The Laws of War and Public Support for Foreign Combatants

Abstract

Are publics in great power democracies more likely to approve of foreign armed combatants that comply with international humanitarian law (IHL)? There is a wealth of evidence that armed combatants with an incentive to seek the support of outside compliance constituencies are more likely to adhere to IHL. Yet a key mechanism underlying these claims—that people in great power democracies are more likely to support armed combatants that comply with IHL—has not been directly tested. We address this question using a series of experiments embedded in nationally representative surveys conducted in three democracies that have frequently been involved in foreign interventions: France, the United Kingdom, and the United States. We find that belligerents—both governments and rebels—that comply with the laws of war are significantly more likely to garner support from publics in likely intervening countries compared to those who do not comply. In all three countries, compliance with international law caused greater approval of armed combatants as well as greater support for economic or military intervention (although support for military intervention remained relatively low in the treatment groups). This lends support to arguments that, to the extent combatants seek support from outside audiences, this can serve as a mechanism by which international law constrains armed combat.

Several recent theories about the relationship between international humanitarian law (IHL) and wartime conduct have a common element: that in some contexts armed combatants comply with IHL to gain the support of the international community, especially audiences in great power democracies.¹ Groups like the Sudanese People's Liberation Army or the Moro Islamic Liberation Front in the Philippines may have been motivated to demonstrate adherence to IHL because they believed it would improve their prospects for international recognition.² Rewards from international approval can be sizable, including not only the rhetorical prize of favorable foreign opinion, but also material benefits like economic or military aid, alongside censure, sanctions, or even military intervention against the adversary.³

But do audiences in great power democracies—in particular their national publics—respond to compliance with IHL by supporting such combatants? There are important reasons to believe so. International law can influence mass preferences and beliefs,⁴ leading to support for sanctioning violators, at least for issues like human rights or the environment.⁵ Reactions to Russia's attack on Ukraine in 2022 and allegations of war crimes appear consistent with this claim. Condemnation of Russia spread in the United States, the United Kingdom, and many other Western countries. And many were willing to match words with deeds, supporting stringent economic sanctions against Russia and providing arms to Ukraine, though opinion was more tempered on direct military intervention.⁶

Yet there are also reasons to doubt the responsiveness of the populations of great power democracies to IHL compliance abroad. Public opinion may prioritize pragmatic security concerns over international legal considerations.⁷ International inaction regarding the Rwandan genocide was at least partly due to apathy toward abuses perpetrated in the country.⁸ Publics in the United States and elsewhere sometimes seemingly show little concern for legal or moral constraints, from supposed taboos against nuclear weapons to prohibitions on targeting civilians.⁹ Several studies indicate that restraints from international legal rules are limited,¹⁰ or can even backfire and entrench opposing views.¹¹

So, are publics in great power democracies more likely to approve of foreign armed combatants that comply with IHL? Addressing this question presents an opportunity to examine a key mechanism underlying observational claims about the relationship between IHL and wartime conduct that has not been directly tested. To address this question we designed a survey experiment that allows us to isolate the effect of information about IHL compliance on attitudes toward foreign state and nonstate combatants. Because democratic publics are better able to influence their leaders, we fielded the survey in three democracies: France, the United Kingdom, and the United States. Alongside being permanent UN Security Council (UNSC) members with a history of foreign intervention, each has substantial institutional and material capacities to grant benefits, or impose costs, on foreign combatants. We also fielded a follow-up US survey providing an even harder test.

Our study yields four key findings. First, individuals in all three countries were significantly more likely to approve of foreign combatants they were told complied with international law. We specifically find that information about a foreign actor's IHL compliance increases respondents’ average approval of them from about 0.35 to 0.40 (on a scale of 0 to 1) to about 0.55 to 0.60, with treatment effects ranging from 21 percent to 55 percent in size, depending on the actor and country sampled. Second, respondents were consistently more likely to support costly policy actions by their home government on behalf of foreign combatants they were told complied with IHL. In most cases, this includes significant increases in support for economic aid (although overall support for military intervention remained relatively low in the treatment groups). Third, using causal mediation analysis, we find evidence of two key pathways by which information about compliance increases foreign approval for combatants: enhancing their reputation, and viewing the way they fight as morally correct. With respect to nonstate actors, there is a third pathway: enhancing the perceived legitimacy of their war aims. Finally, our follow-up US study indicates these findings are robust under more demanding and nuanced contexts with varying levels of US interests at stake in the foreign conflict and potential costs to intervene.

