Intelligence
Content
Trustees of Princeton University
Review: The Function of Intelligence
Author(s): Willmoore Kendall
Source: World Politics, Vol. 1, No. 4 (Jul., 1949), pp. 542-552
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/2008837 .
Accessed: 02/03/2011 12:37
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=cup. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Cambridge University Press and Trustees of Princeton University are collaborating with JSTOR to digitize,
preserve and extend access to World Politics.
http://www.jstor.org
THE FUNCTION OF INTELLIGENCE
By WILLMOORE KENDALL
ShermanKent, StrategicIntelligence,Princeton,PrincetonUniversity
Press, 1949,pp. xiii,240. $3.00.
STRATEGIC INTELLIGENCE can be readvariouslyas:
1) a generalintroduction
to intelligenceworkwhich,say,
thedirectorof almostany sectionof CentralIntelligencemight
give a new recruitto read on his firstday at the office;2) a
memorandumfroman Old Hand at intelligenceworkwho has
thoughtit all over-like thebirdin Peterand theWolf,froma
safe distance-and has a thingor two to tell those of his colleagueswho have stayedon in Washington;and 3) an attempt
by that same Old Hand to make sense-inter alia forhimself
-out of a greatlyexpandedUnitedStates government
activity
in which,as all who knowit can testify,
sense does not leap to
the eye. Because the book is in part each of thesethreethings,
as any one of them.But it is eviit is not perfectly
satisfactory
dentlynot offeredas a learnedtreatiseon any or all of them:
indeed,the gap in the existingliteraturein the fieldis so great
that one can hardly be surprisedat Mr. Kent's failure to
definehis task withprecision.The greatmeritof his book is
materialwhichwill enable
thatit providesa bodyofdescriptive
seriouspublicdiscussionto beginon therelationof intelligence
to policyin a democraticsystem.Since it is Americanpolicyon
whichthe futureof the freeworld seems to depend,it is high
time forthe public debate to commence.
Moreover,Mr. Kent's book is one fromwhichanyoneinterestedin the relation,past,present,and future,betweenUnited
Statesforeignpolicyand social research,can learnmuch-now
fromits manifestcontent;now,perhapsmoreoftenfromwhat
it takesforgranted,whatit says betweenthe lines,and whatit
mighthave said but did not.Therefore,
let us take a look at the
bookin each ofitsthreecapacities.
THE FUNCTION OF INTELLIGENCE
543
I
The beginnerat an intelligenceagencycan getfromStrategic
of the profession
Intelligencean initiationintotheterminology
map in Washington
he is entering-a pictureof theintelligence
how,
functions,
intelligence
(whatorganizationsare performing
grasp
provisional
-and
a
etc.)
mesh
together,
they
theoretically,
of some of the controversialissues because of whichthe lines
on that map lie wheretheylie ratherthan somewhereelse. He
is thatof"describcan learnthatthetaskofforeignintelligence
ing,observing,and reportingupon, and speculatingas to the
futureof... phenomenain foreignlands" (p. 104), whichseems
as he is likelyto findof the task
as good a workingdefinition
agencies.He can, perhaps
by
certain
actuallybeingperformed
best of all, derivefromthe book the elementsof an emergent
"ethics"of the intelligenceprofession-a sense of whyit is imfunctionshouldbe well performed,
portantthattheintelligence
a realizationthat thereare some compromises(with,say, administrativeconvenience,or entrenchedignorance) that he
and his colleaguesmust,in thelongrun,refuseto make,and an
betweengovernmentresearch
insightinto certaindifferences
and academic researchthat make the ethicsof the latternot
strictly
applicableto theformer.
Mr. Kent may,to be sure,awaken in the beginnersome exas, forexample,
pectationsthatare doomedto disappointment,
that
"an
him
intelligenceorganizationis a
when he assures
strangeand wonderfulcollectionof devotedspecialistsmolded
into a vigorousproductionunit" (p. 74) ; but it was two Old
Hands at newspaperworkwhowroteThe FrontPage, and there
is no doubt somethingto be said forinculcatingin beginners
an enthusiasmthat only the more romanticof their seniors
are able to maintain.Speaking as a man who has oftentold
beginnersbadly what Mr. Kent tells them well, the present
writerwelcomesStrategicIntelligencein this one of its three
will make
capacities,and hopes thatthe intelligencefraternity
and
will congooduse ofit. It will save theP-8's precioustime,
tributegenerouslyto the professionaldevelopmentof their
juniorcolleagues.
544
WORLD POLITICS
II
Mr. Kent has lookedat theintelligenceset-upin Washington
and foundit, on balance, good. Thus, StrategicIntelligenceis
who takes pen in hand to expose
not the book of a reformer
and remedya bad situation.On thecontrary:it insiststhatthe
who regularlysubstituteshis own judgmentfor
decision-maker
thatof the "intelligencearm" is, tout court,"turninghis back
by whichwesternman has, since Arison thetwo instruments
totle,steadilyenlargedhis horizonof knowledge-the instrumentsof reason and scientificmethod" (p. 206). It would be
quite unfair,on the otherhand,to suggestthat Mr. Kent puts
himselfforwardas a defenderof the status quo in intelligence
planningand organization.He shows himself,at a numberof
pointsin his argument,genuinelyaware that the existingarrangementsreflect,shall we say, somethingless than the best
thinkingofwhichthe nationis capable. But he has notworked
out in his own mindwhat an ideal set of intelligencearrangements-the ones he would set up if all the resistanceswere removed-would be like,and thushas no standardagainstwhich
he exposes. In
to measurethe magnitudeof the shortcomings
any case, his book is not characterizedby a persistentwillingness to followthroughon adverse criticism,to proceed from
of a basic defectto such questionsas: How graveis
recognition
not
our defectsbe so grave that in so far as it is
this? May
possibleto courtnationaldisasterby mismanagingthe intelligencefunction,we are doingjust thatin the United States?
