LETTER TO: THE EDITOR FROM (Sanitized)

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP83M00171R000300270025-9
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 3
• Document Creation Date: December 16, 2016
• Document Release Date: May 31, 2005
• Sequence Number: 25
• Publication Date: March 5, 1975
• Content Type: LETTER

L NLgk 7b M3002700
TO: The Editor
RONI
Room 6E0/4
Hqs Bldg. CIA
In their very interesting new publication, Review of
National Intelligence, some of my friends on the IC Staff
present a curious finding. They take intelligence analysts
to task for "an old and familiar analytical bias" against
expecting irrational moves by "essentially rational men."
I say, vive le bias! *
25X1
Analysts must consider and test all possible alternatives
in developing situations -- including the most irrational.
Any alternative that would have significant repercussions for
US interests or policies should be duly noted, and the analys.t's
assessment,of its degree of likelihood presented. To give a
dramatic alternative a higher degree of likelihood ecaus it 13Y
is irrational, as the IC Staff's "lesson-learned" seems to
suggest, would, in my opinion, border on the. irresponsible.
An analyst who consistently predicts the irrational -- especially
the extremely irrational, F_ I- is doomed 25X1
to judge wrongly in by far e greatest num e ases. To
allow for the irrational is one thing; to confidently predict
it, quite another.
Those of us who have struggled for years to anticipate
the course of world events are well aware of its irrational
tendencies. Situations usually develop within the bounds of
logic and reason, however, and do not take irrational tacks.
Most international conflicts, for example, are settled peace-
fully and "reasonably"; we should not predict war -- an
essentially irrational alternative -- every time a dispute
arises between two states.
See Principal Findings of the Post-Mortem Report, "An
Examination of the Intelligence Community's Performance Before
and During the 1 and the reprise thereof
in the Review of National Intelligence, February.1975.
FOR ICiAL USE ONLY
Approved For Release 2005/06/09 : CIA-RDP83M00171 R000300270025-9
 Approved For Ree 2005/06/09 : CIA-RDP83M00171 R6300270025-9
FOR OFFICIAL USIL70NLY
I believe the Communit did a fairly good job of reporting
on the incipient crisis in I Jalerting policy makers 25X1
to the possible consequences. con ident intelligence pre-
diction of the attempt would have been
a classic exam estimate which happened to turn
out right.
STAT
Assistant Deputy Director
for Estimates, DIA
f9
-- - kw~rti~rrrr.
0
FOR OFFICIAL USE 01111114A
Approved For Release 2005/06/09 : CIA-RDP83M00171 R000300270025-9
 ApprovedlP,bTSeIekSF-P MG049 1- Pg3 M00030027
furnished a copy of
his letter to the editor, RONI
in re
,
Bob is right, of course. And
his letter points up the note of
'glibness' (some might even say smugness)
that enters these IC post-mortems
(stemming, I believe, from the wonderful
advantage of perfect hindsight coupled
with the pressure to say something).
In the case of
(para. 2 of le ter the only
''rational' reason he could have undertaken
it was if he were himself in danger of
ouster. And that was not the case.
25X1
STAT
Attachment:
5Mar75 letter
from 0 to
Editor/RONI
INFORMATION
Approved For Release 2005/06/09 : CIA-RDP83M00171 R000300270025-9