Skip to content


Defense Intelligence Agency Briefing, 'The 1987-88 Combat in Southern Angola: Lessons Learned'

This document was made possible with support from Leon Levy Foundation

Southern Africa has been a troubled region for almost three decades. Conflicts of national liberation have evolved into bitter civil wars, provoking extensive foreign involvement.


This has certainly been true of Angola, which has been wracked by insurgency and invasion since the early 1980s.


Angola’s recent history affords a variety of lessons which are useful to students of contemporary 3rd world conflict. Here, you see a brief chronology of key events in independent Angola.


12 years after Soviet arms and Cuban troops helped install the Popular Movement for the Liberation of Angola (MPLA), in power, the Marxist regime suffered a near catastrophic military defeat at the hands of the insurgent and South African forces. This briefing will provide a succinct analysis of how a Soviet-designed offensive against Jonas Savimbi’s UNITA in Southeastern Angola ended in disaster, and led to a fundamental change in the regional military balance. Special emphasis will be given to the weapons and tactics that affected the outcome.


Here is our agenda:


Following his defeat at the end of the first phase of civil war in 1976, Savimbi rallied his forces in remote, Southeastern Angola. Reorganizing and rebuilding his force, with significant South African assistance, he gradually increased the tempo of insurgent operations.


By the early 1980s, the UNITA challenge was growing rapidly. In 1983, Savimbi’s forces compelled the Angolan and Cuban garrison to abandon the key town of Cangamba, signifying UNITA’s consolidation of its control over the Southeast, and marking entry into a new phase of the insurgency. As UNITA activity spread through the country, the regime tried to move against Savimbi’s stronghold in the Southeast, attempting to cut the insurgency off at its root.


At the same time, the Angolan government sought to transform its military—FAPLA—into a conventional force along Soviet lines. As a result, a more conventional war developed within the broader context of a nation-wide guerilla conflict.


The Angolan government has launched annual dry season offensives against the UNITA base area in recent years. The early offensives were not particularly successful. But the 1985 offensive made unexpected gains, and government troops penetrated to within 12 miles of Mavinga—a key UNITA logistics hub—before being thrown back. The 1986 offensive was preempted by dramatic UNITA attacks on government forces in Central and Southeastern Angola.


Luanda launched another major, Soviet planned offensive in August 1987 to capture Mavinga. This offensive was larger and more complex than previous operations. The objective was to challenge Savimbi’s territorial control by capturing this key town, and to threaten South African logistic support to UNITA guerrillas operation in Northern and Central Angola.


The offensive seriously challenged UNITA’s base region, and achieved some success before bogging down against intense UNITA resistance. UNITA, with some South African support, counterattacked seizing the initiative and pushing government forces back to their starting position at Cuito-Cuanavale.


As the magnitude of the Angolan defeat became apparent, Cuba increased its commitment of combat troops to Angola, by deploying an additional 15,000 troops to Angola, an unmistakable measure to deflect South African attention from the Cuito front. This was not unlike Cuba’s initial rapid military build-up in Angola in 1975, when Havana sent an expeditionary force to Luanda to secure MPLA victory over competing liberation movements.


Our interest, however, is in the 1987 offensive itself, and in its military and political implications. The 1987 offensive can be broken down into four phases, as shown here.


The preparation phase started at the end of the rainy season in March 1987, and consisted of an extensive logistical buildup and improvements to military facilities, particularly airfield runways. FAPLA transported materiel and supplies to forward staging areas, and increased munitions and fuel storage capabilities.


The second phase comprised an Angolan diversionary attack from the North designed to pin down UNITA forces, as well as to force UNITA to redeploy elements from Mavinga. Elements of three FAPLA brigades moved South along two axes of advance with the objective of occupying the towns of Cassambra and Luvuei, both UNITA garrisons and key towns on the North-South supply route leading into South-Eastern Angola. UNITA effectively countered this advance by ambushing lead elements, harassing the main body and interdicting supply lines, which complicated existing logistical deficiencies and geographic constraints. UNITA attacked these units while they awaited resupply and forced them to withdraw northward. UNITA received no direct support from South African combat forces up to this point.