Let us examinesome of the emphasesthat mighthave led
himto ask suchquestions:
cannotdo itsjob unless
1. Mr. Kentbelievesthatintelligence
it knowsthemindof therecipientsof its reports,and getsfrom
themthe kind of "guidance" that any professionalman needs
fromhis client (pp. 180-182).' He can hardlyavoid the concluthe intellision that wheresuch guidanceis not forthcoming
not beingdone. Is it forthcoming,
then,
gencejob is, in effect,
in thoseareas of nationalpolicythat are of deepestconcernto
us? "The uninitiated,"Mr. Kent replies,"will be surprisedto
1 Curiously,however,he reproduceswithevidentapprovala passage of WalterLippmann's
which urges the reverseof this (p. 200). Mr. Kent's views on KiampfendeWissenschaft
(ibid.) vary sharplyfromchapterto chapter.
THE FUNCTION OF INTELLIGENCE
545
hearthattheelementofguidance. . . becomesrarerand rareras
the job of intelligencemountsin augustness" (p. 182). Does
it not then followthat, as regardsthe great decisions about
foreignpolicy,it is highlyprobablethat (we use his language)
willfailand our statesmenplan in ignorance?
ourundertakings
Evidently,but Mr. Kent does notmaketheinference.
2. One of the thingsthat an intelligenceagencymust do is
to scrutinize,and tryto make sense of,"what goes on abroad"
(p. 152). As is well known,United States intelligenceagencies
make use forthispurposebothof "overt"techniques-that is,
observationand research"
"unromanticopen-and-above-board
(p. 4) -and "covert"or "clandestine"techniquesoffact-gathering. Two choices which any intelligenceagency must make,
are thoserelatingto, first,the allocationof resources
therefore,
betweenthesetwo typesof "collection,"and secondly,the degreeof autonomywhichthe clandestineoperationshall,out of
tenderregardforconsiderationsof "security,"be permittedto
enjoy.Indeed,thewholematterofclandestineoperationsis now
so surroundedby securityrestrictions
as to makeeventhestatementthattheyare surroundedby securityrestrictions
a matter
of doubtfullegality; and the literatureof the subject is not
extensive.
All the moreinteresttherefore
attachesto Mr. Kent's statement-and he is prettyclearly looking rightat Washington
whenhe makes it-that the clandestineforce,"if it allows the
mechanismsof securityto cut it offfromsome of themostsignificantlines of guidance,"ends up destroying
"its own reason
and notcolforexistence"by "collectingthewronginformation
lectingthe right" (p. 167). And the interestbecomes intense
whenthis statementis placed beside the furtherstatementto
the effectthat: "This kind of mis-collection
would be far less
likelyto occur if the operationwere not freeto steerits own
coursebehindthefogofits own securityregulations"(p. 168).2
This tallies preciselywith one of the presentwriter'smajor
criticismsof our presentintelligencearrangements:theyenormouslyexaggeratethe importanceof covertcollection,and yet
permitit to yield shockinglysmall dividends.3Now: either
2 The
italicsare mine.
If you were directorof an intelligenceagency in France, and wished to know "what
goes on" in the United States, which would you do first:take out an airmail subscription
3
546
WORLD POLITICS
the agencyneeds the extra information
it can obtain through
clandestineoperationsafterit has exhaustedthe potentialities
of its overtcollection,or it does not. If it does not, it can do
withoutits clandestineoperationaltogether;if it does, it must
be in a positionto lay down-and enforce-directivesto the
men responsiblefor obtainingit. And if the Washingtonintelligenceagenciesare not in this position,then Mr. Kent is
tellingus somethingthat needs to be said to a much wider
audiencethanthatwhichis likelyto read his book, and somethingwhose implicationsneed to be pointed up much more
sharplythan he has pointedthem up. The whole question of
covertcollectionis, in the presentwriter'sview, one that urgentlyrequires investigationby a Congressionalcommittee
preparedto speakthelanguageof legislativesupremacy,and to
insistthat no democracycan affordto make a simple andin the shortand middletermat least-irrevocable act of faith
in the men called upon to performthis highlyexplosivefunction.
3. "Withoutprofessionalexperts,"says Mr. Kent, "thereis
no intelligence. . .. They are the social and naturalscientists
feelfortheways
and themilitaryexpertswho have a finger-tip
of researchand analysis,who are masters (or dedicatednovitiates) ... oftheirparticularbracketof learning,and to whom
the discoveryof new facts or new relationshipsis a career"
(p. 107).4Again and again,indeed,he showshis awarenessthat
the quality of an intelligenceagency'spersonnelis the major
determinantof the quality of its performance;and if in the
passage just cited he sidestepsthe question,whetherthe intelligencefunctionis today in the hands of men like those he
has described("if theydo not in actual departmentalpractice
of this paragraph,. . . they
measure up to the specifications
to collecthis viewsas to the present
should"), it is not difficult
trendin this regard: "Two forcesof disintegrationare now
workingin concerton themostvaluable people.They are being
nudgedfromwithin[by civilserviceregulations]and beckoned
fromwithout.Their loss is a catastropheto federalintelligence
to the New York Times, or send four ex-Deuxieme-Bureaumen to dispense largesse in
Washington?And now that you have answered that question in the manner that the
presentwriterexpected,how sure are you that you would send the four men at all?
4 The italicsare mine.
THE FUNCTIONOF INTELLIGENCE
547
work" (p. 147).5 But the warningis buriedin a technicaldiscussion in the middle of the book, and has no impact upon
itsgeneralargument.
4. Mr. Kent has otherminorquarrelswith presentintelligencepolicy: he would like a civiliandirectorforCentral Intelligence(p. 100) ; he feelsthat CIA's powersof investigation
vis-a-vis departmental
intelligenceare inadequate (p. 101); he
believesthatWashingtonwould be well-advisedto rethinkthe
relationbetweenresearchpeople in the homeofficeon the one
hand, and "fieldstaff,"includinghere overtas well as covert
collection,on theother(p. 165) ; and so on.
