THE GERSONY REPORT AND RENAMO

• Document Type: FOIA
• Collection: FOIA Collection
• Document Number (FOIA) /ESDN (CREST): 05862684
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 5
• Document Creation Date: March 9, 2023
• Document Release Date: November 13, 2020
• Case Number: F-2018-01308
• Publication Date: May 20, 1988

Approved for Release: 2020/10/29 C05862684
SECRET
� 110irecUroiCenthalhkiligence
1Vash*PIO.C.20505
National Intelligence Council
MEMORANDUM FOR: See Distribution
OCA 1111
Sk_2_
NIC 01790-88
20 May 1988
FROM: Walter L. Barrows
National Intelligence Officer for Africa
SUBJECT: The Gersony Report and RENAMO
1. This memorandum responds to a request by the Senior Director for
African Affairs, NSC, for an Intelligence Community assessment of the
recently released report by Robert Gersony on Mozambican refugees,
particularly its findings about the RENAMO guerrilla group. The request
also asked for an assessment of the relationship between RENAMO and the
South African government. The following, while not a fully coordinated
document, reflects my "sense of the community" after convening a special
meeting of analysts on these questions. I have also inserted my own views,
properly identified.
2. Summary. The Gersony Report contributes toiour understanding of a
murky situation--but analysts are not in accord on its overall intelligence
value. State INR finds it of significant value, particularly for the
insight it provides on RENAMO abuses of civilians. DIA and Army, on the
other hand, contend that it adds only modestly to our information on the
insurgency. CIA stands somewhere in between, believing it to be an
important study that raises serious questions for future analysis. All
agree, nonetheless, that the Report reinforces many of our strong suspicions
about RENAMO--such as its tightly organized and disciplined character -- and
supports our conclusion that the guerrillas are now active throughout
Mozambique. It also prompts us to examine more closely the rebels' mode of
operation and the possibility that they are intensifying force and
intimidation against civilians in response to growing famine and other
shortages.
3. The Report should be read with care, however. It is not based on a
scientifically drawn sample, so extrapolations from refugee interviews to
the overall situation in Mozambique are not justified. Most analysts
believe that certain of its conclusions are overdrawn, particularly its
suggestions that responsibility for atrocities by government forces is
virtually absent and that the relationship between RENAMO and civilians is
bated solely on coercion.
-SE-GRE�T
(b)(3)
Approved for Release: 2020/10/29 C05862684
Approved for Release: 2020/10/29 C05862684
SECRET
4. On a separate but related matter, we have no evidence that South
African officials have encouraged systematic brutality by RENAMO. South
Africa continues to assist RENAMO in important ways, but Community analysts
agree that this assistance is limited and that even a complete cutoff, while
weakening the insurgency, would not cripple it.
5. The Report. Compiling the results of lengthy interviews with nearly
200 refugees and displaced persons at 25 separate camps in Mozambique and
four neighboring countries, the Report reflects a painstaking and
professional effort to shed light on the reasons behind the growing numbers
of Mozambicans who have been dislocated. It draws attention to RENAMO as a
principal cause. In my mind, it makes a useful and provocative contribution
to those of us in the Intelligence Community who have struggled over the
years to discern the nature of RENAMO's obscure war in the bush. Both the
information and the perspective it provides are fresh--by focusing primarily
on recent arrivals at refugee camps to give a near-current glimpse of
developments in the countryside, and by making strenuous efforts to view
these developments through the eyes of the refugees themselves to lend a
vantage previously unavailable in such systematic fashion.
6. The Report is not based on a fully representative sample, however,
which limits our confidence in drawing conclusions about Mozambique as a
whole from the refugee interviews. In particular, extrapolating to 100,000
murders of civilians by RENAMO from the 600 murders reported by refugees
goes far beyond the data and is statistically unjustified. (INR notes that
the number could just as well be too low as too high.) Moreover, in the
judgment of CIA and DIA analysts, broad conclusions about insurgent and
government behavior toward people still in RENAMO-coqtrolled areas cannot be
drawn from the Report. The refugee sample does not necessarily typify these
people, some of whom may support the guerrillas or view government soldiers
as equally threatening.
7. DIA and Army have reservations on two other methodological issues:
Translators could have unintentionally influenced the results of
interviews with refugees. CIA's refugee expert discounts this as
an important factor. I agree; having conducted academic research
at the grassroots level largely through translators, I believe that
a skilled investigator such as Mr. Gersony would quickly detect
systematic distortion.
- -
Interviewees could have been afraid.to speak out against the
government, thereby accounting for the Report's remarkable finding
that virtually all brutality was inflicted on refugees by RENAMO
rather than government troops. Again, CIA's refugee expert
believes that this is not a significant factor. I'm agnostic;
while African villagers are often forthright even to the point of
openly criticizing higher authorities, they also exercise
well-known talents for telling outsiders what is best for them to
hear.
In any case, we need not hold up the Report to excessively fastidious
standards. As with any important source of information, the Report should
SECREI
�--Approved for Release: 2020/10/29 C05862684
Approved for Release: 2020/10/29 C05862684
SECRET
be mined for what is valuable, taking into account possible biases and
comparing its data and conclusions with what we already know or think we
know. In some respects, it converges with our prior understandings about
RENAMO, while in other ways it diverges.
8. Convergence. The Report corroborates our long-standing view of
RENAMO as a well organized and centrally directed insurgency. Discipline is
tight, and its command and control system is effective. The Report also
reinforces our conclusion that RENAMO is now active throughout Mozambique.
9. Analysts disagree, however, over the extent to which the Report's
findings confirm a lack of widespread political support for RENAMO and a
related overall trend toward increasing violence by RENAMO against
civilians. CIA and INR detect such a trend in other reporting over the past
year or so, while DIA is skeptical, contending that the appearance of
intensified violence may result from local intimidation campaigns (such as
that along the Zimbabwe border) rather than a countrywide phenomenon.