These findings make several contributions. First, we provide direct evidence that information about IHL compliance can shape public opinion in great power democracies. Second, our results clarify the role of domestic audiences in enforcing international rules. While much of the public opinion literature on international law focuses on attitudes toward behavior by individuals’ own governments,¹² we show that similar dynamics shape attitudes toward foreign actors. Our finding that information about compliance increases approval of foreign combatants means that public opinion may perform an important function in how outside powers react to combatants’ compliance choices. This finding is consistent with claims in several observational studies that susceptibility to international audiences can shape combatants’ wartime conduct, with external publics serving as important conduits.¹³ We build on prior work showing publics are sensitive to legal and moral concerns in foreign policy.¹⁴ Finally, moving beyond the common focus on the United States,¹⁵ we offer cross-national evidence of differences between nations in the mechanisms underlying public opinion.

The next section provides an overview of existing debates at the intersection of IHL, wartime conduct, and public opinion, from which we generate several testable hypotheses. We then describe the research design and outline a set of cross-national survey experiments. The next section presents the empirical results. We close by discussing the implications of the findings and avenues for future research.

International Humanitarian Law, (Non)compliance, and Public Opinion

The laws of war, or IHL, represent a long-standing humanitarian project seeking to curb excesses of armed conflict by limiting particular means and methods of warfare.¹⁶ Despite IHL's proliferation over the last century, actual compliance varies greatly, partly because of battlefield imperatives and security concerns. Belligerents may have strong incentives to breach these rules, especially as fighting intensifies or the tide turns against them.¹⁷ Mechanisms for enforcing IHL are often viewed as weaker than those in other domains.¹⁸ Domestic compliance constituencies have a hard time promoting restraint when a country's security is at risk.¹⁹ Reciprocity and avoidance of escalation can sometimes constrain combatants; legal agreements may facilitate these dynamics, but are not foolproof.²⁰

The historical record nonetheless suggests the potential of external mechanisms for punishing IHL violations. Neither in conventional interstate and civil wars, nor in insurgencies and other irregular conflicts, are warring parties isolated from outside influences.²¹ From widespread condemnation of the Tatmadaw for killings of the minority Rohingya in Myanmar, to providing weapons to Ukrainian forces fighting Russia's invasion, to outright military intervention decades earlier in Bosnia, other countries have played pivotal roles in many conflicts abroad. A growing literature suggests both government and opposition forces are correspondingly attuned to outside perceptions of their wartime conduct. Governments often depend on other countries for access to arms, foreign aid, and diplomatic assistance, alongside concerns over direct military involvement for or against them.²² The UNSC's denunciation and later authorization of NATO intervention in the 2011 Libyan civil war was motivated in part by the indiscriminate killing of civilians by Qaddafi forces. Rebel groups as varied as the Farabundo Martí National Liberation Front in El Salvador and the Free Papua Movement in Indonesia have likewise frequently employed diplomatic maneuvers to curry favor with foreign countries.²³ The preferences and actions of such third parties can ultimately have an outsized influence on the course and conduct of hostilities.²⁴

Several prominent accounts consider conditions where combatants comply with IHL to appeal to powerful external actors who care about international law. With an eye toward outside legitimacy, rebels seeking “human rights-conscious external sponsorship” appear more restrained in their wartime conduct.²⁵ Similar dynamics may be especially important for secessionist rebel movements whose goals depend on obtaining international recognition. They may use compliance “to signal both their willingness and their capacity to be good citizens of the international community.”²⁶ Yet even nonsecessionist actors may want benefits that come from bolstering their credentials. The Syrian National Coalition actively sought endorsements from Western governments to legitimate their efforts to overthrow the Assad regime.²⁷ As incumbents, government forces may be institutionally more secure but vary in their international sensitivity and corresponding incentives to use restraint.²⁸ Both governments and rebels seek to “win support from domestic and international audiences,” deciding how to fight based on those groups’ preferences.²⁹ Combatants looking for support from “Western international actors” may be less prone to target civilians or engage in other IHL violations.³⁰ Combatants that depend on outside democracies have similarly been shown to be more hesitant to resort to abuse.³¹

Some scholars thus share an emphasis on the crucial role of outside audiences in shaping combatants’ compliance decisions, whether referring broadly to the “international community” or “world opinion,” “Western states” and “democratic third parties,” or “human rights-conscious” “foreign state sponsors.”³² There is thus a wealth of evidence that combatants with incentives to seek support from outside compliance constituencies are more likely to follow IHL. Yet it is not always clear in these accounts who exactly are the relevant actors in these external communities whose attention combatants are vying for, which could include international organizations (IOs), NGOs, country leaders and other national elites, or mass publics.³³

We argue that one of these key audiences is the foreign public, especially in great power democracies. Skepticism certainly exists regarding the public's ability to influence foreign policy, especially with respect to matters of international law.³⁴ Yet a significant body of evidence suggests democratic governments and politicians are fairly responsive to their citizens’ views, including around foreign policy and the use of force abroad, contrary to common pronouncements that “politics stop at the water's edge.”³⁵ Foreign policy issues figure prominently in citizens’ electoral choices, and democratic leaders have historically been mindful of maintaining public consent throughout foreign adventures outside of the immediate threat of the ballot box.³⁶ And when it comes to questions of intervening in foreign conflicts, humanitarian objectives have figured prominently in mass opinion, even as public ambivalence has been cited for the lack of robust action by several Western governments in the midst of particular atrocities.³⁷ Even studies expressing some misgivings about the public in the end acknowledge meaningful mass influence over foreign policies across various countries.³⁸

