproject,whichhe calls a "criticalontologyof ourselves,"mustbe
experimental,must "put itself to the test of reality."26We must explore
the limitsupon us and identifythose we can transgress.But, recognizing
theirdangers,Foucaultrejectsglobal projectsfor a new Man. He is not
seekinga new "program."Instead,appropriatetests are specificstrug-
gles, partial transformations of our actions and thoughts. These
tests" are "workcarriedout by ourselves,upon our-
"historico-practical
selves as free humanbeings."27As this self-critical,self-creativeethos,
Enlightenmentfurthersfreedom.
Foucault describesthis work as the study of "practicalsystems."
Those systemsare the forms of rationalitywhich organizehow we act
and the freedomwe have in acting. ParallelingHabermas'sdifferenti-

23. Ibid., p. 70.

24. Foucault, "What is Enlightenment?" in The Foucault Reader, p. 45.
25. Ibid.
26. Ibid., p. 47.
27. Ibid.
276 Foucault& Habermason Discourse& Democracy

ated modern world, Foucault identifies three areas from which these
systemsarise:"relationsof controlover things, relationsof actionupon
others,relationswith oneself."28Theseare the respectiveaxes of knowl-
edge, power, and ethics. The criticalquestionscorrespondingto these
areasare: "How are we constitutedas subjectsof our own knowledge?
How arewe constitutedas subjectswho exerciseor submitto powerrela-
tions? How are we constitutedas moralsubjectsof our own actions?"29
Foucaultidentifieswhatis at stakein this work:"How can the growthof
capabilities be disconnected from the intensification of power
relations?"30
For Foucault, it is extremelydifficult, perhapsimpossible,to discon-
nect capabilitiesfrom power. This is becausesubjectivity,the sourceof
those capabilities,is alreadysubjugation.His concernis the multipleob-
jectificationswhich make humanssubjects.He identifiestwo meanings
of subject:"subjectto someoneelse by controland dependence,andtied
to his own identityby a conscienceor self-knowledge."31 These suggest
that subjectification and objectification inseparableprocesses.The
are
subjectis constituted as an object of society and self.
Accordingto Foucault,disciplinarypowersubjugatesthe modernsub-
ject. Disciplinarypoweris distinctivebecauseit totalizesas it individual-
izes, or becauseit createscapacitiesas constraints.By subjectingindivid-
uals to constantsurveillance,it forcesthemto scrutinizethemselvescon-
stantly:"He who is subjectedto a field of visibility,and who knowsit,
assumesresponsibilityfor the constraintsof power;he makesthem play
spontaneouslyupon himself;he inscribesin himselfthe powerrelationin
whichhe simultaneouslyplaysboth roles;he becomesthe principleof his
own subjection."32This constitutionof the individualswhomit liberates
is the dark side of the Enlightenment.As Foucaultputs it: "The real,
corporaldisciplinesconstitutedthe foundationof the formal, juridical
liberties. . . . The Enlightenment which discovered the liberties also
inventedthe disciplines."33
Habermasquestionsthis conclusion.He arguesthat culturalrationali-
zation not only allows individualsto adopt reflexive attitudestoward
natureand society, but also frees subjectivityfrom objectification.As

28. Ibid., p. 48.

29. Ibid., p. 49.
30. Ibid., p. 48.
31. Michel Foucault, "The Subject and Power," appendix to Michel Foucault: Beyond
Structuralism and Hermeneutics by Hubert L. Dreyfus and Paul Rabinow (Chicago:
University of Chicago Press, 1982), p. 212.
32. Foucault, Discipline and Punish, p. 203.
33. Ibid., p. 222.
 Nancy S. Love 277

rational positions have explicablegrounds, so rational individualsex-

plain their preferencesin argument.In orderto argue, individualslearn
to considerhow theirpreferenceslook to others,to adopt a third-person
perspectiveon their subjectiveworld. While arguing, they also learn
whichof theirpersonalpreferencesaregeneralizableinterests.Habermas
maintainsthat "evaluativeexpressionsor standardsof value have justi-
ficatoryforcewhenthey characterizea needin sucha way that addressees
can, in the frameworkof a commonculturalheritage,recognizein these
interpretationstheir own needs."34For Habermas, subjects who so
ground their preferenceshave rationallyexaminedand autonomously
chosen them. In contrast, "anyone who is so privatisticin his attitudes
and evaluationsthat they cannotbe explainedand renderedplausibleby
appealto standardsof evaluationis not behavingrationally."35By impli-
cation, such individualsalso cannot achieveautonomy.
In comparisonwith Habermas,Foucault'sanalysisof Enlightenment
seems one-sidedand incomplete.Once Foucault associatessubjectivity
with subjugation,he cannot disconnect our capacities from the con-
straintsupon us. Instead, he disconnectsindividuationfrom freedom;
the self becomesan innerworldproducedby externalstimuli.As Haber-
mas puts it, "genealogicalhistoriographydeals with an object domain
from whichthe theoryof powerhas erasedall tracesof communicative
actions entangledin life world contexts."36Foucaultasks us to examine
how subjectificationlimits us and to resist those limits. But what in us
can criticizeand transgress?Once Foucaultdefines subjectivityas sub-
jugation, wherecan he turn for resistance?
Foucault recognizesthis problem. He defines our task as liberation
from "the type of individuationwhich is linkedto the state" and says,
"We have to promotenewforms of subjectivitythroughthe refusalof
this kind of individuality. .. "37 In such statements, he relies upon indi-
viduality,which is Habermas'sfourth characteristicof modernity.But
Foucault does not provide criteriafor distinguishingbetweentypes of
individualityor forms of subjectivity.38 Althoughhe impliesthat reflex-
ivity can enableas well as constrain,his reflexivesubjectsare conform-
ists, not critics.They conformeven as they criticize,for theircriticismis
constrainedby its objects.

