NORMALLY, THE CIA DOES NOT RESPOND TO ARTICLES WRITTEN ABOUT IT.

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP91-00587R000100470016-5
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 4
• Document Creation Date: December 22, 2016
• Document Release Date: February 18, 2011
• Sequence Number: 16
• Content Type: OPEN SOURCE

Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
FOREIGN POLICY
SPRING 1985
Letters
To the Editor:
Vormally, the CIA does not respond to articles
wwri;ten about it. However, because Allan Good-
rnan s article, "Dateline Langley: Fixing the
Intelligence Mess, " in FOREIGN POLICY 57
(Winter 1984-85), concerns an area of the agency's
activities where we can speak publicly, and because
the article is so inaccurate, we believe the record
should be corrected.
Goodman left he agency in 1980, and his
information concerning it is seriously outdated. A
point-hi.-point rebuttal of all of Goodman s errors
and recommendations would take too much time and
space. Therefore, I will address only the most
egregious inaccuracies.
Goodman states that the intelligence community
does not study its failures and that the results of the
few postmortems that have been undertaken have not
he?n widely disseminated or discussed. But in recent
years the director has assigned a senior group of
distinguished officials the task not only of evaluat
ing some 15 major historical intelligence problems
but also of evaluating retrospectively at 1- and 2-
year intervals virtually every estimate now pre-
pared.
Moreover, the Directorate of Intelligence now has
its own evaluation staff whose principal function is
to conduct retrospective evaluations of CIA assess-
ments on particular subjects. These evaluations
are widely shared with the concerned organiza-
tions. Finally, in this connection, one of the
agency s most popular training courses now is a
course on intelligence successes a ndfailures.
Goodman asserts that analysts learn to be wary of
doing longer-range or in-depth studies and that the
task of writing estimates and think pieces is to be
avoided. He notes that most of these studies are
turned out by members of a special staff and that
promotions of analysts at middle and senior levels
require that they take on management responsibility.
But for more than 3 years now, a substantial
percentage of new analytical- resources has been
dc; oicd to strengthening long-term research. In the
last scar alone, more than 700 long-term research
asctssments were published. Indeed, the structure of
cesztxx;c~ ,been revised in such a way that it is
now more difficult to get analysts to work on short-
range protects and current intelligence than on
longer-term research. The special staff Goodman
cites has not existed for several years, and the agency
now has a number of opportunities for analysts to
rise to GS-IS and even to supergrade level.
Goodman asserts that the intelligence community
posts few analysts abroad and that even short field
trips are hard to come by But one of the benefits
provided by new resources in recent years has been to
permit significant expansion of the number of
analysts assigned overseas. Moreover, for the first
time there are adequate funds for analysts to travel
and work overseas, often for several months at a
time. Goodman ''s statement that analysts are gener-
ally limited to one 6-:reek stretch of temporary duty
every 3 years or so is wrong.
Goodman also states that attempts to reach out to
academics are strongly resisted and dismissed as
cosmetic. But a major CIA initiative in recent
years has been the dramatic expansion of its
contacts not only with academics but also with-
think tanks and the private sector. In 1984 alone
co ne,J,200 analysts attended nearly S00 confer-
ences on substantive issues, many of them spon-
sored by universities. Moreover, analysts are now
required to obtain outside training every 2 years,
either through academic course work or through
attendance at conferences and seminars. Many of
the agency s substantive papers are now reviewed
by academics. Particular emphasis is placed on
seeking out scholars with a different point of view
than the CIA's.
Goodman states that analysts should do more to
distinguish between what they know and do not
know, to identify those judgments based on specific
evidence from those based on speculation, and to
make projections about the future. Goodman suggests
further that policymakers be given some indication
of what to look for in the way of events or
developments that can be used to test CIA judg-
ments.
But one of the principal objectives of new, far
more intensive, substantive review of CIA analysis
is to ensure that its analysts not only are giving
policymakers a more explicit description of their
evidence but also are distinguishing between what
is analysis and what is based on evidence, as well
as stating the agency's .iew of the reliability of
Ccn:
Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
 Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
;'rat e,idcnce. And when analysts speculate, the
reasons behind the speculation are included. A list
of the types of indicators Goodman referred to is
now of en included in national estimates and CIA
assessments.