The presentwriterbelievesthat if all of Mr. Kent's reproofs
wereactedupon,and all his proposalsadopted,theresultwould
in UnitedStatesintelligence
be an improvement
operations.But
this improvement
would, like the infantmentionedin Marx's
famousfootnote,
be verysmall.
III
The mostinstructive
passagesin StrategicIntelligenceare,in
the presentwriter'sopinion,those in whichMr. Kent reveals
what we may call the generaltheoryof the intelligencefunctionto whichhisthinkinghas broughthim-plus thosein which
he records,but on his theorydoes not identifyas such, recognizably pathologicalaspects of existingintelligencearrangements.They are instructive
because Mr. Kent's state of mind
on mostof the problemsto whichsuch a theorymust address
itselfreflects
to a remarkabledegreethatof official
Washington
as the presentwritercame to knowit in the courseof his own
tourof dutyas an intelligence
official.And, so faras he knows,
this is the firsttime that that state of mind has, so to speak,
venturedbeyondthesteelcurtainon 26thStreetto subjectitself
to criticism.
When approachedfromthispointofview,Strategic
Intelligencebecomes a book that everysocial scientistshould
lay in his heart and ponder.Let us notice some of the charofthisstateofmind:
acteristics
5 And he says elsewherethat if we are lookingfor "the encouragingelementin departmentalintelligence,"i.e., the departmentsthat have "realized the importanceof the task,"
shown "a decent respectfor full and accurate knowledge,"and "employedthe rightkind
of professionalpeople," it is to Labor, Commerce,and Agriculture(thus, one infers,not
to State and Defense) that we mustgo (p. 115).
548
WORLD POLITICS
1. It is a stateof mindwhichis-dangerously,in thiswriter's view-dominated by an essentiallywartimeconceptionof
theintelligence
function.Most of the men who occupythe key
positionslearnedtheintelligence
businessduringthewar. They
seemto have acquired a trainedincapacityto put aside, more
than momentarily,
the intellectualhabits appropriateto the
conductof hostilities(or preparationforthe conductof hostilities)againstan actual or potentialenemy,and to giveto the
distinctively
peacetimefunctionsof a governmental
intelligence
agencythe importancetheydeserve.This is not to implythat
researchinto the constituents
of powershould lapse in timeof
peace,or that"strategicintelligence"as thisstateof mindconceivesit shouldbe performed
withany less lovingcare than in
the past; and it is certainlynot to implythatthe possessorsof
this state of mind are in any sense reluctantto place the intelligenceagenciesat the serviceof the quest for (or, as Mr.
Kent likesto put it,thestrategyof) peace. The pointis, rather,
that the veryuse of the term"strategic"as opposed to, say,
"foreignpolicy,"to denotetheentiretask ofintelligence
is itself
and thatthe big job-the carvingout of
profoundly
significant,
United States destinyin the world as a whole, as contrasted
to the conductof United States policytoward a congeriesof
nation-states-tendsto slipthroughtheirfingers.
One illustrationof thisto whichone mightpointis the ease
withwhichMr. Kent moves froma definitionof the role of
intelligencewhich presupposesthat the objectivesof foreign
policyare given (p. 154), as, withrespectto the large issues,
theyare in timeof war,and one that runsin termsof "the objective and impartialexploration"of competingalternatives,
whichis evidentlythe appropriatedefinitionin peacetime (p.
201). Anotherillustrationis his willingnessto acquiesce in a
ofthatrolewhichregardstheknowledge"upon which
definition
we base our high-levelnationalpolicytowardthe otherstates
of the world" as separableinto two parts,namely,that about
otherstatesand thataboutsomethingcalled "our own domestic
scene." Such a definition
puts the firstpart in the hands of a
distinctgroupof officialswhose "research"must stop shortat
the three-milelimiteven when the threadthey are following
runs rightacross it, and yet whichtells itselfit is using the
THE FUNCTION OF INTELLIGENCE
549
scientificmethod.(This ends you up withintelligencereports
thatnever,nevertakecognizanceofUnitedStatespoliciesalternativeto theone actuallyin effect,
suchproblemsbeing"domestic"matters.)
It is this wartimeconceptionof the intelligencefunction
whichleads to a "regionalbreakdown"of the intelligencefunction,and deems knowledgeof the outsideworld as additivein the sense that if six regionaldivision chiefsare gathered
togetherin thename oftheworldsituationtherealso is wisdom
about the latter.It is a state of mind characterizedby a compulsivepreoccupationwithprediction,withthe eliminationof
"surprise"fromforeignaffairs.The shadow of Pearl Harbor
is projectedinto the mists of Bogota, and intelligencelooks
shamefacedover its failureto tell SecretaryMarshall the day
and hourat whicha revolutionwillbreakout in Colombia.The
course of eventsis conceivednot as somethingyou tryto influencebut as a tape all printedup inside a machine; and the
job of intelligenceis to tell the plannershow it reads. With
thisconceptionof intelligenceone does not,and, on the record
at least,cannotdistinguishbetweenwhatwe may call absolute
The latteris what the
predictionand contingentprediction.6
government
needs,especiallyin peacetime.
2. It is a stateof minddominatedby an essentiallybureaucraticconceptionof United States government,
and of the intelligenceproblem.On examiningMr. Kent's discussionof the
relationbetweenthe "producers"and "consumers"of intelligence (pp. 180-206), one recognizesat once that he is concernednot with the relationbetweenhis intelligenceexperts
and the electedofficialswho, as we hope,are still makingthe
actual decisionsabout our foreignpolicy,but withthe relation
betweenhis intelligenceexpertsand what it is now fashionable
to call the "policyplanners."The issue here is fundamental:
ifyou conceivetheintelligence
functionin thismanner,you are
excludingfromits purviewwhatthiswriterwould call its most
crucialaspect-i.e., thatwhichconcernsthe communicationto
6An example of absolute predictionwould be: "General DeGaulle will come to power
this day six months"; or "Japan will attack Pearl Harbor on x-day at y-hour."The contingentpredictionwould read: "The followingfactors,which can be influencedin such
and such a fashionby action fromthe outside, will determinewhether,and if so, when,
GeneralDeGaulle willcometo power."
