Mechanisms of Morality

Abstract

Humanitarian interventions (HIs) are a common aspect of US foreign policy. Policy makers acknowledge the importance of public support for interventions, but scholars remain divided about the extent and basis of that support. Using a series of survey experiments, we evaluate attitudes about HIs, assess whether the public supports these interventions for instrumental or moral reasons, and test which aspects of morality are most salient. The findings indicate that interventions addressing humanitarian crises boost public support, with the basis of that support residing primarily in normative contentions that the United States has a moral obligation to protect civilians. This research advances understandings of morality in foreign policy, mediates debates about the determinants of public attitudes—including when the public makes moral rather than prudent decisions about interventions—and has important policy implications. It suggests a morally motivated public may be more likely to support risky HIs, increasing the likelihood of using force.

Keywords

humanitarian intervention public opinion moral foundations foreign policy

Humanitarian interventions (HIs)—deployments of military force across borders with the predominant purpose of ending the grievous suffering or loss of life of foreign civilians—have become a prominent form of US military action in the post–Cold War period (Pattison 2010, 28). Of the US-led interventions in the last twenty-five years, about half have had primary objectives that were humanitarian in nature. For example, the United States deployed Marines to assist with food shipments in Somalia in 1993 and used air strikes to “minimize and eventually end the human suffering” in Bosnia in 1995 (Clinton 1995). More recently, the United States intervened in Libya in 2011 because Qadhafi had “forfeited his responsibility to protect his own citizens and created a serious need for immediate humanitarian assistance and protection” (Obama 2011). In short, military responses to humanitarian crises have been a common feature of the post–Cold War period.

The path to military action, however, must first be paved with public support. As Samantha Power (2002)—a scholar of human rights and later the US Ambassador to the United Nations (UN)—suggests, when it comes to HIs, “it is in the realm of domestic politics that the battle to stop genocide is lost” (p. XVII). When the public does not support intervention, the president’s ability to act becomes constrained. Conversely, bottom-up pressure from the public can create an impetus for action that would not otherwise take place. Public opinion plays a particularly important role in HIs because rather than responding to security threats, these operations rely on the public’s belief that military action to protect foreign civilians is in the interest of the United States and worth the potential cost (Boettcher 2004). Against the backdrop of widespread attention to humanitarian crises and assertions about the importance of public opinion in decisions to intervene, this article investigates the extent and basis of support for HIs.

One prevailing view in the literature suggests that even if public support is consistent with HI norms, instrumental calculations may still be the primary motivation behind approval. As Goldsmith and Posner (2005, 185) argued in the context of state compliance with international law, “when the instrumental calculus suggests a departure from international law, international law imposes no moral obligation that requires contrary action.” Placed in the context of HIs, this view implies that the humanitarian aspect of these operations is less consequential than the cost or strategic factors that the public associates with humanitarian action. Support for HIs is thus driven by the same instrumental mechanisms as conventional military interventions. Another perspective suggests that the public supports HIs out of a sense of moral obligation to save strangers from harm (Finnemore 2003; Wheeler 2000) not because action overlaps with a narrow instrumental calculus oriented toward improving one’s own strategic or security circumstances. From this view, support for HIs is primarily built through moral mechanisms. While a combination of instrumental and moral mechanisms are present in individual cases of intervention, determining which mechanisms are the most significant drivers of HIs is key to understanding when the public will approve of military action.

To differentiate between the two alternative accounts of public support, we use a series of survey experiments that evaluate attitudes about HIs, assess the relative importance of instrumental and moral mechanisms, and identify the most salient aspects of morality. The findings consistently indicate that interventions intended to address humanitarian crises boost public support. This increased support is driven not primarily by instrumental reasoning but by a normative contention that the United States has a moral obligation to intervene to protect civilians.

Whether and why the public supports HIs matters for multiple reasons. Most fundamentally, understanding attitudes underlying support is important for anticipating the circumstances in which states are likely to use force, since public attitudes create political incentives for continuing or withdrawing from military conflict and acting against these preferences carries political costs (Tomz and Weeks 2013, 850). If instrumental mechanisms drive support, domestic approval will require convincing the public that military action is strategically important or will be low in costs (Gartner 2008; Gelpi, Feaver, and Reifler 2009) and relevant to US security interests (Drezner 2008; Herrmann, Tetlock, and Visser 1999; Mandelbaum 1996). Alternatively, if support operates through the mechanisms of moral obligation, then evidence of a humanitarian crisis should be sufficient on its own to gain public approval for intervention (Finnemore 2003; Wheeler 2000).

The rest of the article proceeds as follows. First, we provide an overview of existing scholarship on HIs. Second, we consider three potential sources of individual attitudes about HIs, including views about cost, strategic consequences, and moral obligation. Third, we outline our empirical strategy, which employs experiments to explore support for HIs and the mechanisms of support. While we initially focus on moral obligation and instrumental factors, we also deduce and evaluate five moral mechanisms that theoretically undergird support for HIs. We then highlight important heterogeneous effects, presenting evidence that partisan identification impacts attitudes about both HIs and the sense of moral obligation. The final section concludes with a discussion of how individuals form opinions about intervention and the circumstances under which HI is more likely.

Attitudes about HIs

The end of the Cold War brought significant changes in how states consider the use of military force. As Voeten (2001, 846) observes, the number of United Nations Security Council (UNSC) vetoes plummeted as the great power paralysis of the Cold War gave way to a more cooperative tenor. Concurrently, the UN increased its peacekeeping efforts and the number of interventions it authorized—from 22 chapter VII resolutions on threats and one peacekeeping mission during the Cold War to 145 resolutions and 15 peacekeeping operations in the eight years after. Several of these resolutions were passed in the service of humanitarian crises, as multilateral coalitions were authorized to use “all necessary means” to restore peace and security in countries such as Somalia, Sierra Leone, and East Timor.¹ These resolutions and operations introduced HIs as a newly legitimate use of military force.

HIs are defined by three characteristics: (1) the use of military force across borders, (2) without the consent of the target state, and (3) with the predominant purpose of preventing the widespread suffering or death of foreign civilians (Finnemore 2003, 53; Pattison 2010, 28). The third characteristic of HIs reflected a new set of commitments that distinguished these cases from security interventions. The purpose of these operations was no longer primarily territorial or strategic but focused on protecting foreign civilians. Such HIs were guided wholly or primarily “by the sentiment of humanity, compassion or fellow-feelings, and in that sense disinterested” (Parekh 1997, 54). Somalia provided the quintessential example of a HI because it was undertaken despite there being “little or no strategic or economic importance to the principal intervener (US)” (Finnemore 1996, 153).²

While the international community endorsed many of these interventions, the decision to intervene falls upon states and often the United States. Power (2002) points out that “the United States’ decisions to act or not act have had a greater impact on the victims’ fortunes than those of any other major power” in part because of its military capaciousness to stand between perpetrators and their victims (p. XX). However, within the United States, these decisions hinge on public debates about whether and why intervention is warranted. With sufficient skepticism about the importance of atrocities being committed abroad, the US government can stand by, whereas a public galvanized to address the humanitarian crisis can prompt American intervention (Power 2002, XVII).

Public attitudes typically reside somewhere between antipathy and fervor, leading to claims that the public is apathetic or ignorant toward foreign policy (Converse 1964; Lippmann 1922). However, others contend that public attitudes toward foreign policy are as stable in the aggregate as attitudes toward domestic policy and influence both presidential approval and election outcomes (Aldrich et al. 2006; Gelpi, Reifler, and Feaver 2007; Holsti 1992; Shapiro and Page 1988). These findings are consistent with evidence from recent interventions, which points to an attentive public concerned with how American resources are deployed. At the outset of the 2011 Libya intervention, for example, 67 percent of Americans reported following American air strikes very or fairly closely,³ and while support was never high, most polls indicated that Obama had the backing of at least a plurality if not a majority of public support.⁴

By many accounts, generating American support for HIs is likely to be an uphill battle. As Drezner (2008) suggests, American political elites may be enamored with the liberal internationalism that would give rise to HIs, but the public embraces a foreign policy rooted in Hobbesian notions of self-defense, violations of state sovereignty, or restraint of an expansionist power. This view is in line with experimental findings that Americans respond strongly to scenarios such as cross-border attacks where force is required to restore national borders (Herrmann, Tetlock, and Visser 1999), which suggest the public is more concerned about security and sovereignty than with humanitarian norms. Indeed, Mueller (1994) notes that rather than being driven by a sense of moral obligation to defend foreign civilians in wartime, individuals can be “remarkably insensitive” to civilians’ fate (p. xvii), especially when those concerns conflict with their state’s narrower security objectives (Boettcher 2004, 347). From this view, public support for HIs is driven by the same instrumental mechanisms as support for other military interventions—the public will approve only when the costs are acceptable and the national interest is at stake. This perspective is compatible with what appeared to be a backlash against HIs in the 1990s, with critics noting that such interventions were at the periphery of American interests. Rather than using military force to promote US values or protect foreign civilians, opponents argued that US interventions should focus more narrowly on issues such as border violations. These critics suggest that one of the reasons a foreign policy focused on humanitarian concerns failed is that “it did not command public support” (Mandelbaum 1996, 16). The public was instead more concerned with containing “potentially dangerous members of the international community” (Mandelbaum 1996, 16-17).