34. Habermas, Theory of Communicative Action, vol. 1, p. 92.

35. Ibid., p. 17.
36. Jiirgen Habermas, The Philosophical Discourse of Modernity, tr. Frederick
Lawrence (Cambridge, MA: The MIT Press, 1987), ch. X, p. 286.
37. Foucault, "The Subject and Power," p. 216; emphasis added.
38. For a more developed discussion, see Dews, Logics of Disintegration, ch. 5, and
"Power and Subjectivity in Foucault," p. 95.
278 Foucault& Habermason Discourse& Democracy

Habermas'sapproachto Enlightenment,specificallyhis concept of

communicativerationality, provides the criterion Foucault lacks. A
liberated subject still conforms, but to "communicativelyachieved
understandings"not to "normativelyascribedagreements."But Haber-
mas's approachis also one-sided.He focuseson capacities,on communi-
cationas individuation,and neglectsconstraints.Moreprecisely,he rele-
gates constraintsto instrumentaldistortionsof communicationand ex-
cludes them from communicationitself. Yet willingnessand ability to
objectifyoneself, to make oneself transparentto others, are the founda-
tions of his rational subjectivity.This looks suspiciouslylike the self-
defeatinghumanistproject:subjectsmasterobjectsby becomingobjects
themselves.What distinguishesthe justificationfrom the normalization
of personal preferences?What remains of otherness within the self?
Withinsociety?39
Habermasalso acknowledgeshis problem. He admits that language
"presents inalienablyindividual aspects in unavoidablygeneral cate-
gories."40For this reason, expressivevalidityclaimscan only be judged
sincereor insincere.He also says,

nothingmakesme more nervousthan the imputationthat because

the theoryof communicativeaction focuses attentionon the social
facticityof recognizedvalidity-claims,it proposesor at least sug-
gests a rationalisticutopian society. I do not regard the fully
transparent-let me add in this context:or indeeda homogenized
and unified-society as an ideal-.41

Still, it remainsan open question "how far the net of intersubjectivity

must be spreadin orderto stabilizethe identityof individualsas well as
that of the social group."42 Habermas'sconcernis the formalprecondi-
tions for democraticdiscourse.He does not addressits content, in this
case, what we should revealand when we should revealit.
Whatthen is the debatebetweenFoucaultand Habermasregardingthe
Enlightenmentand its product,the rationalsubject?Foucaultdoes not
reject Enlightenment;he, like Habermas,would radicalizeit. Nor does
Foucaultrejectmodernity;he abandonsuniversalism,but retainsreflex-

39. For such a Foucauldian critique of Habermas, see: William E. Connolly,

"Discipline, Politics and Ambiguity," Political Theory, 11 (August 1983): 334-35.
40. Jiirgen Habermas, "On Systematically Distorted Communication," Inquiry 13
(1970): 211.
41. Habermas, "A Philosophico-Political Profile," p. 94.
42. Jiirgen Habermas, "Towards a Theory of Communicative Competence," Inquiry 13
(1970): 373.
 Nancy S. Love 279

ivity and differentiation.He also reliesupon subjectivitythoughhis own

analysisprecludesit. It is here, on the issue of subjectivity,that mutually
beneficial debate begins. Foucault must recognize that subjectivity
enables:it offers the possibilityof self-understandingand self-determi-
nation. Criticalreflexivityprovidesa criterionfor distinguishingkindsof
subjectivity.Habermasoffers this criterion,but he does not adequately
address how subjectivity also constrains and thus fails to protect
othernessadequatelywithinthe self and society.

II. Discourse
Habermas'sand Foucault's differencesregardingthe rational subject
shape their debate over democraticdiscourse.Foucault'sideas on dis-
course revolve around the relationship between truth and power.
Accordingto Foucauit,the traditionalphilosophicaldictum, "you shall
know the truth and the truthshall set you free," is hopelesslynaive. So
too is the critiqueof ideology which, following this dictum,juxtaposes
liberatingtruths to powerful illusions. Foucault argues that truth is
neitheroutsideof powernor lackingin powerand hencecannotbe juxta-
posedto it. The productionof truth(andtruthis produced)is thoroughly
imbuedwith power.Foucaultcharacterizestheirinteraction:" 'Truth'is
linked in a circularrelation with systemsof power which produceand
sustain it, and to effects of power which it induces and which extend
it."43Foucaultexploresthesepowerrelationswhichcreatetruthand the
power effects truth creates.
He analyzes discourses as manifestations of a power/knowledge
regimeor a regimeof truth. In "The Discourseon Language,"Foucault
identifiesthreetypes of proceduresproducingdiscourse.44First,rulesof
exclusiondeterminewhat discourseis acceptable.Among theserulesare
prohibitions,e.g., what we can say, when and wherewe can say it, and
who can say what. Also includedhereis the divisionbetweenreasonand
folly which Foucault understandsas a division between common and
idiosyncraticspeech. Here, too, and to some extent encompassingpro-
hibitionsand divisions, the will to truth operates,excludingdesireand
powerfrom discourse.Second,theseexternalrulesarecomplementedby
internalones, which include mechanismsfor identifyingtruth, such as
culturalnarrativesor author'sintentionsto whichit mustconform.They
also include techniquesor proceduresfor acquiringtruth. The distinct