For publics in powerful democracies, there is potential for their governments to take action for or against combatants abroad, which can, in turn, shape foreign conflicts. The United States, Britain, and France have been active in interventions through institutions like NATO or as UNSC members, but also individually, as in France's Opération Licorne in Côte d'Ivoire in 2002. Yet those same countries have chosen not to intervene, or waited to get involved, as killing raged abroad, whether in Rwanda, Bosnia, or initial phases of the war in Ukraine. Examining public opinion in such democracies offers part of an important pathway from below, even if it is not the only one, to understanding conflict dynamics and international law.

In addition to fighting on the battlefield, opposition groups actively pursue a form of “rebel diplomacy” to influence the preferences and beliefs of a variety of external actors, including directly speaking, or “marketing” themselves, to foreign publics.³⁹ This can take the form of working through more traditional channels like lobbyists and interest groups, or more recently through innovative strategies using social media, in either case focusing on the beliefs and preferences of foreign publics.⁴⁰ Reflecting on the reliance of many nonstate armed groups on ordinary people at home and abroad, one study concludes that “as significant as civilian support within rebel-controlled territory is, the support of civilians outside of rebel territory can be equally important.”⁴¹

A similar sensitivity to outside mass opinion can be seen among government forces. Just as rebels seek to “influence public opinion in Western countries,” the changing international normative environment means state forces need to be mindful of “world public opinion” when prosecuting counterinsurgency operations.⁴² Pointing to the need to consider both legal rules and the masses, an analysis of modern military doctrine notes that “no State can afford to be wholly regardless of public and world opinion.”⁴³

There are thus reasons to believe that ordinary individuals in great power democracies can serve as a compliance constituency for foreign combatants regarding IHL. But there is also evidence that members of the public, rather than being sensitive to compliance with international law, follow a consequentialist logic centered on their own national security.⁴⁴ For wartime conduct, several studies have found that public opinion in the United States and other democracies shows little regard for norms of civilian immunity or proportionality during warfare, at least regarding their own forces.⁴⁵ A guiding principle for many US policymakers is to presume a public “resistant to spending money and risking the lives of American soldiers for international efforts that are not directly tied to a narrow reading of the national interest.”⁴⁶ Inconsistent support for humanitarian interventions despite extreme suffering has led to worries over the robustness of norms like the responsibility to protect.⁴⁷ In some contexts backlash against international law means compliance could actually reduce public approval.⁴⁸

On the other hand, a range of research suggests public opinion is informed by international legal appeals, including during wartime.⁴⁹ International law is grounded in a deeper set of normative principles. Just as moral values shape general foreign policy attitudes,⁵⁰ invoking the “law” may activate a particular normative reasoning among audiences.⁵¹ Work in this area has focused on individuals’ attitudes to their own governments, but international law may also shape opinion on foreign actors. Ultimately, to serve as a compliance constituency, members of the public need not already have nuanced understandings of IHL's many rules, nor closely follow every event on foreign battlefields. Rather, the question is whether information about (non)compliance by foreign combatants can affect how these individuals respond. We aim to contribute to this debate by testing the hypothesis:

H1: Information about compliance with international law by foreign combatants increases public approval of those combatants in democratic countries.

Prior research further contends such approval translates into support for more concrete actions on the group's behalf, from the symbolic, like diplomatic representation, all the way to military intervention. For instance, when Libyan rebels gained US recognition, they also received access to billions of US dollars in frozen funds from the Qaddafi regime.⁵² Partly responding to alleged war crimes by Libyan government forces, NATO countries’ air campaign proved vital in turning the tide in favor of rebel forces. Support for interventions can be grounded in moral concerns,⁵³ which may be reinforced by information about IHL compliance. Publics are willing to take costly steps in response to violations of international law, though the specific question concerning IHL remains unexplored.⁵⁴

In other instances, like Rwanda and Bosnia, great power democracies and their publics hesitated to take firm steps.⁵⁵ Economic and military interventions are expensive, and the latter in particular endanger the lives of the public's own soldiers and risks escalating the conflict. One of the biggest concerns over how far NATO and other European countries would be willing to go in response to Russia's invasion of Ukraine was whether their publics would bear the resulting burdens. Drawing out the individual-level implications leads to our second hypothesis:

H2: Information about compliance with international law by foreign combatants increases public support in democratic countries for policies that favor such actors.