550
WORLD POLITICS
the politicallyresponsiblelaymenof the knowledgewhich,to
use Mr. Lippmann'shappy phrase,determinesthe "pictures"
theyhave in theirheads of the worldto whichtheirdecisions
relate.The role of the intelligenceagenciesbecomes,on this
showing, that of mere research assistants to the George
Kennans.
Mr. Kent mightreplyto thiscriticism,as the presentwriter
has heardmanyof his formercolleaguesreply: This is how it
in factworks,and it would be "unrealistic"to discuss the intelligencefunctionin any otherterms.No doubt; but if so,
we do nottellourselvesthatoursis a book aboutthe"knowledge
upon whichour nation's foreignrelations,in war and peace,
mustrest" (p. vii). For the state of mind forwhichthatbook
calls is one whichexplicitlyrecognizeshow small a partof that
knowledgecan or shouldbe providedby a government
operation
(i.e., how large a part of it is and should be providedby professional scholarshipunder non-governmental
auspices, and
howlargea partofit is, whetherit shouldbe or not,providedhoweverbadly-by journalists), and how importantit is to
relateour thinkingabout that small partto our thinkingabout
therest.
3. It is a state of mind characterizedby a crassly empiricalconceptionof the researchprocessin the social sciences.
This, in view of the profoundcommitmentof our intelligence agencies to what we have called the "regional breakdown" of theirproblem,is not surprising.For, if it is regional
unitsyou are building,and it is social scientistsspecializedto
specificcountriesand areas you wish to staffthemwith,what
you end up withis an extremely
highpercentageof historians,
who withthebestwill in the worldcommunicateto the operavices (and virtues) of theirkind of retion the characteristic
oftheintelligence
search.The performance
functionaccordingly
becomesa matterof somehowkeepingone's head above water
in a tidal wave of documents,whose factual contentmust be
"processed"- i.e., in Mr. Kent's language,"analyzed," "evaluated," and exploitedas raw materialfor "hypotheses."The
emphasis,as we have alreadynoticed,is on prediction,which
againstthisbackgroundis necessarilyunderstoodas a matter
THE FUNCTION OF INTELLIGENCE
551
of projectingdiscernibleempiricaltrends into an indefinite
future.
Here also the issue is fundamental:an intelligenceoperation
built upon a conceptionof the researchprocess in the social
sciencesthatassignsdue weightto "theory"as it is understood
in economics and sociologyand, increasinglyone hopes, in
politics,would of coursebe a whollydifferent
affair.It would
recruita considerablepercentageof its personnelpreciselyfor
its theoreticaltrainingand accomplishments;it would free
themfromthe tidal wave of documents;it would enable them
to workunder conditionscalculated to encouragethought;it
would, above all, give them continuous and instantaneous
access (e.g.,by internationaltelephone)to the data that really
matter,namelythe raw data of the developingsituationin the
outsideworld. (The documents,havingcome fromabroad,are
as a matterof courseout ofdate.) This, clearly,is nottheplace
to discussthecomparativemeritsofthetwo conflicting
concepit
is
not
intendedto suggestthat
tions of social research,and
the intelligencefunctionshould be deliveredover entirelyto
that impliedin the foregoingsentences.The point is, rather,
thatcurrentplanningand organizationin the intelligencefield
ignoresone ofthetwo altogether.(It is, fromthispointofview,
thatMr. Kent,forall his numerousreferences
highlysignificant
to the social sciences and social scientists,never employsin
thatconnectionthewordstheoryand theorist.)
4. It is a state of mind characterizedby an uncriticaloptimismregardingthe supplyof the skillsupon whichthe effecoftheintelligence
functiondepends.Mr. Kent,
tiveperformance
in
this
reviewer's
quite rightly
opinion,looks primarilyto the
social sciencesforthese skills.Like most scholarswho served
withOSS duringthewar,however,he tendsto envisagethesolutionoftheintelligence
personnelproblemin termsof attracting
back to Washingtonthe social science scholars who, in the
courseoftheshake-upafterthewar,lostheartand turnedtheir
faces homeward-i.e., in termsof a returnto a Golden Age
situated,like Spain's, in the past. Or, to put the same thing
anotherway,he tendsto assume that"out there"-in the uniin theresearchinstitutes,
versities,
or,one supposes,temporarily
unemployed-themenyou need existand await theircall. One
552
WORLD POLITICS
findsin him,and in the intelligencefraternity
in Washington,
noneoftheanxietiesthatwouldcause mostresponsiblescholars
in the fieldsotherthan History(forwhichwe leave Mr. Kent,
as a professionalhistorian,entirelyfreeto speak) to say to
him: There could be no more dangerouserror.The supplyof
skills you have in mind is hopelesslyinadequate and, failing
drasticnationalaction-for example,a newManhattanProject,
but thistimein the social sciences-to increaseit, will remain
hopelesslyinadequate throughoutthe predictablefuture.It is
inadequate in two senses: We are, and are likelyto remain,
unableto releaseany significant
numberof men to government
intelligence
withoutwreckingtheteachingprogramsuponwhich
the futuresupply of such men must depend; and we have
gravemisgivings-theeconomistsperhapsless than the restof
us, but the economistsalso-about the abilityof our sciences
to supplythe sort of knowledgewhich,in Mr. Kent's phrase,
"our highlyplaced civiliansand militarymen must have."
It is, in fine,not a state of mindwhichis likelyto produce
thebook all of us have beenwaitingforsomeoneto writesince,
twenty-seven
years ago, Mr. Lippmann'sPublic Opinion first
directedour attentionto those picturesin decision-makers'
heads-and to the possibilityof our deliberatelyinfluencing
theiraccuracy.Nor is it a stateof mindlikelyto produceeven
a smallpartof thebook thatwould deal withthe government's
own "high-levelforeignpositiveintelligence."But the publication of StrategicIntelligencemay well make morepossiblethe
writingof thesetwo othercriticallyneededworks.