43. Michel Foucault, "Truth and Power" in Power/Knowledge," p. 133.

44. Michel Foucault, "The Discourse on Language," appendix to The Archaeology of
Knowledge, tr. A. M. Sheridan Smith (New York: Pantheon, 1972), pp. 215-39.
280 Foucault& Habermason Discourse& Democracy

disciplineswhichidentifytruthby the methods,rules, tools, etc. used to

produceit belonghere. Third,conditionsof employmentand appropria-
tion constraindiscourse.Regardingemployment,Foucaultarguesthat
areas of discourseare not equallyopen and penetrable.Rituals, fellow-
ships, and doctrinesselectqualifiedspeakers.Educationdetermineswho
earnsthese qualificationsand therebydistributesspeakersamong kinds
of discourse. These three types of proceduresrespectivelymaster the
powers, control the appearances, and establish the producers of
discourse.They are the multipleconstraintsby whichit producestruth.
Although power affects how discoursesproducetruth, truthalso has
effects of power.Foucaultarguesthat definiterelationsof knowledgeac-
company certaintypes of power. The disciplinarydiscoursecharacter-
istic of our societyhas this doubleepistemologicaland politicaleffect. It
createsa soul to be knownand a subjectionto be maintained.45 We have
alreadyexaminedthe secondeffect, our simultaneousconstitutionas the
subjectsand objects of politics;surveillanceteachesus to scrutinizeand
to normalizeourselves.The first effect, the knowledgerelation,also in-
volves overlappingsubjectificationand objectification.Two modes of
producingtruth accompanydisciplinarypower: scientific and confes-
sional discourse. In another context, we have also already seen the
sciences associated with discipline. They are the human sciences
dedicatedto knowing and developingthe object Man. What remains,
and what is most relevant here, is the subjectiveside of knowledge:
confession.
In volumeone of TheHistoryof Sexuality,Foucaultdescribeswestern
man as the "confessinganimal."46Confession,he argues,is the charac-
teristic discourse of reflexive subjects, a discourse in which subjects
discussthemselves.Confessionproducestruthin two senses. First, indi-
viduals must confess the truth about themselves.Foucault arguesthat
our Christianheritage, now secularized,requiresindividualsto reveal
theirdesiresas well as theirtransgressionsin discourse.He saysthat "we
are forced to producethe truth of power that our society demands,of
which it has need, in orderto function:we must speakthe truth;we are
constrainedor condemnedto confessor to discoverthe truth."47Second,
this discourse itself aims at truth. In confession, we submit our true
desires for examinationand evaluationto determinewhetherthey are
true, whetherthey conform to acceptedand/or acceptablenorms. "In

45. Foucault, Discipline and Punish, p. 305.

46. Michel Foucault, The History of Sexuality: Introduction, vol. 1 (New York: Vintage
Books, 1980), p. 57.
47. Michel Foucault, "Two Lectures" in Power/Knowledge, p. 93.
 Nancy S. Love 281

the end, we are judged, condemned,classified, and.determinedin our

undertakings,destinedto a certainmode of livingor dying, as a function
of the true discourses which are the bearers of specific effects of
power."48Discoursenot only judges desiresbut also administersthem.
In TheHistoryof Sexuality,Foucaultillustrateshow the discourseon sex
displaces, intensifies, reorientsor otherwise modifies desire. He con-
cludes that a multiplicationof discourseinvolves an intensificationof
interventionsof power.49When subjectsknow and control themselves,
they are also known and controlled.
WhereasFoucaultregardsall speechas discourseand regardsall dis-
coursesas implicatedin power,Habermasdefinesdiscoursemorerestric-
tively. Discourseis speech freed from power.50In discourse, speakers
assess validity-claimsbased upon their grounds.This communicationis
rationalbecauseonly "the peculiarlyconstraint-freeforce of the better
argument" operates.5 Habermasspells out the preconditionsfor ra-
tional communication.Linguisticcompetence,i.e., masteryof linguistic
rules, is one. But it is a necessary,not a sufficient,precondition.Haber-
mas regardsspeechas a pragmaticactivity;its purposeis mutualunder-
standing. Achieving understandingalso requirescommunicativecom-
petence, i.e., masteryof an appropriatecommunicativecontext. That
context should approximate, if it cannot achieve, the ideal speech
situation.
Accordingto Habermas,ideal speechinvolvessymmetricalrelationsin
"the distributionof assertionand dispute, revelationand concealment,
prescriptionand conformityamong the partnersof communication."52
Habermaspresentsthese symmetriesas respectivelinguisticconceptions
of truth, freedom,andjustice. Unrestraineddiscussionin whichall opin-
ions can be criticizedleads to unconstrainedconsensus,to his discursive
definitionof truth.Unimpairedself-representation whichincludesrecog-
nition of others'self-representations reconcilescommunicationand indi-
viduation, and therebyrealizesfreedom. Complementarityof expecta-
tions allows universalunderstanding,enforcesuniversalizednorms, and
therebycreatesjustice. Habermasclaimsthat whenthese symmetriesex-
ist, communicationis unconstrainedby its own linguisticor pragmatic
structure.5
Foucault clearlywould contest this claim. Habermas'stheory of ra-

48. Ibid., p. 94.

49. Foucault, History of Sexuality, p. 60.
50. Habermas, Theory of Communicative Action, vol. 1, p. 42.
51. Ibid., p. 26.
52. Habermas, "Towards a Theory of Communicative Competence," p. 371.
53. Ibid., p. 372.
282 Foucault& Habermason Discourse& Democracy