Taken together, the two hypotheses suggest that compliance with international law should shape democratic publics’ views of the conduct of foreign combatants, and how their governments should respond. Yet the empirical record has not told us whether this is true. Proponents and skeptics can each point to historical episodes featuring publics committed to IHL, or indifferent to it. Observational studies offer important indications of the role of external democratic audiences in compliance with IHL. However, they do not directly test underlying mechanisms concerning the attitudes of these audiences, relying instead on proxies like particular attributes of combatants or their connections to third parties. Public opinion research produces contrasting findings, focuses on attitudes toward publics’ own governments rather than those abroad, or examines issues unrelated to IHL.⁵⁶ It remains an open question whether information about compliance with IHL can shape public opinion about foreign combatants in contemporary democracies. We design a strategy to address this question more systematically.

Research Design and Samples

We use a series of survey experiments to test our hypotheses. This method has both advantages and limitations. The main strength of survey experiments is random assignment to treatment, which enables comparison of groups that are similar to each other on average across both observed and unobserved factors. This allows us to infer with greater certainty whether providing information about combatants’ (non)compliance with international law can affect public opinion. Such primes about combatants’ compliance behavior have real-world analogues for mass audiences, where international legal language figures prominently in popular coverage of conflicts and intervention.⁵⁷

Of course, the full reality is more complex, with individuals often receiving conflicting information and misinformation about these issues from multiple sources. Our design is not intended to recreate that tangled reality but to isolate the potential effects of a specific salient type of signal, one that theoretical and observational work suggests should be important. Understanding those effects can help us, in turn, understand why combatants deem such signals (and counter-signals) important enough to send. This approach complements other research tools, especially given the formidable obstacles posed by selection effects and issues of strategic interaction when studying international law.⁵⁸

We fielded our survey in France, the United Kingdom, and the United States. While the United States is the most common site for experimental studies on international law,⁵⁹ we test our hypotheses across multiple countries given research suggesting cross-national similarities but also differences.⁶⁰ We included two European countries to assess common presumptions that Americans and Europeans hold sharply different foreign policy views.⁶¹ These countries’ histories of intervention also allow us to present more plausible scenarios. All three have experienced high-profile debates over intervening abroad both economically and militarily, perhaps most prominently around the 2003 Iraq invasion, when France refused to join the US/UK-led “Coalition of the Willing.” On the other hand, France took the initiative to officially recognize Libya's rebel leadership in 2011, paving the way for others.⁶² More generally, all three are permanent members of the UNSC, so their publics are more likely to see their governments involving themselves in these sorts of decisions. Finally, all are consolidated democracies, where public opinion is more likely to be relevant for policy outcomes.

To provide a common baseline for the survey, respondents were given some background to the scenario and an overall understanding of IHL.⁶³

Around the world in various foreign countries, some opposition groups decide to take up arms and fight against their government. Recent examples include the Shining Path in Peru, the Azawad National Liberation Movement in Mali, and Houthis in Yemen. During the fighting, both opposition and government military forces can differ in the level of violence they choose to use.

To try to limit the effects of armed conflict for humanitarian reasons, countries have collectively created a set of international rules, commonly referred to as the laws of war. These laws impose restrictions on how combatants are allowed to fight during warfare, such as attacks against civilians, abusing prisoners, or using land mines.

We will describe a situation involving an armed opposition group and a foreign government that various countries have faced many times in the past and will probably face again. This is a general situation, and is not about a specific opposition group or government in the news today. Some parts of the description may strike you as important; other parts may seem less important.

Please read the details carefully. After describing the situation, we will ask your opinion about a few policy options.

We adopted an abstract scenario in view of concerns of pretreatment effects from references to historical or current events in opinion research on international law.⁶⁴ We still used details to enrich the vignette and make it more concrete. Alongside the general background prompt referring to a range of real conflicts, we included two sets of contextual treatments varying the government's regime type and the opposition group's war aims. Scholarship shows the importance of such attributes of state and nonstate combatants in shaping their wartime conduct and how they are perceived by outsiders.⁶⁵ We varied whether the foreign government was a democracy or autocracy.⁶⁶ We also randomized the opposition group's objectives: secession or overthrowing the government.⁶⁷

As shown in the online supplement, the effects of compliance with IHL on approval of these actors did not vary significantly based on these contextual treatments. In the rest of our analysis, we therefore focus on the average treatment effects of (non)compliance aggregated across these contextual groups. But greater contextual detail tends to dampen the size of other treatment effects,⁶⁸ so our design should provide a conservative test of the main effects of IHL compliance.

Respondents were randomly assigned additional information, also in random order, regarding whether the government and the opposition group's tactics complied with international law. Government and opposition (non)compliance behavior were allowed to vary separately, resulting in a 2 × 2 × 2 factorial design with eight main experimental groups.⁶⁹ The following is the language for these conditions.

Compliance: “There have been recent reports that military forces from the [Opposition Group / Foreign Government] have been following international laws that limit how combatants are allowed to fight.”

Violation: “There have been recent reports that military forces from the [Opposition Group / Foreign Government] have been violating international laws that limit how combatants are allowed to fight.”

