Foucault on Power: A Problem in Radical Translation?

Author(s): Mark Philp

Source: Political Theory, Vol. 11, No. 1 (Feb., 1983), pp. 29-52
Published by: Sage Publications, Inc.
Stable URL: http://www.jstor.org/stable/191008 .
Accessed: 08/05/2014 10:57

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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 support@jstor.org.

Sage Publications, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Political Theory.

http://www.jstor.org

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 FOUCAULT ON POWER
A Problemin
RadicalTranslation?

MARK PHILP
JesusCollege

I 4IHICHEL
FOUCAULT'S WORK, his studiesof theprison,'of
particularcriminalsand theircrimes,2 of the historicalrootsof
andofa rangeofsocialscientific,4
sexuality,3 medical, andparamedical
discourses,5mayseemat firstglanceto be, at best,onlyobtusely
connectedwiththe concernsof bothliberaland Marxistpolitical
thought.Buttotakethisviewistofailtoappreciate thecentralconcerns
ofhiswork.Norshouldweseehisrecent recognitionoftheimportance
of powerin societysimplyas a minorconcession to theconcerns of
traditional onethatleaveshisworkand objectsof
politicalanalysis,6
studystillessentiallyidiosyncratic,
obscure, andofdoubtful relevance
to the"moreserious"debatesofpoliticaltheory. An interpretation of
Foucaultthatfailsto recognize thathis projectis to undermine and
replacetheclassicalliberalandMarxianformulations ofthenatureof
politics,as an objectof studyas well as of practice,is one that
fundamentally misunderstands hiswork.Of course,hisattempt to do
thismaywellbe idiosyncratic, and itsfavoramongthedisenchanted
radicalsoftheleftmaybe short-lived; Foucault'sworkmaysimply be
thelatestfadwithlittle substance-but
lasting wecannotsimply assume
this.Yetitisquitepossiblethatwithoutanattempt toconsider Foucault
seriouslyoutsidethesomewhat rarifiedatmosphere ofthecirclesand
publications of hisacolytes,thedominant traditionswillsuccessfully

A UTHOR'S NOTE: I am most gratefulto Charles Taylor,Steven Lukes, Jim Tully,

WilliamConnolly,WendyBrown,Michael Brint,and Bob Ashcroft for theircomments
on earlierdraftsof thisarticle.In particular,I would like to thankJohnGrayfor his
continuedencouragement and hisdetailedcriticism.Remainingerrorsare,ofcourse,my
own.

POLITICAL THEORY, Vol. II No 1, February1983 29-52

? 1983Sage Publications,Inc.

0090-5917 83 010029-24$2.65
29

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
30 POLITICAL THEORY / FEBRUARY 1983

consignhimto a possiblyundeservedobscurity. This articleattemptsto

providethe beginningsof an assessmentof his work. But it does so
withoutautomatically acceptinghislanguage,hisstyle,or histheoretical
framework. To thisextentitis an attemptat botha radicaltranslation of
Foucaultand an assessment.As faras I can ascertainthistaskhasyetto
be performed.7
Two immediateproblemswithanysuchattemptare Foucault'sstyle
and language.They are not simplydifficult, theyare deliberatelyso.
Foucault'sveryliterary and ornamental stylereflects
inparthisviewson
languageand knowledge.He sees himselfas attempting to articulatea
new discourse,one that subvertstraditionalconceptsand categories.
His stylethusservesa rhetoricalpurpose-it assistsintheconstruction
ofa newperspective fromwhichtheordinaryobjectsofanalysisand the
acceptedunitiesof thoughtdisappear.Crudely,he givesus a perverse
reformulation and genealogyof our acceptedcognitivegrasp on the
world. His workscan be seen as systematicattemptsto throwinto
questionour unstatedpremises.He takesthebedrockofoursocial and
politicalthoughtand attemptsto dissolveit or ruptureit. Giventhis
approach,thereare evidenttranslationproblems,sinceitcould be said
thatanyattemptto translateis an attemptto recouphisworkwithinthe
guidelinesand upon thebedrockofthedominantlanguagesofpolitical
discourse.We are,then,dealingwitha translation fromthelanguageof
an extremelyreluctantsubject;one who wittingly obstructsany such
attemptand who would repudiateanytranslationoffered. Withsucha
subject should one attemptto translateand to assess? Can any
translationbe adequate? Can any translationbe useful?On the first
point,we can claim that Foucault at least offersus permission;with
reference to his workshe states:"They are, in the finalanalysis,just
fragments and it is up to you or meto see whatwe can makeofthem."8
On the second point,we can suggestthat thoughtranslationsmay
neverbe perfect(indeed,one wondershow one would know if they
wereperfect),theycan be betteror worse.This articleis in some way
an attemptto answerthe thirdquestion.
In the followingdiscussion I shall concentrateon Foucault's
conceptionof power,sincethisseemsto be a pointthatwillallow us to
locate his position clearlyvis-a-visthe more common positionsof
Marxismor liberalism.I shall proceedas follows:I beginbyshowing
how Foucault fitsintoa traditionthatsees poweras a relationship, but
willalso show how his commitments differ fromthosemade byLukes9
and Poulantzas.'0I followthiswithan account of his conceptionof

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 31

power,usinga good deal ofquotationbutalso attempting to renderhis

thoughtrathermoreaccessiblethanitusuallyappearsto be. I thenraise
three problem areas in Foucault's account: his conceptionof the
and hisviewofresistance.Finally,I
political,hisideas about repression,
concludebyaskinghowfarhisaccountofpowercan be consideredas a
contribution to the broaderdebate on the definitionof the concept.

Foucault's view is that power refersto a relationship,but it is a

relationalconceptionof powerof some noveltysinceit differs signifi-
cantlyfromthe formulations of radicalsand Marxists.I willbeginby
offering an accountof relationalviewsof power;I willthenshow how
Foucault's workdepartsfromthe moreabstractviews.
Lukes criticizesParsons and othertheoristsforseeingpoweras a
systemsproperty.Parsonsdefinespowerwithina conservative viewof
the state as the consensuallyvalidated locus of authorityactingin
pursuitof collectivegoals. Such a definitionavoids implicationsof
force,coercion,or conflict;it is conceivedas "a generalisedmediumof
mobilisingcommitments or obligationforeffectivecollectiveaction.""
Arendtholdsa similarview.'2 BothlinkinwithwhatFoucaultrefers to
as the "liberal-juridico" conceptof powerl3-poweris somethingthat
can be held or transferred. I surrendermy powerto the state,in the
contractariantradition,or our cooperationproducespowerthatcan
then be used legitimately to attain certainends, as in Parsons and
Arendt.In theseaccountsofpowerthefocusis,as Lukesputsit,"on the
locution 'power to' ignoring'power over'. Thus power indicatesa
'capacity',a 'facility',an 'ability',not a relationship.Accordingly, the
conflictualaspectsof power-the view thatit is powerexercisedover
people-disappears altogetherfromview."14
It is thisconflictual aspectofpowerthatLukesand Connollywantto
stress.Lukes does so by definingpoweras follows:"A exercisespower
over B when A affectsB in a mannercontraryto B's interests."
Furthermore, he distinguishes histhreeviewsof powerin termsof the
way theyconceiveof interests. '5 "Significantaffecting"
and "interests"
are two key componentsof Luke's view. However,he also adds two
further criteria:(I) thenotionofagency,and (2) thecounterfactual case.
The twoare related.A's exerciseofpoweroverB must,forLukes,be in

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
32 POLITICAL THEORY / FEBRUARY 1983

somesenseintentional,and itmustalso be possibleto statethatwithout

A's action,B would have acted differently.