Review: The Function of Intelligence
Author(s): Willmoore Kendall
Source: World Politics, Vol. 1, No. 4 (Jul., 1949), pp. 542-552
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/2008837 .
Accessed: 02/03/2011 12:37
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=cup. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Cambridge University Press and Trustees of Princeton University are collaborating with JSTOR to digitize,
preserve and extend access to World Politics.
http://www.jstor.org
THE FUNCTION OF INTELLIGENCE
By WILLMOORE KENDALL
ShermanKent, StrategicIntelligence,Princeton,PrincetonUniversity
Press, 1949,pp. xiii,240. $3.00.
STRATEGIC INTELLIGENCE can be readvariouslyas:
1) a generalintroduction
to intelligenceworkwhich,say,
thedirectorof almostany sectionof CentralIntelligencemight
give a new recruitto read on his firstday at the office;2) a
memorandumfroman Old Hand at intelligenceworkwho has
thoughtit all over-like thebirdin Peterand theWolf,froma
safe distance-and has a thingor two to tell those of his colleagueswho have stayedon in Washington;and 3) an attempt
by that same Old Hand to make sense-inter alia forhimself
-out of a greatlyexpandedUnitedStates government
activity
in which,as all who knowit can testify,
sense does not leap to
the eye. Because the book is in part each of thesethreethings,
as any one of them.But it is eviit is not perfectly
satisfactory
dentlynot offeredas a learnedtreatiseon any or all of them:
indeed,the gap in the existingliteraturein the fieldis so great
that one can hardly be surprisedat Mr. Kent's failure to
definehis task withprecision.The greatmeritof his book is
materialwhichwill enable
thatit providesa bodyofdescriptive
seriouspublicdiscussionto beginon therelationof intelligence
to policyin a democraticsystem.Since it is Americanpolicyon
whichthe futureof the freeworld seems to depend,it is high
time forthe public debate to commence.
Moreover,Mr. Kent's book is one fromwhichanyoneinterestedin the relation,past,present,and future,betweenUnited
Statesforeignpolicyand social research,can learnmuch-now
fromits manifestcontent;now,perhapsmoreoftenfromwhat
it takesforgranted,whatit says betweenthe lines,and whatit
mighthave said but did not.Therefore,
let us take a look at the
bookin each ofitsthreecapacities.
THE FUNCTION OF INTELLIGENCE
543
I
The beginnerat an intelligenceagencycan getfromStrategic
of the profession
Intelligencean initiationintotheterminology
map in Washington
he is entering-a pictureof theintelligence
how,
functions,
intelligence
(whatorganizationsare performing
grasp
provisional
-and
a
etc.)
mesh
together,
they
theoretically,
of some of the controversialissues because of whichthe lines
on that map lie wheretheylie ratherthan somewhereelse. He
is thatof"describcan learnthatthetaskofforeignintelligence
ing,observing,and reportingupon, and speculatingas to the
futureof... phenomenain foreignlands" (p. 104), whichseems
as he is likelyto findof the task
as good a workingdefinition
agencies.He can, perhaps
by
certain
actuallybeingperformed
best of all, derivefromthe book the elementsof an emergent
"ethics"of the intelligenceprofession-a sense of whyit is imfunctionshouldbe well performed,
portantthattheintelligence
a realizationthat thereare some compromises(with,say, administrativeconvenience,or entrenchedignorance) that he
and his colleaguesmust,in thelongrun,refuseto make,and an
betweengovernmentresearch
insightinto certaindifferences
and academic researchthat make the ethicsof the latternot
strictly
applicableto theformer.
Mr. Kent may,to be sure,awaken in the beginnersome exas, forexample,
pectationsthatare doomedto disappointment,
that
"an
him
intelligenceorganizationis a
when he assures
strangeand wonderfulcollectionof devotedspecialistsmolded
into a vigorousproductionunit" (p. 74) ; but it was two Old
Hands at newspaperworkwhowroteThe FrontPage, and there
is no doubt somethingto be said forinculcatingin beginners
an enthusiasmthat only the more romanticof their seniors
are able to maintain.Speaking as a man who has oftentold
beginnersbadly what Mr. Kent tells them well, the present
writerwelcomesStrategicIntelligencein this one of its three
will make
capacities,and hopes thatthe intelligencefraternity
and
will congooduse ofit. It will save theP-8's precioustime,
tributegenerouslyto the professionaldevelopmentof their
juniorcolleagues.
544
WORLD POLITICS
II
Mr. Kent has lookedat theintelligenceset-upin Washington
and foundit, on balance, good. Thus, StrategicIntelligenceis
who takes pen in hand to expose
not the book of a reformer
and remedya bad situation.On thecontrary:it insiststhatthe
who regularlysubstituteshis own judgmentfor
decision-maker
thatof the "intelligencearm" is, tout court,"turninghis back
by whichwesternman has, since Arison thetwo instruments
totle,steadilyenlargedhis horizonof knowledge-the instrumentsof reason and scientificmethod" (p. 206). It would be
quite unfair,on the otherhand,to suggestthat Mr. Kent puts
himselfforwardas a defenderof the status quo in intelligence
planningand organization.He shows himself,at a numberof
pointsin his argument,genuinelyaware that the existingarrangementsreflect,shall we say, somethingless than the best
thinkingofwhichthe nationis capable. But he has notworked
out in his own mindwhat an ideal set of intelligencearrangements-the ones he would set up if all the resistanceswere removed-would be like,and thushas no standardagainstwhich
he exposes. In
to measurethe magnitudeof the shortcomings
any case, his book is not characterizedby a persistentwillingness to followthroughon adverse criticism,to proceed from
of a basic defectto such questionsas: How graveis
recognition
not
our defectsbe so grave that in so far as it is
this? May
possibleto courtnationaldisasterby mismanagingthe intelligencefunction,we are doingjust thatin the United States?