A competing camp is more sanguine in terms of public attitudes toward the post–Cold War increase in HIs. That the public would support the use of force for the purpose of “saving strangers” (Wheeler 2000) is not obvious, and yet many examples point to this prospect. In Somalia, despite the scant national interest at stake, on average, a majority of the American public supported military action over the course of this intervention (Eichenberg 2005, 157). In fact, the domestic reaction to images of starvation in Somalia was so significant that it raised concerns the public could create undue pressure for intervention (Robinson 2011). Similarly, while support varied over time, the use of air power to protect safe havens in Bosnia in 1995 (Sobel 1998, 251) and air strikes in Kosovo in 1999 (Pew Research Center, US Politics and Policy 1999) also received a majority of public support. Drawing on evidence of changing norms of intervention, scholars have concluded that support for HI reflects an impulse informed by a sense of moral obligation. This moral impulse leads the public to approve of intervention even if it does not directly promote the national interest (Finnemore 2003; Finnemore and Sikkink 1998, 891-92).

The increased prevalence of HIs after the Cold War prompted lively debate about whether and why the public supports the use of force for these purposes. While one view suggests that HIs are less appealing to an audience that privileges security interests, another suggests that the public is willing to intervene on behalf of strangers. Mediating between the two perspectives forms the basis for the first empirical test (described below) that compares levels of support for HIs with interventions whose motives are not primarily humanitarian.

Mechanisms of Support

Debates over HIs also point to discrepancies about the mechanisms through which support operates.⁵ In line with the view that the public is primarily concerned with security, individuals could support HIs because action serves instrumental state interests (Goldsmith and Posner 2005, 192-205).⁶ If its dominant concern is with the national interest, the public could assess interventions by weighing potential costs with the strategic importance of action—support for HIs would then be based on the assumption that the costs of action are low or the consequences of inaction are high. Alternatively, some scholars suggest individuals support HIs because they see it as a moral obligation (Finnemore 1996; Wheeler 2000). In this case, support for interventions could be driven more by the public’s belief that action is morally right than by costs or consequences. We explore these three mechanisms—two sets of instrumental factors and the normative, moral obligation alternative—below.

Signals of Cost

Consistent with a public primarily concerned with the national interest, prominent theories of public attitudes toward intervention assert that individuals make rational calculations about the wisdom of war. Stam (1996, 59) suggests, “it is the state’s mass public that ultimately decides whether benefits justify the costs.” Gartner (2008) echoes this perspective and suggests the public will support interventions when the benefits are high or the costs (in blood and treasure) are low. In the context of HIs, this could mean that individuals support military action less because of the perceived normative benefits and more because they expect costs to be low. The prospect of HI would act as a cost signal, leading individuals to associate the operation with factors such as multilateral approval, which translate into sharing the burden of casualties and financial costs and thereby make the intervention more attractive.

Underlying this prospect of lower costs is that HIs have invariably had some form of multilateral support in terms of authorization and implementation (Finnemore 1996, 182-83). The only post–Cold War HI without UNSC authorization was the 1999 intervention in Kosovo, which received authorization from the North Atlantic Treaty Organization. HIs have also received material support from allies that limits the depth of the United States’s commitment. One example is the 1994 Haiti intervention, which was UNSC authorized and initially led by the United States. After the United States achieved its objectives in early 1995, it transferred the peacekeeping responsibilities to a robust multinational force. The result was that rather than having an open-ended commitment, the United States could quickly channel its resources elsewhere after the initial intervention (Kreps 2007).

As these examples illustrate, multilateralism has often implied burden sharing. While some suggest multilateral authorization increases support because it changes beliefs about the merits of war or confers legitimacy (Hurd 1999; Thompson 2006), others find that individuals are drawn more to the possibility of reducing the burden in blood and treasure (Jentleson and Britton 1998, 406-07; Milner and Tingley 2013), a mechanism consistent with the rational calculation argument advanced elsewhere in the literature.⁷ This cost signaling mechanism suggests that support for HIs may be driven by characteristics the public associates with such operations—multilateral approval and in turn lower expectations of financial costs and casualties—rather than any genuinely humanitarian impulse.

Strategic Consequences

The public could also support HIs for a different set of instrumental factors consistent with a realpolitik pursuit of the national interest: the strategic context in which crises take place and the negative strategic consequences of inaction. Scholars have argued that failure to respond to foreign crises influences US security interests in two ways—by creating regional spillover effects or harming the US reputation for resolve. First, inaction could perpetuate internal instability, inciting regional spillover violence (Adamson 2006), creating safe havens for terrorists (Piazza 2008), and impeding access to economic resources such as oil (Rotberg 2003). If the US public has internalized these connections, it may support HIs not for their moral desirability but based on a self-interested calculation that global instability could translate into threats to their own security.

A second aspect of the strategic consequences mechanism draws on the potential belief that inaction harms the US reputation for resolve and hampers its ability to achieve future foreign policy goals. While some empirical studies suggest reputation is not a major factor in crisis decisions (Huth 1997; Press 2005), Walter (2009, 19) finds that “governments care a lot about their reputation and it is quite rational for them to do so.” One logic for the importance of reputation, according to Dafoe, Renshon, and Huth (2014, 377), is that interactions between states can come down to “contests of expectations,” such that an actor with a reputation for following through on military threats deters other states from challenging its position. The authors (2014, 381) conclude that leaders and the public “are often concerned, even obsessed, with their status and reputation.” In the case of HIs, American leaders have often asserted that military action is imperative to stop the killing of civilians. For example, President Bush (2004) said in relation to the crisis in Darfur that “it is clear that only outside action can stop the killing…the world cannot ignore the suffering of more than one million people.” Similarly, the Obama (2013) Administration indicated it would act to end the humanitarian crisis in Syria if the regime used chemical weapons. In neither case did the United States respond in the ways anticipated by leaders’ rhetoric. If raising the prospect of humanitarian action and failing to respond creates doubt that the United States will follow through in the future, support for HIs could be attributable to concerns about harming the US reputation for resolve, thereby undermining its ability to achieve future foreign policy goals.

Moral Obligation

Alternatively, rather than instrumental calculations, support for HIs could result from a commitment to normative principles, namely, that using force to resolve humanitarian crises is appropriate and even a moral obligation. Finnemore and Sikkink (1998, 891-92) describe this category of views as prescriptive (or evaluative) norms that dictate a sense of morality or “oughtness” among people, evidenced by a shared belief that it is proper to act a certain way under a certain circumstance.⁸ The presence of a universal definition of humanity following decolonization, the development of human rights, and the post–Cold War environment combined to promote the expansion of HI norms (Finnemore 1996, 160-61), which enable and encourage states to respond to foreign humanitarian crises. Internationally, humanitarian norms challenge the boundaries of state sovereignty and create space for the legitimate practice of HI under international law (International Coalition for the Responsibility to Protect 2015). Domestically, these norms enable interventions by convincing individuals that action is “both morally permitted and morally required” when it comes to “saving strangers” (Wheeler 2000, 310). When civilians are mistreated in a conflict setting, the populace would share the belief that intervention is morally required.

In sum, support for HIs could be a function of attributes that influence individuals’ assessments of whether intervention serves the national interest—low costs or concern about strategic consequences—but could also result from a normative commitment to saving strangers. The second empirical test outlined below differentiates between the effect of instrumental and moral mechanisms and evaluates whether instrumental factors moderate support for HIs.

