Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Iq
NGA Review Completed
Next 1 Page(s) In Document Denied
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85TO1058ROO0100140001-0
Central Intelligence Agency
r.;
% shington.D.C.20505
DIRECTORATE OF INTELLIGENCE
28 February 1985
ANGOLA: IMPACT OF ALTERNATIVE REGIONAL SETTLEMENTS
Summary
The impact of alternative regional settlements on the
situation in Angola varies widely. Calculations by the
affected parties--Angola, South Africa, and UNITA--of the
effects of the different outcomes will continue to influence US
efforts to broker a diplomatic solution.
We examine briefly three alternatives involving a Namibian
settlement and a Cuban withdrawal, and discuss the possibility
and likely impact of reconciliation talks between the MPLA
government and UNITA. Because we believe a settlement probably
would involve the US government in monitoring the withdrawal of
Cuban troops, we also look at a few of the numpro
verification problems that could arise. ~ 7
NGA Review Completed
This typescript memorandum was prepared for the Director of
Central Intelligence by South Africa Branch,
Office of African and Latin American Analysis. It was
coordinated with the Directorate of Operations and the National
Intelligence Officer for Africa. Comments and queries ma
th A C -_ Y be
directed to the thief S
ou
ALA M 85-10026
25X1
25X1
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85TO1O58ROO0100140001-0
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Alternative Settlement Scenarios
A regional settlement that involved Namibian independence under UN
Resolution 435, withdrawal of South African forces from Namibia and some
form of Cuban troop withdrawal from Angola would have significantly
different impacts on the participants depending on the extent of the Cuban
drawdown and the pace with which it was completed. Another alternative
would involve talks between the Angolan government and UNITA intended to
lead to a reconciliation that would end the civil war.*
The withdrawal of South African forces from Namibia is rigidly
specified in UN 435. After implementation and arrival of the UN
peacekeeping contingent, South African troops would be confined to their
garrisons and, in steps over a period of 12 weeks, would be reduced to a
total of 1,500 men. At the end of seven months, when elections are held,
the remaining South African contingent would depart.
1. Complete and Simultaneous Cuban Withdrawal and UN 435
This essentially is the South African position first advanced in May
1982. The proposal calls for departure of all Cuban military personnel--
combat troops, advisers and technicians--from Angola over a period of 12
weeks in such a manner that the number of Cubans and South Africans in the
region was essentially equal once the process got underway.
-- We believe that a Cuban withdrawal at this pace would throw the
Angolan army into disarray and lead to collapse of the regime. At
this stage of the war, compensating for the loss of the Cuban
combat troops, guarding the main garrisons, and more importantly,
the advisers and technicians that run Angola's war machine,
probably would be an insurmountable problem for Luanda. Although
the South African withdrawal from Namibia would deprive UNITA of
the South African deterrent and sustaining long-term support,
UNITA probably would strike quick and hard to settle the issue
before supply shortages cut into its capabilities.
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0 ,, ,:,; q_%,
, Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
-- Luanda has made it clear, both publicly and privately, that
implementation of this alternative is completely out of the
question. We believe there is no likelihood this scenario can be
brought about.
2. Partial Cuban Withdrawal After Completion of South African Withdrawal
This proposal was advanced by Luanda in November and apparently was
vetted by Angola's Cuban and Soviet backers. Luanda said that after
implementation of UN 435--that is, the complete withdrawal of the South
African Air Force from Namibia and the drawdown of South African troops in
Namibia to 1,500 men--20,000 Cuban troops would be withdrawn over a three-
year period. A substantial, but unspecified, residual Cuban force would
remain behind--10,000 by Luanda's count, but probably more like 15,000 by
our estimates.
This alternative, in our judgment, would hurt UNITA more than the
government. Sufficient Cuban military personnel would remain
behind to help guard Angola's key facilities and otherwise support
operations by the Angolan army. Although the Angolan army would
be weakened, the troops to be withdrawn are to be taken primarily
from southwestern Angola where they help defend against South
African incursions and where UNITA is not particularly active.
Soviet weapons deliveries would continue, major urban areas would
be protected, and economically vital oil production would be
secured. In contrast, UNITA would lose the support necessary to
sustain the conflict at its present levels and South African
protection of its sanctuary in the southeast, even if given a
"golden handshake" by Pretoria. Eventually, we believe, UNITA
would lose momentum, although the conflict would continue for some
years as a lower intensity guerrilla war. Lessons elsewhere, such
as the FDN in Nicaragua, however, suggest that Savimbi may prove
surprisingly resourceful in securing at least minimal levels of
support from other donors and continue to pose a threat to Luanda.