Let us examinesome of the emphasesthat mighthave led
himto ask suchquestions:
cannotdo itsjob unless
1. Mr. Kentbelievesthatintelligence
it knowsthemindof therecipientsof its reports,and getsfrom
themthe kind of "guidance" that any professionalman needs
fromhis client (pp. 180-182).' He can hardlyavoid the concluthe intellision that wheresuch guidanceis not forthcoming
not beingdone. Is it forthcoming,
then,
gencejob is, in effect,
in thoseareas of nationalpolicythat are of deepestconcernto
us? "The uninitiated,"Mr. Kent replies,"will be surprisedto
1 Curiously,however,he reproduceswithevidentapprovala passage of WalterLippmann's
which urges the reverseof this (p. 200). Mr. Kent's views on KiampfendeWissenschaft
(ibid.) vary sharplyfromchapterto chapter.
THE FUNCTION OF INTELLIGENCE
545
hearthattheelementofguidance. . . becomesrarerand rareras
the job of intelligencemountsin augustness" (p. 182). Does
it not then followthat, as regardsthe great decisions about
foreignpolicy,it is highlyprobablethat (we use his language)
willfailand our statesmenplan in ignorance?
ourundertakings
Evidently,but Mr. Kent does notmaketheinference.
2. One of the thingsthat an intelligenceagencymust do is
to scrutinize,and tryto make sense of,"what goes on abroad"
(p. 152). As is well known,United States intelligenceagencies
make use forthispurposebothof "overt"techniques-that is,
observationand research"
"unromanticopen-and-above-board
(p. 4) -and "covert"or "clandestine"techniquesoffact-gathering. Two choices which any intelligenceagency must make,
are thoserelatingto, first,the allocationof resources
therefore,
betweenthesetwo typesof "collection,"and secondly,the degreeof autonomywhichthe clandestineoperationshall,out of
tenderregardforconsiderationsof "security,"be permittedto
enjoy.Indeed,thewholematterofclandestineoperationsis now
so surroundedby securityrestrictions
as to makeeventhestatementthattheyare surroundedby securityrestrictions
a matter
of doubtfullegality; and the literatureof the subject is not
extensive.
All the moreinteresttherefore
attachesto Mr. Kent's statement-and he is prettyclearly looking rightat Washington
whenhe makes it-that the clandestineforce,"if it allows the
mechanismsof securityto cut it offfromsome of themostsignificantlines of guidance,"ends up destroying
"its own reason
and notcolforexistence"by "collectingthewronginformation
lectingthe right" (p. 167). And the interestbecomes intense
whenthis statementis placed beside the furtherstatementto
the effectthat: "This kind of mis-collection
would be far less
likelyto occur if the operationwere not freeto steerits own
coursebehindthefogofits own securityregulations"(p. 168).2
This tallies preciselywith one of the presentwriter'smajor
criticismsof our presentintelligencearrangements:theyenormouslyexaggeratethe importanceof covertcollection,and yet
permitit to yield shockinglysmall dividends.3Now: either
2 The
italicsare mine.
If you were directorof an intelligenceagency in France, and wished to know "what
goes on" in the United States, which would you do first:take out an airmail subscription
3
546
WORLD POLITICS
the agencyneeds the extra information
it can obtain through
clandestineoperationsafterit has exhaustedthe potentialities
of its overtcollection,or it does not. If it does not, it can do
withoutits clandestineoperationaltogether;if it does, it must
be in a positionto lay down-and enforce-directivesto the
men responsiblefor obtainingit. And if the Washingtonintelligenceagenciesare not in this position,then Mr. Kent is
tellingus somethingthat needs to be said to a much wider
audiencethanthatwhichis likelyto read his book, and somethingwhose implicationsneed to be pointed up much more
sharplythan he has pointedthem up. The whole question of
covertcollectionis, in the presentwriter'sview, one that urgentlyrequires investigationby a Congressionalcommittee
preparedto speakthelanguageof legislativesupremacy,and to
insistthat no democracycan affordto make a simple andin the shortand middletermat least-irrevocable act of faith
in the men called upon to performthis highlyexplosivefunction.
3. "Withoutprofessionalexperts,"says Mr. Kent, "thereis
no intelligence. . .. They are the social and naturalscientists
feelfortheways
and themilitaryexpertswho have a finger-tip
of researchand analysis,who are masters (or dedicatednovitiates) ... oftheirparticularbracketof learning,and to whom
the discoveryof new facts or new relationshipsis a career"
(p. 107).4Again and again,indeed,he showshis awarenessthat
the quality of an intelligenceagency'spersonnelis the major
determinantof the quality of its performance;and if in the
passage just cited he sidestepsthe question,whetherthe intelligencefunctionis today in the hands of men like those he
has described("if theydo not in actual departmentalpractice
of this paragraph,. . . they
measure up to the specifications
to collecthis viewsas to the present
should"), it is not difficult
trendin this regard: "Two forcesof disintegrationare now
workingin concerton themostvaluable people.They are being
nudgedfromwithin[by civilserviceregulations]and beckoned
fromwithout.Their loss is a catastropheto federalintelligence
to the New York Times, or send four ex-Deuxieme-Bureaumen to dispense largesse in
Washington?And now that you have answered that question in the manner that the
presentwriterexpected,how sure are you that you would send the four men at all?
4 The italicsare mine.
THE FUNCTIONOF INTELLIGENCE
547
work" (p. 147).5 But the warningis buriedin a technicaldiscussion in the middle of the book, and has no impact upon
itsgeneralargument.
4. Mr. Kent has otherminorquarrelswith presentintelligencepolicy: he would like a civiliandirectorforCentral Intelligence(p. 100) ; he feelsthat CIA's powersof investigation
vis-a-vis departmental
intelligenceare inadequate (p. 101); he
believesthatWashingtonwould be well-advisedto rethinkthe
relationbetweenresearchpeople in the homeofficeon the one
hand, and "fieldstaff,"includinghere overtas well as covert
collection,on theother(p. 165) ; and so on.