Moral Mechanisms of Support

Even if the public supports HIs through a sense of moral obligation, a deeper question arises about the foundations of this morality. Despite being frequently invoked in international relations (Liberman 2006; Tannenwald 2005; Tomz and Weeks 2013, 851-52), the concept of morality remains underdeveloped. One challenge with probing how individuals think about morality is that moral judgments may arise from gut instincts that defy easy articulation. The notion of “dumbfounding” suggests that individuals cannot articulate a clear reason that a choice was moral, even if they feel strongly about a particular moral judgment (Haidt, Björklund, and Murphy 2000). Asking open-ended questions that solicit “reasoning-why” responses is thus likely to be fruitless. Seeking ways around the tautology of statements such as “it’s just wrong,” psychologists have performed laboratory studies that gauge physical signs of embarrassment but even these lack clarity regarding the foundations underlying moral intuitions (Haidt 2008).

As Ryan (2014, 380) notes, “morally convicted attitudes are special because they seem to engage a distinctive mode of processing,” perhaps because they “arouse certain negative emotions, engender hostile opinions, and inspire punitive action.” Although scholars have invoked a large array of indicators to capture how individuals ground their attitudes in moral intuitions, many of these are either too limited (e.g., just punitiveness) or too extensive (e.g., thirty-six different values) to help systematically understand attitudes toward a particular foreign policy. Moral foundation theory (MFT) provides a coherent framework by categorizing moral beliefs into five foundations that “describe the types of information that we find morally relevant and the character traits or virtues we use to evaluate others’ behavior” (Clifford 2014, 700): harm/care, fairness/reciprocity, authority/respect, purity/sanctity, and in-group/loyalty (Graham et al. 2013; Haidt 2007; Kertzer et al. 2014). The first two are “individualizing foundations” that focus on the treatment of individuals, while the latter three are “binding foundations” that address community welfare.

While Kertzer et al. (2014) employ thirty indicators associated with the five foundations to understand foreign policy attitudes writ large, many of these indicators are not applicable in the context of HI. Using these foundations and associated indicators as a point of departure, the following sections develop potential individual- and community-driven reactions that individuals might have to HIs given the likely absence of a clear reasoning-why response.⁹ The subsequent analysis is not a direct test of moral foundations but instead uses these foundations as a means of understanding which aspects of HIs generate a sense of moral obligation.

Individual-level Foundations

At the individual level, MFT focuses on concern with the harm/care of others and perceptions of fairness and reciprocity (Haidt 2007). In the context of HIs, individual-level mechanisms can activate a sense of moral obligation based on concern that “harm was done to weak or vulnerable civilians” (Kertzer et al. 2014, appendix). Drawing attention to harm done to civilians is central to the strategies used by government and advocacy organizations to mobilize public support for a wide range of foreign policy and aid campaigns (Keck and Sikkink 1998, 204). Presidents also emphasize attacks on civilians in their pleas for public support, often highlighting the suffering of a single family or individual. Public support that operates through the activation of individual-level mechanisms would be consistent with these personal, emotional appeals.

The individual-level concept of fairness/reciprocity centers on the ideas of fair treatment, denial of rights, and justice. In the domestic context, fairness relates to a justly structured society (Kertzer et al. 2014, appendix). However, in the cases of war and violent repression that often precede HIs, concerns about distributive justice are superseded by threats to individuals’ lives and security. In this context, moral obligation consistent with a justice mechanism can be activated by a sense of retribution (Stein 2015) and desire to hold perpetrators of violence against civilians accountable for their actions. Concerns about justice are in line with Liberman’s (2006, 688) moral punitiveness effect in which “punitive responses to aggression significantly shape public support for war.”

Community-level Foundations

Morality in the context of HIs may also be derived from three community-level concerns: respect/authority, in-group/loyalty, and purity/sanctity. While moral foundations build on individuals’ views of their own communities, we follow Liberman (2006, 689-90) to examine how “under certain conditions, these moral responses are likely to apply to foreign states and peoples, rather than just fellow citizens.” The respect/authority foundations emphasize the importance of fulfilling set roles within society and respecting authority as a means of maintaining order. In the context of interventions, the norms embedded in international society set collective expectations that influence states’ behavior (Katzenstein 1996, 5). Moral obligation derived from the respect/authority foundation could thus be activated by the prospect that the invading country violated international norms and caused disorder within international society. This mechanism speaks not to self-defense or material considerations but to the invasion’s disruption of territorial norms (Zacher 2001) and international law regarding the treatment of civilians (International Committee of the Red Cross 2005).

The in-group/loyalty foundation is consistent with a commitment to protecting a country’s values and identity by responding to attacks on civilians. This moral commitment is often used to establish a country’s distinctiveness and, in the case of the United States, American exceptionalism. For example, this foundation could be activated by Obama’s (2011) claims that US action in Libya was necessary because “Some nations may be able to turn a blind eye to atrocities in other countries. The United States of America is different.”

Finally, moral obligation could stem from concern with the values of purity/sanctity (Haidt 2007; Kertzer et al. 2014), which would be activated when another country is seen as violating the standards of common decency accepted by civilized states. President Obama’s (2013) response to Syria’s use of chemical weapons evoked this concern, stating, “the civilized world has spent a century working to ban” chemical weapons, and their use violated “our sense of common humanity.”¹⁰

The third experiment, described below, differentiates among these morality mechanisms. This analysis allows us to determine which aspects of targeting civilians are most likely to generate public support. If moral obligation is activated by individual-level concerns, public support should be highest in cases with visible evidence of harm done to civilians or easily identifiable perpetrators who can be held accountable for human rights abuses. If morality in the context of HIs stems from the activation of community-level concerns, the public will be most receptive to interventions that help enforce international law or maintain the order of international society.

Evidence of Public Support

As the earlier sections show, the literature on HIs leaves three sets of questions unresolved. First, does the public support HIs more than interventions driven by security interests? Second, is support for HIs driven by a normative sense of moral obligation or by instrumental considerations, such as cost or strategic consequences? Third, through what mechanisms of morality does support operate? One reason these debates remain unresolved is that the evidence used to evaluate attitudes toward HIs and the sources of those attitudes is incomplete in several respects. Some scholars have pointed to leaders’ use of humanitarian justifications for intervention, which imply a common understanding with the public in terms of its views about humanitarian motives. Justifications for intervention are “literally an attempt to connect one’s actions to standards of justice” and, from that, Finnemore (1996, 159) suggests, “we can begin to piece together what those internationally held standards are.” While examining leaders’ statements brings us closer to understanding changes in the purpose of interventions, it is ultimately an indirect reflection of public opinion rather than an explicit measure of attitudes.

Although they provide a more concrete indicator, survey data are collected in the context of a specific intervention and may be susceptible to confounders such as the amount of media attention the crisis received or the number of ongoing domestic or foreign policy challenges. Comparing cases in ways that can hold all relevant attributes constant—and allow for inferences about the effect of humanitarian motives—is therefore a challenge using observational data (Druckman et al. 2011, 21). Moreover, neither observational data nor research that has sought to study differences in support experimentally (Gelpi, Feaver, and Reifler 2009) tells us why individuals might support HIs. In particular, such studies do not distinguish between moral obligation and the alternative mechanisms of cost or strategic consequences. Evidence in support of the latter mechanisms would suggest humanitarian norms are less influential than the instrumental factors with which they correlate.¹¹ In principle, one could collect individual-level data on attitudes about morality versus instrumental factors and examine the mechanisms that underlie public support. However, such data are rarely if ever collected, certainly not on the systematic basis that would be necessary for understanding causal mechanisms.

These limitations demonstrate that existing evidence is insufficient for understanding attitudes about HIs, including the reasons behind these attitudes. An experimental approach helps address these methodological challenges. Randomly assigning individuals to a control or treatment scenario allows us to draw more accurate inferences about the independent effect of the treatment—primarily humanitarian objectives—on support for intervention. Using mediation analysis (Imai et al. 2011) to estimate the effect of humanitarian objectives on each mediator and the effect of each mediator on support for intervention, we can then explicitly test the alternative instrumental and moral mechanisms.¹²

As Druckman et al. (2011, 20) suggest, “Analytically, a single random assignment makes it difficult, if not impossible, to isolate the mediating pathways of numerous intervening variables. To clarify such effects, a researcher needs to design several experiments, all with different kinds of treatments.” With this guidance in mind, we carry out a series of experiments that correspond with the three questions outlined above.

