THE INTELLIGENCE--POLICY RELATIONSHIP

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP93T01132R000100030024-5
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 22
• Document Creation Date: December 23, 2016
• Document Release Date: May 9, 2012
• Sequence Number: 24
• Content Type: MISC

Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93T01132R000100030f~024-5
From arm's length to love-hate
THE INTELLIGENCE-POLICY RELATIONSHIP`
Hans Heymann, Jr.
If we in intelligence were one day given three wishes, they would be
to know everything, to be believed when we spoke, and in such a way
to exercise an influence to the good in the matter of policy. But absent
the Good Fairy, we sometimes get the order of our unarticulated
wishes mixed. Often we feel the desire to influence policy and per-
haps just stop wishing there. This is too bad, because to wish simply
for influence can, and upon occasion does, get intelligence to the place
where it can have no influence whatever. By striving too hard in this
direction, intelligence may come to seem just another policy voice,
and an unwanted one. at that.
Sherman Kent "
In the catechism of the intelligence officer, the thesis that intelligence is
and should be strictly separate from policy is taken as axiomatic. It is as hal-
lowed in the theology of intelligence as the doctrine of the separation of
church and state is in the US Constitution. For much of our early history we
tended to view intelligence somewhat self-righteously as objective, disinter-
ested, and dispassionate, and to regard policy somewhat disdainfully as
slanted, adulterated, and politicized. And we strove mightily to maintain the
much-touted arm's length relationship with policy, believing that proximity to
policy would corrupt the independence of our intelligence judgments. Indeed,
legend has it that members of the Board of National Estimates of the 1950s
and 1960s systematically discouraged analysts and estimators from going
downtown to have lunch with policymakers, for fear that such exposure would
make them policy advocates and tempt them to serve power rather than truth.
Whatever the validity of this legend, such strictures were quite in keeping
with the traditional view of a proper intelligence-policy relationship. By en-
forcing this kind of rigorous separation, the old Board no doubt hoped to
protect the policy neutrality of intelligence; what it did, of course, was to
impose a splendid isolation upon intelligence that assured its eventual policy
irrelevance. The vanishing applause for its product coming from the policy
side caused intelligence to reexamine its assumptions, and a new, unconven-
tional wisdom came to be heard. Its message was that our faith in the arm's
length relationship was misplaced, that no such relationship really ever existed,
and that close ties between intelligence and policy are not only inevitable, but
essential if the policymaker's needs are to be served.
Adapted (min a Presentation at the -(:on(erence on Intelligence Pnlie and Pr(K?ess" at the United States
Air Force .Academy, ('edorado Sprinter. June 1984.
?? "Estimates and Influence,' originally presented ill London. SeptemlN'r I9tifi. snhseeiuentl puhliAled in
Foreign Service Journal. XI.VI. (April I969),
i Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5 ,,
 i
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
A new way of thinking about intelligence and policy began to emerge,
seeing the two communities as awkwardly entangled and intertwined in what
might be described as a competitive and often conflicting symbiotic relation-
ship. Thomas Hughes put it most aptly, when he spoke of the relationship "as
a two-way search: of intelligence in search of some policy to influence and of
policy in search of some intelligence for support." ' Suddenly, out is the com-
forting illusion that intelligence stands outside of and above the policy fray;
that it can load its analytic and estimative ammunition on its wagon and let
the wagon roll down in the general direction of the battle without worrying
where it will come to rest, whether the ammunition is of the right caliber and
how it will be used-to say nothing of whether someone might shoot it back.
And in is the less comfortable notion that intelligence, if it is to be at all
relevant to policy, is very much a participant in the battle; that it must be
attuned to the strategy and tactics being pursued; and that it is by no means
invulnerable to being seesawed and whiplashed in the sociopolitical tug of war
known as the policymaking process.
How this process unfolds in the real world and the intricate ways in which
intelligence interacts with it have, within the past decade, been the subject of
some first-rate analytic writing. Three contributions to this intelligence-foreign
policy literature are particularly worthy of note:
1. One is the observation, vividly illustrated by Thomas Hughes," that
the intelligence community is no more a unitary actor than the policy com-
munity; that it should be seen, rather, as a hydra-headed agglomoration of
competing institutions often at odds with each other, and not necessarily in
predictable patterns. Observing the budgetary, organizational, and substantive
struggles within this community, Hughes notes that
the cross-cutting complexities were striking: position disputes within
agencies, alliances shifting with issues, personal strayings from organ-
izational loyalties, hierarchical differences between superiors and sub-
ordinates, horizontal rather than vertical affinities, and much ad hoc
reaching for sustenance somewhere outside. Thus, while the struggles
within the intelligence community sometimes mirrored simultaneous
struggles in the larger policy community, they did so by no means
invariably and never symmetrically.