We believe both Pretoria and Savimbi would reject this
alternative.
3. Small Residual Cuban Force Remains, Faster Pace of Withdrawal
A possible compromise might leave behind a smaller Cuban residual
force of 3,000 to 5,000 troops, with the Cubans departing at a faster pace
of perhaps two years in conjunction with U.N. 435.
-- The impact of this alternative on either side is more difficult to
assess. UNITA would lose the South African deterrent and
sustaining supply within a matter of weeks, and would be tempted
to intensify the conflict and settle the issue. However, Luanda
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
would have substantial Cuban forces on hand for at least a year
and probably could ride out the initial onslaught. Over time,
however, Luanda's dependence on Cuban manpower is greater, in our
view, than UNITA's reliance on a South African presence in
Namibia, and we expect that Savimbi's forces would begin to
prevail as the Cuban troop withdrawal neared completion.
-- We believe the South Africans and UNITA might accept this package,
provided they were assured on verification and implementation
procedures. Luanda, on the other hand, probably would balk at
such a deal--since it only postpones the MPLA's demise by a year
or two over the first alternative--unless it believed that it
could cheat on the withdrawal schedule.
4. Reconciliation Negotiations between Luanda and UNITA
Left to their own devices, we do not believe that serious
negotiations or talks between the government and the insurgents are
likely, unless the MPLA's military situation deteriorates rapidly. If
such talks are held now, we do not believe they would lead to a
reconciliation. Savimbi demands, as his price for ending the civil war,
complete withdrawal of Cuban forces from Angola and serious negotiations
leading to reconciliation. Moreover, Savimbi's popularity and charisma
frightens almost everyone in the faction-ridden MPLA government.
-- Savimbi reportedly has said that he would explore the possibility
of a partial Cuban withdrawal if it occurred in conjunction with
reconciliation talks. We do not believe, however, that Luanda's
agreement to talk, by itself, would be enough for Savimbi to
accept a package with a sizeable residual Cuban force, since talks
could fail and leave Savimbi militarily disadvantaged.
-- In the unlikely event of successful reconciliation talks resulting
in a government of natio
l
i
na
un
ty, the question of Cuban troops
probably becomes moot. Savimbi is unlikely to agree to any
settlem puts him in Luanda with a Cuban presence there.
Verifying a Cuban Withdrawal
25X1
25X1
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
We believe that of these 35,000 Cuban military personnel, 25,000 are
combat troops serving in ground combat units, air defense missile and
radar units and in an aviation component that includes pilots and their
ground crews. The 25,000 man combat component generally is deployed in or
near major provincial capitals or in larger garrisons. We are more
f'
con dent, or our estimate on combat troop strength
I c l
,000 men serve as advisers to Angolan combat units andatcalllevelsl y
of
the Angolan military establishment. as well as in various tech..;. ,
and
s
pec~aiist positions.
Most speculation about settlement alternatives center on the figure
of 30,000 Cuban military personnel, although we believe all the parties
know there are actually more.
-- Luanda, consequently, will be most interested in limiting
verification to Cuban troop departures from Angola, in an effort
to buttress a negotiated residual force level with the "uncounted"
Cubans.
-- South Africa and UNITA, on the other hand, will want to verify the
residual force level, in effect forcing Luanda to withdraw
"counted" and "uncounted" Cuban military personnel.
Moreover, verification problems would also arise in defining which Cubans
are military personnel and which are civilians, of which we estimate there
are about 6,000. There is a strong likelihood that the "civilian"
contingent would quickly come to be made up of individuals with military
training and experience, as has happened over the past year in
Ali rev.nnl,,% F-- I
Monitoring the departure of Cuban troops would be easier than
monitoring those remaining in a residual force, unless extreme care were
taken to specify locations and functions
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0
Distribution:
Original
-- Director
of Central Intelligence
1
-- DDCI
1
-- SA/DCI/I
A
1
-- Executiv
e Director
1
-- NIO/Afri
ca
1
-- NIC
1
-- DDO/Afri
ca
1
-- DDI
1
-- ADDI
1
1
--BLS
1
-- C/DDI/PE
S
1
-- D/ALA
1
-- ALA Rese
arch Director
2
-- ALA/PS
(one sourced copy; one clean copy)
4
-- OCPAS/IM
D/CB
4
-- ALA/AF
2
-- ALA/AF/S
4
-- ALA/AF/S
ALA/AF/S
(20 February 1985)
Sanitized Copy Approved for Release 2009/12/31 : CIA-RDP85T01058R000100140001-0