Lessons of the Soviet Withdrawal from Afghanistan

  • Middle East Policy

    Middle East Policy has been one of the world’s most cited publications on the region since its inception in 1982, and our Breaking Analysis series makes high-quality, diverse analysis available to a broader audience.

Mark N. Katz

Senior Fellow

After Soviet forces withdrew from Afghanistan in 1988-89, the regime it was defending there fell. This experience contributes to present fears that, if America withdraws from Afghanistan, the regime it is defending will also fall. A closer look at Soviet and Russian actions between 1988 and 1992, though, suggests that this need not have been the result then — and that it need not be the result of an American withdrawal now either.

The Soviet Union invaded Afghanistan in December 1979 to prop up the Marxist regime that had come to power the previous year but which appeared to be on the verge of collapse. Unlike the Soviet invasions of Hungary in 1956 and Czechoslovakia in 1968, however, Soviet forces encountered prolonged resistance that they were unable to defeat. In order to promote his goals of domestic reform and improving Moscow’s relations with the West, Gorbachev withdrew Soviet forces from Afghanistan (which he had termed a “bleeding wound”) between May 1988 and February 1989.

At the time, it was widely predicted that the Marxist regime Moscow had been supporting in Afghanistan would fall within a few months — or even weeks — of the final departure of Soviet forces. The regime of President Najibullah, however, survived until April 1992, over three years after the Soviet withdrawal. Several factors contributed to the regime’s longevity, including the continuation of Soviet military and economic assistance, the mistakes made by some of the mujahideen (the Afghan forces that had fought against the Soviet occupation) as well as their Pakistani supporters, divisions among the various mujahideen groups, and the Najibullah regime’s successful exploitation of these problems. After the downfall of Gorbachev and of the Soviet Union itself in December 1991, though, Moscow’s assistance to Najibullah ended. Without this assistance, Najibullah was unable to continue effectively exploiting the weaknesses of his adversaries. Instead, they were able to exploit his, and so his regime fell. This paper will examine how Moscow’s actions helped Najibullah survive but then contributed to his downfall, as well as how Moscow’s actions affected the other factors influencing the fortunes of the Afghan Marxist regime.

1989-91: Soviet Support for Kabul Continues

Although the Soviet troop presence in Afghanistan ended in February 1989, large-scale Soviet military and economic assistance to its Marxist protégés there continued. As Soviet troops withdrew, they left behind literally all their matériel except for the vehicles needed to transport them back over the border. In addition, as Soviet forces withdrew from Eastern Europe following the downfall of communist regimes there in late 1989, some of this weaponry was transferred to Afghanistan. From early 1989 to late 1991, Soviet assistance to Kabul reportedly amounted to $300 million per month. Perhaps this is not a large figure by today’s standards, but it was a much greater amount than the mujahideen were receiving after the Soviet withdrawal, and was a considerable financial burden on the economically beleaguered USSR. Weaponry that Moscow supplied to Kabul included MiG- 27 fighter jets (the Afghan Marxist regime had an air force with some 200 aircraft plus helicopters). In addition, as Zalmay Khalilzad (whom President George W. Bush appointed as special presidential envoy for Afghanistan and then U.S. ambassador to Kabul) noted in 1991,

Moscow has provided more than thirteen hundred Scud-B missiles, hundreds of shorter range Frogs, several hundred tanks, and sixteen hundred five-ton trucks. To keep Kabul supplied, the Soviets launched the biggest air supply effort in its history, sending some twenty-five or more IL-76 transport planes to Kabul each day for much of 1989 (Khalilzad 1991, 82-83).

Indeed, all this was reportedly more than the Marxist regime could effectively use. To help them, though, Moscow left behind about 300 advisers, some of whom reportedly participated in the firing of the Scud missiles at mujahideen targets. (In addition to a regular army of 55,000 men, the Kabul regime also had the support of a 10,000-strong presidential guard and various militias, including an especially effective one raised and led by the ethnic Uzbek leader, General Abdul Rashid Dostum.)