It should not be astonishing, therefore, to find that policymakers perceive the
intelligence process with as much ambivalence and suspicion as intelligence
makers perceive the policy process and that the interactions among them tend
to be contentious and rivalrous. To quote again from Hughes:
Tom Hughes deserves great credit for lining the first, and surely most art icnlate IC0n111CIAti1 111pp11 ng the
old cenps'elltlollal wlselnnl. His two Fare'ssell Ie'C'lllres as (Ieparling Director of the Rllre'all (If Intelligence and
Research of the l) arluu'nl of State in July 1969 contain the aisive quotation. The leetlures were
subse-ctuently reprinted in Thomas L. I lughes, The Fate of Facts in a World of Men-Foreign Polley and
Intelligence-Making (New York. Foreign Pelhc) .Asseg?iatlan, Headline Series No. 2:33. I)ecenllle'r 1976).
Thomas I. Ilughes, "Tire Passer to Siw'ak and tire Power to Listen in Thomas St Frank and others,
eels., Secrecy arid Foreign Policy (Ness York! Orford Oniversit) Press 1974), 1). 15
~ Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5 -
 I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132ROO0100030024-5
Relationship
Viewed from above by the ranking policymakers, the intelligence
community often seemed cumbersome, expensive, loquatious, prob-
ing, querulous, and at times axe-grinding. Viewed from below by the
intelligence experts, the policy community often seemed determined
to ignore evidence plainly before it-or (even worse) to mistake the
intelligence managers for the experts. Viewed from in between at the
intelligence-policy interface, it looked like controlled chaos-and not
surprisingly, for here was where means and ends were brokered, ju-
risdictional rivalries compromised, contentious controversies
delineated.'
2. Another is the thesis, persuasively argued by Richard Betts," that
intelligence failures, so-called, are more often than not policy failures; or to
put it more gently, that it is usually impossible to disentangle intelligence
failures from policy failures, since (intelligence) analysis and (policy) decisions
are interactive rather than sequential processes. Betts sees the intelligence role
as seeking "to extract certainty from uncertainty and to facilitate coherent
decision in an incoherent environment." In seeking to reduce uncertainty,
intelligence is often forced to extrapolate from evidence that is riddled with
ambiguities. Inability to resolve these ambiguities leads to intelligence prod-
ucts that oversimplify reality and fail to alert the policy consumers of these
products to the dangers that lurk within the ambiguities. Critical mistakes are
consequently made by policymakers who, faced with ambiguities, will substi-
tute wishful thinking and their own premises and preconceptions for the as-
sessments of professional analysts. As Betts puts it:
Because it is the job of decision-makers to decide, they cannot react to
ambiguity by deferring judgment.... When a welter of fragmentary
evidence offers support to various interpretations, ambiguity is ex-
ploited by wishfulness. The greater the ambiguity, the greater the
impact of preconceptions."'
3. A third example is the recent revelation by a former Chief of Israeli
Military Intelligence and Advisor to the Israeli Prime Minister, ' ehoshafat
Harkabi,"" that the tensioned and ambivalent relationship between intelli-
gence and policy is not a uniquely American phenomenon.
These dilemmas and foibles of the intelligence-policy interface are hardly
novel or startling to seasoned intelligence practitioners, especially those senior
officers charged with "brokering" the intelligence-policy relationship-the
communicators and interactors who reside in the twilight zone between intel-
ligence and policy. For them, this is familiar terrain. As managers and stimu-
lators of intelligence production, they know with what difficulty a crisp, lucid
analytic product is extracted from a dissentious community; as participants in
Idem. P19
Richard K. Betts. "Analysis, War and Decision Why Intelligence Failures are Inevitable." World
Politics. XXXI (October 1978)
"' Idem. p. 70
"" Yehoshalat Harkabi. "The Intelligence-I'nlicymaker "f:urglc, in Tlu Jeru,a!enr (hmrlerlu. \nmlw?r
30, Winter 1984. (The article was reprinted in the Son finer 1984 issue of Studies in Inlelligenee. plume 28.
Number 2 )
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132ROO0100030024-5 -F
 I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
the interagency policy process, they observe with what ease that product can
be selectively utilized, tendentiously summarized, or subtly denigrated. But
for these privileged practitioners who move readily from the world of analysis
to the world of action, familiarity with policy does not breed contempt.
Rather, an appreciation of the murky and frenetic policy environment tends to
evoke a certain sympathy for the policymakers' plight.
