Then they collapsed in little more than a week in early August.
But this shouldn’t have been a surprise. This kind of collapse is much more common than the commentary on Afghanistan suggests.
Here’s how to understand what happened.
Armies have unraveled like this before
We have seen this type of unraveling before. In just four days in 2014, two Iraqi army divisions — on paper around 30,000 soldiers, plus 30,000 police — evaporated before an Islamic State offensive in Mosul that fielded fewer than 2,000 fighters. In 1975, South Vietnamese forces collapsed in less than two months in the face of the North Vietnamese spring offensive.
In neither case was the victim annihilated by enemy fire or overrun while defending to the last cartridge. Instead, the vast majority of combatants laid down their arms and either surrendered or deserted as the result of a conscious decision to stop fighting.
Surrender without annihilation is historically ubiquitous: Defeated armies are almost never physically destroyed in battle. Over 13 million Germans served in the Wehrmacht, but fewer than a third had been killed or wounded when Germany surrendered in 1945. Over 5 million Japanese fought in World War II, but fewer than half had been killed or wounded when Japan surrendered.
Nearly 200 years ago, Prussian military philosopher Carl von Clausewitz noted that the aim of combat is not literally to exterminate the enemy, which almost never happens. Instead, the goal is to persuade survivors who could physically continue to resist to stop fighting. Sometimes this happens quickly, after light casualties. Sometimes this happens more slowly, after heavier casualties. But defeat is almost always a choice by survivors rather than an act of brute-force destruction alone.
Why does this happen?
Early surrenders like this happen for a range of reasons, including ethnic divisions, weak ideological commitment, and inefficient unit organization or training. Weak political institutions in Afghanistan played a critical role by promoting corruption and cronyism, which kept elites happy and thus helped prevent a coup, but undermined soldiers’ confidence that their leaders were committed to keeping them alive in the field. All these factors contributed to Afghan forces’ chronically weak motivation.
Yet an ill-motivated Afghan force still maintained a stalemate for years. What converted chronic stalemate into acute collapse was the contagion dynamic that set in after poorly motivated forces encountered an outside shock — the Biden administration’s April announcement of a total U.S. withdrawal. The result was striking, but there is nothing unique to the Afghan army about this dynamic — past research by one of us (Zhukov, with Todd Lehmann) has found similar patterns in battle data spanning 1939 to 2011.
Warfare poses sharp dilemmas for soldiers — what scholars call collective action problems, where people have common interests in an outcome but individuals may not be willing to incur the costs of contributing to the effort.
Success in battle requires that soldiers choose to fight as a unit rather than flee, but individual decisions to fight depend on whether soldiers expect their comrades to do the same. Basic fire and movement tactics, for example, require one team to advance, while another draws attention away from this movement by firing from behind cover. If one team cannot trust the other to resist the natural temptation to flee or surrender, the advancing team will not advance and the firing team will not risk exposing itself to enemy fire.
If one expects one’s comrades to stay and fight, the best chance for survival is to do the same. But if the expectation is that others will flee, rational combatants will flee to avoid being the last man left in the trenches. That can make surrenders rapidly snowball, and spread across an entire force.
How do soldiers gauge whether their comrades will fight or flee? As Lehmann and Zhukov show, soldiers take their cues from their immediate colleagues, but also from their commanders, from soldiers in other units, and from other armies fighting the same enemy. When senior officers abandon their posts, subordinates view future threats to punish insubordination, surrender and desertion as not credible. Soldiers will also use surrender rates in past battles as a noisy signal about their own unit’s resolve. As more battles occur in which friendly forces (or allies) lay down their arms, soldiers are more likely to believe that a similar fate awaits them — particularly if such events happened recently.
The U.S. announcement set off the contagion dynamic
In Afghanistan, the U.S. withdrawal announcement in April was the shock that triggered the contagion dynamic. Before then, chronic motivation problems kept Afghan troops from defeating the Taliban, but the expectation that U.S. airstrikes would backstop them kept the great majority of Afghan soldiers in the field (if often holed up on checkpoints behind fortifications).
But the U.S. withdrawal announcement sent a sudden, nationwide signal of looming abandonment. If anyone then took the opportunity to fall back or surrender, comrades who were now increasingly nervous about others’ commitment began recalculating the odds that everyone else would stay and fight. The first setbacks after the announcement — initially a string of retreats that precipitated the Taliban’s Aug. 6 capture of Zaranj — thus started a chain reaction that quickly gained speed as each retreat encouraged others to give up.
For an entire national military to collapse in weeks may seem surprising. But history shows that surrender can easily become contagious after a sudden crisis — such as the announced withdrawal of a patron’s support.
Stephen Biddle is a professor of international and public affairs at Columbia University, a member of the Arnold A. Saltzman Institute of War and Peace Studies and adjunct senior fellow for defense policy at the Council on Foreign Relations. He has served on a number of U.S. government advisory panels, including Gen. Stanley McChrystal’s 2009 initial assessment team in Kabul, and is the author of “Military Power: Explaining Victory and Defeat in Modern Battle” (Princeton University Press, 2004) and “Nonstate Warfare: The Military Methods of Guerillas, Warlords and Militias” (Princeton University Press, 2021).
Yuri M. Zhukov is an associate professor of political science at the University of Michigan and a research associate professor with the university’s Center for Political Studies. His research focuses on the causes, dynamics and outcomes of conflict, at the international and local levels, using spatial statistics, mathematical/computational modeling and text analysis.