To use thevocabularyof powerin thecontextof social regulationsis to speak of

humanagents,separatelyor together,in groupsor organisations,throughaction
or inaction,significantly thethoughtsoractionsofothers(specifically
affecting ina
mannercontraryto theirinterests).In speakingthus,one assumesthat,although
theagentsoperatewithinstructurally determinedlimits,theynonethelesshavea
certainrelativeautonomyand could have acted differently....Withina system
characterised bytotalstructural
determinism,therewouldbe no placeforpower.16

Poulantzas, against whom these commentswere mainlydirected,

remainedlargelyunrepentant.He retainedand arguedfiercelyfora
relationalviewof power,buthe continuedto avoid referenceto agency
and it is doubtfulthat he could generateany coherentaccount of
counterfactualsto fithis view of structuraldeterminism.

The powerofa classrefers above all to itsobjectiveplaceineconomic,politicaland

ideologicalrelations-a place whichoverliesthepracticesofthestruggling classes
(thatis, theunequal relationsof domination-subordination amongclassesrooted
in thesocial divisionoflabour)and whichalreadyconsistsin powerrelationships.
The place of each class,and henceitspoweris delimited(i.e., at once designated
and limited)bytheplace oftheotherclasses.Poweris not,then,a qualityattached
to a class in-itself,
understoodas a collectionofagents,butdependson,and springs
from,a relationalsystemof materialplaces occupied by particularagents.'7

Puttingaside forthe momentthefraughtquestionof agency,let us

summarizethe pointof therelationalviewof power.Relationalviews
see poweras a termapplyingto a set of relationsin whichthereare
conflictsof interests, goals, desires,preferences,and so on, and where
one side is able to attainitsinterests,goals,desires,preferences,and so
on at the expenseof the other.We have suggestedthatthereare four
componentsthat a relationaltheoryof powermightinvoke,the first
two beingcommonto both Lukes and Poulantzas: significant affect-
ing, a characterizationof interests(goals, desires,preferences, and
so on), agency/intention, and the counterfactual case.
AlthoughFoucault uses a relationalconceptionof power,he rejects
bothLukes'sradicalconflictmodeland Poulantzas'sstructural conflict
view.He rejectstheformer on thegroundsthatitretainsan individualist
accountof agency,and thelatteron thegroundsof itsassumptionofa
generaland organizeddomination.More significantly, he also rejects
both on the groundsthat theyuse the conceptof repression,which

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp/ FOUCAULT ON POWER 33

Foucaultsees as a misguidedlynegativeway of characterizing

the
ofpower.Ofthefourpossible
operation elementsfora relational
theory
mentionedabove,Foucaultadoptsonlythefirst.
To seewhyhedoesso
we needto lookmorecloselyat hisaccount.
Foucaulttellsus whathe doesnotmeanbypower.It is not

a group of institutionsand mechanismswhichensurethe subservienceof the

citizensof a givenstate . . . [nor] a mode of subjugation,whichin contrastto
violencehastheformoftherule... [nor]a generalsystemofdominationexertedby
one groupover another.8

To paraphraseFoucault:The sovereignty of thestate,theformof

the of
thelaw,and overallunity domination arenotgivenattheoutset-
theyareonlytheterminal forms thatpowertakes.ForFoucault, power
mustbe understood in thefirst "as themultiplicity
instance: offorce
relationsimmanent in thespherein whichtheyoperateand which
constitutetheirownorganisation."'9
Sheridanappearsto translate rapportsde forceas "relationsof
power,"but thisgivesus a short-circuited account.On Sheridan's
reading,"powershouldbe understood as themultiplicity
of power
relationsin a particulararea."20But thismerelystatesthatpower
analytically torelations,
refers anditgivesusnobasisfordistinguishing
powerrelations fromotherrelations. The translation
of rapports de
offorce,orforcerelations,
forceas relations avoidsthiscircularity
by
construingpowerrelations as thoserelations
inwhichforceisexercised,
butitleavesundefined whatis to countas an exercise
offorce.I prefer
thislatteroptionand willassumethatFoucaultis sayingthatpower
are,inthefirst
relations instance, relations
inwhichforceis exercised.
Foucaultcontinues hisdefinition: Powermustbe understood

as theprocesswhich,throughceaselessstruggles and confrontations, transforms,

strengthens, or reversesthem(i.e.,therelationsofforce)[and] as thesupportwhich
theseforcerelationsfindinone another,thusforming a chainora system, or on the
contrary, thedisjunctions and contradictionswhichisolatethemfromone another;
and lastly,as the strategiesin whichtheytake effect,whose generaldesignor
institutional is embodiedinthestateapparatus,intheformation
crystallisation of
the law, in the varioussocial hegemonies.2'

We can interpretthisas follows."Power"refers

to a setof force
butitrefers
relations, tomorethanthesumoftheserelations offorce.
It
also refers
to theprocessbywhichtheserelations
becomestableorare

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
34 POLITICAL THEORY / FEBRUARY 1983

changed,to the patternsthat may be formedfromlinkingup sets of

forcerelations,and to the way in whichthese patternsare rendered
functionalto dominationthrougha particularstrategy.

... isthemovingsubstrateofforcerelationswhich,
Power'sconditionofpossibility
byvirtueoftheinequality,constantly engenderstatesof power,butthelatterare
alwayslocal and unstable.22

This "movingsubstrateof forcerelations"accountsfortheubiquityof

power:"Power is everywhere; not because it embraceseverything, but
because it comes fromeverywhere." Insofaras it formsa permanent
structureit is simply"the overalleffect"that emergesfromall these
relations.Power is a "complexstrategicsituation."
Foucault is essentiallysayingthat froma series of "unbalanced,
heterogenous, unstableand tenseforcerelations"patternsare thrown
up, whichare integrated intostrategies,whichgivetheappearanceofa
centralpower.Changes in the forcerelations,and theseare endemic,
produceshiftsin theoverallpatternofpower,whichcan onlyresistsuch
changesinsofaras sets of relationsof forcecan be mobilizedagainst
them.Powerrefersto the"6complex strategicsituation"-to thefieldof
force.It does not constitutethatfieldbut is, rather,the effectof the
patternswithinthatfield.Becausesetsofrelationsofforcefeedback to
conditiontheirmembersand inhibitchange,stablepatterns can emerge.
Power, then, is based on this fieldof relationsof force:"Power is
exercisedfrominnumerablepoints,in theinterplayof non-egalitarian
and mobilerelations."123
Butitremainsunclearjust whatFoucaultrefersto bytheterm"force
relations."He spends a greatdeal of efforton puttingus rightabout
power,butnevergivesus anyaccountofforce.He seemsto relyat times
on resonanceswiththe terms"force"and "power"as theyare used in
physics-and at timeshe sounds verymuch like Nietzschein those
sections of the Will to Power where Nietzsche overstretches the
metaphorbytalkingof powerquanta.24However,I shallreturnto this
problemof forceat a laterpoint;forthepresentit is onlynecessaryto
rememberthatthe conceptof forceis bothfundamental to Foucault's
conceptionof power,and undefined!
Four finalpointsabout Foucault's view need to be made briefly (I
shall be returningto each of themat a laterstage). First,accordingto
Foucault,relationsof powerare notsomethingthatoperateoutsideof
otherrelations(such as economic,sexual,or knowledgerelations)but

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 35

are immanentwithinthese:Theyaretheeffects ofdivisionsthatoccurin

theyare theinternalconditionsofthese
theserelations"and conversely
differentiations."25
Second, Foucault regardspower as comingfrom
below,not above:

The manifoldrelationshipsof forcethattake shape and come into play in the

machineryof production,in families,limitedgroups,and institutions,
arethebasis
forwiderangingeffectsof cleavagethatrunthroughthesocial bodyas a whole.26

Global dominationis,then,theendpointofan analysisofpower,notits

startingpoint.Third,concerning
agencyand intentionFoucaultargues:

Powerrelationsare bothintentional and non-subjective.