Such knowledgeable, involved practitioners, however, represent only a
very small fraction of the intelligence population. The vast majority of that
population-collectors, operators and analysts-is essentially isolated from the
hurly-burly of the policy process. The intelligence services at large, therefore,
are often mystified and frustrated by the policymakers' perennial unhappiness
with their product. Given this puzzlement, it seems worthwhile to try to delve
a little more deeply into the reasons for the unhappiness.
The View from the Bridge
It should be clear from what has been said that policy does not speak with
a single voice..Policies have multiple authors. The numerous players who take
part in policy formulation differ in temperament, education and experience, as
well as in personal and institutional loyalties. Their attitudes toward intelli-
gence, therefore, and their propensity to accept or reject its assessments will
also vary widely. Nevertheless, although generalizations are always hazardous,
we can discern some common attributes and concerns of volicymakers, espe-
cially the "national security principals" '-the key players at the highest levels
of government-that predispose them to react to intelligence offerings in pre-
dictable ways.
First, it is well to remember that the key decision makers are political
leaders who have risen to their positions by being decisive, aggressive, and
self-confident rather than reflective, introspective, and self-doubting. They at-
tribute their success at least in part to their tried and proven ways of thinking,
their simplified models and paradigms that explain to them what makes the
world go 'round. They often regard themselves as their own best analysts and
hence tend to be distrustful of the untested and often counterintuitive judg-
ments of the intelligence professionals.
Second, they have a strong vested interest in the success of their policies
and will, therefore, be disproportionately receptive to intelligence that "sup-
ports" these policies. They bear the burdens of great responsibility and find
themselves perpetually embattled with a host of critics, competitors, and op-
ponents, all eagerly looking for chinks in their armor. They thrive on optimists
and boosters, but encounter mostly alarmists and carping critics.
Festooned in this way, and operating in so hostile an environment, these
highest level consumers of intelligence can hardly be blamed for responding to
its product with something less than boundless enthusiasm. In fact, it can be
documented that every President since Eisenhower, and virtually every Sec-
retary of State since Acheson, has expressed dissatisfaction and irritation with
They include. at a rniniminn, the President, Vice President, National Serurit' .Adsisor. Secretary of
State, and Secretary of Ikdrnsr
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 i
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93T01132R000100030024-5
'II'I
Relationship
intelligence analysis, either in his memoirs or in public or semipublic state-
ments. The best-remembered and widely quoted expostulation was reported to
have been delivered by Lyndon Johnson to his Director of Central Intelligence
at a White House dinner:
Policymaking is like milking a fat cow. You see the milk coming out,
you press more and the milk bubbles and flows, and just as the bucket
is full, the cow with its tail whips the bucket and all is spilled. That's
what CIA does to policymaking.'
Is intelligence at fault for creating this unhappiness? Should it alter its
ways to court greater popularity? Or is the problem integral and endemic to
the intelligence-policy relationship? The answers to these questions may be-
come clearer as we look at some of the concrete ways in which the frictions
arise.
Why Policy Resents Intelligence: Five Ways to be Unpopular
Presidents and their senior advisors will be unhappy with intelligence
when it is not supportive of their policies. They will feel particularly frustrated
when:
1. Intelligence fails to reduce uncertainty-
Policymakers operate under a burden of pervasive uncertainty, much of
it threatening the viability of their policies. They are forever hopeful that
someone will relieve them of some of this uncertainty, and so they look to
intelligence for what common sense tells them should be reserved to augury
and divination. Forecasting, to be sure, is the life's blood of the intelligence
estimator. But there is a world of difference between a forecast (an analytic
judgment resting on carefully defined assumptions) and an oracular prophecy
(secured by divine inspiration). Unfortunately, much of what is expected of
intelligence by policymakers lies in this latter realm.
A good example is the perennial complaint that intelligence failed to
predict a coup d'etat-a coercive regime change or palace uprising-but, of
course, a coup is typically a conspiratorial act that depends for its success on
preservation of absolute secrecy. If intelligence gets wind of such an event, it
means that secrecy has been compromised and the coup is almost certain to
fail.
Intelligence forecasting is actually done quite respectably by the commu-
nity, and can be of real value to the thoughtful policy analyst. When it stays
within its legitimate bounds of identifying and illuminating alternative out-
comes, assigning subjective probabilities to them, and exploring their possible
implications for US policy, the decision maker is well served. But he will rarely
think so. For such a forecast, rather than narrowing uncertainty, will make
him aware of the full range of uncertainty he faces and make his calculations
harder rather than easier. Indeed, much intelligence estimation is and must be
Ilrnry Brandon. The Retreat of An?nean Power (Garden City. N 'i : Doubleday, 1973). P. 103
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93T01132R000100030024-5
 i
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
of this nature. Precisely because it seeks to reflect complex reality, its product
often renders the harassed decision maker's life more difficult.
