PROBABILITIES IN INTELLIGENCE ANALYSIS

• Document Type: CREST
• Collection: Library of Congress
• Document Number (FOIA) /ESDN (CREST): LOC-HAK-538-3-5-6
• Release Decision: RIPLIM
• Original Classification: T
• Document Page Count: 85
• Document Creation Date: January 11, 2017
• Document Release Date: March 9, 2011
• Sequence Number: 5
• Publication Date: November 9, 1973
• Content Type: MEMO

/
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
?111,
MEMORANDUM
NATIONAL SECURITY COUNCIL
T /CODEWORD ATTACHMENTS
MEMORANDUM FOR BRENT SCOWCROFT
FROM: A. W. MarshalOOL
SUBJECT:
Probabilities in Intelligence Analysis
6117
? ACTION
November 9, 1973
With my approaching departure I think I should tell you of some experi-
mental intelligence products I have had prepared in an attempt to get
improved products to HAX. In any case, I would like to have your
reactions to these products.
The genesis of this effort is as follows:
. HAK said that top-level policymaking was essentially a
matter of making complicated bets about the future. He wished
that intelligence would supply him with estimates of the relevant
heIting odds.
MORI/CDF C03358255 Pages
9-33
C05129177 pages 59-74 IN He also said not to ask him what he needed because he
could not tell, but that he would recognize it when he saw it.
? My response has been in a few selected areas to develop
experimental products for his review and reaction.
iState Dept.
review
completed
pages 59-74
Three papers have been produced attempting to give him the betting odds.
They are:
-- The likelihood of Sino-Soviet war.
The outbreak of hostilities in Cyprus.
NASA review completed pages 9-33
The Panamanian reactions in the Pava.ma Canal negotiations.
? Copies of these studies are at Tabs E, F, and G. These studies were sent
to FIAIK who responded favorably, but he indicated more work was required
(see Tab D).
DIA, NSS reviews
completed
T GDS
D ATTACHMENTS
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
IP
The use of numerical probabilities in intelligence estimates has two
possible benefits:
-- Improved intelligence analysis with limited data.
-- More effective communication of analysts' judgments
to consumers.
My feeling is that one of the major failings of current intelligence products
is their attempt to provide only one consensus judgment. Most NSC staff
people feel that a more systematic presentation of several alternatives
along with the weight of evidence or judgment in each would be more
helpful.
Unfortunately, the three studies that are attached stressed improved
? analysis rather than improved communication between the analyst and
consumer. However, an attempt to explore better communication
practices has been made. The results of the Sino-Soviet paper are now
available in three alternative formats. They each show estimates of the
likelihood of hostilities as a function of major variables in the situation
(see Tabs A, B, and C). Each chart would, if included in a product, be
preceded by the following paragraphs:
"Attached is an estimate of the probability of a Soviet attack
on China within the next twelve months. Because of an improved
Chinese military posture and international position, the odds are
fifteen to one against an attack. Of the factors affecting such an
event, five were identified as most influential:
- Changes in the PRC'a internal political situation,
especially the development of pro-Soviet factions.
- The USSR's perception of the PRC's nuclear capability
as a deterrent.
- Serious border incidents.
- Charges in USSR leadership affecting the attitudes of the
key leadership.
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
1111
SECRET
- Changes in USSR external political/military alliances,
which free or divert resources from potential use against the
PRC.
Two of these can significantly increase the probability of attack:
- A deterioration of the PRC' s internal political situation,
occurring upon the death of Mao or Chou and accompanied by
a breakdown in command and control.
- A view, on behalf of the Soviets that the PRC does not
have a credible deterrent."
Questions
? Which, if any, of these alternative presentations improves
upon the traditional product?
? .1,-1, A
.LC'
Tab B
Tab C
-- Which do you like best?
Tab A
Tab B
Tab C
-- Do you have any additional approaches to suggest? For
example, a tick-point, more verbal discussion of probabilities
rather than charts?
SECRET
4ciht. vJi4241.444.&4jciA4
_d1,14iu; 4, eILIAJT8-
CvtJ cfikAat4isok -acJ
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
IP
PROBABILITY OF A SOVIET ATTACK
SOVIETS PERCEIVE
CHINESE NUCLEAR
DETERRENT
SOVIETS DO NOT BELIEVE
CHINESE NUCLEAR CAPABILITY
IS CREDIBLE DETERRENT
AND
BOTH LEADERS DIE, INTERNAL DISCORD,
4*I BREAKDOWN IN COMMAND AND CONTROL
PRO-SOVIET FACTION EMERGES
BOTH LEADERS DIE, INTERNAL DISCORD,
BREAKDOWN IN COMMAND AND CONTROL
PRO SOVIET FACTION EMERGES
BOTH LEADERS DIE, INTERNAL DISCORD,
BREAKDOWN IN COMMAND AND CONTROL
elA EITHER LEADER DIES, INTERNAL DISCORD,
-NU pRir.,,kstWIFT FACTION EMERGES
BOTH LEADERS DIE, INTERNAL DISCORD,
BREAKDOWN IN COMMAND AND CONTROL
EITHER LEADER DIES, INTERNAL DISCORD
EITHER LEADER DIES, INTERNAL DISCORD,
PRO,SOVIET FACTION EMERGES SoP
EITHER LEADER DIES, INTERNAL DISCORD
NO CHANGE INTERNALLY (BOTH LEADERS
LIVE, NO MODERATION IN ATTITUDE
TOWARD USSR, NO INTERNAL DISCORD)
POSSIBLE CHANGES IN INTERNAL
PRC POLITICAL SITUATION
NO CHANGE INTERNALLY (BOTH LEADERS
LIVE, NO MODERATION IN ATTITUDE
TOWARD USSR, NO INTERNAL DISCORD)
?
CURtIENT BEST ESTIMATE
POSSIBLE CHANGES IN INTERNAL
PRC POLITICAL SITUATION
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
PROBABILITY OF A SOVIET ATTACK
? IF SOVIETS PERCEIVE
CHINESE NUCLEAR
DETERRENT
POSSIBLE CHANGES IN INTERNAL
?RC POLITICAL SITUATION
BOTH LEADERS DIE, INTERNAL DISCORD, BREAK-
DOWN IN COMMAND AND CONTROL PRO-SOVIET
FACTION EMERGES
BOTH LEADERS DIE, INTERNAL DISCORD,
BREAKDOWN IN COMMAND AND CONTROL
EITHER LEADER DIES, INTERNAL DISCORD,
PRO-SOVIET FACTION EMERGES
EITHER LEADER DIES, INTERNAL DISCORD
NO CHANGE INTERNALLY (BOTH LEADERS
LIVE, NO MODERATION IN ATTITUDE
TOWARD USSR, NO INTERNAL DISCORD)
IF SOVIETS DO NOT
BELIEVE CHINESE NUCLEAR
CAPABILITY IS CREDIBLE
DETERRENT
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CURRENT
BEST
ESTIMATE
75%
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
PROBABILITY OF A SOVIET ATTACK
SOVIETS PERCEIVE
CHINESE NUCLEAR
DETERRENT
SOVIETS DO NOT BELIEVE
CHINESE NUCLEAR CAPABILITY
IS CREDIBLE DETERRENT
60%
55%
50%
?????
15%
10%
CURRENT
BEST
ESTIMATE
NO CHANGE INTERNALLY
.
480TH LEADERS LIVE, NO
MODERATION IN ATTITUDE
TOWARD USSR. NO
INTERNAL pIscptun
. . ?
?
EITHER LEADER DIES
INTERNAL DISCORD
EITHER LEADER DIES
INTERNAL DISCORD
? PROSOVIET FACTION
: EMERGES ?
- BOTH LEADERS DIE
INTERNAL DISCORD
BREAKDOWN IN -
COMMAND AND CONTROL
POSSIBLE CHANGES IN INTERNAL PRC POLITICAL SITUATION
?
50TH LEADERS DIE
INTERNAL DISCORD
BREAKDOWN IN
COMMAND AND CONTROL
PROSOVIET FACTION
EMERGES
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
J- 1' ei.4.:,.!itkit
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
IP
MEMORANDUM 0 stde S stern
NATIONAL SECUR 1.355X
SECRET/ CODEWORD Information na "4 .517
13 9 3
MEMORANDUM FOR HENRY A. KISS ER.
V Abev
FROM: A. W. MARSHALL ?-
SUBJECT:
Experimental Intelligence Proe4o Improve
the Communication of Uncertainty
Some while ago I initiated a program of experimental intelligence
products. The payoffs have been slim, but a few results are beginning
to appear. The purpose of this memorandum is to show you some
samples of one particular line of products.
They have been designed to respond to:
-- An expressed view of yours that top level policy
making involves making complicated bets on the future,
and that you wished intelligence estimates were designed
to give you the betting odds.
-- Your saying that while you could not describe the
kinds of intelligence products you want, you would immediately
recognize them when you saw them.
Three products have been developed thus far. They are:
-- The probability of Sino-Soviet hostilities in the
next twelve months. (See Tab A)
-- The outlook for hostilities to develop in Cyprus.
(See Tab B)
-- A study of the possibilities of a Canal treaty with
Panama in the next year. (See Tab C)
SECRET/ CODEWORD (XGDS) - 2 3
DECLAS Impdet
BYAUTH H. A. Kissinger
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
v
No Objection to Declassification in Part 2013/09/11 : LO-C-HAK-538-3-5-6
IP
SECRET/CODEWORD
If you have time I recommend you look at these products as a first step
in responding to your needs. The Sino-Soviet study (Tab A) is the best
of the set.
The products:
-- Attempt to produce estimates of betting odds.
-- Are put forward as the beginning of .a survey of the
market to obtain answers to questions such as: Do you find
this new type of estimate useful?, etc.