Goodman states that the blocking of critical
analyses unwelcome to policymakers has been consis-
tent in recent years, citing an estimate on Lebanon
as one example. Not only was the estimate Goodman
cites not blocked-any more than the one he alludes
to relating to Central America-but these estimates
delivered an unvarnished, candid intelligence com-
munity judgment on key issues. No viewpoint was
suppressed in the presentation of those reports. The
charge of the blocking of critical estimates because
they were critical of policy is false; the agency
continues to publish a wide range of estimates
without regard to the political consequences for
policies that may be affected. In view of the strength
of Goodman 's assertion and the centrality of this
issue, we would note simply that neither congression-
al oversight committee has reached his conclusions.
Unlike Goodman, these committees have access to our
assessments.
In sum, many of the policies Goodman advocates
to improve the quality of analysis already have been
implemented. While the agency recognizes that there
is always room for improvement and that there are
still occasional lapses, the policies are in place. And
just as Goodman predicted, these changes in analyti-
cal methods and management have reduced the
failure rate forAmerican intelligence.
George V Lauder
Director, Public Affairs
Central Intelligence Agency
Washington, D.C.
The Author Replies:
I am deeply disturbed that the CIA-as opposed
to other parts of the intelligence community-has
responded to my article apparently by launching
a disinformation campaign.
Notwithstanding Lauder's assurances that CIA
performance has improved substantially since I
left the organization, there is persistent and
disturbing evidence that the U.S. intelligence
community in general and the CIA in particular
remain handicapped by the problems my article
discussed.
In .January 198?, for example, the Senate and
House select committees on intelligence released
public versions of their reports on U.S. intelligence
activities covering the period from January 1, 1983,
to December 31, 1984. Their conclusions arc very
similar to my own. Strikingly, nowhere in eith:r the
61-page Senate report or the 19-page House report is
mention grade of any of the so-called policies and
it:itiall."es to improve the quality of analysis that
Louder describes, despite his assertions that agency
officials "can speak publicly "about them.
Indeed, the Senate report concludes that with
respect to analysis and production "analysts are not
producing enough basic data to meet important
intelligence requirements. Instead, analytic efforts
seem to emphasize short term `current' intelligence
products. " With respect to collection, the report noted
"a growing imbalance between collection and ana-
lytic capabilities `persistent gaps in information
on certain subjects of great importance to national
security'; and the need to "shift away from the
collection of data that is currently proving to be of
diminishing value. "And when it focused specifically
on human intelligence (HUMINT) collection per-
formance, the committee, like my article, found
"collection gaps and dissemination problems" and
urged that "clandestine HUNIMN'T collection by
U.S. agencies be carefully coordinated. "
The House committee's report found "a need for
improved performance on the part of intelligence
collectors and analysts" and "a clear need for better
coordination between the users of intelligence and the
providers of intelligence. " The report also stated that
"care had to be taken lest analytic thought succumb
to pressure to support rather than inform policy"
and that "shortcomings in analysis and collection
continue to appear. "
Possibly in anticipation of these congressional
reports and to rebut my article-which be had seen
in draft-Deputy Director of CIA for Intelligence
Robert Gates submitted an op-ed piece to the
Washington Post that was published on Decem-
her 12, 1984, and that made many of the points
in Lauder's letter. In response, one veteran and
highly respected career officer wrote to the Post in
a piece published January 10, 1985, that Gates's
article was `panglossian. "/Moreover, after listing
all of the initiatives that have characterized his
approach to reform at the CIA, Gates himself
concluded, "1 cannot say this approach would
Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
 Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
ha-.-c enabled us to predict the fall of the Shah [of
Iran] -a major intelligence failure my article
addressed at length because the patterns of collec-
tion, analysis, and management that led to
failure still appear to handicap the CIA and the
intelligence cornrnunity.
The Lauder letter and the Gates article sidestep
the fundamental questions that haz,e been raised
about the quality of the agency's performance by
dismissing my critique and that of others as outdated
and partisan. I did leave the agency in 1980. I have
stayed in touch, however, and based the portion of
my article dealing with the CIA under William
Cas_i, on discussions over the past 3 years or so
with more than 100 analysts, managers, and
policy makers. Some of these discussions were even
arranged by Lauder's office.
It also is important to note that the CIA is only
one part of the U.S. intelligence community.