2. Intelligence restricts their options-
Every new administration comes into office with a national security
agenda of its own, bent upon putting its mark on the nation's foreign policy. It
believes that a significant shift in that policy is both desirable and possible. It
will encounter a foreign policy bureaucracy (including intelligence) that be-
lieves it is neither. Intelligence professionals will greet the administration's
new policy initiatives with cogent analyses showing how vigorously allies will
oppose these new policies, how resolutely neutrals will pervert them to their
own ends, and how effectively adversaries will blunt them. At every step, it
will appear to the policy leaders that intelligence fights them, seeks to fence
them in, and, indeed, helps them fail.
And the pattern persists. As the policy leadership begins to face unex-
pected foreign challenges, its quick responses will often be met with more
intelligence assessments that seem to be saying "it didn't work" or "it will
almost certai->Jy not succeed." The decision makers will conclude that intelli-
gence not only constricts their room for maneuver, but also arms their political
opponents. Worst of all, it constantly and annoyingly reminds them of their
limited capacity to influence events. No matter how well the' interaction may
serve the interests of sound policy, there is no question that it builds tension
between the two sides.
In these encounters, we should acknowledge that intelligence does not
always "know better." There are times when intelligence is unaware that
stated objectives are not the real objectives of policy, and will leave out of its
analysis elements of the picture that may be important to the decision makers.
Presidents paint upon a canvas far broader than the particular segments on
which intelligence tends to focus. Its assessments, therefore, may be quite' valid
for those segments, but may miss broader considerations that Presidents care
about.
A vivid example is provided by the Carter Administration's proposal to
impose sanctions-including a grain embargo-on the USSR, in response to
the Soviet invasion of Afghanistan. The stated objective was to penalize the
offender by imposing political and economic costs on Min. When intelligence
was asked to assess the potential impact of the sanctions package, it responded
with a judgment, the thrust of which was that the sanctions package would not
be an effective instrument. Absent solid participation by our allies, sanctions
would do no serious damage to the Soviet economy nor impair the leaderships
objectives in any significant way. Not surprisingly, President Carter gave the
assessment a rather frigid reception, but its negative judgments turned out not
to be a decisive factor in his calculus. From the ('resident's perspective, the
sanctions package wvas just right. lie considered a highly visible response to
Afghanistan as imperative, but it had to be low-risk. A military undertaking
was ruled out as far too hazardous. Inaction was ruled out, because it would be
read in the rest of the world as a signal of US irresolution and condonement.
The sanctions, though unsatisfying in terms of direct effects, would convey a
strong signal of disapprobation and censure, without engendering worrisome
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
consequences. It would satisfy the popular need to express the nation's sense of
outrage and would portray the President as willing to take the political heat of
angering an important domestic constituency-the farmers-for the sake of a
foreign issue of principle.
It goes almost without saying that intelligence could not then, and cannot
ever, be expected to take such considerations into account.
3. Intelligence undercuts their policies-
Administrations have often found intelligence analyses appearing at times
and in ways unhelpful to the pursuit of policies on which they had embarked.
This can happen in two ways: (1) Through a genuine and protracted diver-
gence of intelligence judgments from publicly stated Administration views of a
given situation, and (2) Through fortuity or inadvertence. An example of the
first phenomenon was provided by the stubborn independence displayed by
the intelligence community in the early phases of the Vietnam escalation in
1964-65, when its national estimates consistently offered up a far more pessi-
mistic assessment of North Vietnamese staying power than was reflected in the
Johnson Administration's public assertions. While this divergence between in-
telligence and policy did not become public knowledge until the appearance
of the Pentagon Papers in 1971, the mid-1960s intelligence performance
evoked considerable disquiet and chagrin among policy insiders at the time.
The days of such protracted differences of view between intelligence and
policy are probably over. In the intelligence-policy environment of the 1980s,
it seems highly unlikely that a divergence of assessment could be sustained for
very long. Congressional oversight and its intimate access to intelligence anal-
ysis would bring any significant disparities quickly to the surface and thus
cause them to be resolved.
The other cause, policy-undercutting by fortuity and inadvertence, is
more likely to survive, as it is a matter of human frailty. Sometimes it is merely
a question of miserable timing-as in the classic case of the intelligence reas-
sessment of North Korean military forces that credited then) with substantially
greater capabilities than had been previously appreciated. The estimate was
fine, but it just happened to ''hit the street'' within a week of President Carter's
announcement of his controversial decision to begin withdrawal of US forces
from South Korea. A pure coincidence, but it caused understandable conster-
nation.