The attached products are not fully developed. More time thus far has
been spent in developing a mode of analysis that will produce estimates
of numerical probabilities attached to various possible future outcomes,
than in developing alternative ways in which such estimates might best
be communicated to top level policyrnakers. I now have the intelligence
people working on the latter problem.
I /awe tried the (Lux. r eatly available products on othr,:r mern:6,-rs of +11.-
?NSC staff. They have:
-- Liked the idea of the numerical probability estimates.
? Found the detailed probability trees intellectually
indigestible.
Do you find these products of any interest?
7:1'a
No
Comment
Have you any suggestions to make as to how to better display, or sum-
marize the analysis results?
Comment
SECRET/CODEWORD
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
MORI/CDF C05473510 Pages
9-33
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassificationin Part 2013/09/11 : DENT
riTTS
DIRECTORATE FOR INTELLIGENCE
ANTFLLIGEN.k:
...PROJECT
AN APPRAISAL
OF THE PROBABILITY OF
N OLITIBL,EAK
ON CYPRUS-
F HOSTILITIES
1 NOVEMBER 1972 - 30 APRIL 1973
,?
A CASE STUDY ILLUSTRATING THE USE OF
PRO3A3 LISTIC PROCEDURES
CLASSIFIED RV ONPKA
EXEMPT f,U3M /,E HAL DECLASSIFICATION
.
SCHEDULE 0iE-xEcuilVE ORDER 1102
E XE MP? ION CA l'EGDRY 2,
DECLA5I.4FY ON DccrmuErt 31, 1980
NO FOREIGN DISSEM
? :cONFi D:ENTIA
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
-CONTENTS
II
HYPOTHESES AND DEFINITIONS
HYPOTHESES
DEFINITIONS
BACKGROUND
U.S. STRATEGIC INTERESTS
DEVELOPMENT OF THE CONFLICT
CURRENT FORCES AT WORK
PROBABILITY ANALYSIS
OVERVIEW
PROBABILITY DIAGRAM
MARKOV MODEL
CONCLUSION: THE PROBABILITY
OF HOSTILITIES
THE PROCESS OF ANALYSIS
SOME OBSERVATIONS 2
THE ANALYSIS
COMMUNICATION
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6 _
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CONFIDEN IAL IP
HYPOTHESES AND DEFINITIONS
HYPOTHESES
CONFIDENTIAL
This analysis develops a probability for each of the following
hypotheses:
I. There will be at least one serious and widespread outbreak of
hostilities on Cyprus during the next six months.
2. There will not be? a serious and widespread outbreak of
hostilities on Cyprus during the next six months.
DEFINITIONS
It is assumed in the following analysis that an outbreak? of
hostilities will almost certainly be-pret-ipiLdicd Loy CAI I 111{,ILICI Alihuuyii
an incident may not directly bring about conflict, it could be expected
to alter the situation on the island. For example, a major clash between
Greek Cypriot factions would likely lead to considerably increased
tension, which in turn would provide a dangerous environment for
any further incidents. This property is incorporated in the analysis?
by defining five separate tension states:
A "normal" situation. It is assumed that
Cyprus is currently in this state.
MGT A moderate state of tension between the
Greek Cypriot and Turk Cypriot communities.
HGT A high state of tension ? between Greek
Cypriot and Turk Cypriot communities.
MGG A moderate state of tension between
opposing Greek/Greek Cypriot factions.
HGG A high state of tension between opposing
Greek/Greek. Cypriot factions.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK:53873-5-6
LunirlIJC.11 I IL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
?
BACKGROUND
Figure 1
origins of the
Greek and
Turk Cypriot
communities
GREECE
TURKEY. :
CYPRUS
U.S. STRATEGIC INTERESTS
The importance of Cyprus stems from its proximity to the Middle
East and the SE flank of NATO, Figure I, and from the U.S. facilities
located on the island. In addition to endangering these facilities, the
outbreak of hostilities on Cyprus could also involve Greece and Turkey
in open conflict, weakening the NATO alliance.
DEVELOPMENT OF THE CONFLICT
During most of its history, Cyprus has been ruled by outsiders.
Its strategic location has been of value to every power that has
dominated the eastern Mediterranean, The Greek character of some
80 percent of the population derives from early Greek colonization
and some 800 years' rule under the Orthodox Christian Byzantine
Empire. The minority Turkish population about 18 percent --
represents a formerly dominant element. The Turks, who came to the
island after it was captured by the Ottoman Empire in 1571, ruled
Cyprus for 300 years, until administration of the island was transferred
to Great Britain in 1878: ?
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Since Greek independence in the early 19th century, Greek
? Cypriots have pressed the claim for enosis -- that Cyprus should become
an integral part of Greece -- first with the ruling Turks and later with
the British. The Turkish Cypriots backed by Turkey have consistently
opposed this union. Today this conflict over the issue of national
identification continues unresolved and is at the heart of the Cyprus
problem.
enosis
? A cartoonist views the conflict between
Greeks and Turks on Cyprus within the
larger context of Greece vs Turkey,
' Taken &um Tile Guardian, Manchester
formation
of the
Republic
London-Zurich
Agreements
:ONFIDENTIAL
The dissolution of the British Empire created the setting for the
Greek Cypriot guerrilla movement of the 1950's. Probably more than
anything else it was the uncertainty of political control that
exacerbated ethnic relations. The situation became acute in 1955 with
the commencement of a guerrilla campaign against the British by the
Greek Cypriot underground. The next four years were marked by
incidents of violence, increasing communal unrest, and growing friction
among the UK, Greece, and Turkey. British-Cypriot negotiations for
limited self-government broke down in 1956 and the leader of the Greek
Cypriot community, Archbishop Makarios, was exiled for a year. By
1958, the Archbishop's interest in enosis had slackened and he was
expressing a desire for an independent Cyprus. In early 1959, Makarios
was invited to London to join in tripartite talks among the UK ?Greece,
and Turkey, which produced the London-Zurich Agreements. Under
these agreements, Cyprus became an independent republic based on
the concept of administrative separation of the two communities
. without geographic partition. Geographic partition as a potential
solution to the Cyprus problem would be extremely difficult to
implement considering the dispersal of the Turk Cypriots, Figure 3.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6 -
No Objection to Declassification in Part 2013/09/11 :
Figure 3
Area of claimed
Greek control
Area of clakned
turkish control
? Village with both Turks
end Greeks
U.K. sovereign base area
To secure the agreement of all parties it was necessary to include a
Treaty of Guarantee which gave Greece, Turkey, and the UK the right
to intervene jointly or individually in the internal affairs of the island
should this be deemed necessary to uphold the agreements. In
December, 1959, Makarios was elected President of Cyprus and took
office the following August, when the island became a state.
The London-Zurich Agreements did not solve the Cyprus problem.
Lack of mutual confidence prevented the tvvu cunenunities from
their cooperating in running the government. The Greek majority were
breakdown determined to rid themselves of the constitutional provisions which
gave the Turkish minority a veto power in matters of foreign affairs,
defense, and key domestic issues. For their part, the Turkish Cypriots
clung rigidly to the constitutional safeguards as their best protection
against the Greek majority. In late 1963, President Makarios presented
13 amendments to the constitution that would have created a unitary
hostilities in state under majority rule -- that is, Greek Cypriot rule. Turkey quickly
1964 rejected these proposals. In the tense atmosphere precipitated by these
events, a minor incident sparked violence which rapidly engulfed the
island. The More numerous.and better organized Greek communal
forces seized the initiative and isolated the Turkish community. The
intercommunal fighting created a grave danger of war between Greece
and Turkey. A British peacekeeping force went into action in January
1964, but failed to prevent new violence. The United Nations
peacekeeping force was activated the following March. It has remained
there since, working constantly to prevent minor incidents from
developing into renewed fighting.
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-576
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
? From 1964 through 1967 numerousetrts were made to find a
? settlement which Greece, Turkey and the Cypriots could all accept.
A number of formulas were considered and rejected. No major crisis
occurred until November 1967, when an attack on two Turkish Cypriot
villages by Greek Cypriots led to a major confrontation between Greece
and Turkey. War was averted by the emergency mediation of
Presidential Emissary Cyrus Vance, working with the three .countries
concerned. The crisis ebbed when the Greeks agreed to withdraw some
7,000 of their forces from the island. It was also agreed that discussions
would be held on the understanding that Cyprus would remain
independent. Negotiations between representatives of the two ethnic
communities on the island began in June 1968, and in an 'expanded
format, they are continuing today.
NFIDENTIAL
CURRENT FORCES AT WORK
The delicate balance in Cypriot affairs suffered a severe jolt with
the return last year of the violence-prone hero of Greek Cypriot
military endeavors against both the British and the Turks, Gen. George
Grivas. Although his clandestine return to Cyprus on about the 1st
of September 1971 has been followed by months of underground
activity, the General has shown little inclination for dynamic action
during his current stay on the island. While his private intentions are
unknown, Grivas remains a dangerous and unsettling factor in the
Cyprus equation, and if he chooses to make a move ,against his old
rival Makarios or against the Turk Cypriots, violence will likely follow.
Negotiations on the Cyprus problem, deadlocked and inactive fnr
almost a year, resumed in the summer of 1972. Participating
representatives from mainland Greece, Turkey and the United Nations
may be able to force a compromise on some issues but the talks are
in danger of stalemate, at any time.
No Obiection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Politically, President Makarmains very strong. If he chooses
to run for re-election in 1973, he will be re-elected. The only other
figure with a strong national following, Gen. Grivas, apparently has
no personal political aspirations. His goal is the union of Cyprus with
Greece, although the means for achieving this? remains clouded,.
As long as Makarios remains in office, he will continue to pursue
a nonaligned and increasingly, independent role for Cyprus and any
proposed settlement which would counter these policies will be
opposed by the Archbishop. Makarios is expected to continue to
dominate all other political forces on the island for the foreseeable
future.
" I have SURVIV66 13 QR?Ck P1Z1Me MMISteRS.