tional communicationremoves only some of the constraintsFoucault

associateswith discourse.Regardingrules of exclusion,linguisticcom-
petenceperpetuatesthe distinctionbetweencommon and idiosyncratic
speech,and excludesthe latter.AlthoughHabermasdiscussesthe trans-
formation of language through persistent, creative usage, he clearly
privilegesstandardspeech acts.54Linguisticcompetencealso privileges
verbalinteractions.Habermasstigmatizesnon-linguisticcommunication
as a pre-linguisticreversion.5Still, Habermasinsiststhat the agendafor
communicationbe open. Anythingwhichcan be verbalizedmay be said.
This responds,in part, to Foucault'sconcernabout rules of exclusion.
Habermasalso addressesFoucault'schargethat discourseis selectively
appropriatedand employed. He stipulatesthat anyone may place any-
thing on the agenda.Habermasacknowledgesthat this conditionwill be
difficultto fulfill. The economicand politicalassymetrieswhichcurrent-
ly distort discoursemust be overcome.56Habermashas been criticized
for failing to explain how to overcome them. But here too the ideal
speechsituationseems to push communicationin what Foucaultwould
regardas an appropriatedirection.
Or does it? For Foucault,afterassymetriesare overcome,once anyone
can say anything,discourseremainsconstrained.Habermasseeks sym-
metry in dialogue because only unconstrainedcommunicationcreates
unconstrainedconsensus,his discursivedefinitionof truth. The will to
truth, to determinethe truth by rational argument,motivates Haber-
masiandiscourse.But Foucaultassociatesthe will to truth,as well as the
accompanyingexclusionof desire and power from discourse,with our
power/knowledgeregime. In "The Discourseon Language,"Foucault
ends his critique of constraints upon discourse by asking whether
uninterrupted discourse is desirable. His answer is no." Even
Habermas'sideal speechsituation,perhapsespeciallythat situation,ex-
pressespower. The will to truth manifest in ideal speech is internalto
discipline.
Habermasiandiscourse constrains in both of the senses Foucault
discusses:individualsmust speakthe truthand speechaims at truth.We
have alreadyexaminedthe first sense: rationalcommunicationdepends
upon speakers' sincerity, in Foucault's terms, upon their willingness
to confess. Habermas'ssecond condition for ideal speech, i.e., mutual

54. Habermas, "On Systematically Distorted Communication," p. 207.

55. Habermas, "Systematically Distorted Communication," p. 215 and "Towards a
Theory of Communicative Competence," p. 369.
56. Habermas, "Towards a Theory of Communicative Competence," p. 372.
57. Foucault, "Discourse on Language," p. 229.
 Nancy S. Love 283

self-representation,does not adequatelyaddressthe constraintsthis in-

volves. Mutual self-representationonly superficiallyreconciles com-
munication and individuation, since individuationalready represses.
Foucault argues that individualsare vehicles of power: "In fact, it is
already one of the prime effects of power that certain bodies, certain
gestures, certain discourses,certain desires come to be identified and
constitutedas individuals."s5For Foucault,transparentsubjectivityis a
problem,not a solution.
Discoursealso constrainsin a second sense: truth is the goal of our
confessions. Habermasasks us to realizethe rationalitypotentialfound
in the validitybasis of speech. To do so, we must rationalizeour social
relations and our culturaltraditions. Social relations are rationalized
when individuals discuss their desires, reach a consensus regarding
generalizableones, and applythat consensusequallyto all. This applica-
tion of general norms to specific individualsis the third condition of
ideal speech;it is Habermasianjustice. Culturaltraditionsare rational-
ized when they are discussed and affirmed or denied. Habermas's
distinctions between "normatively ascribed agreement" and "com-
municativelyachievedunderstanding"is relevanthere. He arguesthat
"the greaterthe shareof prelinguisticallyfixedmotivationswhichcannot
be freely convertedin public communication,the greaterthe deviance
from the model of pure communicativeaction.""9
The questionthat ariseshere is: How rationalcan and should society
be? Regardingsocietalrationalization,we have alreadyseen that desires
cannot alwaysbe verbalized.Can they alwaysbe generalized?Can con-
flicts always be resolved with yes/no decisions? Are those decisions
alwaysapplicableto individuals?Withrespectto culturalrationalization
additional questions arise. Does social identity also rest upon non-
rationalbonds which we cannot, and need not, assess?Does Habermas
respectsuch bonds?Does he distinguishthem from irrationalism?If so,
how? These questions suggest that truth may be more contingentand
ambiguousthan Habermasallows.
Habermasacknowledgesthis problem.He professesrespectfor tradi-
tional understandingsand admits that "actual forms of life and actual
life-historiesare embeddedin unique histories."60 The lifeworld, the
backgroundfor communication,is simultaneouslyascribedand achieved
even in modernsocieties.This also appliesto Habermas'stheoryof com-
municativeaction. He says, "insofaras it refersto structuresof the life-

58. Foucault, "Two Lectures" in Power/Knowledge, p. 98.

59. Habermas,"Towardsa Theoryof CommunicativeCompetence,"p. 373.
60. Habermas, Theory of Communicative Action, vol. 1, p. 70.
284 Foucault& Habermason Discourse& Democracy

world,it has to explicatea backgroundknowledgeoverwhichno one can

disposeat will. The lifeworldis at first 'given'to the theoretician(as it is
to the layperson)as his or her own, in a paradoxicalmanner."61 For this
reason,Habermasdisavowstransparenttruthas well as transparentsub-
jectivity:

Stabilityand absenceof ambiguityare ratherthe exceptionin the

communicativepracticeof everydaylife. A morerealisticpictureis
that drawnby ethnomethodologists-of a diffuse, fragile, contin-
uously revisedand only momentarilysuccessfulcommunicationin
whichparticipantsrely on problematicand unclarifiedpresupposi-
tions and feel their way from one occasionalcommonalityto the
next.62

If truthis so contingent,so ambiguous,Habermasshouldexploreother

forms of communication,e.g., symbolicpolitics, and he should qualify
his claims for communicativerationality.He should stressthe tentative
natureof the consensusreachedin democraticdiscourseand the dangers
of unreflectivelyimposing it upon individuals.63Instead, he leaves
himself open to the chargeof hyper-rationalism.
The ambiguityand contingencyof discourseis not, however,justifica-
tion for abandoningtruthaltogether.This position poses a problemfor
Foucault. Foucaultarguesthat "we are subjectedto the productionof
truth throughpower and we cannot exercisepower except throughthe
productionof truth."64 The associationof truthand freedommasksthis
power. Accordingto Foucault, power requiressuch masks: "Power is
tolerable only on condition that it mask a substantialpart of itself.
Power as a generallimit set on freedomis, at least in our society, the
generalform of its acceptability."65 Foucaultunmaskstruth, revealing
its complicityin power.The problemis that his critiqueof truthalso ap-
plies to his own revelations.
Habermasanalyzesthis problemand accusesFoucaultof overtrelativ-
ism and covert normativism.He characterizesFoucault's relativism:
"Each counter-powermoves within the horizon of power which it at-
tacks, and transformsitself as soon as it is victoriousinto a complexof