At other times it is a matter of inattention-as in the so-called discovery
of the Soviet brigade in Cuba which, it turned out later, had been there, in one-
form or another, all along, but had simply been lost sight of. Issues of this kind,
seemingly unimportant, can suddenly escalate into heated public controversy
and make life difficult for the policy leaders. However minor the transgression,
they will regard intelligence less fondly.
4. Intelligence provokes public controversy-
From time to time, routine differences within the community over how to
interpret ambiguous intelligence evidence turns into heated, and perhaps even
acrimonious debate. When the competing interpretations clearly affect impor-
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
tant policy issues, the internal controversy can easily spill out into the public
arena. In the 1950s and 1960s, when what transpired in the world of intelli-
gence remained largely opaque, such disputes could be easily contained within
the Executive Branch. In more recent times, with the progressive "opening
up of intelligence through the Congress and the media, and through its more
visible involvement with policy, a disputation within the community is soon
drawn into and exploited by the public debate, often in ways that make life
more difficult for the national security policymaker.
Examples of policy-relevant debates that have been stimulated or inten-
sified by intelligence controversy come quickly to mind:
- Whether the Tupolev Backfire bomber is an intermediate-range or an
intercontinental-capable bomber;
- Whether extensive Soviet civil defense preparations add up to en-
hanced "survivability" for Soviet society;
- How significantly Western technology contributes to the growth of the
Soviet economy and its military power;
- Whether Western calculations of Soviet military spending adequately
reflect the real size and burden of Soviet defense;
- To what extent the Soviet natural gas pipeline will aggravate Western
Europe's dependence on imported energy.
This brief sampling is probably sufficient to suggest that the issues in
dispute often bear on strategic, budgetary, arms control, or economic police
decisions importmit to an Administration's overall strategy. To the extent that
intelligence controvers~.helps arm the opposition in such disputes, its contri-
bution is not exactly appreciated.
5.- Intelligence fails to persuade-
Ever,since John F. Kennedy's tour de force in unveiling photographic
intelligence on the presence of Soviet missiles in Cuba to a hushed 11N audi-
ence, successive.atlin inistrat iorrs have sought to emulate that feat. Though the
results have been mixed at lest, hope springs eternal that a release of intelli-
gence findings or a'public display of exotic evidence will enlighten an unin-
formed or risinfgrmed public, win over a cynical journalist, or convince a
skeptical congressman. At one time limited to an occasional State Department
White Paper and a private briefing here and there, the intelligence product
now finds its way into the public domain through more and more channels and
in ever greater volume-most of it, of course, at the instigation and under the
aegis of the policy community. It moves through such vehicles as press confer-
ences, media briefings and backgrounders, testimony on the Hill, formal Ile-
ports to Congress, and official glossy publications.
In a general way, this sea change in public access to intelligence has
undoubtedly had its beneficial impact on public understanding of often com-
plex and murky situations. It is far more questionable, however, whether in-
telligence can be used effectively as an instrument of public persuasion;
whether the marshalling of intelligence evidence on one side or another of a
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
sharply debated issue ever succeeds in gaining solid converts. In a tactical
situation, say, when a heated debate moves toward a crucial vote, a well-
focused, lucid intelligence briefing can often sway a wavering agnostic and
stiffen an irresolute supporter. But the record suggests that the conversion will
not stick, that the gnawing doubts soon return.
Reasons for this phenomenon are not hard to find:
- Time was when public disclosure of intelligence was a rare and nota-
ble event that summoned tip an aura of mystery and miracle, endow-
ing the product with uncommon authority. That is no more. As dis-
closure became ever more routine, the gloss wore off, and an inevita-
ble "debasement of the currency" set in.
- Intelligence assessments, when lifted out of their context, fuzzed and
diluted ("sanitized") to protect sources and methods, lose much of
their authenticity. To the intelligence professional who has built his
mosaic from a welter of carefully evaluated raw data, often accumu-
lated over years, the evidence may be totally compelling. To a public
audience, coming to t9e issue cold and exposed only to the sanitized
version, the evidence will often seem ambiguous and the judgments
inadequately supported.
- Intelligence evidence is brought into public play often in situations of
deep controversy, where the contention usually is not over observable
facts, but over points of principle. The physical things that intelli-
gence is best at recording are often not much help in settling points of
principle. Central America offers a good example: Divergent views of
that threat center on the conceptual question of whether the revolu-
tionary situation in El Salvador is fundamentally endogenous, i.e.,
rooted in and fueled by internal, historic forces, or exogenous, i.e.,
externally stimulated and sustained. That conceptual issue cannot be
resolved by displays of intelligence evidence, however persuasive, that
Soviet arms do indeed flow through Nicaraguan ports to the Salvado-
ran rebels.