. . thereis no powerthatis
exercisedwithouta seriesofaimsand objectives... thelogicis perfectly clear,the
aims decipherable,and yetitis oftenthecase thatno one is thereto haveinvented
themand fewcan be said to have formulated them.27

These claimsrelateto the power/knowledge relationthatis centralto

Foucault'smorerecentwork.The finalclaimthatwe needto regardas
integralto Foucault's conceptionof powerconcernsresistance:

Resistanceis integralto power.The existenceof powerrelationships

dependson a
multiplicity of pointsof resistancewhichare presenteverywhere in the power
network.Resistancesare theodd termin relationsof power;theyare inscribedin
the latteras an irreducibleopposite.28

II

Foucault's account differsfrom that of Lukes and Poulantzas

in at least one major respect.If Poulantzas leaves out agencyand
counterfactuals, Foucault managesto go one step further to heresy:
He removesthe referenceto interests.For Foucault, power does
referto significant butitis too negativeto talksimplyinterms
affecting,
of the suppressionof individual,group,or class interests.Power,for
Foucault,is notsimplya negativeforce,itis productive, and itis in this
assertionthathe breaksfrombothradicaland Marxistviewsofpower.
The productivenatureof poweris closelytiedto Foucault'saccountof
thepower/knowledgerelation.The negativeconceptionofpowerfound
in Marxist and radical writing,where power is seen as repressing,
constraining,distorting,
and so on, implicitlypresupposesthe possi-

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
36 POLITICAL THEORY / FEBRUARY 1983

notmarkedbytheeffects
bilityofsocialrelations ofpower.Foucault
denies this possibility,and he denies it most vehemently
withregard
oftruth:
to theproduction

Truthisn'toutsidepower,or deprivedof power.... Truthis of the world:it is

producedby virtueof multipleconstraints.29
In a societysuchas ours,butbasicallyinanysociety,thereare manifoldrelationsof
power which permeate,characteriseand constitutethe social body, and these
relationsof power cannot themselvesbe established,consolidatednor imple-
mentedwithoutthe accumulationand functioning of a discourse.Therecan be
no possible exerciseof powerwithouta certaineconomyof discoursesof truth
whichoperatethroughand on the basis of thisassociation.We are subjectedto
the productionof truththroughpower and we cannot exercisepower except
through the production of truth. [In our society] power never ceases its
interrogation, its acquisition,its registrationof truth:it institutionalises
and
rewardsits pursuit.
In theend we arejudged,condemned,classified,determined inourundertakings,
destinedto a certainmode of livingor dying,as a functionofthetruediscourses
whichare the bearersof the specificeffectsof power.30

It is, then,onlyat thelevelof discourseand theproductionoftruth

that the strategiesthat integratethe relationsof forceoperate.The
of suchstrategies
intentionality is thusdiscursive,
ratherthanindi-
vidual. Foucault is clearlyissuinga challengeto both radical and
Marxisttheories.AgainstMarxismhe poses theviewthatone cannot
simplyidentify a mode of productionand itsdominantclass and then
deduce fromthiseverything one needsto knowabout theoperationof
powerinthatsociety.Dominationdoes notradiatefromthepeakto the
depths;thisis to be too glib.Rather,we needto see dominationinterms
of a "microphysics" of power:thewayin whichparticularmechanisms
of power, with particularhistoriesand rationales,are colonized,
invested,utilized,and so on, by evermoregeneralmechanisms, which
builtup into formsof global domination.Againstliberalismand the
radicalsFoucault poses theabsenceof an originating subjector actor.
Individualsare theeffect ofpower,theyare itssubjectsand itsvehicles,
not its point of origin.The intentionalityof poweris not individual,
thoughit is articulatedthroughindividuals:

The individualis notto be conceivedofas a sortofelementary

nucleus,a primitive
atom, a multipleand inertmaterialon whichpowercomes to fastenor against
whichithappensto strike,and inso doingsubduesorcrushesindividuals.In fact,it
is alreadyone of theprimeeffects of power thatcertainbodies,certaingestures,

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp/ FOUCAULTON POWER 37

certaindiscourses,certaindesires,come to be identified as
and constituted
individuals....Theindividual
whichpowerhasconstitutedisatthesametimeits
vehicle.
31

Foucault offersus a discursivedeterminism. The power/knowledge

nexus of discourse establishesregimesof truth,which,with their
techniquesand practices,disciplinethe social body. He also suggests
thatthe strategyof power has undergonea mutationin the last two
hundredyears.We beganwitha systembasedon thejuridicalconceptof
power,withthe centralfigureof the sovereign,whichformulated the
dominationof the sovereignin termsof his legitimaterightsand the
legal obligationto obey. Butthelanguageofsovereignty thatwe retain
has obscuredthe fact that sovereignty has givenrise to relationsof
dominationthatare inadequatelycharacterizedand accountedforby
thejuridical model. Althoughwe stillconceiveof power in termsof
sovereignty ("In politicalthoughtand analysis,we stillhavenotcutoff
the head of the king.")32the operation of power has changed and
assumeda disciplinary formwithtwo poles: One

centered on the bodyas a machine;its disciplining,

the optimalising
of its
ofitsforces,
theextortion
capacities, theparallelincrease
ofitsusefulness
andits
intosystems
itsintegration
docility, ofefficient
andeconomic all this
controls,
was ensuredbytheprocedures of powerthatcharacterised thedisciplines:
an
ofthehumanbody.33
anatamo-politics

The other,

focusedon thespeciesbody,thebodyimbuedwiththemechanics of lifeand

as thebasisofthebiological
serving processes:
propogation,birthsandmortality,
thelevelofhealth,
lifeexpectancy withalltheconditions
andlongevity, thatcan
causetheseto vary.Theirsupervision
was effected
through an entireseriesof
and regulatorycontrols:a biopoliticsofpopulation.34
interventions

It is this discursivedeterminism, the view of discoursewith its

techniquesand practicessanctionedby knowledgeclaims and con-
ductedwithinlocalfoyersof power-the consultingroom,theschool-
room, the courtroom,the sick bed, the child's body, the barracks,
and so on-that allows us to viewitspractitioners as conforming to a
strategyofdomination"yetwithouthypocrisy."35 It is also thisviewthat
enables Foucault to offeran accountof poweras significant affecting
while omittingreference to interests,
intention,or counterfactuals. In
thefollowingsection,I shall suggestthatthe omissionof thesefactors

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
38 POLITICAL THEORY / FEBRUARY 1983

does createproblemsforFoucault thatremainunresolved.I shall do

thisthrougha discussionof his conceptionofthepolitical,hisaccount
of repression,and his analysisof resistance.