10. In any case, the Report reinforces our long-held view that RENAMO is
fully able and willing to use brutal intimidation systematically against
civilians when it chooses to for its own tactical purposes. We have ample
reporting, for instance, that it has recently carried out a campaign of
atrocities against villagers inside the Zimbabwe border, most probably to
inflict a cost on Zimbabwe for its military involvement against RENAMO in
Mozambique. We also know that the rebels deliberately attack communal
villages, local officials, and other civilian targets associated with the
central government. Finally, the Report is consistent with reporting on
forced recruitment into RENAMO and widespread use of civilians as unwilling
porters. I believe along with several analysts, however, that the magnitude
of the brutality described in the Report goes well beyond reporting from
other sources and should be viewed cautiously pending evaluation of
additional information.
11. Divergence. The Report's portrayal of RENAMO's relationship with
the rural population as based solely on coercion and exploitation is not
consistent with other sources that suggest RENAMO makes some effort to win
the loyalty of villagers. Over the years we have received occasional
reports from first-hand observers in RENAMO areas who discerned more
reciprocal ties--or at least neutrality--between rebels and villagers. For
instance, several foreign hostages upon their release have reported that
villagers were not cowed by the arrival of RENAMO columns; interaction
seemed amicable. A fairly recent visitor to.the area told me that on one
occasion villagers approached his RENAMO entourage to solicit assistance.
In addition, we know that RENAMO's leadership is concerned about reports of
brutality and has issued orders to prohibit abuses. Finally, RENAMO at
least in some areas espouses a home-spun political formula combining
reverence for local practices and traditional authority with disdain for the
Marxist programs imposed on the countryside by the central government.
Rebel spokesmen also have occasionally appealed to historic tribal rivalries
and--with some justification--charged that the government leadership is
dominated by southerners to the disadvantage of central Mozambicans.
1E-811E-1
Approved for Release: 2020/10/29 C05862684
Approved for Release: 2020/10/29 C05862684
�
SECRET
12. Reporting over the years from a variety of sources, including
interviews with refugees, indicates that RENAMO has not been the only
Important source of brutality against civilians in Mozambique. Government
troops also forcibly recruit conscripts and commit atrocities against
civilians, with their proportionate share of brutality varying from one
locale to another. Moreover, reporting suggests that a share of violence
against civilians can also be attributed to third parties--bandits, splinter
groups, renegade soldiers or rebels, and the like.
13. Speculation. Several possibilities would help explain disparities
between the Report and other sources:
- -
- -
Any trend toward increasing abuse of civilians by RENAMO is perhaps
a response to worsening famine and other shortages aggravated by
the group's rapid growth and geographical expansion. Because the
Report is based on very recent information, it would reflect the
impact of such a trend more heavily than earlier sources.
There could also be a trend toward less brutality by government
forces, perhaps as a result of Maputo's recent efforts to reform
the military or because the army is now less active in the
countryside, preferring instead to occupy garrisons and patrol
roads.
There could be core areas of popular support for RENAMO, or at
least zones where the relationship between rebel and villager
approximates that between government and subject, with some degree
of mutual obligation. Interviews of refugees likely would not have
detected this, since people in such zones would presumably have no
reason to flee. This suggests that for intelligence purposes the
Report's helpful categorization describing different types of
RENAMO control--"tax areas," "control areas," and "destruction
areas"--might be expanded to include, say, "support areas."
I will encourage intelligence collection and analytic efforts to test these
and other hypotheses stimulated by the Report.
14. The South African Connection. The Report does not deal with South
Africa's involvement in Mozambique, but its appearance has raised questions
about Pretoria's responsibility for widespread suffering there. We have no
evidence indicating that Pretoria has deliberately encouraged use of force
and intimidation by RENAMO against civilians', although some South African
officials have been no doubt aware that it takes place.
15. South Africa has continued to support RENAMO since it pledged not to
do so in the Nkomati Accord of 1984. But the scale of this support
diminished considerably after 1984 and is now limited to items such as
communications equipment and training; some materiel support, including
small arms; medicine; military advice and perhaps training; and tactical
intelligence. (DIA doubts that South Africa provides tactical intelligence,
such as targetting information, RENAMO seeming capable of meeting its own
A
St-GRE-T-
Approved for Release: 2020/10/29 C05862684
Approved for Release: 2020/10/29 C05862684
(b)(3)
needs in this area.) Recent intelligence reports detailing links between
the South African military and RENAMO corroborate our view that, while
Pretoria exercises influence over the guerrillas, it certainly does not
control them. Indeed, since 1984, South African influence has waned as
RENAMO has become more independent-minded and self-sufficient. Analysts
agree that should all South African ties with RENAMO be severed, the
insurgency could still sustain itself, albeit at reduced levels.
16. Next Steps. Interviewing refugees has clear intelligence value. In
light of the questions stimulated by the Report, I recommend that:
=1.0
The State Department commission follow-on refugee studies in the
region, from which we expect there would be derivative intelligence
benefits.
-- Country teams in the region undertake more systematic efforts to
glean information from refugees and those who work with them.
The data base supporting the Report be made available to
Intelligence Community analysts.
17. Additionally, I recommend that contacts with the rebel military
leadership for intelligence collection purposes be seriously considered.
Our understanding of the insurgency will necessarily be incomplete unless it
includes a view from the inside. In the meantime, I will work with
Community analysts to identify key collection gaps suggested by our reading
of the Report.
IIMP
Walter L. Barrows
ccror4
Approved for Release: 2020/10/29 C05862684
(b)(3)