During phase three, FAPLA forces conducted the main attack south from Cuito Cuanavale in an effort to seize Mavinga. Angolan maneuver elements consisted of 4 brigades and 2 operations groups reinforced with additional tank and artillery assets. These battle groups advanced cross-country along two primary axes paralleling the Cunzumbia and Cuzizi rivers. FAPLA forces along each axis consisted of roughly three maneuver brigades with attached fire support. FAPLA’s advance was slow-moving and sporadic, averaging approximately 4 miles per day. There were several reasons for this: first, FAPLA forces moved primarily cross country, hoping to mitigate the effects of anti-armor weapons and harassing ambushes. Second, FAPLA grouped its forces tightly together both to enhance combat power and to stay within the air-defense umbrella. The Angolans were concerned with the South African air threat. Advancing Angolan forces were supported by SA-8’s and 9’s. These Soviet-made air defense assets succeeded in forcing South African pilots to take protective measures and reduced their effectiveness. FAPLA maneuver forces tended to halt by late afternoon each day in order to dig in and prepare defensive positions.


Another factor impeding the Angolan advance was the difficulty in providing logistic support to the advancing columns. This lack of adequate transport and ongoing UNITA activity in rear areas hampered this support. Also, the need to heavily defended supply convoys diverted combat forces from the main effort. At any one time, FAPLA committed as much as 1/3rd of the combat forces available for the main attack to convoy security. Additional forces, not part of the attacking force, were used to secure LOC’s in the rear.


Numerous water obstacles also hampered FAPLA’s advance. This was compounded by serious shortages of portable bridging equipment.


UNITA’s response to FAPLA’s advance was multi-faceted beginning during FAPLA’s build-up. Initially, it consisted of raids and ambushes in FAPLA’s rear areas as well as small-unit operations throughout Angola destined to tie down FAPLA forces to prevent them from being deployed to the front. Once FAPLA’s advance began, UNITA conducted retrograde operations, harassing FAPLA forces with artillery, mortar and rocket fire, attacking aircraft with shoulder fired missiles and engaging armor from ambush positions with anti-tank weapons. Savimbi hoped to attrite the attacking forces and force FAPLA battle groups to over-extend their supply lines while he withdrew his forces to more desirable positions.


During this phase, UNITA demonstrated its flexibility. As FAPLA advanced, Savimbi redeployed battalions to the Southeast and reorganized smaller guerrilla forces into battalion size units, permitting them to conduct conventional operations. Furthermore, UNITA’s orderly withdrawal, its systematic destruction of bridges and effective use of obstacles and mines indicated a sophisticated level of command and control.




Then, South African ground elements [excised] drawn from bases in northern Namibia, deployed to Angola and participated with UNITA in operations near Mavinga. In mid-September UNITA and South Africa counterattacked east of Mavinga, effectively destroying one FAPLA BDE. South Africa employed its mechanized forces primarily to block FAPLA’s advance south.


While South African long range artillery fires inderdicted FAPLA logistic support, South African air strikes isolated the battlefield, denying access to FAPLA reinforcements. The South African G5 and G6 155-mm artillery and Valkiri 127-mm multiple rocket launchers played a significant role in this effort. The range, flexibility and power of these systems complemented South African close air support. More important, the G5 and G6 enabled the South Africans to engage FAPLA at standoff range, with the result that FAPLA suffered significant casualties long before they were able to engage South Africa and UNITA ground elements. These extended range 155mm guns were perhaps the single most effective weapon used by either side.


During October and November, FAPLA withdrew under pressure in an orderly manner towards Cuito Cuanavale. They reoccupied positions just east of the Cuito River, which they had prepared in July, just prior to the commencement of maneuver operations. The 1987 offensive had failed.