III

At thetimeof writingthe firstvolumeof his Historyof Sexuality,

Foucaultproposeda conceptionofthepoliticalthatappearsin a modi-
fiedformin thatbook. In a lecturedeliveredat thebeginningof 1976,
he suggeststhat we mightconceive of politicsas war continuedby
othermeans.This

impliesthattherelationsofpowerthatfunction ina societysuchas oursessentially

rest upon a definiterelationof forces that is establishedat a determinate,
historicallyspecifiablemoment,in war and by war.... If it is truethatpolitical
powerputsan endto war. . . thisbyno meansimpliesthatitsuspendstheeffects of
war or neutralisesthe disequilibriumrevealedin the final battle.The role of
political power, on this hypothesis,is perpetuallyto re-inscribethis relation
througha formof unspokenwarfare:to re-inscribeit in social institutions, in
economicinequalities,in language,in thebodiesthemselves ofeach and everyone
of us.36

Furthermore, itimpliesthatall politicalstruggles

shouldbe regardedas
a continuationof war,and thattheend resultcan onlybe a finalbattle.
Foucault admits to having used this war-repression schema in his
previousworks,37 butat theend ofthelecturehe also admitsthathehas
begunto see bothconceptsas inadequateand saysthathe suspectsthat
theymayneedto be abandoned.Butthereis no further discussionofthe
question.Althoughit is unclearwhetherhe gave thelecturebeforeor
afterwritingthe firstvolume of his History,it seems likelythat it
predatesthe book. In thatvolumehe referstentatively to theinversion
of theClausewitzianformula.Aftersuggesting thatpoweris composed
of relationsof forceand refersto a complexstrategicsituation,he asks
whetheror notwe shouldsuggestthatpoliticsis warcontinuedbyother
means. He replies:

If we stillwish to maintaina separationbetweenwar and politics,perhapswe

should postulatethatthismultiplicityofforcerelationscan be coded-in partbut
nevertotally-eitherin theformof"war" or in theformof "politics";thiswould
implytwo different strategies(but one alwaysliableto switchintotheother)for
integratingtheseunbalanced,heterogeneous, unstable,and tenseforcerelations.38

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 39

It seems,then,thatFoucault has modifiedtheinversionof Clause-

witz,but that he has done so in such a way as to suggestthat both
politicsand war are methodsfor integrating relationsof forceinto
Yet at theendofthesamechapter,Foucaultseemsto
coherentstrategies.
have returnedto the originalClausewitzianformula:

It is one oftheessentialtraitsofWesternsocietiesthattheforcerelationships
which
fora long timehad foundexpressionin war,in everyformof warfare,gradually
became investedin the orderof politicalpower.39

I remainperplexedabout what one is supposedto do withsuch a

statement, forthefollowingreasons.It seemsas ifFoucaultis offering
us an accountof theexistenceof forcerelationsin termsofsomebasic
humandrive.He can onlyavoid doingso ifheregardsWesternsocieties
as unique (a) in theirpredilictionforwar,and (b) in thattheytransfer
thisprediliction intopolitics.Evidently,thesavinggraceof Foucault's
originalformulation, at leastforthe Marxist,is theviewthatclasswar
can end in a finalstruggle.In thereformulationitlooksquiteclearlyas if
forcerelationsare hereto stay,thattheyare endemicto humansocie-
ties. Both politicsand war are different waysof organizingthisforce.
But what Foucault now seemsto denyis the possibilityof liberation,
and, as we shall see, thisdoes createdifficultiesforhim.
Foucault's reformulationof his conception of the political is
associatedwithhisattitudetowardrepression.Withinbothliberaland
Marxistconceptsofpowerthereis roomfora theoryofrepression since
both conceivepower in termsof A (some individual,groupor class)
affecting B's (individual,groupsor class) interests.
B's interests
maybe
seen as repressed:somethingthat cannot be fulfilledbecause of A's
exerciseof power. Such repressionmay be conceivedof in a variety
of ways-psychologicallyor psychoanalytically, linguistically(as in
Habermas'snotionof systematically distortedcommunication), physi-
cal, or social-and thesevarious formsmay be combinedin a num-
ber of ways. Thus some Marxistshave seen economic exploitation
as effectiveat theleveloftherepressionofdesire.40Furthermore, at least
withinthe Marxistperspective,the repressionis seen as being,in an
importantrespect,temporary-in that the objectivenatureof class
interestswilleventuallycondense,througha systemsimperative, intoa
practicalpoliticalstruggle.Liberaltheoriststendto ignorerepression.
Conflictsof interestare seen in termsof conflictsin policy prefer-
ences,whichare settledthroughestablishedproceduresthe resultsof
which participantsaccept as legitimate.Policy preferences are thus

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
40 POLITICAL THEORY / FEBRUARY 1983

subordinated to the higherinterestin maintaining thedecisionpro-

cedure.A broaderliberalview,suchas thetwo-dimensional viewthat
Lukesidentifies, mayhave roomto recognize thata decisionpro-
ceduremayexcludecertainpotentialconflicts of interestfromthe
policyagenda,andthatthismayleadto dissatisfaction withorprotest
againstthatprocedure. The moreradicalthree-dimensional viewof
powerdoes havemoreroomfora conception of repressionsinceit
acceptsthe categoriesof latentconflictand objectiveinterests-
objectiveinterestsbeingdefinedas whatindividuals wouldwantifthey
wereunconstrained intheirchoices.
Foucault'sattitude towardrepressionhasto be seenintermsofhis
stresson thepositivefunction ofpowerandhisassociatedviewofthe
as a discursive
"6subject" product.Ifthesubjectisa productofdiscourse
itcannotbeclaimedthatheorshehasanyobjective interests.
Therecan
be no positionof exteriority forsubjectsfromdiscourse(to use
Foucault'sratherawkwardformulation) and his rejectionof the
determining effectof theeconomicstructure or "theinternal social
dynamicsof societiesand theirmodesof production"4' leavesno
ofrecourse
possibility to anynotionofobjective interests.
Giventhis,
therecanbenorealconception ofrepression.
Consequently, Foucaultis
trueto formwhenhe rejectstheconcept:

I believe that the notion of repressionremainsa juridical-disciplinarynotion

whateverthecriticaluse one wouldmakeof it.To thisextentthecriticalapplication
of thenotionof repressionis foundto be vitiatedand nullifiedfromtheoutsetby
the two-foldjuridicaland disciplinaryreferenceitcontainsto sovereigntyon the
one hand and to normalisationon the other.

If we acceptFoucault'sconception of powerand discourse as the

etherinwhichsubjectivity isformed,andifwehaveonlya viewofpower
as significant
affecting, arewe facedwitha viewofhumansocietyas
operating according to discursive
lawsand discursive
rationalities-a
viewofdiscourse as a hermetically
sealedunit;a domination
thatcannot
be escaped?It would seem that Foucaulthas produceda major
difficulty.
Consider, forexample:

It seemsto methatpoweris"alwaysalreadythere,thatone is neveroutsideit,that

thereare no marginsforthosewho breakwiththe systemto gambol in.43

But,iftherearenomargins,
isitpossible
tobreakfromthesystem
atall?
We seemto have a dead end. Foucaultoffersus threeconflicting

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 41

solutions.The firstis theleastsatisfactory sinceit relieson material

publishedaroundthetimehefirst useda conceptofpower.44 As suchit
is fullof references to elements thathe now rejects:He refersto
something liketherepressed45 and to theexistence of placesoutside
power;46he talks about those people upon whompoweris "ex-
and aboutindividuals
ercised"47 in a struggle thatconcerns theirown
interests;48he acceptsthe proletarian struggleas a revolutionary
movement49 andhetalksofa discourse thatis againstpower,50andof
thewaythatpoweris alwaysexercised at theexpenseofthepeople;5'
and,finally, he refersto poweras setting up a system of blocksand
prohibitions.52All thesereferences sit uncomfortably withhislater,
moredelphic, utterances.
In particular,hisclaimsaboutthestruggle for
liberation
suggest thatthisworkrestson a muchmore"naive"viewof
power-theproductive function ofpower,theintricate power/ knowl-
edgerelation, therejectionofan originating subject,andtheassociated
conceptoftherepressed, areall absent:

If thefightis directedagainstpower,thenall thoseon whompoweris exercisedto

theirdetriment, all who findit intolerable,can beginthe struggleon theirown
terrain.... In engagingin a strugglethatconcernstheirown interests, whose
objectivestheyclearlyunderstandand whose methodsonlytheycan determine,
theyentertherevolutionary process.... Women,prisoners, conscriptedsoldiers,
hospitalpatients,and homosexualshavenow beguna specificstruggleagainstthe
particularisedpower,the constraintsand controls,thatare exertedover them.
Such struggles ... are radical,uncompromising and non-reformist,
and refuseany
attemptat arriving at a newdispositionofthesamepowerwith,at best,a changeof
masters.53

The secondsolutionto theproblemof breaking

fromthesystem
directly thestatement
contradicts quotedaboveabouttherebeing"no
margins forthosewhobreakwiththesystem to gambolin"-and it
appearsthreepagesbeforethatstatement!