The impact that intelligence can have on public perceptions is further
constrained by the understandable tendency of people to reject bad
news-what social psychologists used to call "cognitive dissonance.'
Many of the issues on which intelligence is brought to bear publicly do
indeed have unhappy implications. Acceptance of the had news
means having to draw costly, risky, or generally unsettling conse-
quences. A classic example is the case of "Yellow Rain," the discovery
of lethal toxins being used under Soviet tutelage in Southeast Asia and
Afghanistan. In spite of the overwhelming weight of confirmatory
evidence accumulated over eight years, extensively published, briefed
and shared worldwide, the findings continue to be challenged and
contested, sometimes with offerings of bizarre scientific counter-
explanations that defy common sense. The extreme reluctance to ac-
cept the evidence at face value cannot be attributed simply to the fact
that intelligence can never meet the rigorous laboratory standards for
evidence that scientists like to insist upon. The explanation for the
itf 3
Declassified in Part Sanitized Copy Approved for Release 2012/05/09 : CIA RDP93T01132R000100030024 5
 I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
Relationship
continued questioning must surely lie in the unpleasantness of the
implications, insofar as they seem to raise doubts about the viability of
arms control agreements.
In sum, for all the reasons enumerated above, policy leaders are bound to
develop a rather ambivalent view of the support they can hope to get from
their intelligence community. From what has been said, it should be clear that
the resulting "love-hate" relationship is endemic to the situation and that there
is not much that intelligence can do, or should do, to alter it. Indeed, a greater
effort to "serve policy well" could lead to even greater ambivalence and dis-
cord on the part of those we seek to serve. Which takes us back to Sherman
Kent's admonition in the leitmotif at the beginning of this paper:
By striving too hard in this direction, intelligence may come to seem
just another policy voice, and an unwanted one at that.
66
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
offprint
THE
JERUSALEM
QUARTERLY
30vinter 1984
~ Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
The Intelligence-Policymaker
Tangle
Yehoshafat Harkabi
The publication of the Kahan Commission report, with its
indictment of the IDF Intelligence Chief, reopened the debate on
the relationship between the Intelligence services and their
clientele, the policymakers. The formal description of how
Intelligence supplies the policymakers with information and
evaluations as a for molding policy 1s slms .'.:c and
incomplete. The relations between these echelons are complex and
WOwn-ridden, as evident when one looks bey onrmof al
hierz~structures and processes at the influence of informal
relations on the workings of administrative bodies.
The study of the functioning of Intelligence services, which has
greatly developed in recent years, does not focus only on how the
Intelligence service produces its reports - information gathering
and analysis. It also deals with the crucial area where the
usefulness of the service is put to trial: namely, the transmittal of
the Intelligence service's product to the policymaking bodies, the
'interface' between Intelligence and policy.
Intelligence is not an autonomous operation whose raison d'ttre
Has in itselL Intelligence activities depend on having a clientpIw rn-
ser_ve. However, its clients are not n v recentive t
Intelligence, for what they often look for is not so much data on the
basis o w ' to shape policy but rather support for m-e-formed
political and ideological cone ns. The Intelligence service finds
? Y. Barkrabi FS Haattr PrOfQWW of Inttnadmal Itdadom and Middii Eastern
Smdio at tbs Hebrew University of .Jenasaism. He unred as ChW of Jrasa
Military IntalilIgaoa (1955-19591 wd as Advise for Intallijeam to bads
Prima Minister I1977). Thin ardda was arlgtnally pubbabed kin Hebrew) in tba
BWIedn of the Hebrew Udv.*y Facuky of Sodal Scenes
(M Jwumb G wt *y. Number 30. Winer 19649
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93T01132R000100030024-5
itself in difficult straits, for it is aware that many of its efforts will
not be utilized or appreciated, and the use made of its assessments
and reports will differ from its expectations. Matters get worse the
more ideologically motivated is the regime, for then policy is d
more on the basis of ideological inputs than on the basis of
Intelligence reportings on reality, which to the extent that they
contradict the ideology may be discarded, and the Intelligence
service ends up frustrated.
Policy can be judged according to the extent of its 'sensitivity' to
Intelligence - will it change if a certain evaluation requires such a
change? As a concrete example, what intelligence reporting could
V induce a change in Israel's present policy on Judea and Samaria?