61. Fred Dallmayr discusses this tension in Habermas's concept of the lifeworld in "Life-
world and Communicative Action: Habermas," in Critical Encounters (South Bend, IN:
University of Notre Dame Press, 1987), p. 73-100.
62. Habermas, Theory of Communicative Action, p. 100-101.
63. See Connolly, "Discipline, Politics, and Ambiguity."
64. Foucault, "Two Lectures" in Power/Knowledge, p. 93.
65. Foucault,Historyof Sexuality,p. 86.
 Nancy S. Love 285

power whichprovokesa new counter-power."66 That is, Foucault'sno-

tion of power/knowledgeregimesdoes not permit privilegedcounter-
powers.It underminesthe validitynot only of rulingdiscourses,but also
of his counter-discourse.Yet Foucaultneeds to distinguishamong dis-
courses based upon their complicity in power. Otherwisehe cannot
answerHabermas'squestions: "Why is strugglepreferableto submis-
sion? Why ought dominationto be resisted?" Habermasargues that
"only with the introductionof normativenotions could he [Foucault]
beginto tell us whatis wrongwith the modernpower/knowledgeregime
and why we ought to oppose it."67
Foucault,however,does not introducea normativefoundationfor his
counter-discourse;he merely assumes one. Ironically, his normative
assumptionsincludethe traditionalassociationbetweentruth and free-
dom whichhe attacks.By unmaskingtruthas power, Foucaultprovides
a new truthwhichfreesus.68Foucaultknowsthat he needsto distinguish
among kinds of discourse.He calls for "a new politics of truth," for
"changes"in the "political, economic, and institutionalregimeof the
productionof truth.""69 But his associationof truthwithpowerprecludes
the liberatingdiscoursehe desires.
Habermas'sand Foucault'sdebateon discourseparallelsthat on sub-
jectivity.Foucaultneeds to recognizethat discourseenables;he requires
criteriafor distinguishingbetweenpowerand validity-claims.Habermas
provides such criteria with his ideal speech situation. But Habermas
neglects how discoursecan also constrain. He is insensitiveto power,
especiallyto the power of truth.

III. Democracy
The Habermas-Foucault debateregardingthe liberatingpotentialof dis-
course centers on the distinction between power and validity claims.
Foucaultmaintainsthat this distinctionmaskspoweras truthand Haber-
mas arguesthat it providesa normativefoundation for critique.Since
Habermas'sdistinctiondependson the democratizationof discourse,I
concludewith their debateon dominationand democracy.

66. Jiirgen Habermas, "The Genealogical Writing of History: On Some Aporias in

Foucault's Theory of Power," Canadian Journal of Political and Social Theory, 10 (1986):
4-5. Habermas develops this argument in The Philosophical Discourse of Modernity, chs.
IX-X, pp. 238-93.
67. Ibid., p. 7.
68. See Charles Taylor, "Foucault on Freedom and Truth," Political Theory 12 (May
1984): 173-74.
69. Foucault, "Truth and Power" in Power/Knowledge, p. 133; emphasis added.
286 Foucault& Habermason Discourse& Democracy

Habermasrarelydiscussesdomination.He treatsthe topic most exten-

sively in the article, "Hannah Arendt's CommunicationsConcept of
Power." Therehe adoptsArendt'sdistinctionbetweenforce and power.
Forceexistswhen "the purposive-rationalactor,who is interestedonly in
the success of his action, . . . dispose[s] of means with which he can com-
pel a subjectcapableof choice, whetherby threatof sanctions,persua-
sion, or by cleverchannelingof choices."70Habermasarguesthat force
accompanies strategic actions which are attempts to instrumentalize
another'swill. He concedesthat strategicaction allows genuineagree-
ments, e.g., bargainsand compromises,but only as meansto individual
ends. Agreementis not the telos of strategicaction, and the compromises
or bargainswhich resultare not unconstrained.
In contrast,"the fundamentalphenomenonof poweris not the instru-
mentalizationof another'swill, but the formationof a commonwill in
communicationdirectedtoward reachingagreement."71The power of
the consensuscommunicationcreatesrestson the rationalvalidityclaims
immanentin speech, on the "forceless force" of the better argument.
Althoughthis poweris an end in itself, it still has its uses. Arendtargues
that "power serves to maintain the praxis from which it springs."72
Communicativeaction producesan intersubjectivelysharedlife-world,
i.e., a commonwill, whichis embodiedin politicalinstitutions.The legit-
imacy of these institutionsarises from unconstrainedcommunication.
They must protectit to perpetuatethemselves.
Habermasdoes supplementArendt'sanalysis,but herconceptof com-
municativepower remainsfundamental.73Communicativepower pro-
vides Habermas'snormativefoundation for criticizingforce. He says,
"with the communicationsconcept of power, we can make the institu-
tionalizationof relationsof force comprehensibleas a transformationof
force into a power outfittedwith the appearanceof legitimacy."74That
is, the concept of communicativepower allows him to distinguishbe-
tween power and validity-claims.
Yet FoucaultregardsHabermas'sdistinctionbetweenforceand power