The presentwriterbelievesthat if all of Mr. Kent's reproofs
wereactedupon,and all his proposalsadopted,theresultwould
in UnitedStatesintelligence
be an improvement
operations.But
this improvement
would, like the infantmentionedin Marx's
famousfootnote,
be verysmall.
III
The mostinstructive
passagesin StrategicIntelligenceare,in
the presentwriter'sopinion,those in whichMr. Kent reveals
what we may call the generaltheoryof the intelligencefunctionto whichhisthinkinghas broughthim-plus thosein which
he records,but on his theorydoes not identifyas such, recognizably pathologicalaspects of existingintelligencearrangements.They are instructive
because Mr. Kent's state of mind
on mostof the problemsto whichsuch a theorymust address
itselfreflects
to a remarkabledegreethatof official
Washington
as the presentwritercame to knowit in the courseof his own
tourof dutyas an intelligence
official.And, so faras he knows,
this is the firsttime that that state of mind has, so to speak,
venturedbeyondthesteelcurtainon 26thStreetto subjectitself
to criticism.
When approachedfromthispointofview,Strategic
Intelligencebecomes a book that everysocial scientistshould
lay in his heart and ponder.Let us notice some of the charofthisstateofmind:
acteristics
5 And he says elsewherethat if we are lookingfor "the encouragingelementin departmentalintelligence,"i.e., the departmentsthat have "realized the importanceof the task,"
shown "a decent respectfor full and accurate knowledge,"and "employedthe rightkind
of professionalpeople," it is to Labor, Commerce,and Agriculture(thus, one infers,not
to State and Defense) that we mustgo (p. 115).
548
WORLD POLITICS
1. It is a stateof mindwhichis-dangerously,in thiswriter's view-dominated by an essentiallywartimeconceptionof
theintelligence
function.Most of the men who occupythe key
positionslearnedtheintelligence
businessduringthewar. They
seemto have acquired a trainedincapacityto put aside, more
than momentarily,
the intellectualhabits appropriateto the
conductof hostilities(or preparationforthe conductof hostilities)againstan actual or potentialenemy,and to giveto the
distinctively
peacetimefunctionsof a governmental
intelligence
agencythe importancetheydeserve.This is not to implythat
researchinto the constituents
of powershould lapse in timeof
peace,or that"strategicintelligence"as thisstateof mindconceivesit shouldbe performed
withany less lovingcare than in
the past; and it is certainlynot to implythatthe possessorsof
this state of mind are in any sense reluctantto place the intelligenceagenciesat the serviceof the quest for (or, as Mr.
Kent likesto put it,thestrategyof) peace. The pointis, rather,
that the veryuse of the term"strategic"as opposed to, say,
"foreignpolicy,"to denotetheentiretask ofintelligence
is itself
and thatthe big job-the carvingout of
profoundly
significant,
United States destinyin the world as a whole, as contrasted
to the conductof United States policytoward a congeriesof
nation-states-tendsto slipthroughtheirfingers.
One illustrationof thisto whichone mightpointis the ease
withwhichMr. Kent moves froma definitionof the role of
intelligencewhich presupposesthat the objectivesof foreign
policyare given (p. 154), as, withrespectto the large issues,
theyare in timeof war,and one that runsin termsof "the objective and impartialexploration"of competingalternatives,
whichis evidentlythe appropriatedefinitionin peacetime (p.
201). Anotherillustrationis his willingnessto acquiesce in a
ofthatrolewhichregardstheknowledge"upon which
definition
we base our high-levelnationalpolicytowardthe otherstates
of the world" as separableinto two parts,namely,that about
otherstatesand thataboutsomethingcalled "our own domestic
scene." Such a definition
puts the firstpart in the hands of a
distinctgroupof officialswhose "research"must stop shortat
the three-milelimiteven when the threadthey are following
runs rightacross it, and yet whichtells itselfit is using the
THE FUNCTION OF INTELLIGENCE
549
scientificmethod.(This ends you up withintelligencereports
thatnever,nevertakecognizanceofUnitedStatespoliciesalternativeto theone actuallyin effect,
suchproblemsbeing"domestic"matters.)
It is this wartimeconceptionof the intelligencefunction
whichleads to a "regionalbreakdown"of the intelligencefunction,and deems knowledgeof the outsideworld as additivein the sense that if six regionaldivision chiefsare gathered
togetherin thename oftheworldsituationtherealso is wisdom
about the latter.It is a state of mind characterizedby a compulsivepreoccupationwithprediction,withthe eliminationof
"surprise"fromforeignaffairs.The shadow of Pearl Harbor
is projectedinto the mists of Bogota, and intelligencelooks
shamefacedover its failureto tell SecretaryMarshall the day
and hourat whicha revolutionwillbreakout in Colombia.The
course of eventsis conceivednot as somethingyou tryto influencebut as a tape all printedup inside a machine; and the
job of intelligenceis to tell the plannershow it reads. With
thisconceptionof intelligenceone does not,and, on the record
at least,cannotdistinguishbetweenwhatwe may call absolute
The latteris what the
predictionand contingentprediction.6
government
needs,especiallyin peacetime.
2. It is a stateof minddominatedby an essentiallybureaucraticconceptionof United States government,
and of the intelligenceproblem.On examiningMr. Kent's discussionof the
relationbetweenthe "producers"and "consumers"of intelligence (pp. 180-206), one recognizesat once that he is concernednot with the relationbetweenhis intelligenceexperts
and the electedofficialswho, as we hope,are still makingthe
actual decisionsabout our foreignpolicy,but withthe relation
betweenhis intelligenceexpertsand what it is now fashionable
to call the "policyplanners."The issue here is fundamental:
ifyou conceivetheintelligence
functionin thismanner,you are
excludingfromits purviewwhatthiswriterwould call its most
crucialaspect-i.e., thatwhichconcernsthe communicationto
6An example of absolute predictionwould be: "General DeGaulle will come to power
this day six months"; or "Japan will attack Pearl Harbor on x-day at y-hour."The contingentpredictionwould read: "The followingfactors,which can be influencedin such
and such a fashionby action fromthe outside, will determinewhether,and if so, when,
GeneralDeGaulle willcometo power."