I shall SURVIVe the founteenth too. I shall
ask foR alb Nom Moscow."
anchBishop Makantos
1971
CONFIDENTIAL
No Objection to Declassification in Part_2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
111IP
PROBABILITY ANALYSIS
a probability
diagram
State at beginning of first
3-week period
Figure 4
NORMAL STATE OF TENSION
OVERVIEW
If serious hostilities are to develop cm Cyprus, they are almost
certain to escalate from an incident (although there is an outside chance
that hostility can develop without an incident). Many minor incidents
occur daily; however, this analysis will only consider those that, at
a minimum, have a clear potential for violence.
Figure 4 indicates that an incident can either lead directly to
hostilities, in which case the dot at the end of the branch indicates
that the analysis terminates, or it can bring about a changed state
of tension which in turn may promote hostilities at a later time. In
either case, there are several alternative sequences of events which could
mediate the impact of an incident. The analysis below therefore begins
with the stepwise development of a probability diagram that traces
out these alternative sequences of events. Using the probability
diagram, the likelihood is calculated of going from one tension state
to another, or to hostilities, in a three-week period. This is a sufficiently
short period to capture the dynamics following an incident and long
enough for there to be a reasonable chance of an incident occurring.
Once these probabilities have been calculated, they are
incorporated into a Markov model which is used to chain together
a sequence of probability diagrams in order to assess the probability
of moving from the cui twit state of Tension into hostilities as a function
of an increasing interval of time.
NFIDENTIAL
INTERVEANING
OT,71,0)
EVENTS
State at end of first
3-week period
,14,50'1/4?
.1/40
(H)
Remains in normal state of tension
(N)
Escalates to moderate state of ten-
sion between GCs and TC's(MGT)
Escalates to high state of tension
between GC's and TC's (HGT)
Escalates to moderate state of
tension between G/GC's (MGG)
scalates to a high state of tension
between G/GC's (HOG)
Remains in normal state of tension
(N)
4,
4> 0 Escalates to moderate state often-
ion between GC's and ICs (MGI)
('-H)Escalates to hi-h state of tension
between GC's and ICs (HT)
Escalates to moderate state of
tension between G/GC's (MGG)
Escalates to a high state of tension
between G/GC's (HOG)
No Obiection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6 .
110(The reader who is not con d with the details of the analysis
? may wish to skip the next two MEtions of this chapter and proceed
directly to the final section titled, "Conclusion -- The Probability of
Hostilities".)
GC's vs TC's
(INTERCOMMUNAL)
to Figure 6
NORMAL STATE
OF TENSION (N)
GREEK/GC's
(INTER-GREEK)
to Figure 7
NO INCIDENT
.99
MGT .005
0
.005
G
OTHER
61"
N .98
01
HOT 0
GG 01
Figure 5
instigation of
incident
PROBABILITY DIAGRAM
G 0
The probability diagram in Figure 5 identifies three types of
incidents which might lead to hostilities. The first two branches
describe the most important types: (I) those between Greek Cypriots
(GC's) and Turk Cypriots (TC's), called intercommunal incidents, and
(2) those between antagonistic factions of GC's, called inter-Greek
incidents. Based on an analysis of reported incidents over the past
six months, it is about twice as likely that any given incident wil4
be intercommunal rather than inter-Greek. Figures 6 and 7 which
follow will develop the implications of the intercommunal and
inter-Greek incidents respectively.
The third branch of Figure 5 describes all other kinds of incidents
and is included to exhaust all possibilities. An incident between Soviet
and Turkish vessels off the coast of Cyprus, during a Turkish troop
rotation, is an example of this category. This type of incident is very
unlikely to occur and very unlikely to lead to hostilities so, although
it is included in this analysis, it will not be discussed further.
9
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
111
intercommunal
incidents
TC initiated
? Figure 6 continues the GC vs TC branch of Figure 5 to further
categorize the type of intercommunal incident according to who
? instigates it. Although the TC's could initiate an? intercommunal
incident, as shown by the top branch of Figure 6, they are
outnumbered, outgunned and hold generally to .a defensive posture;
? any incident they perpetrate therefore is unlikely to be major. More
likely, it would be a situation from which they would quickly withdraw
and one that they would be anxious to resolve.
MAJOR
N 50.
IC INSTIGATE
T3$
08
02
19
to
MAJOR
, 01.
50
Figure 6
81 ORTVAS.
07
02
MINOR
30
AIWA
rc
Go
H
02
01
'
50
NOT OR AS
07
02
GC INSTIGATE
-GC's N TCr
MUTUAL
QNPDE.NTIAL
OTHER
MI
08
01
0
ea,
TI
5
07
.02
.01
MINOR
+H
14 3g
.02
AS
.06
.02'
40
70
08
02
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
?
No Objection to Declassification in Part 2013/09/11 :
On the second branch of Figure 6 the GC's instigate the
GC instigated.
intercommunal incident. They are more aggressive and have initiated
several incidents in the past. The most significant example is the attack
on TC villages which brought on the 1967 crisis; evidence also points
to GC instigation of several lesser incidents. While it is unlikely that?
any given GC instigated incident will be in the form of an attack, if
an attack should occur then Grivas would most likely be responsible
for it. Furthermore, an attack by Grivas could be expected to be of
major proportions, while any GC attack conducted by forces other
than the Grivasites would stand a good chance of being contained,
smoothed out, and eventually negotiated out of the picture. The
category designated "other" covers all other types of GC-instigated.
intercommunal incidents, and it is here we could expect the vast
majority of incidents to fall. Most of these incidents would be of little
significance and could be quickly contained and negotiated away.
The third branch of Figure 6 describes incidents that are mutually
instigated by both sides. These appear to be brought on by the peculiar
circumstances on the island, where opposing forces often stand toe
to toe, rather than by the design of either side. It follows that such
an intercommunal incident would be of minor proportions with both
sides being well disposed to quash an incident that neither instiriatPd
to begin with.
mutually
initiated
non-Cypriot
initiated
resultant
hostilities or
tension states
The fourth branch of Figure 6, designated "other", makes the
breakdown exhaustive. An intercommunal incident brought on by Arab
terrorists on Cyprus is an example of this type of incident. Here too
both sides would be disposed to contain the incident and it would
likely have a minor impact on hostilities.
The terminal branches on Figure 6 show that every intercommunal
incident will culminate either in hostilities or in one of the five tension
states. As expected, the high probabilities of hostilities follow the major
intercommunal incidents, particularly those instigated by the GC's.
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CONFIDEN IAL IIP
an inter-Greek
incident
12
CONFIDENTIAL
Figure 7 is an analysis of the inter-Greek problem; the second
of the three kinds of incidents contained in Figure 5. The return of
Gen. Grivas to Cyprus, the conflict between Makarios and the Athens
leadership, and the several clashes between opposing GC factions all
lend substance to fears that the spark to set off explosive violence
will come from inter-Greek fighting rather than an intercommunal
clash. Open inter-Greek fighting on the island could quickly spill over
into TC areas and develop into full intercommunal hostilities.
The top branch of Figure 7, some form of clash between the
Greek and GC factions, is by far the most likely kind of inter-Greek
incident. Over the past several months there have been several instances
of clashes, mainly between the GC factions supporting Grivas and those
supporting Makarios. There are other, overlapping factions, e.g., those
identified with the Cypriot Bishops, who have attacked Makarios on
religious grounds, and also forces identified with the Athens leadership.
Clashes between GC factions do not constitute a direct threat to
Makarios, however, and an incident of this type would most likely
be of minor proportions leaving Cyprus in a normal state of tension.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-.5.-9 IDENTIAL
Grivas
Athens
The second branch of Figure 7 addresses an inter-Greek incident
instigated by General Grivas moving directly against Makarios. The use
of force is characteristic of Grivas, and there is a high probability that
if he strikes, it will be a violent blow. The two principle alternatives
are (I) Assassination of Makarios, and (2) Coup d'etat. It is also possible
that Grivas will not use force at all. He may (3) rely upon a political
front organization to challenge Makarios or upon the dissident Cypriot
Bishops.
On the third branch of Figure 7, the inter-Greek incident is caused
by Athens moving against Makarios. A move by Athens is less likely
to involve force than is a move by Grivas. In the unlikely event that
Athens uses force, the options are assassination, direct military
intervention, work through Grivas, or some other means. The most
likely alternative is for Athens to use the strong military forces at
its disposal on the island rather than forces brought in from mainland
Greece. Athens has a regular Greek Army contingent of almost 1000
? men on Cyprus, and the Greek Cypriot National Guard is officered
by mainland Greek Army officers. Another option open to Athens
is to use the Grivas forces, however, whatever connections exist
between Athens and Grivas are obscure, making this option unlikely.
Other contingencies, not readily foreseen, are included for
completeness.
It is much more likely that Athens will not resort to force in
an effort to oust Makarios. Options under this category are: (I) Athens
supports the Cypriot bishops in pressing their case from the religious
angle. The prospect that this will prove effective is slight, but it is
a continuing form of pressure which represents an effort to unseat
the Archbishop from his temporal office. (2) Athens uses the Grivas
organization as a political tool to confront Makarios on purely political
grounds. (3) Athens uses the threat of force. In view of the resident
Greek forces on the island and the Greek officer corps of the Greek
Cypriot National Guard, Athens can apply significant pressure on
Makarios. This is the most likely of the above options. Other measures
(4) which Athens may take in pursuit of this goal are less likely.
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CONFIDE
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Figure 8 displays the entire probability diagram a synthesis of
Figures 5, 6, and 7. It shows that although there are many different
Nays in which events Can occur during the three-week interval, there
ire only six possible end states - hostility or one of the five states
A tension.
the entire
three-week
diagram
:ONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
(nNF1DENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
4111 IP
.70
.03
.02
20
GRIVAS MOWS
AGAINST MAKAMOS
ASSASSINAY
MAKARIOS
05
09
06
30
6.30
50
sA;504ArioN)(\st.