Further, the policies and initiatives Lauder refers
to apply to only one part of the CIA, namely, the
Directorate of Intelligence (DDI). The DDI has
been reorganized during the Reagan administra-
tior and some of its internal procedures have been
changed, as my article noted. After discussing
these specific changes with numerous analysts,
managers, and policy consumers, 1 concluded in
print that the reforms have been slow to take hold,
that their impact on the quality of intelligence has
been marginal, and that they have not been
widely adopted throughout the intelligence com-
munity. The community is still plagued by
fragmented intelligence.collection and reporting.
Such fragmentation-coupled with the politici-
zation of analysis-still leads policy consumers
to be critical of the product, and also leads to
shocking failures that demonstrate persistent
problems with the way the American intelligence
system is organized and managed.
Thus Lauder's letter overstates the significance of
the actions taken by the DDI and their impact both
on the community as a whole and on the quality
of CIA analysis in particular. To cite specifics:
> I stand by my judgment that the intelligence
community does not study its failures. The DDI does
this to some extent but, just as in the pre-1981
period, the senior review group referred to by
Lauder that is charged with this task examines
only a very small part of what even the DDI
produces-60 or so estimates annually versus
thousands of community judgments contained in
briefings and memoranda. The work of this group
is meticulous but not widel accepted throughout
the community, in part because some members
lack intelligence experience. Nor did Lauder point
to even one way in which the group's findings
have actually affected the way an anahst's
routine or crisis work is done. Lauder also neglects
to mention that the course on intelligence successes
and failures has been criticized for holding back
on providing students with the details, especially
the sensitive intelligence reports, on which both
successes and failures appeared to turn.
> The present DDI may emphasize research
more than ever before, but depending on defini-
tions, 700 "long-term research assessments" per
year strikes me as about average for the past 8 or
more years. Yet the community as a whole has not
followed suit, as the Senate oversight committee
also noted. Further, the special staff of the
National Intelligence Council to which 1 referred,
and on whom the principal burden of writing
estimates and think pieces falls, still very much
exists, and its members are worried about how
well they will be absorbed back into the system.
Analysts and managers alike told me that there
are still too few opportunities for analysts to rise
to the GS-1S and senior intelligence service ranks.
And the Senate intelligence committee report
asserts that the Defense Intelligence Agency has
"been significantly handicapped in its ability to
recruit and reward outstanding analysts and
other intelligence specialists. "
> I did not say that the CIA has failed to
provide analysts with an opportunity to attend
academic conferences. I did say that substantive
consultations with academics, when they proved
critical of agency work, were resisted by many
managers. Lauder's letter does not dispute this
point at all. Moreover, Lauder ignores the
chilling effect of several recent firings of intelli-
gence officers for speaking out at academic
conferences on the motivation of other analysts to
attend or participate in such conferences.
> As Lauder knows, more explicitly and frankly
explaining to policymakers the evidence behind
intelligence analysis is not being done on a signifi-
cant scale beyond national intelligence estimates.
The House oversight committee's recent report refers
Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
 Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5
to intelligence community concerns about this con-
tinuing problem.
> Politicization of reporting and analysis has
taken place and in my judgment has vitiated the
effect of many of the initiatives Lauder highlights.
Indeed, the effects of politicisation cannot be glossed
over by saying that eventually a product sees the
light of day even if it is critical of administration
police. As Lauder knows, senior career off cers have
complained, and some have resigned in response to
the pressures exerted by the director of central
intelligence to reshape national intelligence estimates
for political effect. In fact, the details of one such
episode were printed in. the Washington Post in a
January 2, 1985, article by John Horton titled
"Why 1 Ouit the CIA. "As Lauder also knows, the
newly composed oversight committee is likely to
agree to several senators' requests to investigate
the specific charges of politicization mentioned in
my article.
The record, therefore, only rein forces concern that
the intelligence community is not functioning effec-
tively. I do not doubt that Lauder and his associates
will continue to work to improve CIA performance.
But I am skeptical of their efforts in the current
politicized environment and, as I noted, previous
internal agency reorganizations and reforms have
not significantly improved the quality of analy-
sis. America still lacks a centralized intelligence
collection and analysis system. If it fails to
develop one now, disaster lies ahead.
Allan E. Goodman
Georgetown University
Washington, D.C.
Sanitized Copy Approved for Release 2011/02/18: CIA-RDP91-00587R000100470016-5