Experiment 1: Preliminary Assessment of Support

The first survey experiment examined differences in support between an HI scenario with the primary objective of protecting civilians and an intervention for more narrowly defined security objectives. It also tested the feasibility of the morality, cost, and strategic mechanisms. We recruited American adults via the Internet and randomly assigned them to baseline and treatment conditions.¹³ The baseline scenario involved a foreign leader invading another country to gain power, akin to Jentleson’s (1992, 50) foreign policy restraint (FPR) scenario in which military force restrains an adversary from aggressive action. The HI scenario emphasized the key distinguishing characteristics of HI outlined earlier, namely, that the primary objective was to prevent the suffering or death of innocent foreign civilians. Both intervention scenarios introduce the use of military force across borders without the consent of the target state and are differentiated by the predominant objective for military action—restraining an aggressive state (FPR) or protecting civilians (HI). Respondents were then asked whether they supported intervention.¹⁴

To elaborate, both scenarios first told participants that “An authoritarian leader abroad has deployed troops to a neighboring country” before varying whether the primary reason for military action was that the invaded state was unable to defend itself or its own civilians. We then told respondents, “the US president has called on the US military to intervene” either to “expel the invaders” or to protect “civilians, many of whom are women and children.” Each of these scenarios controls for the emergence of international hostilities (as opposed to internal conflict). While recent HIs have involved leaders’ violence against their own civilians, this is not a precondition for HIs, as cases such as the 1995 intervention to end Serbian attacks on Bosnian civilians illustrate. This trade-off in terms of the representativeness of recent HIs is necessary to ensure that the two conflict scenarios are the same along all dimensions other than the primary objective of military action.

To investigate the mechanisms of support, we also asked questions corresponding to the three sets of alternative explanations—morality, costs, and strategic consequences—outlined above. First, we gauged morality by asking whether the United States had a moral obligation to intervene. Second, to assess the cost signal mechanism we probed individuals’ expectations about burden sharing and multilateral authorization, the likelihood of US casualties, the cost of the operation, and the military strategy.¹⁵ Third, for the strategic consequences mechanism, we queried individuals about the potential consequences if the United States did not intervene, ranging from creating a breeding ground for terrorists to generating reputational costs.

In this initial assessment, we investigated differences in support between the baseline and the humanitarian treatment. In the baseline condition, just over a majority (52.6 percent) of the respondents supported intervention whereas in the treatment, 80 percent favored intervention, more than a 27 percent increase in support, which was statistically significant at the 1 percent level.¹⁶ These levels of support are similar to recent responses from the Chicago Council on Global Affairs survey, which found that 71 percent of respondents supported using US troops “to stop a government from committing genocide and killing large numbers of its own people” (Smeltz, Daalder, and Kafura 2014, 24). The lower levels for our baseline condition about expelling an invading state approximate the levels of support for using troops “if North Korea invaded South Korea” (47 percent). Thus, the magnitude of support reported in the treatment condition is comparable with levels of public opinion for other similar interventions, if not a bit larger. One reason may be that the sample contains many liberals who are more sensitive to the humanitarian obligation, a possibility we test more systematically in a subsequent section.

In our preliminary investigation of the mechanisms of support for HIs, the most important factor was the perception that the United States had a moral obligation to intervene. As Table 1 shows, a significantly larger percentage of respondents receiving the humanitarian treatment reported a sense of moral obligation than respondents in the FPR scenario. Compared to moral obligation, the impact of the humanitarian scenario on perceptions of costs and strategic consequences was more diminished, with only the prospect of burden-sharing registering as potentially relevant. Individuals in the treatment condition were generally more likely (p = .0503) to expect that the United States would receive multilateral assistance in carrying out the intervention. This initial analysis did not yield evidence that other instrumental factors—such as the likely military strategy, expected cost and casualties, and the strategic consequences of inaction—vary systematically between the baseline and treatment conditions.¹⁷ In sum, the preliminary analysis suggests that individuals are more likely to support scenarios that present protecting civilians as the primary objective than scenarios involving narrower security interests. Additionally, the evidence indicates that respondents support HI scenarios because they change their perceptions of the morality of intervention and, to a lesser extent, of the likelihood of burden sharing.

Table 1.

Effect of Humanitarian Intervention on Potential Mechanisms.

Mechanism	Foreign policy restraint (%)	Humanitarian intervention (%)	Effect of humanitarian intervention (%)
Norms
Moral obligation	45.5	66.3	20.8*** (10.0, 31.5)
Cost
Burden sharing	45.5	60.7	15.2** (0.2, 30.3)
High casualties	55.8	51.9	−3.9 (−15, 7.1)
Financial cost	35.7	33.3	−2.4 (−12.9, 8.2)
Strategic
Terrorism	48.6	48.1	−0.5 (−14, 12.9)
Spillover	41.9	41.5	−0.4 (−14.2, 13.3)
Economic loss	46.2	49.5	3.3 (−10.1, 16.7)
Reputation	46.1	51.7	5.6 (−7.2, 42.7)

Note: Table reports tests of proportions. Nine-five percent confidence intervals are reported in parentheses.

*p < .1; **p < .05; ***p < .01.

Experiment 2: Context and Mechanisms

Skeptics might counter that the first experiment lacks external validity (Druckman et al. 2011, 27). While the scenario reflects the rhetoric about massacres that is common in presidential speeches, real-world interventions also discuss the trade-offs associated with military action.¹⁸ As Boettcher (2004, 338) observes, “a humanitarian crisis will involve a number of value dimensions and some of those dimensions will force value trade-offs.” Taking these trade-offs into account, the second experiment investigates the possibility that, while they are not the primary mechanisms, costs and strategic consequences moderate support for HIs.

We again used FPR and HI scenarios, but this time expanded the descriptions of the primary objectives for intervention. In the FPR scenario, respondents were told that the US intervention would counter the invading state’s attempt to “gain power and resources” and was necessary because the country under attack was “unable to defend itself.” In the HI scenario, respondents were told that intervention would counter the “massacre of civilians, including innocent women and children” and was necessary because the country under attack was “suffering a humanitarian crisis.” As before, both scenarios involved the coercive use of force across borders and manipulate the character of the intervention by varying the primary policy objective. To probe the conditions under which individuals support HI and develop richer scenarios, we then introduced two contextual variables—multilateralism (help) and reputational costs—which create the experimental conditions outlined in Table 2. We included the burden-sharing aspect of multilateralism because it was the only significant instrumental mechanism in the preliminary analysis and it has been found to influence how individuals perceive the prudence of intervention (Grieco et al. 2011). Additionally, we included information about reputational costs because reputation has been linked closely with individuals’ attitudes about the use of force (Dafoe, Renshon, and Huth 2014).¹⁹ If individuals continue to offer high levels of support for HI irrespective of these costs, we can be more confident of the impact of moral obligation on support.

Table 2.

Experimental Conditions.

Baseline scenarios	Contextual variables
Baseline scenarios	Help		Reputational costs
Foreign policy restraint (1)	Low (2)	High (3)	Low (4)	High (5)
Humanitarian intervention (6)	Low (7)	High (8)	Low (9)	High (10)

Beyond these contextual factors, we again probed the potential mechanisms underlying support for intervention. The survey was administered to 1,500 American adults through Yougov, a well-established polling firm.²⁰ The full survey instrument is provided in the Online Appendix.

Results from Experiment 2

Consistent with the preliminary analysis, the HI scenario received significantly higher support than the alternative scenarios. In this second experiment, the magnitude of support for the HI scenario was more modest (73.4 percent) and support for the baseline scenario was higher (62.7 percent), but the difference remained statistically significant (p < .05). A closer look at the categorical measure of support—ranging from strongly oppose to strongly support—reveals that receiving the HI scenario decreased opposition and increased support across all categories (see Online Appendix E5). The effect of presenting humanitarian concerns as the primary objective is thus not limited to individuals who are on the fence about intervention. Additionally, the tests of proportions reported in Table 3 show that the HI scenario received higher levels of support across all values of the help and reputational cost variables. This difference is statistically significant, with the exception of the high reputational costs condition.

Table 3.

Effect of Multilateral Help and Reputational Costs on Support.