" .04
,N1/412,
H 35
30
33
OTHSR
ASSASSINATION
.40
.40 n
USE MILITARY
FORCE
t44.40
.2o
ROM 15
Ossece 4k05 .
).1 .02
50
N, 0
H PoT ..1,3
.3o
.3s
OTHER
" 30
00
02
40
.20
N 7,73r
USE GRIVAS
POLITICALLY
NOT use FoRce
70
05
0
20
05.
THRSATEN
FORC
OTHER
20
as
15
CONFIDENTIAL
?rr
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
ONF1DENTIAL
Because of this, the large probability diagram, which is
useful for analysis of the many possible implications of an incident,
can be summarized :by the simple diagram of Figure 9. It displays
the-lxPeotad Probability of terminating in each of the six possible
end states, There is a ..one percent chance of moving into hostilities
during the three-week interval but Cyprus is far more likely to continue
in a normal state of tension.
?
State at beginning State at end
of 3-week interval of 3-week interval
:CQN.FIDENTIAL
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-
CONFIDE,NT1AL.:
MARKOV MODEL
Figure JO extends the analysis for a second three-week period;
here too :the process -terminates in the same six end states if Cyprus
is in a normal state of tension at the end of the first three-week interval,
then the probabilities shown in Figure 9 are also applicable for the
extending second three-week period. However, if Cyprus has escalated into a
beyond three higher state of tension, different probabilities are required, in this case
weeks ? probabilities which imply a higher likelihood of moving into hostilities.
State at beginning of first 3.weelc paled, State at, end 'of fin+ 3-efeek
Petiod,beginning of second
3.wee1t'pastiod
'F4furf tO N (Currant nonnal state of tension
MGT
- State at end Olio:peel
3-week petkof
MGT '0491
G, 117 ?
.01 MOT
HOT-
MOO '
MGT
T
MOO
HOG
MOT
MGT
S.
MOO
HOG
17
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CONFIDENTIAL IP
Figure 11
? Example, if Cyprus were in a state of moderate
Greek-Turk tension at the beginning of any
? 3-week period, there is a 10 percent chance of
moiling too state of high Creek-Turk tension
at the end of that period.
Sl*mead of hm 3"..mok Whoa
STATES
1
N
2
MGT
3
HGt
4
MGG
S
HOG
H
1 N
.91
.04
.01
.02
.01
.01
2 ANGr
.51
.
..10
.02
. .01
.06
i '
3 MT
.16
.33
.33
.03
.01
?
.10
4 MOO
.40
.06
.02
-,:
.29
.09
.06
3 HGG
.13
.07
.02
.34
.34
.08
a summary
matrix
CONFIDENTIAL
State at beginning of fi
State at end of first 3-week period
? STATES
1
N
2
MGT
.-
3
HOT
4
MGG
5
HOG
H
1 N
.91
.04
.01
.02
.01
.01
2 MGT
.51
.30
.10
.02
.01
.06
,
?
? 3 HOT
.
.16
.35
_
.35
_
. 3
,_
.01
.10
4 MGG
.48
.06
.02
-
.29
.09
.06
.
.
5 HGG
.15
.07
.02
,
.34
.
.
.34
.08
Figure II displays the transition probabilities -- the probability of
moving from one state to another in any three-week period assessed
for all possible states. For example, it shows that there is a 91 percent
chance of remaining in the normal state at the end of any three-week
period if Cyprus was in the normal state at the beginning of the period;
a 4 percent chance of moving to a state of moderate Greek-Turk
tension, and so forth. Implicit in this matrix is an important underlying
assumption; namely, that the probability of moving from one state
to another is dependent only on the state at the beginning of any
period, not on how that beginning state was reached. For example,
suppose Cyprus was first in a normal state of tension, then in a state
of high Greek-Turk tension and then in a state of moderate Greek-Turk
tension; the probability that it will next move to the normal state,
or to any other state, is only dependent upon it currently being in
a state of moderate Greek-Turk tension, not how it got into that state.
This assumption makes it unnecessary to re-assess the transition
probabilities for extensions of the probability diagram of Figure 10
beyond six weeks. Thus, calculation of the probability that Cyprus
will be in a particular state for any given multiple of the basic
three-week period can be ?easill .accomplished.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-
Once the transition probabilities have been assessed, it is possible
to reduce the six-week probability diagram of Figure 10 to the simpler
diagram shown in Figure 12. This figure shows that at the end of six
weeks there is a 3 percent probability of moving into hostilities but
again, the most likely outcome is that Cyprus will remain in a normal
state of tension. One reason that the probability of moving into
hostilities is higher than at an end of three weeks is that there is twice
as much time for hostility to develop. A second, more important,
reason is that the increased time permits escalation to take place. That
is, it is possible to arrive at hostility not only by moving there directly
from a normal state of tension but also, indirectly, by escalating from
a normal to a higher state of tension and then moving into hostilities.
State at beginning
of first 6-week period
N (Current normal state of tension)
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
1111
1ONFIDENTIAL
and
de-escalation
The process of escalation and de-escalation is fully incorporated
only by extending the time horizon many periods. The abbreviated
state diagram, Figure 13, shows how this escalation and tie escalation
can occur. Each arrow refers to a single three-week period. The
probabilities attached to the connecting arrows come from the matrix
in Figure II. The ?red arrows show escalation toward hostility or
increased tension, the green arrows show movement toward decreased
tension, and the yellow arrows imply that the state doesn't change.
Tension between the Greeks and Turks is shown on the left and that
among Greeks on the right. The arrows which connect these
left-and-right-hand states are not shown. Any possible sequence, such
as escalation followed by de-pscatatipn, can be represented by chaining
together a sequence of three-week transitions. For example, Figure 14
shows such a sequence for a period of nine weeks. Beginning in the
normal state, it would be possible first to transition to moderate
Greek-Turk tension (l);.next, to remain at moderate Greek-Turk tension
(2); 'then to escalate to high Greek Turk tension (3); and finally to
de-escalate to a normal state (4).
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK--5-43-1--61DENTIAL
The process described by Figure 13 is called a Markov chain.
? Because of the particular transition probabilities given in Figure II, this
Markov chain, which represents the forces at work on Cyprus, indicates
? that there is a high probability of Cyprus being in low states of tension
? in the future. First, it is more likely (91 percent) to remain in a normal
de-escalation state than it is to move into a higher state of tension. Second, if it
is more enters a higher state it has only a 30-36 percent chance of remaining
likely there and, if it leaves, the most likely transition is a de-escalation to
a lower state of tension.
Figure 14
(4)
(2)
HGT
(3)
21
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11' : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
w.39NFID.ENTIAL:-.
Figure 15 shows, as a function of time, the probability that Cyprus
? % time in each will be in the normal state; assuming that the process begins in the
? tension State normal state and has not yet reached hostilities. Beyond approximately
four months, the figure shows that there is about an 85 percent chance
that Cyprus will be in the normal state and only abOut a 15 percent
chance that it will be in a higher state of tension.
22
CONF1DEN'TIAL
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CONFIDENTIAL??
Figure
1.+mrimi
1 Nov 72
? CONCLUSION :
THE PROBABILITY OF HOSTILITIES
The goal of this analysis is to assess the probability that hostilities
will begin on Cyprus. The red curve in Figure 16 shows how that
probability increases with time. This curve is derived from the matrix
14% chance in Figure ii. it thaws that there is about a 14 percent chance that
% of hostilities hostilities will begin within a six-month period and that it 15 about
within 6 as likely as not (48 percent) that hostilities will occur within two years.
months These assessed probabilities are based on starting in the current, normal,
state of tension. If, on the other hand, Cyprus were now in a high
state of tension, there would be a much greater chance of hostilities
beginning. For example, if there were now a high state of tension
between GC's and TC's the probability of hostilities would be given
?? by the dashed curve in Figure 16. In that case, there would be about
a 30 percent chance of hostilities within six Months and 60 percent
within a year and a half.
'40
?
sto*
3
months year. 18 months 2 rail,
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
ANALYSIS SOME OBSERVATIONS
inconsistencies
adjustments
THE ANALYSIS
The use of structured analysis procedures improved on a
completely intuitive analysis in two major ways. First, it encouraged
the analyst to think through his problem in a clear and logical manner.
Factors affecting the likelihood of hostilities were explicitly identified
and uncertainties surrounding the occurrence of future events were
formally incorporated into the analysis. Second, the use of probability
diagrams and the Markov model provided a means for the analyst to
play his intuitive assessments with respect to hostilities against the
implications of the model to insure !Nice! consistency. For example,
the analyst a) made a direct assessment of the probability of hostilities,
then b) assessed the probability of hostilities indirectly through the
probability diagram, c) assessed the percentage of time in a tension
state, and d) judged a time in the future such that he would be
indifferent as to whether hostilities would occur before or after that 1.it:r
time.
As it happened, these separate assessments were incompatible with
each other given the underlying structure of the model which the
analyst felt was correct. Consequently, it was necessary to adjust the
different assessments, constrained both by the logic of the model and
by the analyst's knowledge of Cyprus. It is not the case that the analyst
assessed tha probability of hostilities and then constructed a model
that was compatible with that assessment. On the contrary, given the
feedback from the model, he adjusts all of his assessments until
convergence was achieved between the implications of the model and
his perception of the situation.
COMMUNICATION
In addition to sharpening the analyst's perception cif the structure
of his problem, these analytic procedures also enhance the
communication .of the likelihood of hostilities and the reason for that
likelihood. In the first place, the use of a percentage, 14 percent, to
describe the likelihood of hostilities is certainly more meaningful than
the use of a verbal qualifier such as "unlikely". In addition, the use
of the probability diagram and Markov model communicates a
quantitative justification for the probability assessment so that if the
reader does not agree with that assessment, he not only knows that
he disagrees, but where and by how much. The use of the analytic
model thus serves to focus dialogue and debate in specific areas instead
of on the assessment as a whole.