70. Jiirgen Habermas, "Hannah Arendt's Communications Concept of Power," Social

Research 44 (Spring 1977): 3-4.
71. Ibid., p. 4.
72. Ibid.
73. According to Habermas, the acquisition, maintenance, and employment of power
may involve force and violence, but only communication can generate the legitimacy upon
which these other activities rely. (Jiirgen Habermas, "A Reply to my Critics" in Habermas:
Critical Debates, ed. John Thompson and David Held (Cambridge, MA: MIT Press, 1982),
p. 269).
74. Habermas, "Hannah Arendt's Communications Concept of Power," p. 4.
 Nancy S. Love 287

as primarilya semanticone.7 Foucaultdefines a power-relationas "a

mode of action upon otheractions" and maintainsthat "to live in socie-
ty is to live in such a way that actionupon otheractionsis possible-and
in fact on-going." He concludesthat "a societywithoutpower-relations
can only be an abstraction."76 Communication,as a way of actingupon
others, is alwaysimplicatedin power.77Accordingto Foucault,the dis-
tinction betweenforce and power, the idea of power without coercion,
rests on a juridicalor negativeconceptof power. This conceptof power
is centralto democratictheory. Democratictheoristsjuxtaposerightsto
sovereignty;the problem of rights becomes a problemof sovereignty.
Foucault argues that this discourseof right, like that of truth, masks
power.It "efface[s]the dominationintrinsicto powerin orderto present
the latter at the level of appearanceundertwo differentaspects:on the
one hand, as the legitimaterightsof sovereignty,and on the otheras the
legal obligation to obey it."78 Foucault reveals how right, like truth,
originatesin and extendsdomination.
To do so, he substitutesthe problemof domination/subjugationfor
sovereignty/obedience.This substitutionis necessaryto explainthe in-
timate connectionsbetweendemocraticpolitics and disciplinarypower.
Accordingto Foucault,disciplineis a non-sovereignpower;it circulates
in the capillariesof societiesconstitutingtheircitizen-subjects.A theory
of sovereigntywhichbasespoliticallegitimacyandpoliticalobligationon
individualrightsconcealsthis power. Fromthe perspectiveof sovereign-
ty, institutionsand laws based on publicrightlook democratic,not dis-
ciplinary.But Foucaultarguesthat the democratizationof sovereigntyis
itself grounded in discipline: "The juridical systems . . . have enabled
sovereigntyto be democratisedthroughthe constitutionof a publicright
articulatedupon collectivesovereignty,whileat the sametime this demo-
cratisation of sovereignty was fundamentally determined by and
groundedin mechanismsof disciplinarycoercion."79Disciplinenormal-
izes individuals, insuring that democraciescohere or, in Habermas's
case, that discoursecreatesconsensus.Foucaultconcludesthat "it is not
throughrecourseto sovereigntyagainstdisciplinethat the effects of dis-
ciplinary power can be limited, because sovereigntyand disciplinary
mechanisms are two absolutely integral constituents of the general
mechanismof power in our society."80

75. Michel Foucault, "Politics and Ethics: An Interview," in The Foucault Reader, p.
378.
76. Foucault, "Subject and Power," p. 222.
77. Ibid., p. 217.
78. Foucault, "Two Lectures," in Power/Knowledge, p. 95.
79. Ibid., p. 105.
80. Ibid., p. 108.
288 Foucault& Habermason Discourse& Democracy

Habermas'sconcept of communicativepower is such a recourseto

sovereigntyagainstdiscipline.He providesa democraticjustificationfor
politicalpower,groundinglegitimacyand obligationon a processof will-
formation in which all participate freely and equally. Habermas
acknowledgesthat force, or what Foucaultmightcall discipline,accom-
panies even legitimatepoliticalorders:

A sphereof valueto whichsociallyinfluentialideasbelongcan gen-

erallyonly be incompletelyembodiedin a legitimateorder.Thiscan
be seen in the force that is built into the structureof action norms
despite their "consensual character." Norms require sanctions,
either externalsanctions (disapprovalby membersin the case of
conventions,an organization'scoerciveapparatusin the case of
legalnorms)or innersanctions(suchas shameand guiltin the case
of ethicalnorms.)8'

But, although Habermasacknowledgesthe power of norms and the

presenceof normalization,he does not limit consensusaccordingly.
When Habermasdoes discuss the constraintsconsensuscreates, his
concernis less the generalizationor normalizationof particularinterests
thanthe particularismof presumablygeneralinterests.Habermasinsists:
"In no sensedo I begin fromthe basisthat in all, or even in the majority
of politicaldecisions,legal or administrativeregulations,a generalinter-
est is at stake."82He arguesthat whereparticularinterestsare at stake,
bargainingand compromiseare entirelyappropriateand speculatesthat
the "pluralismof life-forms"and the "individualismof life-styles"will
even increasein a socialistsociety.83
But how does Habermasdistinguishbetweena trueand a false consen-
sus? Betweenparticularand generalinterests?Habermasdoes not pro-
vide a prioricriteriafor makingthese distinctions.They resultfrom the
processof democraticwill-formationitself. Participantsin rationalcom-
municationdecidewhatthe limitsof consensusshouldbe andwhenthose
limits should be revised. This seems appropriateas long as the institu-
tional preconditionsfor democraticdiscourseexist. Habermasacknowl-
edgesthat institutionaltransformationsare requiredhere.84Thosetrans-

81. Habermas, Theory of Communicative Action, vol. 1, p. 190.

82. Habermas, "A Philosophico-Political Profile," p. 95.
83. Ibid., p. 96.
84. Habermas, "Arendt's Communications Concept of Power," p. 223. Lorenzo
Simpson has argued that because Habermas leaves these decisions to democratic discourse,
his ideal speech situation does not necessarily discriminate against particular interests.
What Simpson neglects is Habermas's failure to specify the institutional transformations
 Nancy S. Love 289

formationswill furtherdevelop the principlesof bourgeoisright, pre-

sumablyto protectindividualsfrom oppressivenormsas well as particu-
lar interests. Otherwisea false consensus which excludes and/or sup-
pressesparticularinterestsis not only possible but also likely.
But what transformationsin political institutions are necessary?
Habermasonly hints that a differentparty system and anotherkind of
separationof powers are required.What transformationsin economic
relations are also needed? Habermassuggeststhat capitalismmust be
abolished and a capacity for economic self-organizationdeveloped.
Again, how? And how are these transformedpolitical and economic
systems anchored in our communicatively-achieved understandings?85
Until Habermasaddressesthesequestions,his theoryof democraticwill-
formationremainsdangerouslyvague.
These questionsrevealthe tentativenatureof Habermas'sdistinction
betweenpower and validityclaims. Habermasalso acknowledgesthis.
He admitsthat participantsin discoursemust