When the Soviets withdrew, much of the anti-foreign-presence motivation for many Afghans to fight with the mujahideen disappeared. Indeed, some mujahideen groups themselves appeared tainted for being so very close to Pakistan. Soviet assistance also allowed Kabul to effectively compete with Pakistan and the various mujahideen groups based there in paying off local commanders and tribal leaders inside Afghanistan. While Pakistan tied its support to various mujahideen groups not only to whether or not they fought against Kabul but whether they did so in the manner specified by the ISI, Kabul gave support to various groups just in exchange for not fighting against it. As Khalilzad noted at the time, “Najib’s offer is more attractive than ISI’s; while ISI wants it clients to fight and risk their lives, Najib is willing to pay if the commanders agree not to fight” (Khalilzad 1991, 81).

Further, the groups Pakistan supported were not always effective. In March 1989, some 15,000 Pakistani-backed mujahideen forces attacked the town of Jalalabad. Their unwillingness to accept prisoners, though, meant that the defending government forces had strong motivation to fight on. Although the mujahideen laid siege to the town, the Kabul government was able to resupply and reinforce its garrison by air, launch a counterattack, and break the siege by mid-May 1989. This was a major humiliation for Pakistan and its allies.

In addition to military assistance, though, Moscow provided Kabul with key economic assistance. As mujahideen forces approached Kabul and interrupted the supply of food and other consumer goods into the city, the Soviets airlifted these commodities to the Afghan capital.

At Moscow’s urging, the Kabul regime attempted to broaden the basis of its support by downplaying its Marxist nature, appointing non-Marxists to visible positions and trying to appeal to nationalism. According to contemporary accounts, though, these efforts were not particularly successful, as the Marxist regime — especially President Najibullah — was extremely unpopular with the Afghan population. Indeed, while outwardly broadening the base of the regime, it appears that Najibullah actually narrowed it by increasing reliance on his hard-core supporters.

But while Najibullah and his regime lacked popular support, the mujahideen themselves frequently provided Afghans with strong incentive either to support the Marxist regime or to see it as the lesser of two evils. The mujahideen’s efforts to impose an economic blockade on Kabul as well as their periodically shelling it did nothing to endear them to the citizens of the capital. Even worse, when the mujahideen captured the town of Khost in March 1991, they not only looted it but killed all the government forces they had captured instead of holding them prisoner. This action not only created fear in other towns; it also made clear to irresolute government forces that defecting to the mujahideen was probably not an option for them.

Mujahideen groups also fought among themselves, and this was something that the Marxist government was able to exploit. As was noted in the International Institute for Strategic Studies’ Strategic Survey 1989-1990,

…by early 1990, although the mujahideen continued to control the bulk of the countryside, most of them had in effect ceased being mujahideen, in the sense that they were no longer fighting against the central government, but were instead attempting to work out compromises with Kabul which would ensure their local power, particularly against their former fellow comrades in arms. Most local commanders had reverted to the traditional relationship between local powers and a weak central state that has shaped Afghan history since the eighteenth century. The central state is seen less as an enemy than as a referee which can help to promote the interests of the local group. This development was expected to play a decisive role after the collapse of the Najibullah regime, not before. That it has come into play so soon is a result of the unexpected adroitness of the regime, aided by the ineffectiveness of US-Pakistan policies (IISS 1990, 160).

At the time, Khalilzad seemed to suggest that the Afghan Marxist regime might even come out on top in the ongoing conflict, when he noted that the Kabul regime “…is likely to increase its efforts to reach out to make deals with commanders and the supporters of the former king at the expense of the majority of the Peshawar-based leadership. Should it succeed, it can reduce the fighting in the country” (Khalilzad 1991, 84).

1992: Russian Support Ends

But, of course, the Afghan Marxist regime did fall in April 1992. Once again, Russian actions appear to have played a key role in bringing this about. Shortly after the failed August 1991 coup attempt in Moscow and under very different political circumstances, Moscow and Washington agreed to stop aiding their respective Afghan allies as of January 1, 1992. Not only did Moscow end its arms supply to Kabul, but it also stopped providing it with food and fuel. By contrast, although Pakistan had agreed to stop aiding the mujahideen, Saudi aid to them via Pakistan continued.