	FPR (%)	Humanitarian intervention (%)	Effect of humanitarian intervention (%)
Baseline	62.7	73.4	10.7** (0.2, 21.4)
No multilateral help	61.2	73.3.3%	12.1 (1.8, 22.4)**
Multilateral help	73.4	82.7	9.3 (0.3, 18.8)*
Effect of multilateral help	12.2 (1.8, 22.6)**	9.3 (0.17, 18.8)*
No reputational costs	66.4	75.6	9.2 (1.1, 19.5)*
Reputational costs	72.9	77.5	4.6 (−5.3, 14.5)
Effect of reputational costs	6.5 (−4.0, 16.9)	1.8 (−7.8, 11.4)

Note: Table reports results from tests of proportions. Ninety-five percent confidence intervals are shown in parentheses.

*p < .1; **p < .05; ***p < .01.

While suggestive, the comparison of proportions in Table 3 does not directly test whether help or reputational costs significantly moderate the effect of the HI scenario. We conducted formal moderation analysis (Baron and Kenny 1986) to evaluate whether there was a significant difference in the magnitude of the effect of primarily humanitarian objectives between the high and low help/reputational cost scenarios. As the logistic regression results presented in Table 4 show, the interactions between the HI scenario and help as well as between the HI scenario and reputational costs are not statistically significant. These results offer no evidence that help or reputational costs moderate the effect of receiving primarily humanitarian objectives on support.²¹

Table 4.

Moderation Analysis.

	Multilateral help	Reputational costs
Humanitarian intervention	.56** (.24)	.45* (.26)
Help	.56** (.25)	—
Reputation	—	.31 (.25)
HI × Help	−.01 (.38)	—
HI × Reputation	—	−.21 (.37)
Constant	.46*** (.17)	.68*** (.18)
Log likelihood	−351.09	−344.18
Pseudo R²	.023	.007
N	611	596

Note: Table reports results of a logistic regression model. Standard errors are in parentheses. HI = humanitarian intervention.

*p < .1; **p < .05; ***p < .01.

Having shown that support for HIs is not conditional on burden sharing or the political stakes of the operation, we return to the question of how the humanitarian scenario increased support.²² To differentiate between alternative mechanisms, we conducted mediation analysis using Hicks and Tingley’s (2011) statistical package and the procedure outlined by Imai et al. (2011). This approach uses regression models to estimate the effect of the treatment on the potential mediator and the effect of the mediator on support for intervention. These estimates are then used to compute the average casual mediation effect and the percentage of the total effect mediated by each mechanism (Imai et al. 2011, 773).²³ Consistent with the preliminary test, the mediation analysis confirmed that moral obligation mediates more than 50 percent of respondents’ support for HI. Support for the humanitarian scenarios was thus driven primarily by concerns about morality rather than by participants’ assumptions about costs or security gains from the intervention, which were not significant mediators (see Online Appendix). In the following section, we make further use of mediation analysis to unpack the story about morality that has emerged thus far.

Experiment 3: Mechanisms of Morality

Having determined that HI scenarios elicit higher levels of support through the mechanism of moral obligation, the third experiment provides a preliminary investigation of the most influential aspects of morality.²⁴ Respondents were randomized into either the FPR or humanitarian scenario exactly as presented in the second experiment.²⁵ Individuals were then asked to indicate which of the following factors were most important in their decision to support intervention: the five moral foundations discussed above (Online Appendix A), two security factors (whether their own security or that of the target state would be affected) and a cost factor (the likelihood that the United States would put troops on the ground). Based on these responses, we again conducted mediation analysis²⁶ to illustrate the effect of the treatment operating through each factor, summarized in Table 5.²⁷ This analysis gauges whether each factor is a significant mediator of support and compares how much of the total effect is transmitted through different mechanisms, highlighting their relative influence.

Table 5.

Moral Mediators and Support.

Explanation	Mechanisms of support	Indicator	Treatment effect (%)	Average causal mediation effect	Percentage of total effect mediated
Morality	Individual level	Harm to civilians	12.12***	.032 (.005, .062)	20.09
	Individual level	Perpetrators able to get away with human rights abuse	0.5	−.002 (−.024, .015)	2.1
	Community level	Disorder within international society	−1.3	−.006 (−.02, .007)	−4.4^a
		Betray values and identity	1.6	−.003 (−.02, .009)	−2.03^a
		Violate expectations of civilized country	2.6	−.001 (−.012, .01)	1.3
Cost signals	Expectation of cost	Boots on the ground	1.1	−.001 (−.005, .001)	−0.8^a
Strategic signals	Security concerns	Own security	−18.7***	−.011 (−.0002, .028)	−7.8^a
Strategic signals	Security concerns	Allies’ security	−10.3**	−.02 (−.04, −.003)	−15.3^a

Note: Treatment effect column reports the results of comparison of means tests of the effect of the humanitarian intervention treatment on the presence of each potential mechanism. Mediation analysis was conducted on each mediator individually using the procedure from Hicks and Tingley (2011). Ninety-five percent confidence intervals are in parentheses.

^aSome values are negative because the effect of the mediator acts in the opposite direction of the effect of the humanitarian treatment.

*p < .1; **p < .05; ***p < .01.

Disaggregating moral obligation into the five morality mechanisms shows that the individual-level concern with harm accounts for the largest percentage of support. Concern that harm would be done to weak or vulnerable civilians explains more than 20 percent of the heightened support for HI. The significance of this mediator is consistent with the comparison of means tests, outlined in the second column of Table 5, which shows that respondents in the humanitarian condition reported significantly higher levels of concern with harm done to civilians. These findings indicate that information about an HI heightens individuals’ concern with foreign civilians, which in turn increases their support for the intervention.

Although scholars have long focused on the effect of a country’s own casualties on support for intervention, the expression of concern for other civilians is consistent with a growing body of work that shows domestic unease with the casualties of foreign civilians (Kreps 2014; Walsh 2015). Less activated was the other individual-level indicator that captured whether perpetrators would be able to get away with violations of human rights. This finding suggests that the judicial and institutional logic common among advocacy networks (Keck and Sikkink 1998, 204) is likely to be less effective in bolstering support. Additionally, respondents were not significantly moved by community-level indicators. Contrary to conventional wisdom, the results also provide little evidence that attitudes toward HIs depend on assumptions about cost or security, as the perception that intervention would involve placing troops on the ground did not significantly mediate support. Together, these results offer suggestive evidence that public preferences are more consistent with achieving positive outcomes for others—the protection of civilians—than either punitive, negative consequences for others or even positive outcomes for one’s own country. In this sense, the public expresses attitudes in line with sincere humanitarian impulses rather than aspects of morality that could serve longer-term ulterior motives.

Partisan Identification and Public Attitudes

Until this point, the analysis has assumed the treatment effects are roughly uniform across respondents. However, a number of scholars have found that partisan identification influences attitudes about foreign policy issues, including the use of force, international law, and crisis decisions (Berinsky 2007; Brutger 2016; Grieco et al. 2011; Kriner and Shen 2014; Wallace 2013). It can also influence how individuals approach morality (Jost et al. 2003) and moral foundations (Clifford 2014, 707-8; Haidt and Graham 2007; Ryan 2014). On the other hand, some scholars have found that “partisan considerations remain in the background” (Levendusky and Horowitz 2012, 2) when individuals make foreign policy judgments. We therefore investigate whether partisanship affects the influence of the HI scenario, individuals’ sense of moral obligation, and the most salient aspects of morality.

As Table 6 illustrates, in all three experiments, regardless of the intervention scenario, Republicans exhibited high levels of support for the use of force. However, receiving the HI scenario did not significantly increase Republicans’ support, suggesting that they were not moved by the prospect of humanitarian motives but rather offered high levels of approval for all forms of intervention. In contrast, humanitarian objectives significantly increased Democrats’ and Independents’ support of intervention across all three experiments.²⁸

Table 6.

Effect of Humanitarian Intervention by Party ID.

