IS THE CIA'S ANALYSIS ANY GOOD?

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP99-01448R000301300003-9
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 2
• Document Creation Date: December 23, 2016
• Document Release Date: May 22, 2013
• Sequence Number: 3
• Publication Date: December 12, 1984
• Content Type: OPEN SOURCE

Declassified and Approved For Release 2013/05/22 : CIA-RDP99-01448R000301300003-9
art Wt_E APPEARED _
CIA r - Z
?Ei;_.
Robert M. Gates I
WASHINGTON POST
12 December 1984
Is the CIA's Analysis: Any food?
Yes.
The Central Intelligence Agency was
created to provide comprehensive, all-source
collection and analysis of information so that
we might prevent strategic surprises like
Pearl Harbor and be forewarned of other
developments adverse to American interests.
Granted, the effort is immense and complex.
Recent press accounts prompt some to ask: Is
it any good? Is it honest and objective.)
There is little question that maintaining the
quality of CIA assessments became much
more difficult in the late 1960s and the
1970s. Collection capabilities declined. Our
analytical effort on the Third World had been
significantly reduced by the early 1970s-
just when problems there were multiplying.
By 1980 the number of analysts working on
the Soviet economy (including defense indus-
tries) had declined from over 300 to fewer
than 50. There was little money for analysts
to travel abroad or to meet with nongovern-
ment experts at home. Many academics were
unwilling to talk to us and share ideas. From
1973 to 1977, moreover, CIA had five differ-
ent directors, and from 1975 to January,
1982, there were six chiefs of the analysis di-
rectorate-the Directorate of Intelligence.
Much has changed in the past five years.
The resource picture began to improve in
1979, thanks initially to the House and Senate
oversight committees. We have since made
impressive strides toward rebuilding the
corps of analysts. New resources for the en-
tire intelligence community have greatly im-
proved the collection of information across
the board. We have also undertaken sweeping
measures to improve the quality of analysis.
The directorate of intelligence was reor-
ganized in 1981 to bring political, economic
and military experts together in regional of-
fices. We have dramatically expanded our
contacts outside government, drawing on an
extraordinary number of experts in universi-
ties, think tanks and business for information
and ideas. We require all CIA analysts to have
outside training every two years.
CIA has strengthened longer-term analytical
research, long put at risk by the pressures of
day-to-day reporting. In the first nine months of
this year we issued some 700 research studies
for nearly every department of government.
For the first time there are adequate funds for
analysts to travel and work overseas as well as
to consult with again-cooperative academic and
other experts at home.
CIA assessments now are subjected to
.more rigorous internal review than ever be-
fore. Every manager at every level reviews
all substantive assessments that come out of
his organization. We often offer drafts for
comment (though not consent) to senior mili-
tary commanders, embassies and experts in
other agencies. Many of our assessments are
reviewed by nongovernment experts.
We not only offer our best estimate of what
will happen in a given situation but also in-
form our readers of other possible though less
likely outcomes-and the implications of
each. I cannot say this approach would have
enabled us to predict the fall of the shah in
1978-79, but I believe that that outcome now
would certainly be addressed as a possibility.
We are more candid now with our readers
about the level of our confidence in our judg-,
ments and the reliability of our sources. We
also make more of an effort to lay out our evi-
dence. Using the example of the fall of the
shah, under present practice we would have
acknowledged the paucity of information on
internal Iranian affairs and the self-serving
nature of some of our sources.
We now evaluate past CIA assessments and
national estimates to see how they have held
up over time. The directorate of intelligence
has for the first time its own independent
evaluation staff. We voluntarily share these
evaluations with the House and Senate over-
sight committees.
We organize special task forces of agency ex-
perts and outside specialists to do competitive
analysis and to ensure we are examining all
aspects of key problems. We submit our work
on important issues, such as the Soviet econ-
omy, to panels of outside experts for scrutiny.