assume that in the inescapable pragmatic presuppositionsof

rational discourse only the non-coercivecoercion of the better
argumentgets a chance.But they know ... that eventhe presuppo-
sition of an ideal speechsituationis only necessarybecauseconvic-
tions are formed and contestedin a mediumwhich is not "pure"
nor removedfrom the world of appearancesin the mannerof the
platonicideals.86
Withthis, Habermasadmitsthe "everlastingimpurity"of discourse,but
maintainsthat ideal speechis an appropriateideal. Pursuitof that ideal
must, however, include clarificationof the institutionalpreconditions
for consensus. Appropriatedistinctionsamong interests, appropriate
limits on consensus,dependon those preconditions.Withoutthem, par-
ticularinterestsremainvulnerable.
Aware of this danger,FoucaultquestionsHabermas'sdistinctionbe-
tweenpowerand validityclaims. But, ironicallyFoucaultalso revealsits
importance.Foucaultarguesthat power is a relation, a mode of action
upon other actions. This makes power relations, despite Foucault's

which are necessary to make inclusiveness a realistic possibility ("On Habermas and Par-
ticularity: Is There Room for Race and Gender on the Glassy Plains of Ideal Discourse,"
Praxis International 6 (October 1986): 328-40).
85. On this issue, see Anthony Giddens "Reason without Revolution in Habermas and
Modernity, ed. Richard Bernstein (Cambridge, MA: MIT Press, 1984), pp. 95-121 and
"Labor and Interaction" in Critical Debates, pp. 149-61.
86. Habermas, "Entwinement," p. 30.
290 Foucault& Habermason Discourse& Democracy

denials, virtuallysynonymouswith social relations.Although Foucault

arguesthat a society withoutpower relationsis an abstraction,he con-
cludes that this makes their analysismore, not less, necessary:"For to
say that there cannot be a society without power relationsis not to say
eitherthat those whichare establishedare necessary,or, in any case, that
powerconstitutesa fatalityat the heartof society such that it cannotbe
undermined."87 But whatpowerrelationsareunnecessary?And how can
they be undermined?Again, Foucault cannot answer these questions
withoutdistinguishingamongkindsof powerbasedon a normativestan-
dard.Foucaultconcedesthat a notionof consensualpowerprovidessuch
a standard."The idealof a consensualpoliticsmay indeedat a givenmo-
ment serveeven as a regulatoryprincipleor betteryet as a criticalprinci-
ple with respect to other political forms."88But power persists. For
Foucault,one can only ask "whatproportionof nonconsensualityis im-
pliedin sucha powerrelationand whetherthat degreeof nonconsensual-
ity is necessaryor not and then one may questioneverypowerrelationto
that extent."89What Habermascalls consensualpower, remains,to a
greateror less extent, consensualdiscipline.
Latein life, Foucaultbeganto describeconsensusdifferently,in terms
closer to Habermas's.He made the following remarkson the logic of
dialogue:

The personaskingthe questionsis merelyexercisingthe rightthat

has been given him: to remainunconvinced,to perceivea contra-
diction,to requiremoreinformation,to emphasizedifferentpostu-
lates, to point out faulty reasoning,etc. As for the personanswer-
ing the questions,he too exercisesa rightthat does not go beyond
the discussionitself; by the logic of his own discoursehe is tied to
what he said earlier,and by the acceptanceof dialoguehe is tied to
the questioningof the other.90

Here, the violation of rights, not the pursuitof truth, constitutescoer-

cion. Elsewhere,Foucaultelaborateson this "new right": "One should
turn ... in the directionof a new right, one whichmust indeedbe anti-
disciplinary, but at the same time liberated from the principle of
sovereignty."91This is a right to intervenein politics which is neither

87. Foucault, "Subject and Power," p. 223.

88. Foucault, "Politics and Ethics," p. 378.
89. Ibid., p. 379; emphasis added.
90. Foucault, "Polemics, Politics, and Problematizations" in The Foucault Reader, p.
381.
91. Foucault, "Two Lectures" in Power/Knowledge, p. 108.
 Nancy S. Love 291

delegatednor commissionedby government.92 But despitethesegestures

toward anti-sovereignright and anti-disciplinaryconsensus, Foucault
continuesto define poweras domination.He rethinksnegativepoweras
positive power, but he lacks a theoreticalfoundation for affirmative
power.
One-sidednesspersists, then, in each philosopher'sviews on democ-
racy and again it poses problems.After neglectingthe constraintsfrom
which consensusoriginates,i.e., rationalityand subjectivity,Habermas
now neglects those it imposes. He fails to establishadequatelimits on
democratic discourse. Foucault, who focuses upon how democracy
dominates,fails to distinguishkindsof power, a distinctionnecessaryto
defend his anti-sovereignright and anti-disciplinarypower.

IV. Conclusion
How then does the Habermas-Foucaultdebate conclude?What does it
revealabout the problemsof democracyand modernity?I have argued
elsewherethat the philosophiesof Habermasand Foucault cannot be
combinedin a harmoniouswhole.93Each precludesthe other'sposition.
For this reason,it is not surprisingthat they talkedpast one another.In-
deed, WilliamConnollyarguesthat they confrontscholarswitha choice:
democracyor nihilism.94Still, if Habermasand Foucaulthad heardeach
other (and if othershearthem), perhapsthe choice need not be so stark.
Could their positionscoexist in a creativetension?Othershave reached
this conclusionand have proposedbalancingtheir positions.
Connollyhimselfproposesthe first seeminglybalancedalternative.He
says, "it may be possibleto articulatea visionof democraticlife that con-
sciously maintainsthe tension betweenthese two tendencies,affirming
the legitimacyof limits and conventionsessentialto democraticpolitics,
while otherwiseexposingand opposingthe moderndrift towardration-
alization, normalization, and dependency."95With this, Connolly
moderatesFoucault's nihilism. He admits the necessityof social con-
straintsand distinguishesessentialfrom non-essentialones.
A second balance is offered by RichardRorty. Accordingto Rorty,