U Does the rigidity of a political position make
r G r ' In ence? An ideological regime may revel in exotic covert
"JiM A' operations, encourage them, and still keep Intelligence
evaluations at arm's length. Nor is there simple transitive
between the quality of the Ia as of policy.
te ence no guarantee of good whcy and vice versa.
Even Intelligence portrY reality correctly and its evaluations
were accepted, policy, also includes other components, such as
goals, objectives, and assumptions about causal relations between
policy and outcomes, which are not necessarily Intelligence
products.
Policymakers too have their legitimate complaints against
Intelligence, UM-itsupplies them with a motley catchall
collection of information, containing everything but what
a a e timeat it expresses itself in equivocal and
reserved language that leaves them ; or still worse, that
its evaluations not reliable and excessively opinionated.
The Intelligence service should enter the policymaking process
twice: first, by providing data and assessments of the situation,
which will contribute to the shaping of policy; and secondly,
after the policy has been formulated, Intelligence should also
evaluate theely reactions of adversaries and third parties to
that policy and its success or failure. Owever, it often happens
that statesmen ;ZM from seeking the Intelligence service's
opinion on this, for basic reasons. For by making such a request
of the Intelligence elevate it to the position of
policy. Thus a taugle is created whereby the In arm
which is a body. subordinate becomes an arbiter a kin of
supervisor over its masters. What is more, the statesmen may
harbor suspicions that the Intelligence services may cite the
difficu= and wweaknesses of their policy. Not fortuitously has
the Inte
negativistic', a discourag-
ing factor, for it may tend more to poin w d wbacks than call
attention to opportunities. Hence, Kissinger stigmatised the
Intelligence service for pushing towards 'immobilism'.
The Intelligence service itself will not volunteer for the role of
v
/S 5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93T01132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
policy-monitoring, fearing that it may mar its relations with its
superiors. the policymakers, and may cause it to collide with
conceptions sacred to than, or with their dreams. For example,
once the idea of getting the Phalanges into action in Beirut
became a desire, almost an obsession, among the Israeli policy.
makers. a presentation of the hazards of such a policy placed the
Intelligence in an uncomfortable position. Similarly, it may be
supposed that an organization like the KGB would be inhibited
from presenting evaluations that clash with Marxism and with
Soviet policy. The Intelligence service, therefore, will not volunteer
to serve as a traffic signal light flashing red and green alternately to
the advancing policy carriage.
There is an exaggerated tendency to present the Intelligence
service as if it were an aninstitutioa for the sounding of tocsins. The
Intellige ce service is~pnm l institution for the provision of
information which is mlead to knowledge and un
and is not m y a warming mechThe principal
line of defense against surprises is'understanding', not 'warning'.
Warning is in order in times of emergency and before the onset of
calamity - but those are few and far between. And if indeed the
Intelligence service is expected to warn about impending dangers
stem ming o sic ion initiated the enenr , it is hard to ex
that it aL w1
an institution that warns unous
.aLbe outcomes of our own po cy, or our home-made surprises. That is
an important difference, which it seems, the Kahan Commission
was not alive to. Certainly the Intelligence service would do well
were it itself, on its own, to point out the probable consequences of
policy, but it is advisable that the chiefs-of-state understand the
Intelligence's reluctance to become overseers, august or meek, on
their policy and address it with explicit queries, as an invitation for
the Intelligence's intervention. People are not aware of how
complicated and difficult is the Intelligence service's work of
collecting, analyzing and evaluating information. The Intelligence
service will not willingly seek out additional troubles for itself. It is
not sheer squeamishness
In short, the Intelligence service is an institution more for the
giving of answers than for sounding warnings, especially about
z-liicy. It is the task of the leaders to put questions to it, an
they o not ask let it not be said that they assumed that the service
would inform them of its own accord. True, since the Intelligence
service provides reports on an ongoing routine basis, the
impression might be formed that it offers its opinions on every
relevant issue automatically. That is an error, and it would have
been helpful to Israeli policymaking had the Kahan Commission
been alert to it and drawn attention to these aspects.
It may be argued that the Intelligence service does not- fully
discharge its duty by providing the policymakers with information
and assessments, and that precisely because its product may be
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
critical for policy, the service must we to it that its reports are
property understood. However, the Intelligence service will
refrain from testing whether the policymakers have properly
understood the material that has been passed on to it, that it will
shrink from taking the role of a pedantic teacher correcting
misunderstandings an the part of the policymakers. Indeed, a
pretension on the part of the Intelligence service to be the
policymakers' 'mentor' is liable to be counterproductive.