Word choices for these conditions were influenced by two concerns. First, to make the treatments as similar as possible, only one word—“following” versus “violating”—differs between them. This minimal contrast should bias against finding significant differences in outcomes, while allowing us to better infer the source of any effects detected. Second, we refer to “international laws” generically, rather than to specific legal instruments. This was because individuals are generally less familiar with the detailed requirements of specific international treaties or customs, and because naming specific laws (the Geneva Conventions, for example) risks raising unmeasured concerns or priors in participants’ minds.⁷⁰

Respondents were then asked a pair of items (in random order) evaluating each actor's actions: “Do you approve, disapprove, or neither approve nor disapprove of the way the [Opposition Group / Foreign Government] is fighting the war?” Answers used a five-point scale from “strongly approve” to “strongly disapprove.”

Along with these main outcome items, we asked questions to assess possible mechanisms underlying the relationship between international law and public opinion, which remain unresolved from prior experimental and observational studies. As with previous questions, the reputation and morality questions were asked twice, once for each combatant. But because the government is the incumbent and generally seeks to maintain the status quo, we asked about legitimacy of war aims only for the opposition.

Reputation: “Do you think the [Opposition Group / Foreign Government]'s conduct has helped or hurt the [group / country]'s reputation in the world?” (Answers: Helped a lot / Helped somewhat / Neither helped nor hurt / Hurt somewhat / Hurt a lot)

Morality: “Do you think the [Opposition Group / Foreign Government]'s conduct is morally acceptable?” (Answers: Yes/No)

Legitimacy: “How legitimate do you think are the aims of the Opposition Group?” (Answers: Very legitimate / Somewhat legitimate / Neither legitimate nor illegitimate / Not very legitimate / Not at all legitimate)

To better assess whether respondents would be willing to move beyond attitudinal affirmation to more costly actions, we asked about three types of policy interventions by their country's government toward either combatant: verbal statement of (dis)approval; economic sanctions/aid; and military intervention. The verbal statement and economic policy questions allow respondents to express support for a more positive or negative (or neither) position. Respondents could thus support a statement by their country's leader condemning both sides, or condemning one and praising the other, or some other combination. To do so, we asked a pair of each of these policy items, once with respect to each armed combatant (with the text adapted to respondents’ home country). Responses to the two military intervention items were combined, as the results section describes.

Verbal statement: “To what extent would you support the [US President / UK Prime Minister / French President] making a speech about this issue, where 1 is strongly supporting a speech condemning the [Opposition Group / Foreign Government], 7 is strongly supporting a speech praising the [Opposition Group / Foreign Government], and 4 is saying nothing about the [Opposition Group / Foreign Government]?” (Respondents chose a whole number from 1 to 7)

Economic policy: “To what extent would you support the [US President / UK Prime Minister / French President] using economic policy to address this issue, where 1 is strongly supporting imposing economic sanctions on the [Opposition Group / Foreign Government], 7 is strongly supporting sending foreign aid to the [Opposition Group / Foreign Government], and 4 is doing nothing economically about the [Opposition Group / Foreign Government]?” (Respondents chose a whole number from 1 to 7)

Military intervention 1: “To what extent do you approve or disapprove the [US President / UK Prime Minister / French President] sending military forces to the country?” (Strongly approve/ Somewhat approve/ Neither approve nor disapprove / Somewhat disapprove / Strongly disapprove)

Military intervention 2: “If the [US President / UK Prime Minister / French President] sent military forces, would you prefer them being sent to support…” (The Foreign Government / Neither side (neutral) / The Opposition Group)

We fielded this experiment, using the online survey firm Lucid, to diverse national samples in March (United States), April (United Kingdom), and August (France) of 2019. Lucid used quota sampling based on gender, age, and location (as well as race/ethnicity for the United States) so that participants were broadly representative of their relevant national adult populations.⁷¹ Lucid samples have been shown to correspond well to national populations, and studies using Lucid have generated experimental results matching those using national benchmarks.⁷² Table 1 summarizes the national sample sizes and experimental groups.

Table 1. Experimental groups

Note: Sample sizes are 2,009 for the US, 2,051 for France, and 2,198 for the UK.

Results

We begin with descriptive results for the main compliance treatments. Figure 1 shows the average approval for combatants depending on whether the actor followed or violated IHL.⁷³ Across country samples, average approval of either actor is 0.55 to 0.60 in the compliance groups, and 0.35 to 0.40 in the violation groups, providing an early indication that our respondents were sensitive to this information.