	Percentage of respondents who support intervention
	Democrats	Republicans	Independents	Overall
Experiment 1
Humanitarian	83 (75, 91)	86 (74, 98)	67 (51, 83)	80 (74, 86)
FPR	45 (34, 56)	81 (66, 95)	46 (30, 62)	53 (45, 60)
Difference	38 (25, 51)***	5 (−13, 23)	20 (−1, 42)*	27 (17, 37)***
Experiment 2
Humanitarian	78 (73, 83)	77 (71, 83)	78 (72, 83)	76 (74, 79)
FPR	68 (62, 73)	78 (72, 84)	62 (56, 69)	68 (64, 71)
Difference	10 (3, 18)***	1 (−10, 8)	15 (7, 23)***	9 (4, 13)***
Experiment 3
Humanitarian	88 (81, 94)	81 (71, 92)	71 (56, 86)	83 (78, 88)
FPR	66 (57, 75)	68 (55, 81)	52 (37, 67)	64 (57, 70)
Difference	22 (11, 33)***	14 (−3, 30)	19 (−2, 39)*	19 (11, 28)***

Note: Table reports results from tests of proportions. Ninety-five percent confidence intervals are in parentheses. Experiment 1 N = 314, experiment 2 N = 1,500, and experiment 3 N = 408. FPR = foreign policy restraint.

*p < .1; **p < .05; ***p < .01.

The results also indicate that Republicans think differently about moral obligation, although the evidence is less clear-cut. Table 7 illustrates differences in concern with moral obligation based on party identification. As the table shows, in the first two experiments, Republicans were more likely to support intervention out of moral obligation regardless of the treatment condition. Consistent with studies pointing to conservatives being more likely to embrace “rigid morality” (Jost et al. 2003, 347), the humanitarian objectives did not significantly increase their sense of obligation. Alternatively, Democrats report significantly higher levels of moral obligation in the HI scenario. Independents were also influenced by humanitarian objectives, but their overall perception of moral obligation was consistently lower than Republicans and Democrats. This finding is in line with previous work showing that the intensity of partisan identification—irrespective of which end of the spectrum—is associated with a higher propensity for moral conviction (Ryan 2014).²⁹

Table 7.

Moral Obligation by Party ID.

	Democrats	Republicans	Independents	Overall
Percentage of respondents who agree the United States has a moral obligation to intervene
Experiment 1
Humanitarian	72 (62, 81)	77 (63, 92)	42 (25, 58)	66 (59, 74)
FPR	39 (28, 50)	68 (51, 85)	41 (26, 57)	45 (38, 53)
Difference	33 (18, 47)***	9 (−12, 31)	0 (−22, 22)	21 (10, 32)***
Experiment 2
Humanitarian	76 (71, 81)	76 (70, 82)	72 (66, 78)	73 (69, 76)
FPR	69 (64, 75)	75 (69, 82)	59 (53, 65)	66 (63, 70)
Difference	7 (−1, 14)*	1 (−8, 10)	13 (4, 21)***	6 (1, 11)***
Percentage of respondents who cited the harm foundation as most relevant
Experiment 3^a
Humanitarian	73 (64, 81)	63 (50, 76)	68 (54, 83)	69 (63, 76)
FPR	64 (55, 73)	43 (30, 57)	57 (42, 71)	57 (50, 64)
Difference	9 (−4, 21)	20 (1, 38)**	12 (−9, 32)	12 (3, 21)***

Note: Table reports results from tests of proportions. Ninety-five percent confidence intervals are in parentheses. Experiment 1: N = 314, experiment 2: N = 1,500, and experiment 3: N = 408. FPR = foreign policy restraint.

^aIn experiment 3, instead of moral obligation, respondents were asked whether their support for intervention was driven by concern about the harm being done to weak or vulnerable civilians.

*p < .1; **p < .05; ***p < .01.

Additionally, party identification influenced which moral mechanisms were activated by respondents’ sense of moral obligation. Compared to Democrats and Independents, fewer Republicans viewed the harm/care foundation as important to their support for intervention (see shaded section of Table 7), consistent with studies that show conservatives rely more heavily on authority, loyalty, and sanctity foundations compared to liberals who are more inclined to lean on care and fairness foundations (Clifford 2014, 707; Jost et al. 2003, 342). However, receiving the humanitarian treatment did have a positive and statistically significant effect on the frequency with which Republican respondents listed harm as important, whereas the treatment effects among Democrats and Independents were not statistically significant.³⁰

These preliminary findings suggest that humanitarian objectives may be particularly effective in bolstering support for military action when presidents must convince Democrats and Independents to support intervention, such as when a Republican president is in power. From the strategic perspective of leaders, HIs have the dual benefit of maintaining high support and a sense of moral obligation among Republicans while building support among Democrats and Independents.

Discussion and Conclusion

The United States has increasingly undertaken HIs in the post–Cold War environment. Such interventions are enabled in part by the prospect of multilateral authorization that came with the end of the US-Soviet rivalry (Voeten 2001), but our analysis suggests that another driving factor may reside at the domestic level. Using a series of experiments, we find that humanitarian objectives boost support for intervention, and that the basis of this support stems from the perception that protecting civilians is a moral obligation.

Our preliminary experiment demonstrates that humanitarian objectives attract high levels of support and that moral obligation explains significantly more of this support than instrumental mechanisms such as costs or strategic consequences. Our second experiment confirms heightened support for HI scenarios and determines that this support is not conditional on burden sharing or reputational costs. We do not mean to suggest that instrumental factors are not relevant to individuals’ support, but that normative factors and morality have a sizable impact on how people think about military action in general and HIs in particular. Drawing on the primacy of moral obligation as a mediator of support for HIs, the third experiment disaggregates morality into five mechanisms. We offer suggestive evidence that instead of retribution or community-level factors, individuals’ support for HIs operates most saliently through concerns about civilian harm.

The findings provide evidence across time and samples that the public supports HIs out of moral obligation. One cannot entirely eliminate the risk that unobserved confounders correlated with the treatment influence the results, but the article takes three steps to mitigate this concern. First, the experimental design controls for the presence of an interstate conflict and the US intention to take coercive action. Second, follow-up questions and experiments (see Online Appendix) provide no evidence that the treatment effect is confounded by perceptions of threat, costs, benefits, or difficulty. Third, sensitivity analysis, outlined in the Online Appendix, indicates unobserved confounders would have to explain a sizable portion of the variation in support for the effect of morality to be zero. Combined, these analyses increase our confidence in the importance of moral obligation.

This research offers several contributions. First, whereas previous studies relied on leaders’ rhetoric as indirect evidence that individuals held views consistent with humanitarian norms, the experimental treatment offers more direct exposure to these views (Finnemore 2003). Second, by differentiating among mechanisms of support, we contribute to the literature on the determinants of public opinion in the post–Cold War period. A number of scholars have found that HI can garner public support (Eichenberg 2005, 144; Jentleson and Britton 1998) implying that support is driven by factors such as prospects for success or the legitimacy that stems from multilateralism, but our research points to an underappreciated factor through which support operates: the sense of moral obligation. Third, we advance the more recent literature on moral foundations (Clifford et al. 2015; Kertzer et al. 2014; Ryan 2014), developing and investigating five moral mechanisms in the context of HIs. Fourth, our analysis of partisanship extends the literature on how party identification affects attitudes toward foreign policy (Berinsky 2007; Brutger 2016; Levendusky and Horowitz 2012). We find that Democrats are most likely to respond to humanitarian objectives and to see HI as a moral obligation.

The findings also have implications for how intervention frames can affect the relationship between the public and war. One potentially troublesome dimension of the finding that the public overwhelmingly favors HIs speaks to the potential for humanitarian pretext that some scholars have raised as a cautionary flag (Goodman 2006). To the extent that leaders can manipulate public support by referencing humanitarian crises, they may be better able to gain backing for the uses of force that are intended for instrumental national interest rather than the protection of civilians. A public motivated by a sense of moral obligation may be more willing to support otherwise risky interventions when they are presented primarily in humanitarian terms that focus on weak or vulnerable civilians, which indirectly increases the likelihood of conflict.

While we have taken a first step toward understanding the basis of support for HIs, this is not the last word. Future research should consider the limits to the persuasiveness of humanitarian claims. Are there circumstances under which the disjuncture between the humanitarian justification and the real motive is large enough that the public finds the basis for HI not to be credible? How do HIs fare relative to a larger array of military interventions? Do the populaces of other countries tend to support HIs at the same rates and for the same reasons? Additionally, our scenarios focused on the massacre of women and children, but subsequent studies should test different types of humanitarian crises to investigate whether they activate different aspects of morality.