Before continuing, let’s briefly consider some implications of these events.


An assessment of the losses during the 1987 offensive indicates that UNITA inflicted five times as many casualties as it sustained. Some FAPLA combat units may have suffered losses of 12–25 percent of the entire army—the single greatest defeat suffered by Luanda in the 12-year war. In some units the FAPLA attrition may have been higher, especially during the decisive clashes in September-October. UNITA also captured large quantities of combat equipment.


FAPLA’s offensive failed for a variety of reasons: Soviet planning and Angolan strategy, by concentrating on large-scale conventional offensives, may have been unwise. A more fundamental failure, however, was FAPLA’s failure to exploit the advantages of its numbers and more sophisticated equipment. Equally important was UNITA’s superior tactical skill; as was the fact that FAPLA seriously underestimated UNITA’s anti-armor capability. Another critical event was UNITA’s destruction of the key Cuito bridge along the only supply route between Cuito Cuanavale and Mavinga. Also, the necessity for FAPLA to conduct assault crossings of a number of rivers along both fronts permitted UNITA to attrite FAPLA during crossing attempts. Finally, FAPLA’s poor morale and discipline was an important factor in the failure of the operation. This was illustrated by the commander of a brigade-sized battle group. During the decisive battle, he fled, abandoning his troops.


Returning to the military situation, from November-March, FAPLA-held Cuito Cuanavale was under intermittent attack and virtual siege. FAPLA reinforced by the Cuban troops stubbornly held on to its positions on the east bank of the Cuito River. Due to the continued air and ground threat, FAPLA redeployed aircraft from Cuito to the more secure base at Menongue. The tactical command post was also relocated some 12 miles Northwest of Cuito.


UNITA and South African forces could probably have captured Cuito in the November to March period. However, neither was willing to take the heavy casualties that would have resulted from a full-scale effort. South African probes of defensive positions were firmly rebuffed. Thus Pretoria was unable to do more than eliminate the possibility that FAPLA could use Cuito as a forward staging base for a 1988 government offensive.


UNITA exploited its tactical victory at the operational level, continuing conventional and unconventional operations during the subsequent rainy season with numerous attacks on towns, government forces, and lines of communication throughout Angola. In December, UNITA routed a FAPLA brigade at Munhango and destroyed the key bridge over the Cuanza River. These successes isolated the FAPLA garrison at Cuemba, enabling UNITA to tie down six FAPLA brigades attempting to resupply and defend the garrison. UNITA’s area of control temporarily expanded to a nearly 200-km section of the Benguela railroad from Luena to Cuemba, facilitating UNITA’s transport of personnel and supplies to the north.


Following FAPLA’s 1987 defeat, Angolan President Dos Santos reportedly met with President Castro and requested a qualitative as well as quantitative increase in Cuban military assistance. Starting in late 1987, the Cubans deployed approximately 15,000 additional combat troops to Angola. [Excised]. Most of these personnel deployed southward toward the Namibian border in March 1988. By June, Cuban forces in South Western Angola consisted of 5 tank brigades with 10 to 12,000 troops and some 500 tanks.


While UNITA and South Africa had won the battle in the southeast, the Cuban deployment to the southwest ended Pretoria’s military dominance of southern Angola. Castro’s move was a strategic coup because he challenged the South Africans militarily without actual combat. [Excised]. Despite routing the Cubans in a 27 June clash, Pretoria was not able to reassert its military ascendancy in southern Angola.


Although we cannot be certain as to what the various parties may have learned from the 1987 offensive, we have identified several areas in which FAPLA could improve its performance on the battlefield. Based on our analysis, FAPLA success would require significant improvement in the ability to synchronize the employment of various combat arms. The Angolan Army’s slow advance during the ‘87 offensive not only permitted UNITA to control the tempo of the battle, but also necessitated significant increases in logistic support. Improving the discipline and morale of FAPLA soldiers would also improve FAPLA’s capability. FAPLA’s inability to withstand the combined UNITA/South African counterattack at the Lomba River may have been due to FAPLA’s poor morale and discipline, as it was superior to UNITA/South African combat power. FAPLA’s past reliance on tanks to spearhead attacks proved ineffective. The use of heavily armored forces complicated logistic operations and degraded mobility. Fairly heavy brush in southern Angola limits the ability of tanks to engage at standoff range, making them vulnerable to attack by more mobile and lightly armored vehicles.