25
(Reverse Blank)
CONFIDENTIAL
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Secret
DIRECTORATE OF
INTELLIGENCE
Intelligence Memorandum
Panama: 1973 - The Year of the Treaty?
State Dept review completed pages
59-74
MORI/CDF C05129177 Pages
59-74
Secret
1c3i.
28 November 1972
No. 2438/72
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
EJ Li
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
CENTRAL INTELLIGENCE AGENCY
Directorate of Intelligence
28 November 1972
INTELLIGENCE MEMORANDUM
PANAMA: 1973?The Year of the Treaty?
The current round of canal negotiations which began in June 1971 has
been stalled for the past eight months, in part because General Torrijos was
unwillinglgsgmelf_on_suktantallsges_Qt_al1ow?1.1rJaBs to move
from an explorato to a bar,ainin shase. Having solidified his position and
_legitimized his rule Torrijos can now begin to focus on t ie can issue.
1:.'-7'&?mi-n7the Pariamanian and US positions is substantial, but Panama
apparently believes that the US can be pressured into offering additional
concessions. Negotiations will probably resume in Panama in the near-future.
Torrijos has approached the negotiations with ambivalence and appre-
hension. Panama's position as he perceives it is one of moral superiority, but
material and tactical inferiority. He believes that international opinion sup-
ports Panamanian aspirations for control of the Canal Zone and that even
the US public, if it could only be informed, would regard the current
situation as anachronistic. On the other hand, aside from his ability to
embarrass Washington, Torrijos realizes that Panama has relatively little
bargaining leverage. Most of the give is on the US side, most of the get is on
the Panamanian side.
Because Torrijos does not approach the negotiations as an equal with
equivalent points to concede, he seems deeply_mpiAmisjaf,X24.1wrrol
? negotiating ntos.ess. He apparently sees the canal issue more as a political
? rather than a diplomatic erobleni. As such, it would require a broad agree-
ment on prmciples bloc ed out at the to4L.w.tl the details left for the
diplomatic tec tructans to handle.
Despite suggestions from the US and his own foreign policy advisers
that he develop a completed position on all issues, Torrijos thus far has
Note: This memorandum was prepared by the Office of Current Intelligence and co-
ordinated within CIA. - ?
bk,U.K P., 1.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
0 411P 4
? SECRET ?609
retained maximum flexibility and has only limned some of the basic in-
gredients of the new treaty. There are, of course, certain points that Torrijos
regards as essential to any agreement: a definite termination date for a new
treaty after which the canal would be turned over to Panama; full Panama-
nian jurisdiction over the Canal Zone, including an immediate end to trials of
Panamanians in US courts. On a number of other issues, such as compensa-
tion, a status of forces agreement, and land use, Torrijos has not yet made up
his mind, but these issues appear to be negotiable.
A major impediment to an agreement is the inability of Torrijos' aides
to persuade him that a US offer is a good one or that it is firm and final.
Afraid of being outmaneuvered, he is probabl unwil
of the US position that does not 'come dirpctlyjrprn_t4e thest levels of
government Also, he has tended to believe that threats and prTsi7u-e7W?ni t
bnz-711?reaTif-concessions. Over the past few months Panama has attempted
to lay the groundwork for the possible application of such pressure.
Panamanian ambassadors around the world have publicized Panama's side on
the canal issue, Panamanian representatives have sought support at various
international conferences, and a major effort has been made to have the
United Nations Security Council meet in Panama City in March.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-53R_R
25X1
25X1
Oft
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
SECRET
Preparations for Negotiation
To negotiate or not to negotiate has never been the problem. Instead,
when to negotiate, where to negotiate, how to negotiate, and what to
negotiate have been the questions. It took the Torrijos government 2 year to
find answers to the first two questions and another year to resolve the third.
The fourth question still appears to be the object of continuing discussion, if
not confusion.
? Torrijos inherited the canal problem at a most difficult time. Panama-
nian dissatisfaction with the 1903 treaty had culminated in bloody rioting in
January. 1964. A number of Panamanians and US soldiers were killed, and
subsequently Panama broke relations. with the US and complained to the UN
and the OAS. In April, relations weie resumed, and representatives of both
countries initiated discussions on means of eliminating sources of conflict. In
December, President Johnson announced that the US was prepared to
negotiate a new treaty, and for nearly three years an arrangement that would
satisfy Panamanian aspirations without impairing US security interests was
sought. Finally, in June 1967, Presidents Johnson and Robles announced
agreement on three draft treaties: one on the present canal, one on US
military base rights and status of forces, and on a new sea-level canal.
Pan r a e4, ? Itch Jtadqtj?y come,
nder intense political fire. The Robles government became involved in the
bitter y oug i campaign preceding the May 1968 presidential election, and
Arnulfo Arias, who won the election, barely had time to adjust his presiden-
tial sash before he fell victim to the military coup that brought Torrijos to
power in October 1968.
drnft ParLaa.? They voided the
hated perpetuity clause, and called for the cession of the lock canal to
Panama by the year 2000 and the reversion of a proposed sea-level canal by.
the end of 2067. Panama would have promptly received land and water areas
no longer needed for defense or canal operations. The fixed annuity of about
$2 million was done away with, and, in its place, Panama was awarded a
royalty on tonnage passing through the canal which at present transit levels
would have yielded approximately $24.4 million in fiscal 1972. Nevertheless,
the agreement was criticized and later rejected because instead of giving
Panama full sovereignty over the Canal area, a foreign enclave under a joint
US-Panamanian administration in which Panama was a minority partner was
? to be continued. It was also hit because it did not commit the US to build a
sea-level canal and because it perpetuated the large US military presence.
btA....1:12, I
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
?
25X1
? Without concerning himself at all with the merits of the case, Torrijos
seems to have immediately grasped three points First, he could not accept
the drafts as they were. Presenting himself as a true nationalist and reformer,
he had to put his own stamp on a treaty and had to get a better deal than
had the oligarchs. Second, some mending of fences with the US was required
before Washington would even discuss the canal issue with him. The US, he
sensed, was not ready to cozy up to a government that had so blatantly
flouted the popular will and that lacked even minimum constitutional
underpinnings Third, the canal issue was too complex and far too important
to negotiate without some degree of planning and preparation. Torrijos had
only the vaguest conception of the underlying issues, even less idea what he
wanted to achieve, and virtually no time to devote to the matter. He first
had to consolidate his power and learn how to run a government
By mid 1969 he had appaiently begun to feel more secure in his
position and confident enough to tackle the canal issue. A canal advisory
group was appointed to study the 1967 drafts, and the government-con-
trolled press began to focus on the canal issue. When Tonijos was in New
York in September 1969, he told US officials that he would like to resume
the negotiations as soon as possible.
Torrijos at that point had develoned no clear negotiating goals, strategy,
or timetable. Although Torrijos occasionally mentioned that he was under
domestic pressure to resume negotiations, the public showed little interest in
the canal issue. The people were seized instead with gaining a fix on where
the Torrijos government was heading, particularly in internal affairs.
Torrijos' initial push for negotiations may, therefore, have been less an
interest in getting talks going as a tactic to get the US to deal with and
? accept his government. Rather than excite hopes that he would quickly
succeed where other governments had failed, Torrijos seemed merely to want
to convince the public that the US would negotiate with and support his
? Provisional Junta Government.
Torrijos' concern over the direction of US-Panamanian relations in-
creased during the next few months. A coup attempt by military rivals in
December 1969 was blamed on the US by Panamanian officers who were
frantically trying to divert Torrijos' suspicions. In February 1970 an aide to
deposed president Arias claimed the US had supported his efforts to organize
a coup; when the aide was unable to complete his plans, he fled to safety in
the Canal Zone. In July 1970 the three colonels who had led the December
coup attempt escaped from jail with the help of a US Navy civilian employee
SECRET
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
oft
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
SECRET
residing in the Zone., In August, a Canal Zone policeman who was a
Panamanian national confessed that he had used the Zone as a refuge while
plotting against the junta and that he had acted as a link between plotters in
Miami and anti-government elements in Panama. ToxijoLbitterly ?merited
the availabilit of th- ? ac ? .3?..e.ulor...his_opp.on nts
negotiating objectives?full Panamanian jurisdiction over the Zone?began to
aaalesce.?TraTibly more tlirn?Fier IrcirriI5iTiewed the resumption of active
negotiations as necessary proof of US acceptance of his regime.
In August 1970, as the media began a drumbeat of propaganda against
the Zone, Torrijos formally rejected the 1967 drafts, indicating that the
government had its own idea on the canal agreement. The frost in US-
Panamanian relations persisted, and there were hints that Panama would
0{-- denounce the 1903 treaty before the United Nations General Assembly. ?
Calmer heads prevailed, however, and instead of a confrontation, there was,
in October;an amiable meeting between President Nixon and Panamanian
President Lakas, who was visiting the US in conjunction with the 25th
anniversary celebration of the UN.
Despite this meeting, US-Panamanian relations remained somewhat
roiled as Torrijos made the most of every opportunity to exploit problems
with the Zone and to extract maxitnnrn propaganda mileage. lithe govern-
ment's agitation Over accumulated irritants was the stick used to prod the US
toward negotiations, the Torrijos government did not forget the carrot.
Panamanian negotiating objectives, to the extent that they were formulated,
were couched in very broad and seemingly reasonable terms. Whether
deliberate or not, there seemed to be a slurring of differences, an implication
that agreement could be reached hi. short order and a hint that if desires for
jurisdiction could be met, then perhaps Panama would be accommodating on
other points. In late June 1971 negotiations were formally resumed.