There is indeed always somethingin the social body, in classes, groups and
individualswhichinsomesenseescapestherelationsofpower,something whichis
byno meansa moreorlessdocileor reactiveprimalmatter, butrathera centrifugal
movement, an inverseenergy,a discharge.Thereis ... a certainplebianqualityor
aspect.... Thereis [sic] plebsin bodies,in souls,in individuals,intheproletariat,
in the bourgeoisie,but everywherein a diversityof formsand extensions,or
energiesand irreducibilities.54

withPoulantzas'soutrageat thismove.55
One sympathizes The
of"plebs"seemsat bestarbitrary,
introduction at worstitcontradicts

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
42 POLITICAL THEORY / FEBRUARY 1983

Foucault's whole conceptionof power as constitutive of subjectivity

(and its associatedclaim that"thereis no positionof exteriority
from
discourse"). If we ignore for the moment what seems to be an
uncharacteristic move(and mostFoucaultiansdo), we are leftwithhis
account of resistanceas the only way out of the hermetically sealed
discursiveworld:
Thereare no relationsofpowerwithoutresistances; thelatterare all themorereal
and effective becausetheyare formedrightat the pointwhererelationsof power
are exercised;resistanceto powerdoes nothaveto comefromsomewhere elseto be
real,noris itinexorablyfrustratedthroughbeingthecompatriotofpower.Itexists
all themorebybeingin thesame place as power;hence,likepower,resistances are
multipleand can be integratedintoglobal strategies.56

Poulantzasis also extremelyirritatedbythismove;he suggeststhat

theresistances assertion"havingno foundation,
are a "gratuitous "a pure
affirmationof principle":

For if poweris alreadythere,ifeverypowersituationis immanentin itself,why

should thereever be resistance?From wherewould resistancecome, and how
would it even be possible?57

Poulantzas regards Foucault as having no basis for evading the

conceptualtrap of a dominationthatcannot be eluded. If everything
takes place within power-even this resistance-then there is no
breakingfreefrompower.Foucault'sreferences to the"plebs"are seen
as a movemade in desperation.However,muchas wemightsympathize
withPoulantzas'sire,sinceFoucault'sargumentis at bestobscure,the
situationis not quite the way he sees it. That it is not does not,I will
suggest,help Foucault much,but it is important to criticizehimat the
rightpoints.
First,letme clarify thenatureoftheargument. Thereseemto be two
points at issue: (I) If Foucault is simplyoffering us an explanatory
accountofthefunctioning ofhumansociety,hehas failedto indicatethe
criteriabywhichrelationsofforceare to be identified and he has failed
to explainwhyresistanceis generatedbytheoperationofpower;(2) ifhe
is not simplyofferingan objectivistaccount/explanationof social
order,itwouldseemthathe requiressome basisforjustifying, as wellas
explaining,resistance-if we do not resistour disciplinarysociety,
should we?
Foucault has, then,to answerat least the explanatoryquestion;in
addition,ifheisto retainhisradicallaurels,he mustanswerthequestion

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 43

To tryto show how farhe does eitherofthesethingsI

ofjustification.
willofferan accountof Foucault's"theory"of resistancebylookingat
fourwaysin whichhecouldgroundeitheran explanatory orjustificatory
account.I willsuggestthatthereare elementsoftheformerbutnotthe
latter,and I will discussthe implicationsof thisabsence.
One way to avoid Poulantzas'sclaim thatresistanceis a gratuitous
assertionagainst the ubiquityof power is to stressthat power is
predicatedupon forcerelations-"everyrelationofforceimpliesat the
same timea relationof power"59-and;

Machiavelliwas among the few... who conceivedof thepowerofthe Princein

termsof forcerelationships.... Perhapswe needto go one stepfurther
... and
decipherpowermechanismson the basis of a strategythatis immanentin force
59
relationships.

Stressingthis,we mightthen claim that forceand resistanceare

related analytically(that to force A to x is just to overcomeA's
resistanceto x) and thus that, since power is predicatedon force
relations,resistanceis alwaysevidentat pointswherepoweris. Butthis
onlymovestheprocessfurther backto theill-defined
terms"force"and
"resistance."However,althoughFoucault would probablynot accept
thisaccount,thereis one thatis similarthathemightwellrecognize.We
mightsuggestthatresistanceis the"other"ofpower-just as sicknessis
the"other"of health,madnessis the"other"ofreason,and devianceis
the "other"of normality.In the Historyof Sexuality,forexample,
Foucaultshowshowsexualityitselfis producedthroughthetechniques
ofdiscourse;and in Disciplineand Punishheshowsina similarwayhow
thebodyof theprisoneris subjectedto rigorousdiscursivecontrol-in
boththe individualis made a subjectby his or hersexual or criminal
identity.In bothcases, however,(and herethe"plebs"findan echo) we
can see Foucault as positinga prediscursive unformedprimalbodily
matter,whichis workedup in discourseintoa discursive subject.In this
workingup an "other"is created-in creatinga subject,thoseelements
of this primalmatterthat cannot be incorporatedin the discursive
identityare subjugated.A biopoliticsof healthproducesdisease; a
discourseof fidelityproducesof itselfits subjugatedother,infidelity;
and as we search for normalsexualitywe uncoveronly ever more
perversion.In thissense,an "other"is alwaysinscribedin discourse-it
is an inescapableconceptualconditionof possibility.
But, it remainsunclearhow farthepower-resistance relationreally
parallelsthisprocess.First,is the "other"of a discoursenecessarilya

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
44 POLITICAL THEORY / FEBRUARY 1983

