CHALLENGE MECHANISM

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP80M01082A000800020002-4
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 3
• Document Creation Date: December 15, 2016
• Document Release Date: March 25, 2004
• Sequence Number: 2
• Publication Date: November 27, 1974
• Content Type: MF

Aooroved For Release 2004/03/31 oIA~'E D 80M01082A0008000206e
IC 7 4-2137
27 November 1974
IEMORANDU T FOR:
SUBJECT: Challenge Mechanism
1. 1 think your paper is a valiant effort to explore the
feasibility of institutionalizing a challenge mechanism. Certainly
the comments of your highly qualified contacts are informed
and persuasive- -and a little frightening. In spite of the effort to
be objective. their message comes through very clearly. They
appear to believe that the producer of finished intelligence is
almost always right and that he should be left alone to get on
with the job. Destroying ONE was an "institutional" mistake
and saddling the producer with a challenge mechanism would
only compound the error. Given your premises and the back-
grounds of your interlocutors, your rather pessimistic conclusions
are hardly surprising.
2. It seems to me that the development of a viable challenge
mechanism would be more manageable and realistic if the term
''institutionalising" were taken less literally and if the purpose
of a challenge mechanism were more clearly defined. Taking
the latter point first, i1 is, my view that the primary purpose of
a challenge mechanism should be to assist production analysts
to overcome three occupational hazards to which. according to
our post morterns, they are generally. and sometimes seriously.
subject. These are:
?Preconceptions: the tendency to discount information
that runs counter to long-held convictions`.
--Reinforcing consensus: the tendency for divergent
views of individual analysts to be submerged in
sea of conventional collective wisdom;
--The current intelligence syndrome: a myopic view
of the wood because of too great a focus on the current
intelligence trees.
Approved For Release 2004/03/31 :
80M01082A0008000 -
25X1
25X1
 Approved For Release 2004/03/31 : CIA-RDP80MO1082A000800020002-4
3. Under this definition, a challenge mechanism is not
a means of presenting dissenting views to the customer. (This
is a different problem and is already being acted on in various ways.)
It is a means rather of reminding, nudging, alerting the analyst--
telling him. in fact, that in spite of his widely acknowledged
expertise he may have overlooked or unconsciously suppressed
something important.
4. Who, then, is going to perform the function of saving
the analyst from the built-in weaknesses of his profession?
Clearly the answer is not to try to set up a competing "challenge"
group that in knowledge. background, acumen, and access to infor-
mation has the same attributes as the analysts being challenged.
The challengers would inevitably suffer the same weaknesses as
the challenged. Clearly `what is needed as challenger is an individual
or group that does not have preconceptions (at least not the same
ones as the pros). is not burdened by the consensus of colleagues.
and does ,n, read every scrap of current intelligence. To be a
responsible and believable critic, the challenger would presumably
have to know a great deal about the substance of the matter he is
critiq-ding. The validity of his challenge, however, would rest
not primarily on the scope of his knowledge but rather on the different
perspective he would bring to the problem at issue. a perspective
untrammeled by the occupational hazards of the professional analyst.
S. Where would you find such people? In my view. they are
scattered throughout the community and in relatively large numbers.
They are the "old China bands, ""the old Near East hands, " the former
Kremlinologists, or even the bright, young case officer just returned
from the held--people who can be expected to have all kinds of
interesting and perhaps
least currently, shainto ve no direct role in the intelligence
problems but who,
production of finished intelligence.
6. How then do you arrange to transmit to the production analyst
the different perspectives and possibly useful insights of these out-
siders? This gets to the problem of "institutionalizing" and manage-
ment style, particularly the DCI's. I think a possibly useful analogy
is what Mr. Colby has done on the (small w) warning problem. Instead
of thinking in terms of new bureaucratic entities. now lines
in of c mrnu-the
nication. and new assignment of responsibilities (as many
community did, as revealed by the attitudes of U5IB representatives
Approved For Release 2004/03/31: SMV OM01082A000800020002-4
 Approved For Release 2004/03/31 : CIA-RDP80M01082A000800020002-4
at the first meeting of the ad hoc committee set up to review the
problem), Mr. Colby said simply that when the community believed
that the policy maker needed to be warned, he (the DCI) would warn
him through the medium of an Alerting Memorandum. The NIOs
would be responsible for preparing the memorandum for the DCI's
release, but any element in the community could propose, or even
initiate, it.
7. In the case of a challenge mechanism, the person to be
warned is, of course, the production analyst rather than the policy
maker. But in other respects the two systems would appear to me
to have a great deal in common. Any member of the community
(and perhaps even consumers of intelligence) would be encouraged
to prepare a "challenge" memorandum whenever, in his view, a
particular product was significantly misguided or weak. These
memoranda would be addressed to the producer--the NIOs in the case
of NIEs, the Editor of the NIB in the case of articles . therein, etc.
8. The production analyst would be under no obligation to
modify his views as a result of "challenge" memoranda, any more
than Dr. Kissinger is under any obligation to take seriously an
Alerting Memorandum sent to him by the DCI, Whether. in either
case, the message "gets through" will in the last analysis depend
on its quality.
9. As for institutionalizing this concept, I would suggest that
all that is needed (as in the case of Alerting Memoranda) is for the
DCI to explain it to USIB and say "Let's do it. " I think that at a
minimum it would be an interesting experiment.
25X1
25X1
Distribution:
x
1
1
RD Chrono
25X1 IC/Re istry
.mjs
Deputy Chief, PRD
SECRET
Approved For Release 2004/03/31 : CIA-RDP80M01082A000800020002-4