550
WORLD POLITICS
the politicallyresponsiblelaymenof the knowledgewhich,to
use Mr. Lippmann'shappy phrase,determinesthe "pictures"
theyhave in theirheads of the worldto whichtheirdecisions
relate.The role of the intelligenceagenciesbecomes,on this
showing, that of mere research assistants to the George
Kennans.
Mr. Kent mightreplyto thiscriticism,as the presentwriter
has heardmanyof his formercolleaguesreply: This is how it
in factworks,and it would be "unrealistic"to discuss the intelligencefunctionin any otherterms.No doubt; but if so,
we do nottellourselvesthatoursis a book aboutthe"knowledge
upon whichour nation's foreignrelations,in war and peace,
mustrest" (p. vii). For the state of mind forwhichthatbook
calls is one whichexplicitlyrecognizeshow small a partof that
knowledgecan or shouldbe providedby a government
operation
(i.e., how large a part of it is and should be providedby professional scholarshipunder non-governmental
auspices, and
howlargea partofit is, whetherit shouldbe or not,providedhoweverbadly-by journalists), and how importantit is to
relateour thinkingabout that small partto our thinkingabout
therest.
3. It is a state of mind characterizedby a crassly empiricalconceptionof the researchprocessin the social sciences.
This, in view of the profoundcommitmentof our intelligence agencies to what we have called the "regional breakdown" of theirproblem,is not surprising.For, if it is regional
unitsyou are building,and it is social scientistsspecializedto
specificcountriesand areas you wish to staffthemwith,what
you end up withis an extremely
highpercentageof historians,
who withthebestwill in the worldcommunicateto the operavices (and virtues) of theirkind of retion the characteristic
oftheintelligence
search.The performance
functionaccordingly
becomesa matterof somehowkeepingone's head above water
in a tidal wave of documents,whose factual contentmust be
"processed"- i.e., in Mr. Kent's language,"analyzed," "evaluated," and exploitedas raw materialfor "hypotheses."The
emphasis,as we have alreadynoticed,is on prediction,which
againstthisbackgroundis necessarilyunderstoodas a matter
THE FUNCTION OF INTELLIGENCE
551
of projectingdiscernibleempiricaltrends into an indefinite
future.
Here also the issue is fundamental:an intelligenceoperation
built upon a conceptionof the researchprocess in the social
sciencesthatassignsdue weightto "theory"as it is understood
in economics and sociologyand, increasinglyone hopes, in
politics,would of coursebe a whollydifferent
affair.It would
recruita considerablepercentageof its personnelpreciselyfor
its theoreticaltrainingand accomplishments;it would free
themfromthe tidal wave of documents;it would enable them
to workunder conditionscalculated to encouragethought;it
would, above all, give them continuous and instantaneous
access (e.g.,by internationaltelephone)to the data that really
matter,namelythe raw data of the developingsituationin the
outsideworld. (The documents,havingcome fromabroad,are
as a matterof courseout ofdate.) This, clearly,is nottheplace
to discussthecomparativemeritsofthetwo conflicting
concepit
is
not
intendedto suggestthat
tions of social research,and
the intelligencefunctionshould be deliveredover entirelyto
that impliedin the foregoingsentences.The point is, rather,
thatcurrentplanningand organizationin the intelligencefield
ignoresone ofthetwo altogether.(It is, fromthispointofview,
thatMr. Kent,forall his numerousreferences
highlysignificant
to the social sciences and social scientists,never employsin
thatconnectionthewordstheoryand theorist.)
4. It is a state of mind characterizedby an uncriticaloptimismregardingthe supplyof the skillsupon whichthe effecoftheintelligence
functiondepends.Mr. Kent,
tiveperformance
in
this
reviewer's
quite rightly
opinion,looks primarilyto the
social sciencesforthese skills.Like most scholarswho served
withOSS duringthewar,however,he tendsto envisagethesolutionoftheintelligence
personnelproblemin termsof attracting
back to Washingtonthe social science scholars who, in the
courseoftheshake-upafterthewar,lostheartand turnedtheir
faces homeward-i.e., in termsof a returnto a Golden Age
situated,like Spain's, in the past. Or, to put the same thing
anotherway,he tendsto assume that"out there"-in the uniin theresearchinstitutes,
versities,
or,one supposes,temporarily
unemployed-themenyou need existand await theircall. One
552
WORLD POLITICS
findsin him,and in the intelligencefraternity
in Washington,
noneoftheanxietiesthatwouldcause mostresponsiblescholars
in the fieldsotherthan History(forwhichwe leave Mr. Kent,
as a professionalhistorian,entirelyfreeto speak) to say to
him: There could be no more dangerouserror.The supplyof
skills you have in mind is hopelesslyinadequate and, failing
drasticnationalaction-for example,a newManhattanProject,
but thistimein the social sciences-to increaseit, will remain
hopelesslyinadequate throughoutthe predictablefuture.It is
inadequate in two senses: We are, and are likelyto remain,
unableto releaseany significant
numberof men to government
intelligence
withoutwreckingtheteachingprogramsuponwhich
the futuresupply of such men must depend; and we have
gravemisgivings-theeconomistsperhapsless than the restof
us, but the economistsalso-about the abilityof our sciences
to supplythe sort of knowledgewhich,in Mr. Kent's phrase,
"our highlyplaced civiliansand militarymen must have."
It is, in fine,not a state of mindwhichis likelyto produce
thebook all of us have beenwaitingforsomeoneto writesince,
twenty-seven
years ago, Mr. Lippmann'sPublic Opinion first
directedour attentionto those picturesin decision-makers'
heads-and to the possibilityof our deliberatelyinfluencing
theiraccuracy.Nor is it a stateof mindlikelyto produceeven
a smallpartof thebook thatwould deal withthe government's
own "high-levelforeignpositiveintelligence."But the publication of StrategicIntelligencemay well make morepossiblethe
writingof thesetwo othercriticallyneededworks.