Figure 1. Average approval of combatants (with 1 being “strongly approve” and 0 being “strongly disapprove”)

We follow up with statistical tests that estimate the treatment effects of IHL compliance or violation information (in terms of percent change) on individuals’ approval of the combatants. The upper panel in Figure 2 shows the effects for foreign government compliance; the lower panel reports the corresponding effects for opposition group compliance. In all cases, the effect is positive and statistically significant: compliance with IHL robustly improved individuals’ attitudes toward each combatant.⁷⁴ Audiences thus appear to evaluate the compliance behavior of foreign state and nonstate combatants in similar ways, which is consistent with several past studies.⁷⁵ The treatment effects ranged from about 21 to 55 percent. In the US sample, compliance increased average approval of the foreign government from about 0.42 to 0.57, a treatment effect of about 37 percent. In the France sample, compliance increased average approval of the foreign government from about 0.43 to 0.52, a treatment effect of about 21 percent. In the UK sample, compliance increased average approval of the foreign government from about 0.38 to 0.56, a treatment effect of about 49 percent. In the US sample, compliance increased average approval of the foreign opposition from about 0.43 to 0.58, a treatment effect of about 35 percent. In the France sample, compliance increased average approval of the foreign opposition from about 0.43 to 0.54, a treatment effect of about 25 percent. In the UK sample, compliance increased average approval of the foreign opposition from about 0.33 to 0.55, a treatment effect of about 55 percent. There is evident cross-national variation in treatment effect sizes. UK respondents were most sensitive to information on compliance or violation, followed by the United States, and lastly France. Overall, these results provide strong support for H1.⁷⁶ Equally important, we found no evidence that these effects are conditional on the conduct of the other actor (see the online supplement). While some work argues reactions to one group's (non)compliance may be affected by their opponent's actions,⁷⁷ our results suggest respondents are evaluating each side's behavior on its own terms.

Figure 2. Effects of compliance on approval of combatants, by country (with 95% confidence intervals)

To test H2, we analyzed the effects of information about IHL compliance/violation on individuals’ support for actions of varying costs taken by their own government. In the first two instances, higher/lower values indicate stronger approval for actions that are helpful/harmful to the relevant combatant: for verbal statement supporting a speech praising as opposed to condemning, and for economic policy supporting foreign aid as opposed to sanctions. We combine respondents’ answers to the pair of military-related questions into one binary item, as follows. For each actor, respondents are coded as supporting a military intervention on behalf of that actor if in military intervention 1 they strongly or somewhat approved of military intervention, and in military intervention 2 they backed military intervention in support of that actor.⁷⁸

The upper panel in Figure 3 shows the treatment effects (in terms of percent change) of foreign government compliance on individuals’ support for policy actions by their home government. The lower panel shows the corresponding treatment effects for opposition group compliance. In all cases, the effect is positive, and in all but three instances statistically significant: compliance generally increased individuals’ willingness to support policy responses. With respect to verbal statement, in the US sample foreign government compliance increased average support from about 3.6 to 4.1 (on a scale of 1 to 7), a treatment effect of about 13 percent. In the France sample, foreign government compliance increased average support from about 3.7 to 3.9, a treatment effect of about 6 percent. In the UK sample, foreign government compliance increased average support from about 3.5 to 4.0, a treatment effect of about 15 percent. In the US sample, foreign opposition compliance increased average support from 3.8 to 4.1, a treatment effect of about 8 percent. In the France sample, foreign opposition compliance increased average support from about 3.8 to 4.0, a treatment effect of about 4 percent. In the UK sample, foreign opposition compliance increased average support from about 3.5 to 4.0, a treatment effect of about 16 percent.

Figure 3. Effects of compliance on support for policy interventions, by country (with 95% confidence intervals)

With respect to economic policy, in the US sample foreign government compliance increased average support from about 4.5 to 5.0 (on a scale of 1 to 7), a treatment effect of about 13 percent. In the France sample, foreign government compliance increased average support from about 4.6 to 5.0, a treatment effect of about 8 percent. In the UK sample, foreign government compliance increased average support from about 4.2 to 5.0, a treatment effect of about 19 percent. In the US sample, foreign opposition compliance increased average support from 4.6 to 5.1, a treatment effect of about 10 percent. In the France sample, foreign opposition compliance did not significantly increase average support. In the UK sample foreign opposition compliance increased average support from 4.3 to 4.9, a treatment effect of about 14 percent.

Recall that military intervention is a binary variable, so we can describe the percentages of respondents who supported or opposed intervention. In contrast to the other two outcomes, support for military intervention, on behalf of either foreign combatant, was generally low across our samples, probably because of the costliness of such interventions. Thus, while we do find significant treatment effects, they are relative to low baselines. In the US sample, foreign government compliance did not significantly increase support for military intervention. In the France sample, foreign government compliance increased support from about 14 percent to 17 percent, a treatment effect of about 24 percent. In the UK sample, foreign government compliance increased support from about 12 to 16 percent, a treatment effect of about 33 percent. In the US sample, foreign opposition compliance increased support from about 8.5 to 16 percent, a treatment effect of about 88 percent. In the France sample, foreign opposition compliance did not significantly increase support for military intervention. In the UK sample, foreign opposition compliance increased support from about 5 to 7 percent, a treatment effect of about 45 percent.

