Syrians inspect the damage after reported regime airstrikes on the town of Muhambal, in Idlib province, on Saturday. (Amer Alhamwe/AFP/Getty Images)

When Syria and Russia intensified their Idlib offensive in May, there was condemnation from United Nations officials and other countries, as well as renewed media scrutiny. Cluster and barrel bombs targeted hospitals and residential areas, killing 300 civilians and displacing 300,000 in just five weeks.

Attacks on civilians are not unique to Syria — consider, for instance, the actions of the Israeli, Ukrainian and Saudi Arabian governments. Policymakers and scholars tend to agree that targeting civilians backfires, because those who survive often start cooperating with insurgents, buttressing the rebellion. Here’s the puzzle: Why do governments launch counterproductive indiscriminate attacks?

Most explanations imply incompetence — such as a lack of intelligence or internal discipline. My co-written research suggests that this could instead be a deliberate strategy. A survey in postwar Ukraine revealed that civilians who are safe from government attacks tend to believe that rebel provocation exonerates the government from blame for targeting civilians. However, victims of attacks do not view rebel provocation as absolving the state from blame. Ultimately, even competent governments could choose this strategy as long as they avoid directly targeting their supporters.

Indiscriminate violence can have vast political implications

When government forces kill civilians, aggrieved citizens blame the government for the loss of life and property. Civilians may then collaborate with rebels, potentially shaping the tide of the insurgency. Whom civilians blame for this type of collateral violence could have major operational implications in war.

Insurgent groups exploit this pattern by attacking government forces or holdings from densely populated areas. Their logic is straightforward: Any government response will inevitably harm noncombatants, which helps insurgents secure greater cooperation from aggrieved civilians.

Why do governments take the bait and opt to counterattack in urban or residential areas?

One possible answer is that civilians themselves may see government targeting of other civilians as justified. Governments work hard to create the perception of the dangers of rebel provocation. For instance, the Russian permanent representative to the United Nations justified the bombings of Syrian civilians, saying that “terrorists must not be allowed to … use hundreds of thousands of civilians in Idlib as a human shield.”

Similarly, the Ukrainian military said it abstained from bombing separatists in June 2014 because “militants rushed into residential buildings for cover,” thus suggesting that the rebels were willing to put civilians at risk. Or consider, for example, that the Israel Defense Forces publicized the discovery of Hamas’s BM-21 “Grad” rocket launchers inside a school building in 2014.

These examples illustrate that governments try to portray their attacks as direct responses to provocation from insurgents. That is, governments seem to believe that the population differentiates between provoked and unprovoked attacks on civilians — and if there are civilian casualties, a government response to a provoked attack is less likely to backfire politically.

How we did our research

To test this intuition, we surveyed individuals from all geographic regions of Ukraine except Crimea and the rebel-occupied territories of Luhansk province. Our sample included individuals who indicated that they had no direct experience of bombing in the previous year and those who did (7 percent out of 2,022 respondents). Most of the targeted respondents resided in Donetsk province, in both rebel- and government-controlled territory. Why Ukraine? This seemed a good test case, as rebels regularly attacked government positions from residential areas.

For our main question, respondents needed to select the answer that best described their thoughts when they saw a destroyed apartment building in Slovyansk, a city in Donetsk province that sustained heavy warfare. The first option in the range of answers was that government targeting of residential areas is a crime regardless of whether the army was retaliating for a rebel attack that originated from the residential neighborhood. The second option was that, before blaming the army, one should consider whether the army was retaliating for a rebel attack that originated from the residential neighborhood. Respondents could also refuse to answer or choose “Hard to say.”

The geographic concentration of ethnic Russians and Russian speakers in eastern Ukraine meant that individuals who experienced bombing were more likely to speak Russian and to identify as ethnic Russians. To ensure that respondents with and without exposure to government attacks were indeed comparable, we also analyzed our sample without those respondents who spoke primarily Ukrainian or identified as ethnically Ukrainian — and we also accounted for other demographic characteristics.

Here’s what we discovered

Respondents who had experienced a bombing event in the previous year were 50 points more likely to pick the first option, “It’s a crime regardless,” than those who lived in safety and had not experienced any such event. Even when we compare only the residents of Donetsk province, all of whom lived in proximity to the war zone, there was a 20-point gap in willingness to attribute blame regardless of provocation among individuals exposed to bombing.

These findings suggest that previous analyses of civil conflict looked primarily at the potential victims of attacks. Civil warfare in a given country, especially a separatist conflict, often includes geographically concentrated violent events. Citizens of the same country become “observers” or “recipients” of violence because of their different levels of exposure to violence. We found that civilians living in relative safety tended to act as rather cynical endorsers of government attacks on other citizens. If the domestic public — which keeps the government accountable — is insulated from violence, governments probably can avoid public disapproval if their indiscriminate response produces civilian casualties.

What does this suggest? We think it’s possible that even competent governments could choose this strategy — as long as they believe they can insulate their core supporters from violence.

Anna O. Pechenkina is an assistant professor of political science at Utah State University (anna.pechenkina@usu.edu).

Don’t miss anything! Sign up to get TMC’s smart analysis in your inbox, three days a week.