President Trump speaks during a meeting with Canadian Prime Minister Justin Trudeau in the Oval Office on Thursday in Washington. Trump said Thursday that "Iran made a very big mistake" in shooting down a U.S. drone but suggested it was an accident rather than a strategic error. (Evan Vucci/AP)

On Thursday, the Iranian military downed an unmanned American surveillance drone. After weeks of President Trump ramping up his rhetoric against Iran and its leaders, many wondered whether he would use this event as a pretext for military action against the Islamic Republic.

According to Trump’s Friday tweets, he thought about authorizing airstrikes but decided against it last minute. What explains the president’s decision not to use force?

Trump had at least two choices

One possibility is that the president did not have to use force because he had a range of other alternatives. Scholars of foreign policy often talk about “substitutability,” which is really just a way of saying that leaders have a menu of choices to select from when making foreign policy decisions. In this case, Trump had (at least) two options to retaliate against Iran’s aggression: escalating with airstrikes or sticking with (and increasing) sanctions. The choice to not use airstrikes is expected because they tend to be used in low-salience crises, but choosing sanctions is unexpected in an urgent, high-stakes (potential war with Iran) confrontation because sanctions (especially ones against autocratic states) often take a long time to produce results.

This choice stands in contrast with Trump’s decision to employ airstrikes against Syria in April 2018. In that case, the stakes for the United States were much lower. Syria, unlike Iran, has been destabilized by its ongoing civil war. It does not have Iran’s capacity to strike back against the United States. In the Iran case, unlike in Syria, the United States faces the very real possibility of a regional war. This may explain his choice of a more restrained option.

There may also be a strategic logic

Another reason Trump may have pulled back on using airstrikes against Iran relates to strategy. Air power can be used in two primary ways: denial and punishment. Denial airstrikes target the military capabilities of an opponent to limit its ability to respond. Punishment airstrikes attempt to break an opponent’s will to fight by inflicting pain on the civilian population. In our research, we find evidence that denial strategies are historically more effective. Autocratic targets are also less likely to be vulnerable to coercion via punishment.

Iran’s action in shooting down an American drone demonstrates a desire to deny American capabilities and limit access to intelligence. Trump may have worried that the proposed American strikes look more like punishment, given the high expected casualty numbers. Knowing that such strikes are unlikely to be militarily effective, he may then have determined that gains were unlikely to outweigh the negative public response to Iranian casualties. Leaders of democratic countries are more likely to be concerned about casualties than their autocratic counterparts.

Or Trump may have worried that an airstrike would be disproportionate

A final and related concern that might have influenced Trump’s decision-making is the risk of collateral damage. Trump’s choice to call off airstrikes may also be explained by his discussion of the risk of a disproportionate response, in which collateral damage and casualties can increase the expected cost of airstrikes and make them less appealing to leaders.

In his tweet, Trump invoked the classical just war theory of proportionality, which generally refers to the military benefits obtained from an attack being large enough to justify the harm caused by the attack. Trump’s tweet did not specify what the target sites in Iran were, but it implied that killing 150 Iranians was too large a cost to justify the potential benefits of this attack. Notably, Iran also made a public announcement stating it had the ability to shoot down a U.S. military plane but chose not to do so. Similarly to Trump, the Iranians seem to be signaling a desire not to escalate the crisis by avoiding human casualties.

Beyond military casualties that could escalate a crisis, our work shows that using air power in conflict, whether in conjunction with ground forces or on its own, leads to increased civilian casualties. Though Trump’s tweet did not specify whether those 150 expected casualties would be military or civilian, any time air power is used, civilian casualties are a potential cost. Though aerial technology has grown more precise, the immediacy of airstrikes, as well as their large distance from the target, can lead to increased collateral damage. Civilian casualties can lead to negative attention from the U.S. public and the international community, as has been the case with U.S. uses of air power in the Middle East and Africa.

Restraint won out

Though airstrikes are often thought of as a low-cost option, their potential for creating casualties and escalating a conflict made them too costly of an option in this case. Recent reports by human rights groups and the publication of civilian harm information by the nongovernmental organization Airwars, as well as an unprecedented review from the Pentagon of civilian casualties in U.S. military interventions in recent years, demonstrate strong public concern about the collateral deaths of civilians.

Airstrikes are more likely to result in collateral damage than other forms of military force. Given the popularity and prominence of air power as a foreign policy tool, leaders are perhaps beginning to take this unintended consequence into account.

Susan Hannah Allen is an associate professor at the University of Mississippi whose research focuses on coercion in the international system. Find her on Twitter @lady_professor.

Carla Martinez Machain is an associate professor at Kansas State University whose research explores military effectiveness and military policy. Find her on Twitter @carlammm.