resistant"other"?Are we to assume that mere"otherness"simplyis

resistance?Second, is resistancethe "other"of power or of force?If
poweris to be understoodin termsofchainsor relaysofforcerelations
organizedthroughdiscourse,then,althoughthe"shape" ofthe"other"
is determined bythe"shape"ofthediscourse,theexistenceofa resisting
otheris actuallya functionoftheseforcerelations.Thus theaccountof
resistanceas the"other"of powerhas no explanatoryforcebecauseno
accountis givenof itsoriginsin thebasic unitofpower-namely,force
relations.Furthermore, it seemsthatifFoucaultis positingthisprimal
matterand suggesting thatitwillalwaysprovidean "other"to discourse,
thenhe is no closerto justifying resistance,sincethereseemsno good
reason whythe "other"should be welcomedas a liberator.Afterall,
each liberationinvolvescreatinga new"other"(unless,ofcourse,there
is sometrue,repressed andtherecannotbeanyextradiscursive
subjectivity)
grounds for assessing the respectiveclaims of subjects and their
"other."In morepracticalterms:Ifde Sade is thesubjugated"other"of
normalizingconceptionsofsexuality,thenwho willbe the"other"ifde
Sade triumphs, therebyproducinga newsexuality withitsowntyranny-
its own subjugated"other"?And, above all, have we any groundsfor
thinkingit mightbe preferable? Indeed, can we have transdiscursive
preferences? Althoughtheconceptofthe"other"is suggestive, itseems
insufficiently specifiedto provide us with an adequate explanatory
accountof resistance,and itseemsto havelittleto offertheproblemof
justifying resistance.
A secondaccountofresistancewouldsee B as resisting becauseofthe
conflictbetweenA's demandsand B's discursively constituted identity
and its associatedpatternof interests. But Foucault seemsto discount
interestsand individualwills.60 Furthermore, thereis a riskofcircularity
sinceit is difficult
to see howdiscursively constituted willscan formthe
basis forresistanceto discourse.Certainly,therecould be no basis for
sayingthatB shouldresist,sinceB would simplybe swappingone dis-
cursiveidentityforanotherand thereseemsno basis forhavingpref-
erencesbetweendiscursiveidentities.
A thirdpossibleaccountofresistancewouldrefertoequality:"Power
is exercisedfrominnumerable points,intheinterplay ofnon-egalitarian
and mobilerelations."6'All powerrelationsare relationsof inequality
thatarise in "economicprocesses,knowledgerelationships, sexual re-
lations . . . in families,limitedgroups and institutions."Such relations,
as faras we can makesenseof Foucault'saccount,involverelationsof

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 45

force.The problemthatarisesis essentiallyone ofwhetherrelationsof

forcearisein relationsofinequality, orwhether theyarethebasisforthe
constitution and continuationofsuchinequalities.If Foucaultgoesfor
the formeroption,basingrelationsof forceon existinginequalitiesin
material,sexual,economic,and knowledgerelations,itwouldseemthat
he mightbe able to groundresistancein somethinglike an interestin
equalityorjustice-as in resistanceto exploitationon thebasisthatitis
unjust. But such a notionis inordinately problematicalforsomeone
withFoucault'sviewofknowledge.It seemsthatFoucaultwouldinfact
see inequalitiesas beingproducedand maintainedby force.But this
means thatwe stillhave no account of resistanceunless Foucault is
arguingthatforce,and thuscounterforce, is an endemicfactofhuman
lifein society.Thereare twounderlying problemsforFoucaultwiththis
formulation: One is thatit entailsa commitment to theexistenceof a
universalcharacteristic of humannatureor humansociety,one that
preexistsdiscursiverelations,and thissitsill withhis nominalistand
relativistpretensions;thesecondis thatthiscan neverprovidethebasis
of
fora justification resistance-likeHobbes's warofall againstall,the
accountis completelynaturalistic.
For a finalpossibleanswerto thequestionofresistancesin Foucault
we can referto his claim that
poweris tolerableonlyon conditionthatit mask a substantialpartof itself.Its
successis proportionalto itsabilityto hideitsown mechanisms.Would powerbe
acceptableifitwereentirely cynical?For it,secrecyis notinthenatureofan abuse;
it is indispensableto itsoperation.... Would[thoseitdominates]acceptitifthey
did not see it as a merelimitplaced on desire,leavinga measureof freedom-
howeverslight-intact?62

The factthatthejuridico-politicalconceptionof poweras exercised

in law remainsdominantin an era when disciplinarypower has far
exceeded it in scope and effectis seen by Foucault as a part of the
necessarymaskingof theeffects ofpower.Ifwe saw thedisciplinesand
theirmultipleregimesof truthas merelythe operationof forceand
powerwe would resist-as prisonershave resisted-at thelevelof the
body.63This suggeststhat resistanceis an inevitableresponseto the
experienceofthe materiality ofthemechanismsofpower.Coercion,in
itsmanifest form,is necessarily
productiveofresistance.Buttheaura of
doom thatPoulantzas sees as associatedwithsuch a viewneedsto be
questioned.My resistanceto yourcoercionrendersyourpoweroverme
unstable.It maybe stabilizedbylinkingitto a netwokofotherrelations

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
46 POLITICAL THEORY / FEBRUARY 1983

offorceto forma patternofstrategic domination,but,equally,thismay

be overthrown bylinkingpointsofresistanceand breakingthechainof
effective coercion.This is surelywhatFoucaultmeansby"shifting and
mobilerelations."Relationsofforceneedto be integrated intostrategies
to be effective in maintainingcoercion.
But evenin thisformulation Foucault is stillnottellingus anything
about whypeople resistthe materiality of power;we are back to the
assertionthattheywillresistit,butwiththeadded rider,"particularly if
it is visible."Furthermore, I failto see thatFoucaulthas done anything
to take us away fromthe conceptionof politicsas war continuedby
othermeans,and thisis a naturalisticaccount. Much of Poulantzas's
critiqueremainsvalid in thatFoucault has notyetprovidedus withan
accountof whypeople resist,nor has he shownwhytheyshouldresist.
But, contraPoulantzas,the problemlies less withthe conceptionof
powerthanwithitsrootdoctrineofforce.Nevertheless, we areleftwith
a viewof resistance as an incoherent and libidinalresponse-aneruption
entwinedand reworkedwithinthediscourseagainstwhichiterupts.To
resolvethis problem,Foucault must eitherabandon all claims to a
practical political orientation,or he must invoke a philosophical
anthropology, whichcan thenbe used as a basis fortheformationof
coherentpoliticalgoals.
Foucaultdoes nottakethefirst option,althoughhedoes occasionally
implya strongskepticismto politicalgoals. At the end of the first
volumeofhisHistoryofSexuality,hesuggeststhatone day"perhaps,in
a different economyof bodies and pleasures"people will no longer
understandor shareour currentobsessionwithsexuality-"ouraustere
monarchyof sex," whichentailsan endlesssearchforits secret,and
whichironicallyinvolvesus in believingthat"our 'liberation'is in the
balance."64But, if our liberationis not to be foundin a scrupulous
examinationof sex, if thatexaminationis indeeda conditionof our
imprisonment, we still retain some conception of liberation,and
Foucault seems to believethatwe should do so. His genealogiesare
critiques;theirverylanguageimpliesthatthereis some alternative:

Knowledgedoes not slowlydetachitselffromitsempiricalroots,theinitialneeds

fromwhichit arose, to becomepurespeculationsubjectonlyto thedemandsof
reason; its developmentis not tied to the constitution and affirmation
of a free
subject; ratherit createsa progressiveenslavementto its instinctiveviolence.
Wherereligiononce demandedthe sacrificeof bodies, knowledgenow calls for
experimentation on ourselves,callsus tothesacrificeofthesubjectofknowledge.65

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 47

Foucault'sstudiesofthisprogressive enslavement to thewillto truth

owe much to Nietzsche. Both show how the thingswe trustin,
civilization,science,truth,welfare,and a Christianconcernforour
fellowman or woman,have theirrootsin theiroppositesand continue
to dependon theseoppositesfortheirsupport.Yet Foucaultcan only
accuse thewillto truthof tyranny if,counterfactually, thereis another
possibleset of statesthatcould have been,or could be, broughtabout
withoutthis tyranny.Withoutthis,"tyranny"becomessimplymeta-
phorical,and is as usefulforpoliticalthinkingas thinking oftheEnglish
weatheras tyrannical. Butthemostcommoncandidatesupon whichto
base some conceptionof alternativepossibilitiesare ones thatinvolve
some reference to nature,the will,or to interests.Yet:

In generaltermsI believethatpoweris not builtup out of 'wills'(individualor

collective)nor is it derivablefrominterests.66

And if poweris not builtfromwillsor interests, and resistancesstem

frompower,thenresistancecannotbe said to be on thebasis of these
factors either. Yet Foucault does seem to have some alternative
conception-how else could he write:"The rallyingpoint for the
counter-attack againstthedeploymentof sexualityoughtnotto be sex
desire,but bodies and pleasures."67
In summary, on thequestionofresistance,
we can seethatFoucault's
methodologymakes sense largelyin termsof Nietzsche'sconceptof
genealogy:It is theundercuttingof theknowledgeclaimsofthesocial,
medical,and paramedicalsciencesby referenceto theirobscureand
frequentlyiniquitous origins,and by revealingthe normativeand
delusivenatureof the truthsthat theyproduce. The polemicaland
politicalpurposesof such an analysiscan also be understood:It is an
attemptto break up areas of congealed knowledgeby a careful
delineationof thestructuralcrackswithinthem,and bypullingtherug
of epistemologicalvalidityfromundertheirfeet.Nietzschemighthave
seen himselfas givinga helpfulpush to the crumblingbuildingsof
But to claima morepracticalpoliticalmotivation,
Christianity. and to
be something morethana nihilistic thegenealogist
anarchist, mustsurely
have some conceptionofthestepsto be takenonce thedusthas settled.
There mustbe a conceptionof new horizons,somethingthatcould be
generated,forexample,by recourseto Nietzsche'swillto power.68 But
Foucault rejects this aspect of Nietzsche,and he also rejectsthe

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
48 POLITICAL THEORY / FEBRUARY 1983

possibility
ofa society
freedfromtheeffects
ofpowerandforce. Butifhe
does both thesethingscan he offerus any reasonforrejecting
power,andcanhesuggest
disciplinary whyitsremoval wouldnotresult
ina stillmorecrushingtyranny?Theproblem Foucaultfacesisthathis
accountofresistance and forcehas becomea naturalistic
one;conse-
quently, we mayhelpnaturealong,but,thusfar,he has givenus no
indicationwhereit mightlead us. Certainlyhe givesno reasonsfor
believingthatthereisanythingbetter
(inanysenseoftheword)forusin
thefuture.Andallthismakeshisstatusas a radicalsomewhatquestion-
able.

IV

In his discussionof power,Lukesarguesthat"talkof interests

providesa licenceforthemakingofnormative
judgementsofa moral
and politicalcharacter. . . different
conceptionsof whatinterests
are
are associatedwithdifferentmoraland politicalpositions."69 Lukes's
doubtsaboutthevalueofPoulantzas's schemaforpowerareassociated
withthisviewof the essentially contestable character of the term
"power."He suggests thatPoulantzasoffers us an accountofstructural
determinism-in C. WrightMills'slanguage, "'fate"-not an accountof
power,becauseitprecludes seeingan exerciseofpoweras associated
withchoice,andthuswithresponsibility. To dissociatepowerfrom the
languageofchoiceand responsibility is to offera naturalistic
descrip-
tionofpoliticaleventsthatprecludes us fromseeingitas a spherefor
humanaction.Whileitistruethatstructural Marxism altogether denies
thatit makesnormative judgments, and also rulesoutan analysisof
humanaction,itisgrossly stipulative
toclaimthattheyarenotoffering
us a conceptofpowersincetheydo acceptthekeyfactors ofsignificant
affectingandinterests.
AsGrayhasshown, powerisa theory-dependent
term,70andthemeta-theoretical
grounds forarbitratingbetween theoreti-
cal paradigmsremainsufficiently complexand obscureto deter-
minetheory choiceandto render theexclusion ofa paradigm fromthe
debate,as Lukesdoeswithstructural Marxism, at beststipulative.
The
samecanbesaidtoapplytoanyattempt toruleoutFoucault'sconcept
ofpower.
Havingsaid this,however,we can go further, withGray,and
recognizethatrivalparadigmsinvokerivalphilosophical
anthropologies,

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp/ FOUCAULTON POWER 49

in whichaccountsofhumannatureand actionare given.In effect, rival

paradigmsinvolvedifferent commitments to philosophicalpositionsin
thephilosophiesof mind,language,action,and so on. Gray'scriticism
of structuralMarxismpointsout thatit suppressesitsanthropological
assumptions,but that the commitments that it would have to have
would underminethe determinism of the approach. The argumentis
complex,butwe neednotpursueitforourpresentpurposes.The points
thatGray makesthatare of relevanceto our evaluationof Foucault's
workare thataccountsofpowerand social structure involvea rangeof
metaphysical commitments, thatthereis roomforcriticism wherethere
is a hiatusbetweenthosecommitments and theaccountdeveloped,and
thatalthoughempiricalevidenceand debateinthephilosophyofmind
underdetermine suchcommitments, theynevertheless limittherangeof
commitments we can make.
My accountof Foucault's conceptionof powerhas suggestedthat,
althoughhe maintainswhatappearsto be a viewofsignificant affecting,
he avoids thequestionof conflictsof interests.He substitutes forthis
questionthe claim that poweris productiveof resistance.This is not
necessarilyincoherent.However,he does radicallyunderspecify such
termsas "force,"whichleaves us feelingthathis claimthatresistance
occurs is simplyan assertion,unbackedby argumentand lackingany
explanatoryaccountto back it up. But even ifwe wereto acceptthat
Foucault's account of forceand resistancecould be formulatedin an
explanatoryform,wearestillleftwiththequestionofjustification. Here
hisviewbreaksdown,and itbreaksdownlessintoincoherence thaninto
an impenetrable silence.He refusesto elaboratetheanthropology that
evidently does underliehiswork.Counterfactually, ifhe holdsno such
anthropology,thenhe is offering us a naturalisticdescriptionof the
operationof powerthatcannotformthebasis fora politicallyrelevant
critiqueofpractice.Ifwe are rightto see Foucaultas seeinghimself as a
radical,we mustrecognizethathe remainssilentand elusiveabout the
philosophicalpositionfromwhichhe is working.
Thereis one wayin whichFoucaultmightclaimto be able to remain
both silent and politicallyradical. He might wish to argue that
discussionsof powerand of"Man," exceptwheretheyare genealogical
in form,contribute to thewillto truth,and as sucharefurther instances
of the dominationby thatwillof Westerncivilization.But thissimply
postponesthe real question. If power is ubiquitous,if it comes from
everywhere, if forceis everywhere, if,thatis, Foucault has rejecteda
view of a societywithoutconflict,thenare all societiesequallyvalid?

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
50 POLITICAL THEORY / FEBRUARY 1983

The repudiationof the will to truthseems to entaila repudiationof

evaluation:We willactfromresistanceand revengebutwe can onlyever
installa new regimeof powerand a new tyrannyof truth.Foucault's
resistanceto representation-"theindignity of speakingforothers"
and his recognitionthatknowledgeis investedwiththe techniquesof
powerleaves himsilentabout thefuture.All he can offeris a tolerance
fortheviolenceofthepeople'sjustice7'and a fewhintsabouttheneedto
change.72Buthe can onlyoffer us a critiqueofourdisciplinarysocietyif
thereis some preferableform,and whilehe clearlydoes notacceptthat
we can eliminateforce,power,and conflict, he does seemto believethat
thereare waysofdealingwiththesethatwe shouldprefer, and anysuch
conceptionmustreston a particular seriesofphilosophical
commitments.
Refusingto statesuchcommitments, refusingto talkofthefuture, and
refusing to offerforour reflectionpossibilitiesforthefutureleaveshis
work enigmaticand only suggestive.It also allows himto elude the
debatingchamberon powerand itsnature,togetherwithdebatesin the
philosophiesof mind actionand language,whichsupportthe former
debate. But in eludingthesehe eludes also the possibilityof rational
discoursebetweentheoretical frameworks: One eitherspeaksFoucault's
language,or one is condemnedbyit. Recognizingthisshouldhelpus to
recognizethatifthereare problemsin understanding Foucault,these
arisenotbecauseofthelanguageinwhichFoucaultwrites,butbecause
ofwhathe willnotsay.Thatis,itisthereluctanceoftheanthropological
subjectto talkaboutthekeyconcernsofhisculturethatmakeshiscon-
ceptionof the worlddifficult to grasp.