Supplemental material

Supplemental Material, JCR_Appendix - Mechanisms of Morality: Sources of Support for Humanitarian Intervention

Supplemental Material, JCR_Appendix for Mechanisms of Morality: Sources of Support for Humanitarian Intervention by Sarah Kreps, and Sarah Maxey in Journal of Conflict Resolution

Supplemental material

Replication_Materials - Mechanisms of Morality: Sources of Support for Humanitarian Intervention

Replication_Materials for Mechanisms of Morality: Sources of Support for Humanitarian Intervention by Sarah Kreps, and Sarah Maxey in Journal of Conflict Resolution

Footnotes

Authors’ Note

Authors are listed in alphabetical order and contributed equally to the article. Replication materials will be made available on Sarah Maxey’s website by the time of publication.

Acknowledgments

We are grateful to the editors and two anonymous reviewers for their helpful comments and suggestions. The authors would also like to thank participants at Stanford’s Center for International Security and Cooperation, the University of Pennsylvania’s Experiments and International Relations workshop, Ohio State University’s Mershon Center, Cornell’s dissertation colloquium, the 2015 American Political Science Association meeting, and the 2016 International Studies Association meeting for useful feedback. We especially benefited from the comments of Scott Clifford, Josh Kertzer, Katy Powers, Tim Ryan, Kenneth Schultz, Rachel Stein, Mike Tomz, Geoffrey Wallace, Jessica Weeks, Keren Yarhi-Milo, and Amy Zegart.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors gratefully acknowledge the support of the Cornell University Institute for the Social Sciences.

Supplemental material

Supplemental material for this article is available online.

Notes

References

Adamson

Fiona

. 2006. “Crossing Borders: International Migration and National Security.” International Security 31 (1): 165–99.

Aldrich

John H.

Gelpi

Christopher

Feaver

Peter

Reifler

Jason

Sharp

Kristin Thompson

. 2006. “Foreign Policy and the Electoral Connection.” Annual Review of Political Science 9 (1): 477–502.

Baron

Reuben M.

Kenny

David A.

. 1986. “The Moderator–Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations.” Journal of Personality and Social Psychology 51 (6): 1173–82.

Berinsky

Adam

. 2007. “Assuming the Costs of War: Events, Elites, and American Public Support for Military Conflict.” Journal of Politics 69 (4): 975–97.

Berinsky

Adam

Huber

Gregory A.

Lenz

Gabriel S.

. 2012. “Evaluating Online Labor Markets for Experimental Research: Amazon.com’s Mechanical Turk.” Political Analysis 20 (3): 351–68.

Boettcher

William

. 2004. “Military Intervention Decisions Regarding Humanitarian Crises.” Journal of Conflict Resolution 48 (3): 331–55.

Brutger

Ryan

. 2016. “Domestic Politics of International Compromise.” Working Paper. Accessed April 7, 2017. http://web.sas.upenn.edu/brutger/files/2016/10/Domestic_Pol_Intl_Comp_Full_Version_10-14-16-1obr4wf.pdf.

Bush

George W

. 2004. “Statement on Violence in Darfur, Sudan.” September 9. Accessed April 7, 2017. http://georgewbush-whitehouse.archives.gov/news/releases/2004/09/text/20040909-10.html.

Clifford

Scott

. 2014. “Linking Issue Stances and Trait Inferences.” Journal of Politics 76 (3): 698–710.

10.

Clifford

Scott

Jerit

Jennifer

Rainey

Carlisle

Motyl

Matt

. 2015. “Moral Concerns and Policy Attitudes: Investigating the Influence of Elite Rhetoric.” Political Communication 32 (2): 229–48.

11.

Clinton

William J

. 1995. “The President’s Radio Address.” The American Presidency Project. June 3. Accessed April 7, 2017. http://www.presidency.ucsb.edu/ws/index.php?pid=51446.

12.

Converse

Philip E.

1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, edited by Apter

David

, 206–61. New York: Free Press.

13.

Dafoe

Allan

Renshon

Jonathan

Huth

Paul

. 2014. “Reputation and Status as Motives for War.” Annual Review of Political Science 17 (1): 371–93.

14.

Drezner

Daniel W.

2008. “The Realist Tradition in American Public Opinion.” Perspectives on Politics 6 (1): 51–70.

15.

Druckman

James

Green

Donald

Kuklinski

James

Lupia

Arthur

. 2011. Handbook of Experimental Political Science. New York: Cambridge University Press.

16.

Eichenberg

Richard

. 2005. “Victory Has Many Friends: US Public Opinion and the Use of Military Force.” International Security 30 (1): 140–77.

17.

Finnemore

Martha

. 1996. “Constructing Norms of Humanitarian Intervention.” In The Culture of National Security, edited by Katzenstein

Peter

, 153–85. New York: Columbia University Press.

18.

Finnemore

Martha

. 2003. The Purpose of Intervention. Ithaca, NY: Cornell University Press.

19.

Finnemore

Martha

Sikkink

Kathryn

. 1998. “International Norm Dynamics and Political Change.” International Organization 52 (4): 887–917.

20.

Gartner

Scott Sigmund

. 2008. “The Multiple Effects of Casualties on Public Support for War: An Experimental Approach.” American Political Science Review 102 (1): 95–106.

21.

Gelpi

Christopher

Feaver

Peter

Reifler

Jason

. 2009. Paying the Human Costs of War: American Public Opinion and Casualties in Military Conflicts. Princeton, NJ: Princeton University Press.

22.

Gelpi

Christopher

Reifler

Jason

Feaver

Peter

. 2007. “Iraq the Vote: Retrospective and Prospective Foreign Policy Judgments on Candidate Choice and Casualty Tolerance.” Political Behavior 29 (2): 151–74.

23.

Geys

Benny

. 2010. “Wars, Presidents, and Popularity: The Political Cost(s) of War Re-examined.” Public Opinion Quarterly 74 (2): 357–74.

24.

Goldsmith

Jack L.

Posner

Eric A.

. 2005. The Limits of International Law. Oxford, UK: Oxford University Press.

25.

Goodman

Ryan

. 2006. “Humanitarian Intervention and Pretexts of War.” The American Journal of International Law 100 (107): 107–41.

26.

Graham

Jesse

Haidt

Jonathan

Koleva

Sena

Motyl

Matt

Iyer

Ravi

Wojcik

Sean P.

Ditto

Peter H.

. 2013. “Moral Foundations Theory: The Pragmatic Validity of Moral Pluralism.” Advances in Experimental Social Psychology 47 (1): 55–130.

27.

Grieco

Joseph M.

Gelpi

Christopher

Reifler

Jason

Feaver

Peter

. 2011. “Let’s Get a Second Opinion: International Institutions and American Public Support for War.” International Organization 55 (2): 563–83.

28.

Haidt

Jonathan

. 2007. “The New Synthesis in Moral Psychology.” Science 316 (5827): 998–1002.

29.

Haidt

Jonathan

. 2008. “The Emotional Dog and Its Rational Tail: A Social Intuitionist Approach to Moral Judgment.” In Reasoning: Studies of Human Inference and Its Foundations, edited by Adler

Jonathan

Rips

Lance

, 1024–52. New York: Cambridge University Press.

30.

Haidt

Jonathan

Björklund

Fredrik

Murphy

Scott

. 2000. “Moral Dumbfounding: When Intuition Finds No Reason.” Unpublished Manuscript. Accessed April 7, 2017. http://faculty.virginia.edu/haidtlab/articles/manuscripts/haidt.bjorklund.working-paper.when%20intuition%20finds%20no%20reason.pub603.doc.

31.

Haidt

Jonathan

Graham

Jesse

. 2007. “When Morality Opposes Justice: Conservations Have Moral Intuitions That Liberals May Not Recognize.” Social Justice Research 20 (1): 98–116.

32.

Herrmann

Richard

Tetlock

Philip

Visser

Penny

. 1999. “Mass Public Decisions to Go to War: A Cognitive-interactionist Framework.” American Political Science Review 93 (3): 553–73.

33.

Hicks

Raymond

Tingley

Dustin

. 2011. “Causal Mediation Analysis.” Stata Journal 11 (4): 1–15.

34.

Holsti

Ole R.

1992. “Public Opinion and Foreign Policy: Challenges to the Almond-Lippmann Consensus.” International Studies Quarterly, Mershon Series: Research Programs and Debates 36 (4): 439–66.

35.

Hurd

Ian

. 1999. “Legitimacy and Authority in International Politics.” International Organization 53 (2): 379–408.

36.