Negotiations?Phase One
? The Panamanians approached canal negotiations with an almost reli-
gious fervor. They were searching for the promised land and were beyond
the point where minor changes or a more cosmetic packaging of the 1903
treaty structure would suffice. It was not out of character, therefore, that
when the Panamanian negotiators repaired to the bar ainin table they were
read to stake out a maximurn?Tatiorion broad issues but were wea on
. It quickly became apparent, for example,
that Panama saw no possibility for a trade-off between increased jurisdiction
and a longer time frame for the treaty. Although the .government had
25X1
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
Ll -
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
rejected the 1967 drafts, the drafts were clearly regarded as a bargaining
floor. Similarly, all of the concessions that Panama had previously won were
considered to be closed to further negotiation. The only question, as the
Panamanians saw it, was how the terms could be further improved.
The Panamanians immediately insisted that anew treaty must have a
faxed termination date. When th?ey that they indicated they We=
1-7FIMM-d=15Fjposal providing for US control of the present canal for
SO years, the present canal with an added lane of locks another 35 years, and
alternatively a sea-level canal for another 40 years?even though the final
termination date was earlier than provided for in the 1.967 drafts. The
Panamanians also demanded an end to the Canal Zone government and the
establishment of Panamanian control ovefall normal governmental responsi-
bilities such as police, fire, hospitals, schools, and. the mails. Furthermore,?
they demanded that Panama be granted full jurisdiction in the Zone within
five years, that all US commercial activities be turned over to private
ownership within two years, and that all US criminal jurisdiction over
Panamanians be ended as soon as a new treaty went into effect. They hinted
that US base rights treaties with Spain should set a standard for compensa-
tion.
Panama nlonger v_kmecjilae. jeaLoisampe.usatian_groyided by the ,.
1967 drafts?approxamate.,2_)_miltuan....-..as?adequate; they suggested
Inste nount vo ama insisted that the US
cede substantial land and water areas, particularly near Panama City and
Colon and that Panama have the final say on secondary uses of the land
retained by the US for the operation and defense of the canal. Panama
objected to the denial of the canal to nations at war with the US, arguing
that the canal should be completely neutral. Although not prepared to
discuss defense arrangements in depth, the Panamanians proposed that their
own forces play a role in canal defense and that any other functions, such as
hemispheric defense, be separately and specifically negotiated.
? Even on the issue of canal expansion?virtually the only quid pro quo
Panama could offer in return for the myriad concessions sought from the
US?Panama remained hard nosed. The negotiators opposed a US request for
a straight 25-year option during which a decision on expansion could be
made. Instead, Panama argued that the US should have a 15-year option,
renewable for two five-year periods and that the US should pay for each
renewal.
The junta played down public commentary on the progress of the talks.
Nevertheless, the canal issue was kept before the public eye and efforts were
bl!,(Atil I
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
An.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
made to garner declarations of support. The official mood was alternatively
optimistic and threatening as the Torrijos government studied the US posi-
tiOn and assayed Washington's willingness to compromise.
Late in the summer of 1971, when preparations were beginning for a
massive celebration of the third anniversary of Torrijos' seizure of power,
Panamanian concern over the progress of the negotiations was reflected in
rumors that there would be anti-US demonstrations or? even incursions into
the Zone. As the anniversary approached, however, Panama a arentl ha
arsalLthaughts. The rumors died down, and at a mammoth rally - orrijos
limited himself to a thinly veiled threat to lead the people into the Zone if
the negotiations failed.
By November, the air of expectancy about a new canal treaty had
shifted to pessimism. Once again the government began to create a more
tense and uneasy atmosphere, and its relationship with the Canal Zone
government noticeably cooled. Rumors of a student incursion into the Zone
again were generated, and Torrijos warned that failure to achieve a satisfac-
tory agreement would lead to a "spontaneous" explosion.
Despite these ill-considered attempts to give the negotiations a nudge,
the talks in Washington were proceeding in a businesslike and constructive
manner. In December 1971, however, a decision point was reached. The
Panamanian negotiators had articulated and then refined their position and,
in turn, had received a detailed US offer. There had been a frank and open
exchange as both sides explored the issues. The Panamanians had little room
for maneuver, however, and returned home for new instructions.
The Panamanian negotiators were probably surprised to find that
Torrijos viewed the progress of the negotiations with a jaundiced eye. He was
unenthusiastic about the US offer and was not willing to give the negotiators
. additional flexibility. He was apparently uninformed on details and was not
yet prepared to make the hard decisions that were necessary. In short, he did
na_tapriLthr?amotatians_witiunach_semit.of..urgp,,n.cLabm.the Panaman-
ian negotiators returned to Washington after the Christmas holidays, it was
clear that no basic policy decision had been made.
From January through July 1972, the Panamanian negotiators at-
tempted to persuade Torrijos to issue new instructions so that they could
make a reasonable counteroffer. But, aside from four sessions in March, the
talks have been stalled because the Panamanian team was not in a position to
participate in the negotiations.
SECRET
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
?
1
1
REINVINDICACION
PATRIOTICA, VA ! !
The vindication of the fatherland is on the march!
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 :LOC-HAK-538-3-5-6
t..,4
SECRET
? Part of the difficulty was caused by a dispute within the Panamanian
foreign policy apparatus. With Torrijos playing an essentially negative role?
criticizing US proposals rather than suggesting alternative solutions?the
negotiators were forced to compete with a number of individuals and groups
basically espousing an idealistic, nationalistic line. The Foreign Relations
Advisory Council, for example, composed of a number of experts, weighed
Sin with an extremely chauvinistic position. Foreign Minister Tack, though
somewhat more reasonable than the council, also advocated an uncom-
promising stance.
Periodially the US was told that a comprehensive Panamanian position
was being drafted. In June, for example, when US negotiators visited
Panama, Tonijos claimed that a paper would be ready within ten days. In
July, one of the Panamanian negotiators indicated that a position paper,
which he described as "flexible" and "really forthcoming," was almost
completed. It was becoming clear by this time, however, that Torrijos was
preoccupied with domestic matters, particularly the August legislative elec-
tions, and was not interested in pursuing negotiations until after the US
elections in November.
The 'Domestic Situation
In October 1971, during the anniversary celebrations, Torrijos had
presente his audience. He spoke about the canal, and he
promised that there would be national elections by the following August.
Given the government's limited capacity and Torrijos' own temperament, it
was clear that both paths could not be traveled at the same time. It was
probably about this time that Tonijos decided to concentrate on the
elections rather than on the negotirMir'''-??"*"
There were cogent reasons for such a choice. The extralegal nature of
his regime had always been a sore point with Torrijos. He had promised
elections before, but each time he had allowed the date to slip by. He felt
also that there was? a direct link between his desire to legitimize the
government and his search for a new treaty relationship with the US.
Although he spoke of ratifying a treaty through a plebiscite, he believed that
the US would not sign a treaty before constitutional government had
returned to Panama. He also felt that his bargaining position in the negotia-
tions would be strengthened by an election.
Torrijos perceived the election as little more than a vote of confidence
on his rule. Panama's many political parties did not participate in the
t i
No Objection to Declassificationi" I.
nPart'2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
IP
SECRET
balloting for the 505-member assembly, which was skewed in favor of the
rural areas where literacy and sophistication were low and Torrijos' pcpu-
larity high. The candidates, carefully screened by the government, were
mainly local figures without experience in politics.
The assembly was elected in August 1972, convened on 11 September,
and remained in session only one month. It had only two rubber-stamp
functions: to elect a president and vice president, and to approve a new
constitution. The National Guard provided escort officers to each provincial
delegation to ensure that delegates were properly informed of Torrijos'
wishes.. Provisional President Lakas. and junta member Sucre were elected
constitutional president and vice president. Torrijos, in a "spontaneous"
gesture of affection, was granted' full powers to continue running the
government. The constitution was approved with minimal alternations.
The biggest publicity splash for the assembly was its resolution calling
on Torrijos to demonstrate Panama's sovereignty over the Zone by refusing
the $2-million annuity from the US. The resolution was not binding, and the
regime claimed that it came as a complete surprise, but Torrijos said in
response to reporters' questions that he had no choice but to follow the
expressed will of the people.
Negotiations?Phase II
Panamanian leaders have often behaved as though the negotiation of a
new canal treaty was of the most critical and sublime urgency?at least until
the United States agreed to negotiate. Then negotiations have gone into low
gear. Delay and procrastination have become the order of the day, promised
positions never appear, and promised compromises never materialize. When
agreement finally seems possible, fear sets in and second thoughts develop.
This paradox can be explained rather simply. From a strategic point of
view, the abrogation of the 1903 treaty and the development of a new canal
arrangement are the most important foreign policy goals of any Panamanian
government. But tactically there is always dissonance between the Panama-
nian diplomat's sense of what is possible to achieve and the Panamanian
politician's sense of what Panama's nationalists will accept. In short, --
? Panama seems in some ways EngnITSE17767377-vith US intransigence than
with US assent If the US disagrees with a Panamanian position it can be
pilloried, but if it accedes, then the Panamanian position itself becomes
suspect. The government wonders whether it understood the full significance
of the official position, or indeed, if it asked for enough.
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-_R
t
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
11/
...ti WIPES DE13EN
ESPERAR HASTA EL ono
SILO, VER QUE
LE vonmos PAR!
...and you will have to wait until the next century to see what we can give you!
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LO?C-HAK-538-3-5-6
Niiee
SECRET
This situation stil obtains adv. The government is divided and unsure
of its negotiating tactics. OrrijOS is afraid of giving away something lie
should not and of failing to get something he should. Ile does want anew
lisityjny, evermijno _many ways the time is ripe. TEra7in?aii?altis
domestic political necessities taken care of, and noVv, after four years in
power and with the people convinced of his concern for their well-being, he
is in a position to redeem some of his promises and substantiate some of his
rhetoric.