Shortly after this in February 1992, Najibullah (a Pushtun) apparently tried to bolster his authority over the Uzbek militia chieftain Dostum by appointing a fellow Pushtun as a commander in the northwestern Uzbek heartland. But if this was his intention, it backfired in April 1992, when Dostum defected from the government and joined forces with long-time anti-Soviet Tajik mujahideen commander Ahmad Shah Massoud (whose relationship with both Pakistan and the Pushtun mujahideen groups it supported was adversarial). Non-Pushtun forces from the north and Pushtun forces from the south then rushed to capture Kabul, with the Marxist regime splitting along ethnic lines and either joining forces or making deals with their ethnic kin. Najibullah resigned and sought sanctuary inside the UN compound in Kabul after his attempt to reach the airport was blocked (by his erstwhile ally Dostum, according to some). The Islamic State of Afghanistan was declared, but the mujahideen remained divided. After a short, sharp battle for control of Kabul, the Dostum-Massoud alliance prevailed over their Pushtun opponents, for the time being. Russia appeared to play no role in these events.


Six observations can be made about the events described here:

First, even after the withdrawal of Soviet forces was completed in February 1989, Soviet military and economic assistance enabled an unpopular regime to remain in power in Afghanistan — at least, in the major population centers — for over three years.

Second, despite continuing to receive significant aid from Pakistan and other nations, the mujahideen were unable to overthrow the Afghan Marxist regime so long as that regime was receiving significant aid from the Soviet Union.

Third, opposition to the Afghan Marxist regime appeared to decline after Soviet troops withdrew. Further, while they had not performed effectively during the period of Soviet occupation, the effectiveness of Afghan government forces increased significantly after the Soviet withdrawal.

Fourth, after Soviet assistance to Kabul ended at the beginning of 1992, the Afghan Marxist regime’s strength declined rapidly.

Fifth, the collapse of the regime in April 1992, though, was not due just (or perhaps even mainly) to the actions of the Pakistani-backed Pushtun mujahideen. Indeed, the immediate downfall of the regime was precipitated by the defection of the previously pro-regime Uzbek militia leader, Dostum, to the side of the non-Pushtun opposition to the regime.

Sixth, as the collapse of the regime approached, the most salient division in Afghanistan was not Marxist vs. Islamist, but Pushtun vs. non-Pushtun.

History, of course, is not destined to repeat itself. These six observations from the 1989-92 period, however, may have salience for the present as well as the short- and medium-term future. They suggest the following:

First, even after the withdrawal of ISAF forces is completed by the end of 2014, American and allied military and economic assistance to the current less-than-popular Karzai government may enable it to remain in power in the major population centers.

Second, the Taliban are not destined to return to power, despite the likelihood that they will continue to receive Pakistani assistance so long as the Kabul government continues to receive significant aid from America and its coalition partners.

Third, opposition to the Karzai government may actually decline after the American and coalition withdrawal. Once they are responsible for their own survival, the effectiveness of Afghan government forces may quickly increase.

Fourth, if American and allied support for it ends, the Kabul government’s strength is highly likely to decline rapidly.

Fifth, under these circumstances, ethnic divisions within the Kabul government leadership are likely to become exacerbated. It is highly likely that the non-Pushtun officer corps would seek to oust the Pushtun president, Hamid Karzai, and his entourage.

Sixth, even if (indeed, especially if) the Taliban manage to seize control of Kabul once again, the most salient division within Afghanistan is once again likely to be Pushtun vs. non-Pushtun.



IISS, Strategic Survey, 1989-1990. (Brassey’s/International Institute for Strategic Studies, 1990).

Zalmay Khalilzad, “Soviet-American Cooperation in Afghanistan,” In Mark N. Katz, ed., Soviet-American Conflict Resolution in the Third World (United States Institute of Peace Press, 1991).

  • Middle East Policy

    Middle East Policy has been one of the world’s most cited publications on the region since its inception in 1982, and our Breaking Analysis series makes high-quality, diverse analysis available to a broader audience.

Scroll to Top