Respondents in all samples were significantly more likely to support military intervention on behalf of at least one combatant that complied with international law, even if absolute levels of approval for such intervention remained fairly modest: in the United States, the opposition group; in France, the foreign government; and in the United Kingdom, both. Comparing the treatment effects across actors, we found that in all but one case, the relative effects of compliance of foreign government and opposition forces were statistically indistinguishable, suggesting respondents’ sensitivity to legal information may not have depended on which actor was engaging in (non)compliance. The one exception concerns US support for military intervention: the effect of compliance on support for intervening on behalf of opposition groups was significantly larger than the effect based on the foreign government's conduct. That is, Americans’ support for military intervention on behalf of foreign governments is less sensitive to whether such governments did or did not comply with international law. Overall, these results provide support for H2.

To help uncover possible mechanisms underlying these relationships, we examine likely causal mediators of our results, using procedures proposed by Imai and colleagues.⁷⁹ We have already mentioned these candidate mediators, covered by follow-up questions in our survey, which are intended to probe possible rationales for our respondents’ answers: reputation, morality, and legitimacy. These factors may have mediated the effects of the compliance treatments on individuals’ approval of the combatants. For each potential mediator and actor, we estimated a model with the mediator as the dependent variable, estimated a model with approval of the actor as the dependent variable, and then estimated the average causal mediation effect (ACME).

Figure 4 reports the results—the top panel for the foreign government, the bottom for the opposition.⁸⁰ Reputation and morality were significant mediators with respect to both actors in all three country samples. That is, respondents were more likely to approve of combatants who followed international law partly because they believed such compliance improved those actors’ reputations, and partly because they believed compliance was morally correct. This does not imply, however, that there might not be other, unmeasured mediating factors. The ACME for France is significantly smaller for both actors and both mediators, suggesting especially in this sample that other unmeasured mechanisms may be at work. With respect to the third mediator—the legitimacy of the opposition group's cause—the estimated ACMEs are positive, but they are not significant in the US sample. More importantly, ACMEs for this mediator are generally smaller than for the other two. That is, while legitimacy concerns might shape the relationship between compliance and outside audiences’ approval, they are less important than reputation and morality.

Figure 4. Causal mediation analysis, by country (with 95% confidence intervals)

Follow-Up Study

While our design looks to inject sufficient contextual detail about the conflict and combatants to resonate with respondents, there may still be concerns that a harder test is warranted, in view of common critiques that international legal principles are set aside when core national interests are at stake, or there is a prospect of significant costs to one's own country.⁸¹ Taking this possibility into account, we fielded a follow-up study in March 2024 in the United States (n = 549) using the same survey firm, Lucid, but with two new contextual treatments, which explicitly varied (1) the extent the foreign conflict affected US security and economic interests (high or low), and (2) the potential costs to the United States should it choose to get involved (high or low).⁸² Limiting the number of moving parts to maintain sufficient statistical power, we fixed the attributes of the combatants from the two contextual treatments from our original survey (regime type and opposition aims), but kept all other elements of the design identical.⁸³

We begin with descriptive results. Figure 5 shows the average approval for combatants depending on whether the actor followed or violated IHL for each of the experimental groups. Average approval of either actor is 0.55 to 0.60 in the compliance groups, and 0.35 to 0.45 in the violation groups. This first look suggests that foreign combatants’ compliance with IHL continues to matter a great deal, even across changing interests and costs. We follow up with statistical tests that estimate the treatment effects of information about IHL compliance or violation and information about US interests and costs on individuals’ approval of the combatants.

Figure 5. Average approval of combatants by experimental group, 2024 (with 1 being “strongly approve” and 0 being “strongly disapprove”)

Figure 6 reports the effects of information on government and opposition compliance with international law. The first row of each panel reports the effect across all survey respondents, while the remaining rows do so by subsample, each of which received different information on US interests and costs. Across all groups, information about foreign combatants’ compliance with IHL consistently increases US respondents’ approval of that actor, suggesting the results of our 2019 study are unlikely to be an artifact of a particular sample or point in time. With respect to the foreign government, compliance increased average approval from about 0.39 to 0.54 across all groups, a treatment effect of about 38 percent. When US interests were low, average approval increased from about 0.43 to 0.53, a treatment effect of about 25 percent. When US interests were high, average approval increased from about 0.35 to 0.55, a treatment effect of about 54 percent. When US costs were low, average approval increased from about 0.40 to 0.51, a treatment effect of about 27 percent. When US costs were high, average approval increased from about 0.38 to 0.57, a treatment effect of about 50 percent. With respect to the foreign opposition, compliance increased average approval from about 0.46 to 0.57 across all groups, a treatment effect of about 25 percent. When US interests were low, compliance did not significantly increase approval. When US interests were high, average approval increased from about 0.42 to 0.59, a treatment effect of about 41 percent. When US costs were low, average approval increased from about 0.46 to 0.56, a treatment effect of about 21 percent. When US costs were high, average approval increased from about 0.45 to 0.58, a treatment effect of about 30 percent.