Finally, the skill and dedication of analysts in
CIA and elsewhere in the intelligence com-
munity are exceptional-perhaps never better.
While some of the criticism in the press of
our capabilities and acumen is justified, most
of it is grossly inaccurate. I urge the reader to
consider the access and motives of sources of
criticism-and to be alert to later retractions.
Meanwhile, I have hundreds of letters, cables
and messages, from the president on down,
commending our work. Various news organi-
zations report that policy-makers and mem-
bers of Congress acknowledge that the qual-
ity of assessments has improved markedly.
CIA was created in part to ensure that in-
telligence assessments would be prepared by
people with no stake in approval of weapons
programs, defense budgets or particular poli-
des. Perhaps the strongest cultural trait coin'
mon to all CIA analysts is a very deep sensi-
tivity to the dangdrs of ,politicization. Indeed,
sometimes we bend perhaps too far toward an
adversarial relationship with policy-makers to
avoid even the appearance of being suborned.
There is no question that policy-makers
have always been intensely interested in the
outcome of our assessments, especially on
contentious issues. If they were not, it would
mean we were working on the wrong prob-
lems or were irrelevant. Beyond our natural,
visceral independence-contrariness, some
would say-a num-
ber of safeguards exist:
Approval of CIA's assessments
rests with intelligence professionals. I
have been with CIA nearly 20 years. I am
the final approving official for all of CIA's
daily production of current intelligence that
goes to the president and senior government
officials. The director of Central Intelligence
(DCI) first sees it at the same time as the
policy-maker. I also approve all longer-range
assessments.
Our assessments go to the two congressional
oversight committees. I am confident they
would not hesitate to act promptly if they de-
tected a policy slant In addition, both Foreign
Relations, Armed Services and Appropriations
committees receive a great number of our as-
sessments.
A variety of other groups, both independent
of CIA (the president's Foreign Intelligence Ad-
visory Board), and inside the Agency. also
evaluate CIA assessments and estimates:
Directors of CIA have always played an ac-
tive role in the preparation and approval of
national estimates, which are produced by the
entire intelligence community. Similarly, our
directors always have had strong views on the
major substantive issues we analyze. John
McCone, President Kennedy's DCI, believed
the Soviets would send missiles to Cuba in
1962 long before the intelligence analysts
agreed. However, national estimates also are
reviewed by the heads of a dozen other intelli-
gence organizations. The estimate that re-
cently was alleged, in the press, to be slanted
went through many drafts and even then
Cottnued
Declassified and Approved For Release 2013/05/22 : CIA-RDP99-01448R000301300003-9
 Declassified and Approved For Release 2013/05/22 : CIA-RDP99-01448R000301300003-9
Z
nearly half the community's intelligence agen-
cies dissented-and the dissent was spelled
out on the fast page.
Finally, perhaps the best guardians of the
integrity of the process are the caliber and
honesty of the people involved. We are not
cowards. We present assessments unwelcome
to policy-makers day in, day out on a broad
range of issues, and we have for a long time. I
believe most policy-rakers would attest that,
especially on controversial issues, intelligence
assessments are more likely to be trouble-
some than supportive.
Our assessments are not produced .in an
ivory tower atmosphere. The debates and
clash of ideas sometimes are rough. No one's
views-from the director to the newest ana-
lyst-are protected from challenge. It is not a
place for delicate egos or mediocrity or peo-
ple with special agendas.
But, however hot the debate or pointed the
questions during the drafting, the final product
is as honest and accurate as humanly possibke.
Despite imperfections, CIA and the intelligence
community produce the best, most comprehen-
sive and most objective intelligence reporting in
the world. We are working every day to make it
better, and however surprising it may be to our
critics, we believe they contribute to this proc-
ess, and so we listen to them.
The writer is chairman of the National Intelli.
gence Council and CIA's deputy director for intel-
ligence.
Declassified and Approved For Release 2013/05/22 : CIA-RDP99-01448R000301300003-9 -