Huth

Paul

. 1997. “Reputations and Deterrence: A Theoretical and Empirical Assessment.” Security Studies 7 (1): 72–99.

37.

Imai

Kosuke

Keele

Luke

Tingley

Dustin

Yamamoto

Teppei

. 2011. “Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies.” American Political Science Review 105 (4): 765–89.

38.

International Coalition for the Responsibility to Protect. 2015. “The Responsibility to Protect.” In The Responsibility to Protect. Accessed April 7, 2017. http://responsibilitytoprotect.org/.

39.

International Committee of the Red Cross. 2005. “Humanitarian Law, Human Rights and Refugee Law: Three Pillars.” ICRC Resource Centre. April 23. Accessed April 7, 2017. https://www.icrc.org/eng/resources/documents/statement/6t7g86.htm.

40.

Jentleson

Bruce

. 1992. “The Pretty Prudent Public: Post Post-Vietnam American Opinion on the Use of Military Force.” International Studies Quarterly 36 (1): 49–73.

41.

Jentleson

Bruce

Britton

Rebecca L.

. 1998. “Still Pretty Prudent: Post-cold War American Public Opinion on the Use of Military Force.” Journal of Conflict Resolution 42 (4): 395–417.

42.

Jost

John T.

Glaser

Jack

Kruglanski

Arie W.

Sulloway

Frank

. 2003. “Political Conservatism as Motivated Social Cognition.” Psychological Bulletin 129 (3): 339–75.

43.

Katzenstein

Peter

. 1996. The Culture of National Security. New York: Columbia University Press.

44.

Keck

Margaret

Sikkink

Kathryn

. 1998. Activists beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press.

45.

Kertzer

Joshua D.

Brutger

Ryan

. 2016. “Decomposing Audience Costs: Bringing the Audience Back into Audience Cost Theory.” American Journal of Political Science 60 (1): 234–49.

46.

Kertzer

Joshua D.

Powers

Kathleen E.

Rathbun

Brian C.

Iyer

Ravi

. 2014. “Moral Support: How Moral Values Shape Foreign Policy Attitudes.” Journal of Politics 76 (3): 825–40.

47.

Kreps

Sarah

. 2007. “The 1994 Haiti Intervention: A Unilateral Operation in Multilateral Clothes.” Journal of Strategic Studies 30 (3): 449–74.

48.

Kreps

Sarah

. 2014. “Flying under the Radar: A Study of Public Attitudes towards Unmanned Aerial Vehicles.” Research and Politics 1 (1): 1–7.

49.

Kriner

Douglas

Shen

Francis

. 2014. “Responding to War on Capitol Hill: Battlefield Casualties, Congressional Response, and Public Support for the War in Iraq.” American Journal of Political Science 58 (1): 157–74.

50.

Levendusky

Matthew S.

Horowitz

Michael C.

. 2012. “When Backing Down Is the Right Decision: Partisanship, New Information and Audience Costs.” Journal of Politics 74 (2): 323–38.

51.

Liberman

Peter

. 2006. “An Eye for an Eye: Public Support for War against Evildoers.” International Organization 60 (3): 687–722.

52.

Lippmann

Walter

. 1922. Public Opinion. New York: Macmillan.

53.

Mandelbaum

Michael

. 1996. “Foreign Policy as Social Work.” Foreign Affairs 75 (1): 16–32.

54.

Milner

Helen

Tingley

Dustin

. 2013. “The Choice for Multilateralism: Foreign Aid and American Foreign Policy.” Review of International Organizations 8 (3): 313–41.

55.

Mueller

John

. 1973. War, Presidents, and Public Opinion. New York: Wiley.

56.

Mueller

John

. 1994. Policy and Opinion in the Gulf War. Chicago, IL: University of Chicago Press.

57.

Obama

Barack

. 2011. “Remarks by the President in Address to the Nation on Libya.” Office of the Press Secretary. March 28. Accessed April 7, 2017. https://www.whitehouse.gov/the-press-office/2011/03/28/remarks-president-address-nation-libya.

58.

Obama

Barack

. 2013. “Remarks by the President in Address to the Nation on Syria.” Office of the Press Secretary. September 10. Accessed April 7, 2017. https://www.whitehouse.gov/the-press-office/2013/09/10/remarks-president-address-nation-syria.

59.

Parekh

Bhikhu

. 1997. “Rethinking Humanitarian Intervention.” International Political Science Review 18 (1): 49–69.

60.

Pattison

James

. 2010. Humanitarian Intervention and the Responsibility to Protect: Who Should Intervene. New York: Oxford University Press.

61.

Pew Research Center, US Politics and Policy. 1999. “Continued Public Support for Kosovo, but Worries Grow.” Pew Research Center for the People and the Press. April 21. Accessed April 7, 2017. http://www.people-press.org/1999/04/21/continued-public-support-for-kosovo-but-worries-grow/.

62.

Piazza

James

. 2008. “Incubators of Terror: Do Failed States Promote Transnational Terrorism?” International Studies Quarterly 52 (3): 469–88.

63.

Power

Samantha

. 2002. A Problem from Hell: America and the Age of Genocide. New York: Basic Books.

64.

Press

Daryl

. 2005. Calculating Credibility: How Leaders Assess Military Threats. Ithaca, NY: Cornell University Press.

65.

Robinson

Piers

. 2011. “The CNN Effect Reconsidered: Mapping a Research Agenda for the Future.” Media, War, & Conflict 4 (1): 3–11.

66.

Rotberg

Robert

. 2003. State Failure and State Weakness in a Time of Terror. New York: Brookings Institution Press.

67.

Ryan

Tim

. 2014. “Reconsidering Moral Issues in Politics.” Journal of Politics 76 (2): 380–97.

68.

Shapiro

Robert

Page

Benjamin

. 1988. “Foreign Policy and the Rational Public.” Journal of Conflict Resolution 32 (2): 211–47.

69.

Smeltz

Dina

Daalder

Ivo

Kafura

Craig

. 2014. Foreign Policy in the Age of Retrenchment: Results of the 2014 Chicago Council Survey of American Public Opinion. Chicago, IL: Chicago Council on Global Affairs.

70.

Sobel

Richard

. 1998. “Trends: United States Intervention in Bosnia.” Public Opinion Quarterly 62 (2): 250–78.

71.

Stam

Allan

. 1996. Win, Lose, or Draw: Domestic Politics and the Crucible of War. Ann Arbor: University of Michigan Press.

72.

Stein

Rachel M.

2015. “War and Revenge: Explaining Conflict Initiation by Democracies.” American Political Science Review 109 (3): 556–73.

73.

Tannenwald

Nina

. 2005. “Stigmatizing the Bomb: Origins of the Nuclear Taboo.” International Security 29 (4): 5–49.

74.

Thompson

Alexander

. 2006. “Coercion through IOs: The Security Council and the Logic of Information Transmission.” International Organization 60 (1): 1–34.

75.

Tomz

Michael

. 2007. “Domestic Audience Costs in International Relations: An Experimental Approach.” International Organization 61 (4): 821–40.

76.

Tomz

Michael

Weeks

Jessica

. 2013. “Public Opinion and the Democratic Peace.” American Political Science Review 107 (4): 849–65.

77.

Voeten

Erik

. 2001. “Outside Options and the Logic of Security Council Action.” American Political Science Review 95 (4): 845–58.

78.

Wallace

Geoffrey P. R.

2013. “International Law and Public Attitudes towards Torture: An Experimental Study.” International Organization 67 (1): 105–40.

79.

Walsh

James Igoe

. 2015. “Precision Weapons, Civilian Casualties, and Support for the Use of Force.” Political Psychology 36 (5): 507–23.

80.

Walter

Barbara

. 2009. Reputation and Civil War: Why Separatist Conflicts Are So Violent. Cambridge, UK: Cambridge University Press.

81.

Weinberg

Jill

Freese

Jeremy

McElhattan

David

. 2014. “Comparing Data Characteristics and Results of an Online Factorial Survey between a Population-based and a Crowdsource-recruited Sample.” Sociological Science 2014 (1): 292–310.

82.

Wheeler

Nicholas

. 2000. Saving Strangers: Humanitarian Intervention in International Society. Oxford, UK: Oxford University Press.

83.

Zacher

Mark W.

2001. “The Territorial Integrity Norm: International Boundaries and the Use of Force.” International Organization 55 (2): 215–50.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.24 MB

0.00 MB

0.35 MB