Figure 6. Effects of compliance on approval of combatants, 2024 (with 95% confidence intervals)

Interestingly, the treatment effects of compliance were significantly smaller (p < 0.05) with respect to both actors when US interests were low than when US interests were high. In other words, information about compliance with IHL increased approval of the foreign actor by a greater amount when respondents were told the US interests in the conflict were greater. This is consistent with several studies showing that legal rules are important even under more trying circumstances.⁸⁴

Conclusion

Are publics in great power democracies more likely to support foreign combatants that comply with IHL? Important recent theoretical developments seeking to explain wartime conduct assume that they are. That work further argues that compliance choices are driven in part by armed combatants’ seeking approval and assistance from various compliance constituencies, including publics in great power democracies, as well as states and NGOs.⁸⁵

We build on this research by directly examining the relationship between information about compliance with IHL and public opinion in three powerful democracies. We find that such information significantly and consistently increases public approval for foreign combatants, as well as for government policies to assist them. We also uncover at least part of why individuals in democratic countries are more likely to approve of foreign combatants that comply with IHL: they believe that such compliance enhances these actors’ reputations, and that compliance is the morally correct choice, though with some important cross-national variation. Moreover, with respect to nonstate combatants, we find that a third mechanism—perceived legitimacy of war aims—also mediates the relationship between IHL compliance and public approval, though to a lesser extent.

In addition to providing a better understanding of the relationship between IHL compliance and public opinion in great power democracies, our findings shed light on the ways IHL may constrain armed combatants. The spread of IHL has been argued to shape the incentives of various combatants.⁸⁶ Although enforcement is costly and inconsistent, actors often comply, when we might expect otherwise. How and why does IHL influence the choices of combatants? Our findings suggest an important role for public opinion in powerful democracies. Democratic publics are more likely to approve of, and support costly assistance to, foreign combatants that comply with IHL. This may result in more restraint in some situations in which combatants seek the support of those publics, than might occur in the absence of IHL. This provides evidence of an external role for domestic audiences in the enforcement of international rules, specifically in the challenging area of armed conflict. Our results also contribute to findings on the role of “ordinary” actors in international law's functioning.⁸⁷

Yet there are reasons to interpret our results less optimistically. We did not use deception, but outside our controlled experimental setting actors regularly attempt to mislead and misinform. Returning to our opening example of the war in Ukraine, Russia engaged in a concerted propaganda campaign before and after its 2022 invasion, especially toward subsequent allegations of war crimes and other IHL violations.⁸⁸ If individuals are susceptible to accurate information about compliance, they may also be susceptible to misinformation. This might lead them to support actors they have been misled into believing are “following” IHL, in which case public opinion would not contribute to compliance but perhaps support abuses. In view of this important empirical baseline that publics can be influenced by legal appeals in foreign conflicts, we hope future work will explore how information and misinformation compete in shaping public opinion during wartime.

Our findings suggest other avenues for future work. First, we focused on the average effects of compliance with IHL on public opinion, while incorporating a few relevant contextual features. Our follow-up US study built on this by varying the level of US interests in the foreign conflict and the cost of intervention. Yet many other factors may condition the receptivity of domestic audiences to international legal appeals, such as whether the offending foreign combatants are formal or ideological allies, or the potential agenda-setting role of domestic or other foreign elite voices that could cloud perceptions of wartime behavior or the precise meaning of legal rules. Additional research should assess the resilience of the effects we found, especially when confronted by counteracting legal and nonlegal forces, which can help broaden our understanding of how mechanisms involving the public work in bolstering (or limiting) international law.⁸⁹

Second, while we focused on three democracies that have at times acted alone, many of their interventions have been multilateral, requiring buy-in from other countries. Smaller democracies have played pivotal diplomatic and military roles, but it remains unclear whether their publics are similarly susceptible to compliance information.

Third, while publics in autocracies are generally thought to hold less sway over their leaders, they may be sensitive to information about foreign policy behavior, including international law.⁹⁰ Mass-level influences may still operate in ways that make it fruitful to explore in greater detail the international legal preferences of publics in nondemocracies.⁹¹

Fourth, not only do states sometimes provide assistance to armed combatants, but sometimes so do IOs and NGOs. These actors may represent important additional compliance constituencies for rebel groups and governments.⁹² In the future, we hope to explore whether and how combatants’ compliance with international law affects elites working in such organizations, which would help deepen our understanding of whether belligerents base their strategic decisions in part on these institutional actors’ preferences. Alongside elites within IOs and NGOs, attention should also be devoted to more traditional national policymaking elites because research has suggested important areas of similarity (but also differences) between the foreign policy preferences of elites and mass publics.⁹³