92. For a more sympathetic interpretation of Foucault's "new right," see: Tom Keenan,
"The 'Paradox' of Knowledge and Power: Reading Foucault on a Bias," Political Theory
15 (February 1987): 5-37.
93. Nancy S. Love, "Dialectics and Politics," Polity 19 (Summer 1987): 693-705. For a
discussion of the Marxian and Nietzschean roots of this incompatibility see my Marx,
Nietzsche, and Modernity (New York: Columbia University Press, 1986).
94. Connolly, "Discipline, Politics, and Ambiguity," p. 333.
95. Ibid.
292 Foucault& Habermason Discourse& Democracy

Habermascriticizespost-structuralistsbecause they do not develop a

theoryto directsocialchange.Rortyacceptsthis characterization but not
the criticism.He suggests"splittingthe difference"betweenHabermas
and post-structuralistsby developinga "de-theoreticizedsense of com-
munity." Rorty says that "if one had such a de-theoreticizedsense of
community,one could accept the claim that valuing 'undistortedcom-
munication' was of the essence of liberal politics without needing a
theory of communicativecompetenceas backup."96From this perspec-
tive, philosophybecomesan unimportantexcursusin the questfor com-
munity. "Those who want beautiful social harmonies," Rorty con-
cludes, "want a postmodernistform of social life, in whichsocietyas a
whole assertsitself without botheringto grounditself."97
My concernis that these "balances"begin with an unacknowledged
and problematicchoice. Connolly and Rorty side with Foucault. Con-
nolly revealshis choice by distinguishingonly amongconstraints-some
essential and some non-essential.Habermas,who suggeststhat those
essential"limitsand conventions"arereallycapacities,is banishedfrom
the debate. Connolly claims that his theory of communicativeaction
fostersconfessionand normalization.Rorty'schoiceis equallyclear.He
arguesthat Foucault'shistoricalnarrativesshould supplantphilosophi-
cal metanarratives."Such narratives would not unmask something
createdby powercalled 'ideology'in the name of somethingnot created
by power called 'validity' or 'emancipation.'They would just explain
who was currentlygettingand using powerfor what purposes,and then
(unlikeFoucault)suggesthow some other people might get it and use it
for other purposes."98Rorty's "unlike Foucault," revealsthe problem
their choice poses. It is not enoughto say that "some" limits and con-
ventions are essential or to suggest that "some" other people obtain
power for "other" purposes.What limits and conventions?What peo-
ple? What purposes?How will these decisionsbe made? These crucial
questionsremain.
An initialchoicemust, I agree,be made. But I wouldchoosedifferent-
ly. We needa theoreticalframeworkwhichprovidesa normativefounda-
tion. Otherwise,ours is a different,moredangerous,balancingact. Con-
nolly and Rorty are, I suspect, suspendedupon Zarathustra'stightrope
over the abyss. Habermasprovidessuch a frameworkby distinguishing
among kinds of rationality, subjectivity,and democracy.Rationality

96. Richard Rorty, "Habermas and Lyotard on Postmodernity" in Habermas and

Modernity, ed. Richard Bernstein (Cambridge, MA: The MIT Press, 1985), p. 173.
97. Ibid., p. 175.
98. Ibid., p. 173; emphasis added.
 Nancy S. Love 293

allows criticism and conformity; subjectivityincludes autonomy and

identity;democracyinvolves participationand cooptation. The former
areour capacitiesandthe latterarethe constraintsupon us. Accordingto
Habermas,we can use our capacitiesto limit those constraints.He also
suggestshow to do this, i.e., throughdemocraticdiscourseundercondi-
tions approximatingidealspeech.Thisis a procedural,not a substantive,
norm as befits a modernworld. It providesa foundationfor answering
the questions Connolly and Rorty raise. By defendingtheir positions,
they implyits legitimacy.They enterHabermas'ssphereof communica-
tive rationality.
Foucault often gesturestoward such distinctionsbetween capacities
and constraints,but he fails to ground them. As Habermasrecently
argued, Foucault's total critique self-destructs.9 We are always con-
strainedby what we criticizesince it is part of what we are. Ironically,
this Foucauldianinsight revealsa problemFoucaultfails to overcome.
He remainsnegatively-boundto the discourseof modernity.Habermas
maintainsthat a paradigmonly loses its force when it is determinately
negatedby anotherparadigm.By showingthat we are more than con-
straints, that we also have capacities, Habermasrecovers that other
paradigmfrom our past.
But the debate does not end here. Habermasdoes not incorporate
Foucault'sposition. Although Habermasprovidesa theoreticalframe-
work, he does not achievea balance. Foucaultrevealsthat Habermas's
distinctionsare less clear than they seem. Communicativerationalityis
not necessarily"anotherparadigm."Habermasis also constrainedby
the discourseof modernity. His "capacities" are often implicatedin
disciplinarypower. This suggeststhat Habermas'sparadigmshouldalso
be debated;communicativerationalityshould itself be communicatively
achieved.A morereflective,a morecritical-a Foucauldian-considera-
tion of communicativerationalitywould addressthe limitsof consensus
and the ambiguityof truth.
The Foucault-Habermas debate,then, focusesthe tasksof democratic
theory. It raisesthe questionsa balancedtheoryof democraticdiscourse
must consider:When does subjectivitybecome subjection?When does
communicationbecome confession? When does democracy become
domination?By asking these questions, democratictheoristscan avoid
tippingthe scales.Theycan acknowledgethe achievementsand avoidthe
liabilitiesof modernity.

99. Habermas, Philosophical Discourse of Modernity, chapter XI.