There are no indications that FAPLA has learned from its 1987 failure. There have been no comprehensive changes in command, nor has there been reorganization at the tactical level. Furthermore, we have seen no evidence that FAPLA morale and discipline have improved significantly since the offensive. Soldiers are still inadequately fed, clothed and equipped, and desertions continue at a high rate.




UNITA, on the other hand, has adapted much more effectively to the combat environment it faces in Angola. By 1987, UNITA had clearly recognized its limitations to counter the initial stages of a full-scale conventional FAPLA offensive. FAPLA’s greater firepower makes this extremely hazardous. Instead, by conducting orderly retrograde and delaying operations, UNITA allowed FAPLA’s inherent logistic and command and control deficiencies to take their toll. Ultimately, UNITA could establish viable defensive positions and assemble a counter attack force, while attriting the advance with guerilla attacks.


During the initial stages of the 1987 offensive, UNITA’s combined retrograde and unconventional warfare operations cut FAPLA logistic support and effectively retarded the movement of FAPLA maneuver units along main avenues of approach. When FAPLA was in a tactically vulnerable position, UNITA effectively counterattacked and seized the initiative.


Cuba apparently realized that deploying Cuban combat units to the areas of most likely conflict would ensure FAPLA’s survival as a combat force. Realizing that it was not sufficient to merely secure bases and provide advisers, Cuba had to demonstrate its willingness to deploy its forces on the front line. Cuba’s reinforcement and redeployment of its forces in Angola close to the Namibian border, effectively demonstrated this.


South Africa, concerned with the deployment of Cuban troops to the southwest in early 1988, now proved interested in Cuba’s willingness to consider withdrawing its forces in exchange for implementation of UN Security Council Resolution 435. South Africa reassessed its position in Angola to determine whether or not active involvement in Southern Africa was a vital national interest. Apparently, in the near term, South Africa has decided it is not.


In conclusion, Angola suffered a strategic defeat during its 1987 offensive. Although FAPLA was able to retain Cuito Cuanavale, the number of casualties suffered—as much as 15 percent of its army—as well as UNITA’s ability to maintain the integrity of its territory in southeastern Angola, threatened to significantly alter the balance of power in the region. It was this development that led to the Cuban build-up in southern Angola, which redressed this military balance, challenging South African military dominance along the Namibian border. This, in turn, provided new impetus to peace negotiations and resulted in the December 1988 Accords among South Africa, Angola and Cuba.


Analysis of the military campaign waged in Southern Angola in 1987-88 as well as the lessons learned from this exchange. Document indicates that FAPLA failed to learn from its mistakes, while UNITA adapted much more effectively to the combat environment it faced in Angola. Although FAPLA managed to hold on to Cuito Cuanavale, its large losses led to a Cuban build-up in Southern Angola, which challenged the existing military balance in the region. This provided new impetus to peace negotiations and resulted in the December 1988 Accords among South Africa, Angola and Cuba.

Document Information


National Security Archive, SA02547. Included in "Southern Africa in the Cold War, Post-1974," edited by Sue Onslow and Anna-Mart Van Wyk.


The History and Public Policy Program welcomes reuse of Digital Archive materials for research and educational purposes. Some documents may be subject to copyright, which is retained by the rights holders in accordance with US and international copyright laws. When possible, rights holders have been contacted for permission to reproduce their materials.

To enquire about this document's rights status or request permission for commercial use, please contact the History and Public Policy Program at [email protected].

Original Uploaded Date



Record ID



Leon Levy Foundation