It may come to pass that senior Intelligence functionaries may
differ with the policymakers' policy. Their critical stance vis-d-vis
the adopted policy may be based on an evaluation of the historical
trend, yet they may not be able to adduce factual proof for their
position. In most instances, the error of the policy line emerges in a
clearly decisive way only in the long range, for the feedback circuit
in such matters is slow. In the short range a mistaken policy line
does not necessarily entail outcomes that refute it. It may then ap-
pear to the policymakers that their course is succeeding. and that
the facts abet it. Hence. the Intelligence service cannot use such
facts to validate its criticism of policy, for in a confrontation with
the policymakers it can avail itself only of facts whose message is
clear and evident; and thus its assessments of long-range trends
may not, in such cases. be serviceable for it. The Intelligence
criticism of policy may then appear as arbitrary and irksome, even
as stemming from lack of sympathy towards the policymakers
themselves. Thus, here too, the Intelligence service may choose to
withhold counseL Later, when the error of the policy becomes
clear, there will be those who will protest that the Intelligence
service should have warned in time about the mistaken policy, and
an inquiry commission may even find the service culpable.
The Intelligence service is aware that it treads on precarious
ground and is liable to be singled out for blame in any error, since in
every political or military decision there is an assumption on the
situation or a component of knowledge, the lack of which can be
imputed to Intelligence. For instance, a commander can decide to
outflank and attack from the left, not because the Intelligence
service advised him to do so. Were decisions based only an
Intelligence data, decisions and policy would simply 'follow' from
it and there would be no need for policymakers. If his attack fails,
the commander can shift the blame to Intelligence by contending
that it did not warn him that the left flank was strong and could not
be gushed. Any military action can fail, either because our troops
were not good or because the enemy's troops were. There is no
institutionalized body whose job is to evaluate our troops, and thus
it is easy to transfer the blame for a military failure to Intelligence,
which, as it were, slighted the enemy's ability. The Tatelli
ence
service has been frequently described as the staff's 'w ' ' boy'.
In many fields a human error of evaluation or judgement is
Thus,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5 ,,,
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
considered as agenuating ci cumrtances; however, it is the fate of
Intelligence that ertvr devaluation is always enshrined in Is
bill d' sot.:Was the popular saying has it that 'to err is
human';c an aLwa superhuman_trafection In apected of
Intelligaocs W. ana . living. among our own people with no
W Siam of aoosssion to knowledge and still we stunned -by
domestic political develcpmonts. But if Intelligence does not mo-
cesdulky forecast a political dent in a foreign counts y,
brows are wrinkled: how is that possible? What inefdencyt
After the oe service has failed in reporting an some
information err evaluation, it is likely to take out insurance for itself
by way of enlarging the quantity of its reports and including
everything in them, so that it may not be found wanting in
reporting. It will than tloo the policymakers with Iatell aooe
r gore. However, over-repeating may be detrimental for the
Intelligence service Mumma as important items maybe lost in the
multitude of the lees important and trivial ones. This, what will
eventually prove important does not always immediately catch the
eye. The statesman may be able to defend themselves against
overabundance of Intelligence reporting, by employing an aide to
sift and summarise the material for them . Such an aidefilstherole
of Intailigance waiter' who marks for his superior what is worth
his attemion. What is significant in the eyes of the Intelligence
waiter' and the Inteligenos service is not necessarily identical.
Despite thevitakrole such anasdstaotfulSLforhismaster, such an
intermediary arrangement may also complicate things, for the
Intelligence service does not know what information has reached
the policymakers, of who they an aware and of who not.
Furthermore, tateamen may tend to look or rather browse over
Intelligence material, often at the and of an exhausting day when
they are fatigued or half drowsy.
Presumably, it In good that the chiiaf the Intelliaeaoe service be
an dose terms with the policymakers and have their trier.
E ft=, such boom companionship too has its draw
'hue, the more be is a part of the inner Byzantine court that
develo as a matter course state Ma'greatarishb_
influence: and his
indepe and gradually stceumbs to the
2299MINIL,7HO is than unable to detech
festivities of pacymakiOg just like the other sal[
gratified members tithe courtwho bask Jim their connections with
power. Thomas Hughes urged that Intelligence should give the
pokTmakers ' utmot support with utmost reservation. That sure-
ly is no simple combination.
In its reports the Intelligence service must differ lute between Y
state? of fact and evaluanoos concerning the Future, w
are alw a menu of cogdect re. It is an error to present as
evaluation of future fiends as if they were facts. The desire of the
A
Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/05/09: CIA-RDP93TO1132R000100030024-5
I>
IatelUgance pmplo to present a cb nvt unqualified oploion is
caenmsndaW bgtit may mjdmd them to preseirt thou hypotheses .
about,: tba4tlrajm.#