Torrijos is likely to see the canal issue as the most fruitful pursuit at
this time. He tends to concentrate on only one problem at a time, and since
the likelihood of achieving quick victories in agriculture or social develop-
ment is remote, he probably views ?domestic problems as less promising and
more intractable than the treaty issue.
The canal problem, moreover, represents to some extent a constraint on
Ton-ijos. He likes to portray himself as a "revolutionary." Whether he has in
mind moderate reform with a slightly nationalistic tinge or ambition to
pattern himself on the Peruvian or even the Cuban model, he realizes that it
could be counterproductive to be very "revolutionary" before he has com-
pleted negotiating a treaty with the US. The government bac also been
somewhat strapped for funds, and a new treaty would substantially increase
revenue. Instead of the approximately $2 million Panama now gets each year
from the US, it could expect to receive $20-25 million at the very least. This
money and the increased bon-owing capacity could make a substantial
difference in the ability of the regime to promote social and economic
reform.
The negotiations also represent something of a constraint on foreign
policy. Torrijos, sensitive to his country's traditional reputation as a virtual
protectorate of the US, wants to strike an independent pose. Yet so long as
the negotiations are under way, he has to factor in the effect which any
other foreign policy initiatives would have on the talks.
Tonto? interest in a treaty does not, of course, derive solely from a
desire to be free of certain constraints. The negotiations provide him with an
opportunity to win lasting fame and a prominent place in the pantheon of
Panamanian patriots. In a sense, replacing the 1903 treaty structure would
be the most revolutionary act possible, and Torrijos feels that he is the man
who can do it.
This does not mean, however, that he would settle for just any treaty.
Not only does he sec time on his side, but getting the "wrong" treaty?one
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
25X1
JOS.
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
IIIP
SF CR. FT
giving too little to Panama?would be far worse? for his reputation than
getting no treaty at all. The next election is not until 1978, and there is no
public clamor of the sort that would in :pose a deadline or even a rigid
timetable on his conduct of the negotiations.
Given Torrijos' lack of confidence in dealing with the mass of details
involved, his negotiating approach will probably continue to concentrate on
broad issues and principles. He is likely to keep on ? la in a reactive role
refining US ?ro osals r
s own detailed osition. This
approach could work to the advantage of the US. If Torrijos were orced to
develop a comprehensive standjt_Itygpld probably become the definitive
one. Given Torrijos' fear of committing himself and of beingoutmaneuvered
he wou
included all of the nationalistic items sugzestecl b
then in very lax t to bac
TmraCtilto Jigfor...len; He prefers things that the people
can readily see and understand, not those that require careful study before
they can be identified as a "good deal" for Panama. He wants quick and
concrete changes when the treaty is ratified rather than concessions that will
be implemented in a decade or two. This, IL would weal, is why he has put
so much emphasis on a rather brief transition period before turning jurisdic-
tion in the Zone over to Panama.
If Torrijos' basic objectives are met, if he can get fairly complete
jurisdiction over the Zone in a relatively short period of time, and if he can
significantly shorten the duration period that was embodied in the 1967
drafts, then a treaty agreement may be possible in 1973. The Panamanians,
to be sure, would bargain hard on the other issues, some of which have not
yet really been explored. Nevertheless, then would be a flexibility on such
questions that does not now exist on the duration and jmisclietion issues.
Torrijos recognizes that he is in a weak bargaining position. He has little
power to force his views on the US and has little to offer. Apart from the
right to build a sea-level canal, the US has all that it wants and far more than
it would have under a new treaty. Although Torrijos himself is responsible
for the delay in negotiations, paradoxically he sees the US at fault for the
lack of progress. He seems to feel that he cannot get the US to focus on
Panama and that the negotiations are on a back burner in Washington.
As a corollary, Torrijos apparently believes that only when'the US sees
Panama as a problem will the treaty receive the kind of attention it deServes
No Objection to Declassification -in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
SECRET
and that only by being tough on all points can he maintain a credible
bargaining position. Panama's questioning of the US interpretation of canal
neutrality, for example, was probably designed to make a continuation of
the status quo seem like a concession. Similarly, its attempt to redefine the
US defense role seemed directed at making US retention of military rights
seem like a quid pro quo. Torrijos apparently believes, moreover, that
pressure tactics will work, that a hard, perhaps even an uncompromising,
stance is the only way to win further concessions. Finally, there is the
feeling, born of insecurity and self doubt, that only by squeezing a little bit
more can he be sure that he has, in fact, pushed the US to its final position.
During much of 1972, Panamanian moves seemed designed to increase
its nuisance potential and to force the US to take the Torrijos government
more seriously. This concern with creating the "proper" psychological
,climate for negotiations was reflected in numerous attempts at bluster and
threat, in statements at international conferences drawing attention to the
canal issue, and in efforts by Panamanian ambassadors to mold public
sentiment in their favor. The most significant undertaking, however, was the
campaign to have the United Nations Security Council meet in Panama City.
Panama has been actively lobbying for such a meeting since September
and seems to have the necessary support (9 of 15 affirmative votes). It will
probably seek a vote on the question in January. Panama wants the meeting
held in March when it is scheduled to chair the council. This would be a big
opportunity for Torrijos to develop a world reputation and enhance the
prestige of his government. Panama will be on the SecuritiSzacil on
tatiLPecember 1973. anstaarijos?will-not-be-easily-persuaded-tc-gixe.W3
this culu4..._...ne opportunity.
? Panama has admitted that a major purpose of the meeting would be to
focus world and particularly US public attention on the canal issue. It
apparently hopes that if it can build up the meeting as a strong possibility,
the US may feel compelled to defuse the event by offering generous terms to
.win agreement on a treaty before March.
Apparently in line with this strategy, the ToLeijouragua ta.atiati2Lrr,ied
the October it is prepared tjuataLtlejLtiations but in
. Foreign Minister Tack, however, based
Panama's proposal on an earlier letter from the US, which had restated the
American position. Tack chose to interpret the letter as indicating a new
willingness by the, US to compromise. TaCk's reply as well as other govern-
ment statements seem to indicate that Panama intends to stand pat on the
- 10 -
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
ank, geA
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
? SECRET
basic issues of jurisdiction and duration, seeking but not offering com-
promises, and relying on the US to supply the initiative.
Torrijos' efforts to demonstrate to Washington that he could be tough
have always been balanced by signals that he was really quite reasonable.
When Cuba seized two US-owned Panama-flag ships that allegedly were
involved in illegal activity against Cuba, Torrijos tried to be helpful in
securing the release of the crew. He pointedly avoided, moreover, causing
any problems during the US presidential campaign. In sum, Torrijos has
exhibited a sense of pragmatism and realism. He has kept his "yankee
baiting" within controlled limits and, while occasionally causing the US
some discomfort, has always backed off when the situation threatened to
become serious.
? Thus, while believing that a Security Council meeting in Panama would
give his government bargaining leverage, even if negotiations do not bear
fruit by March, it is unlikely that Torrijos will automatically carry out the
implied threat to embarrass the US at the meeting. During a meeting,
Panama would probably seek to conduct itself with maturity and restraint
lending credibility to its assertions that it is prepared to assume respon-
sibility for control of the Canal Zone. Of course, even if Panama were on its
good behavior, a Security ro,,rrcil session in Panama ?,vould not be p"'nlcss
for the US. Panama would probably lobby extensively for its position behind
the scenes. There might be "guided tours" along the Canal Zone borders and
unfavorable comments regarding the status quo from foreign journalists and
diplomats. While the government would certainly try to prevent any ir-
responsible incidents, it might feel compelled to allow some public expres-
sion of nationalist sentiment for a new canal arrangement. After March,
Torrijos would evaluate the progress of the negotiations, particularly any
new US offers on jurisdiction and duration. He would then re-examine his
tactics and decide whether to step up the pressure or move toward serious
negotiations.
- I -
I
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
WORKING PAPER
4 December 1972
No. 2441/72
PROBABILITY ANALYSIS OF THE CHANCES OF
A CANAL TREATY WITH PANAMA BY END OF 1973
? The probability analysis that follows developed
as a side product during the preparation of a CIA
? Intelligence Memorandum on the Panama Canal Negotia-
tions.* The analyst, attempting to refine and sup- 'JO
plement research techniques, usega_this method in an tl
?tIZSELta_pin down some of the relevant factors 5aa-- G
to project their impact over the next ear. He bene-
fite rom iscussion with colleagues, but the resultr?
are not formally coordinated.
This working paper may be of some use as point-
ing the way Lu techniques that could be experimentally
applied to other intelligence problems. It represents
only the author's position.
*Panama': 10973.--21he Year of the Treaty? 28 Novem-
ber 1972iNo. 2438/72
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
eft
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
SWIRRT
In an attempt to sharpen the analyst's percep-
tion of the structure of the problem quantitative 4'S'
. . . .
2M?tifiCatiOn is substituted for the probability
assessments that are usually expressed by verbal
qualifiers such as "likely" and "unlikely." Key
factors affecting the prospects for reaching an
agreement on a new canal treaty in 1973 are ex-
plicitly identified and displayed by the use of
a probability diagram. It is important that this
format not imply a false precision. No attempt
is made to do more than present analytical judg-
ments in quantitative form. The reader by follow-
ing along can decide whether he agrees or disagrees
with the probability assessment. If he disagrees
he can more easily than in a textual presentation
identify those assumptions he disputes and pinpoint
where and by how much he would alter the assessment,
thereby focusing more sharply on the problem.
The probability diagram is a structured pro-
cedure to encode the expert judgment of an analyst
or group of analysts. It details the factors (and
the linkage between the factors) that bear on an
analytical judgment and expresses opinions about
the likelihood of their occurrence in quantitative
form. The first step in constructing the diagram
? is to select the factors that influence the proba-
bility of a treaty agreement. Then, proceeding
through the diagram, the analyst makes probability
assessments at each node, assuming that all of the
preceding branches represent events that have al-
ready occurred. Computations are made by using
the rules of probability theory; the product of
the branch probabilities yields a path probability.