NOTES

1. Michel Foucault, Disciplineand Punish(London: Allen Lane, 1977).

2. Michel Foucault,ed. 1, PierreRiviere(Harmondsworth:Penguin,1978).
3. Michel Foucault, The Historyof Sexuality: Vol. 1 An Introduction(London:
Allen Lane, 1979).
4. Michel Foucault, The Orderof Things(London: Tavistock,1970).
5. Michel Foucault, The Birthof the Clinic (London: Tavistock, 1973); idem,
Madnessand Civilisation(London: Tavistock,1967).Furtherreflections on thelastthree
mentionedworksareto be foundin Foucault'smethodological treatise,TheArchaeology
of Knowledge(London: Tavistock,1973).Thereare threecollectionsof Foucault'sessays:
Language,Counter-Memory, Practice,ed. D. F. Boucahard(Oxford:Blackwell,1977);
Michel Foucault: Power, Truth,Strategy,ed. M. Morris and P. Patton (Sydney,
Australia: Feral, 1979); and Michel Foucault: PowerlKnowledge,ed. C. Gordon

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
 Philp / FOUCAULT ON POWER 51

(Brighton: Harvester, 1980). Furtheressays by Foucault, on Foucault, and by

Foucaultianscan be foundin Ideologyand Consciousness,nos. 4, 5, 6 and 7. See also J.
Donzelot, The Policingof Families(London: Hutchinson,1979); thisis a major work
withinFoucault's methodologicaltradition.
6. This developmentis firstindicatedin Michel Foucault, L'ordre du Discours
(Paris: Gallimard,1971);("Ordersof Discourse,"Social ScienceInformation10,2[19711,
pp. 7-30). Foucault discussesthe changein PowerlKnowledge,ch. 6.
7. The most accessiblepieces,to mymind,in descendingorderare: H. V. White,
"Michel Foucault," in J. Sturrock,ed., Structuralism and Since (Oxford: University
Press, 1979); J. Minson,"StrategiesforSocialists?Foucault's Conceptionsof Power,"
Economyand Society9, 1 (1980),pp. 1-43;C. Gordon,"Afterword," in Power/Knowledge.
8. Foucault, PowerlKnowledge,p. 79.
9. S. Lukes, Power:A Radical View(London: Macmillan,1974).
10. N. Poulantzas, Political Power and Social Classes (London: New LeftBooks,
1973); idem,State, Power,Socialism(London: New LeftBooks, 1979).
11. Lukes, Power,p. 28.
12. H. Arendt,On Violence(London: Allen Lane, 1970),Part 2.
13. Foucault, Power/Knowledge, p. 88.
14. Lukes, Power,p. 31.
15. Ibid., pp. 34-35.My referencesto Lukesdo nottake intoaccounttheverymuch
broaderaccountof powerhe givesin his paper"Powerand Authority," in T. Bottomore
and R. Nisbet,eds., A Historyof SociologicalAnalysis(London: Heinemann,1979).In
thispaper he allowsa lesserstresson intention.See W. Connolly,The Termsof Political
Discourse(Lexington,MA: D. C. Heath, 1974),ch. 3.
16. Lukes, Power,54-55.
17. Poulantzas,State, Power,Socialism,p. 147.
18. Foucault, Historyof Sexuality,p. 92.
19. Ibid., p. 92.
20. A. Sheridan, Michel Foucault: The Will to Knowledge(London: Tavistock,
1980),p. 183. See Michel Foucault, La Volontede Savior (Paris: Gallimard,1976),pp.
121-129.
21. Foucault, Historyof Sexualityl,pp. 92-93.
22. lbid., p. 93.
23. Ibid., p. 94.
24. F. Nietzsche,The Willto Power,Kaufmannand Hollingdale,trans.(New York:
Vintage,1968),sections634-660.
25. Ibid., p. 94.
26. Ibid.
27. Ibid., pp. 94-95.
28. Ibid., pp. 95-96(thisparaphrasesFoucault'sargument).
29. Foucault, Power/Knowledge, p. 131.
30. Ibid., p. 93.
31. Ibid., p. 98; emphasisadded.
32. Foucault, Historyof Sexuality,pp. 88-89.
33. Ibid., p. 139.
34. Ibid.
35. Ibid., p. 95.
36. Foucault, PowerlKnowledge,p. 90

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions
52 POLITICAL THEORY / FEBRUARY 1983

37. Foucault,Disciplineand Punish.

38. Foucault,History ofSexuality,p. 93
39. Ibid.,p. 102.
40. Forexample,H. Marcuse,Erosand Civilisation (London:Sphere,1969).
41. G. Therborn, TheIdeologyof PowerandthePowerofIdeology (London:New
LeftBooks,1980),p. 42.
42. Foucault,Power/Knowledge, p. 108.
43. Ibid.,p. 141.
44. Foucault,Language, Counter-Memory, Practice,pp. 205-217.
45. Ibid.,p. 207.
46. Ibid.,p. 211;"theanti-judicial is a struggle
struggle againstpower.
47. Ibid.,p. 216.
48. Ibid.
49. Ibid.
50. lbid.,p. 209.
51. Ibid.,p. 211.
52. Ibid.,p. 207.
53. Ibid.,p. 216.
54.Foucault,Power/lKnowledge, pp. 137-138.Foucaultalsomentionsthe"plebs"in
Practice,p. 158,but in a verydifferent
Language,Counter-Memory, way.
55. Poulantzas, State,Power,Socialism.pp. 150-151.
56. Foucault,Power/Knowledge, p. 142.
57. Poulantzas, State,Power,Socialism, pp. 148-149.
58. Foucault,Power/Knowledge, p. 185.
59. Foucault,History ofSexuality,
p. 97.
60. Foucault,Power/Knowledge, p. 188.
61. Foucault,History ofSexuality,
p. 94.
62. Ibid.,p. 86.
63. Foucault,Discipline and Punish,p. 30.
64. Foucault,History ofSexuality,
p. 159.
65. Foucault,Language, Counter-Memory, Practice,
p. 163.
66. Foucault,Power/Knowledge, p. 188.
67. Foucault,History p. 157.
ofSexuality,
68.SeeA.C. Danto,Nietzsche as Philosopher(NewYork:Columbia University Press,
1965),chs.5, 8.
69. Lukes,Power,p. 34.
70.J.Gray,"Political Power,SocialTheory andEssential inD. Miller
Contestability,"
andL. Siedentop, eds.,TheNature ofPoliticalTheory(Oxford:
OxfordUniversity Press,
forthcoming). Gray'scriticisms
ofstructuralMarxism areindebtedtoWilliamConnolly,
Appearance andReality inPolitics
(Cambridge: Cambridge Uhiversity
Press,1981);see
thefirsttwochapters inparticular.
71. Foucault,Power/Knowledge, Ch. 1.
72. Ibid.,throughout; idem,Language,Counter-Memory, PartIll.
Practice,

Mark Philpis ResearchFellowin Social Scienceat JesusCollege,Oxford.He has

published work on the philosophyand sociology of welfareand is currently
engagedin workon WilliamGodwinand Britishradicalismin the 1790s.

This content downloaded from 169.229.32.137 on Thu, 8 May 2014 10:57:15 AM

All use subject to JSTOR Terms and Conditions