This analysis develops the probability of a
canal treaty in 1973 based on three alternative US
policy approaches. These are not exhaustive of US
0 tions, but are mergly_offered_to_ea a
pro i ies. Under the first option, the US
?..a4-apts an aggressive approach through 1973 undeterred
-'by Panamanian "toughness." The US takes the initia-
tive in presenting proposals and then re-fines these
S1.C.12 Ti.r1
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
_ rft
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
IIP SECRET
proposals in light of Panamanian criticism. In ad-
dition, it Agnumpd thp US adopts a position
that comes closer to meeting Panamanian aspiratiga"--
-
on duration and jurisdiction.
?
The second option assumes that the US demon-
strates a v'eryatje attiErin-177==1"-Mmmm-ar-
iations and is prepared to offer at"-773-T,3:23,4/2_4,
certain concessions outside of .IEJ-Eagotiating frame-
work that win Panamanian good will
and demonstrate the good faith of the US. These
concessions might include joint police patrols in
the Zone, flying the Panamanian flag on vessels
transiting the canal, or transferring certain fa-
cilities within the Zone to Panama. The US under
this option does not take an activist position on
the negotiations, and makes
in....LULAtaad-caL.4Ww.allon_anAjurisdiction.
The third option assumes that the US is ready
tajral_but that after it makes a comprehen-
sive offer to Panama, the ball is in the Panamanian
court. Under this approach the US stands back and
leaves the initiative to Panama.
Outside the negotiairHg
framework and does not sulstantiallyH modify current
pstions on duration and jurisdict.on.
co,
an
Under each of these policy options, an assess-
ment is made of the probabilitx_thaLl'atELI11.10111.1_
r.emain Lp_,Q,Nex_tirhrough the end of 1973 and the
probability of a decision being made in. January to
Securitv Then an
assessment is made of the probability that events
occurrin rior to Ma h will i g_Eanam4_a_pd-s=rve
a itu e toward the US. Such events might include
a-ErgE:Tgra-E17777Egi meeting, some advance in the
negotiations, or some interim concessions. The
probability of events occurring prior to March,
? events such as coup plotting linked to activities
in the Canal Zone or unfavorable publicity on nar-
cotics, which might sour Panama's attitu5le toward
the US is then considered. Next the probability
of Positive or negative events that might cancel '
RRCT FT
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1::
25X1
eft, erA
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
grriz -wr
each other out and the possibility that no signifi-
cant events occur and that Panama's attitude toward
the US is neither particularly positive nor negative
is evaluated. In light of these factors, Panama's
probable behavior at the meeting--whether it will
adopt a vitriolic anti-US tone or a non-aggressive
tack--is assessed.
Under each of these conditions, the ossibility
of Torrijos becoming personally involved in e
tia ing process is evanaer--TEZET-Erga- on the
presence or absence of Torrijos' personal involve-
ment and in light of all of the previous factors
an assessment of the probability of signing a treaty
in 1973 is made. Finally, an over-a11 probability
of reaChing a treaty in 1973 based on each of the
three policy approaches is computed. The assess-
ment in all three cases concludes that the chances
that Torrijos will remain in power through the end
02=7.3_,Lte_zezy brgE.- If Torrf3761-Is re15.75147-tnM
resulting power-Fialignments would impair Panama's
ability to concentrate on anot .UYEty-na-r-trere-Vu'rifd
be . mh,
bility that Torrijos_(43_years old) will die or be
removed _from power next year. A71wo perbefft
bility is assigned only because Torrijos travels
extensively through the interior on Panamanian-main-
tained aircraft that use unimproved landing sites
and because there is an outside possibility of as-
sassination.
.44444.?
/4---472;
ce?M/ --t
1,02-z-m4 ?Lot/ X":" C?
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
rit
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
US Policy. Posture A
Torrijos .Becomes
Security Council Net Effect Panama Behaves Personally
Torrijos In Power Meeting Is Of Events Maturely InvoIwo.' Treaty By End
To End Of 1973 Held In Panama Before Meeting at Meeting In Negotiations of 1973
Positive
BO
Yes 175 Neutral
15
70 Y"
^ Yee
Yes ^ No ?
Yes
74
10
No
74
35
30
No GS. No
Yes
Yea
84
0
No ?
Yes
Yes
76
20
4.,Z341
10\
No25
05
No
' Yes
75
?
No ?
Yes
. No
Yes
0.."'3%?
No
YesYes
80
10 No
Yea
No
75 No
Go Yes
Yes
Yes
Yes
No
02
No
00 Probability of specific oceurance assuming events have reached this point.
00 Probability of a treaty agreement from that point forward.
Nate: Numbers am in percent.
,01% 12 72 OA
60
No
or...noun,
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
e%
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
SECRET
?
? Policy Posture A--US actively pushes the nego-
tiations. This course of action yields the highest
probability of reaching a treaty agreement in 1973.
A fairly high probability_(90 percentl as-
ska=i_12-thQ-EnlaPect
Council meeting in Panama even if the US takes the
ur-7E3=?tive.11.7?trlie negotatiOns. The meeting has
positive value to Panama quite independent of the
canal issue, substantial momentum has developed for
such a meeting, and newspaper publicity that has
already appeared would make a reversal of policy
rather difficult. If as a result of US representa-
tions Panama does reverse course and there is no
Security Council meeting, this would indicate Tor-
rijos' belief that an agreement is possible in a
relatively short time. There is consequently a
very high probability that Torrijos would become
personally involved in the negotiations and a high
probability that a treaty will be signed in 1973.
? After a decision is made to have a Security
Council meeting in Panama, Panamanian behavior dur-
ing the meeting will depend in part on the net im-
pact of events occurring before the meeting, but
even if untoward incidents impair US-Panamanian
relations, there is still a 60 percent probability
that Panama will adopt a mature attitude toward the
meeting.
/f the US pushes the negotiations after the
Security Council meeting, the probability is high
that Torrijos will become personally involved in
the negotiations. If so, there is a high probability
of reaching agreement; if he does not participate,
there is a very low probability. Paradoxically,
the probability that Torrijos will become personally
involved is somewhat higher if Panama behaves in a
nationalistic rather than a less aggressive manner
during the meeting. The explanation is twofold:
Torrijos would emerge from the meeting as a hero
capable of protecting Panamanian interests, and he
would regard continued US willingness to negotiate
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
? _ Oh'
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
0, SECRET
under. tough Panamanian conditions as a signal that
an agreement was possible. Xnteresting too is the
judgment that if the net impact of events before
the meeting is negative and Panama still behaves
in a mature manner, there is a higher probability
that Torrijos will become personally involved in
the negotiations than if the net impact of previous
events were positive or neutral. Torrijos would
regard Panama's good behavior as a signal to the
US of a particular interest in concluding a treaty
and would reason that the US would reward him for
his restraint.
SECRET
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
25X1
25X1
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Security Council
Torrijos In Power Meeting Is
To End Of 1973 Held In Panama
US Policy
Net Effect
Of Events
Before Meeting
Posture B
Torrijos Becomes
? Panama Behaves
Maturely
at Meeting
Yes
No.
Personally
Involved
In Negotiations
Yes
Treaty By End
of 1973
Yes
cfr
50
g5
No
50
?
Yes
No
Yes
DO
70
No
Yes
40
34
73'0"%r
?
No
is
Yes
No
00 Probability of specific occurance assuming events have reached this point.
CO Probability of a treaty agreement from that point forward.
Nom Nuinliers ore in prrreni.
fit 741,; 1Z/,.
owr.nts-r ?
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
US Policy Posture IC
Torrijos Iteeoznes
Security Council Net Effect Panama Behaves Personally
Torrijos In Power Meeting Is Of Events Maturely Involved Treaty By End
To Eta 01 1973 Reid In Panama Before Meeting at Meeting In Negotiations of 1973
30 YoS
00 Probability of specific occurance assuming events have reached this point.
00 Probability or a treaty agreement from that point forward.
Note: Numbtom nreP impcon.e.
12 Grit
'
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
f7.2
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
Ill SECRET lip
Policy Posture 13--US offers concessions outside
the negotiating framework, but does not actively push
the negotiations. This posture yields a much smaller
chance of a treaty in 1973.
Under these conditions there is an even greater
probability that Panama would decide on a Security
Council meeting, but only a five percent greater
? probability that Panama would act nationalistically
at the meeting. This posture which leaves the
initiative in the negotiations to Panama has a
much smaller chance of getting Torrijos personally
involved in the negotiations. But the probability
of his involvement would rise if Panama behaves
maturely at the meeting since Torrijos could feel
that Panama deserves .a reward from the US for its
? good behavior and that the US would be more recep-
tive to Panamanian demands. On the other hand, if
Panama adopts a mature attitude and Torrijos is per-
sonally involved in the negotiations, the chance of
a treaty is not as high as when Panama had first
acted nationalistically. In the former case, Tor-
rijos might feel that he had been too forbcaring
and? should apply pressure before agreeing to the US
? position; in the latter case, Torrijos would have
? already put the screws on and would, therefore, be
more willing to accept the US position at face value.
SECRET
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3-5-6
25X1
?
No Objection to Declassification in Part 2013/09/11 : LOC-HAK-538-3-5-6
1111?SECRE't
?Policy Posture C--US offers no interim conces-
sions and leaves the initiative in the negotiations
to Panama. Here the probability of a treaty drops
to less that 10 Percent-
The probability of a Security Council meeting
is over 95 percent, and there is a greater chance--
though still less than 50 percent--of Panama, adopt-
ing a nationalistic attitude at the meeting. The
chances of Torrijos' personal involvement in the
negotiations are negligible, and a treaty agreement
in the 1973 time frame is virtually impossible.
SECIth
No Objection to Declassification in Part 2013/09/11: LOC-HAK-538-3__R