Secessionist social services reduce the public costs of civilian killings: Experimental evidence from the United States and the United Kingdom

Abstract

How do international audiences evaluate the legitimacy of secessionist insurgencies? Although secessionists often propagate their behavioral choices, such as state-building and non-violence, to international audiences in the hopes of generating support, scholars know little about the effects of this information. In this article, we use survey experiments in the United States and the United Kingdom to examine how international audiences respond to two commonly used strategies of secession: civilian killings and social service provision. We find that international audiences view secessionists who avoid civilian killings and provide social services as more legitimate than secessionists who kill civilians and do not provide services, respectively. Further, we show that service provision can allow secessionists to reduce—and, in some cases, eliminate—the public costs of civilian killings. These findings have important implications for ongoing secessionist conflicts across the globe.

Keywords

Civil war terrorism rebel diplomacy rebel governance

In January 2018, a commander of the Syrian Kurdish Women’s Protection Units published an op-ed in the New York Times imploring the United States to support the Kurds in their fight against Turkish military inquests (Derik, 2018). The commander appealed to the US public by noting that the Kurds had built “a haven for anyone fleeing the terror of the civil war” and developed their own own democratic institutions. She emphasized that the Kurds support refugees and have “established local councils so that all can participate in the decisions affecting their neighborhoods and communities. We hold independently monitored elections and ensure that women and all ethnic groups are strongly represented in governance.” By publishing an editorial in one of the highest circulating American newspapers, Kurdish rebels in Syria appealed directly to the American public—not just American diplomats—emphasizing their commitment to inclusive democratic institutions and refuting Turkish claims that Kurdish fighters were “terrorists.”

These state-building and terrorism-avoiding strategies—and efforts to publicize them—reflect the wartime actions of numerous contemporary and historical secessionist insurgencies (Stewart, 2018; Huang, 2016a; Fazal, 2013). Secessionists believe that engaging in and publicizing these behaviors will increase international legitimacy (Stewart, 2018; Huang, 2016a; Jo, 2015; Fazal, 2013), which may promote the secessionists’ ultimate goal of international recognition (Seymour, 2017; Heraclides, 1990). However, existing research focuses overwhelmingly on how secessionist behaviors affect the opinions of local civilians in the area of conflict (but see Huff and Kruszewska, 2016). As a result, we know little about the extent to which secessionists’ strategic behaviors affect international public opinion.

In this article, we examine how international audiences respond to information about two commonly used strategies of secession: avoiding civilian predation (Fortna, 2015; Fazal, 2013; Kalyvas, 2006) and social service provision (Stewart, 2018; Arjona, 2016; Mampilly, 2011). We focus on secessionist insurgencies because these groups are especially reliant on the international system for recognition (Huang, 2016a; Jo, 2015; Fazal, 2013; Coggins, 2011; Mampilly, 2011). Building on previous research into foreign policy attitudes, we argue that secessionists can cultivate legitimacy among international audiences by avoiding killing civilians and by state-building. Further, we argue that state-building allows secessionists to reduce the public costs of killing civilians among international audiences.

We test our expectations with survey experiments in the United States and the United Kingdom. The experiments, which resemble news media coverage of international insurgencies (e.g., Derik, 2018; Hammer, 2016; Fuller, 2012; Wax, 2012; Gray, 2007), randomize the information provided about real life secessionists’ record of violence against civilians and state-building before measuring multiple dimensions of legitimacy (Marquez, 2016). Consistent with our argument, we find that secessionists who avoid killing civilians and who provide social services are viewed as more legitimate than groups that target civilians and do not provide services. Moreover, we find that violent and non-violent behaviors interact to shape international evaluations of legitimacy: while international audiences view groups that use indiscriminate violence as less legitimate than groups that avoid violence, this negative violence effect is reduced—and, in some cases, eliminated—when services are provided and as these services become more inclusive.

These results make a number of important contributions. First, they improve our understanding of the effectiveness of historical and contemporary attempts by secessionist rebels to cultivate support by highlighting their tactical choices (see, for instance, Bob 2005).¹ Second, our findings help us better understand possible constraints on policymakers. Although citizens are unlikely to make vote choices based on overseas secessionist conflicts, research suggests that citizens do have consistent preferences about foreign policy (Kertzer and Zeitzoff, 2017), preferences that constrain and shape foreign policy decision-making (Levendusky and Horowitz, 2012; Fearon, 1994).

Our findings also have important implications for ongoing secessionist conflicts across the globe. Specifically, they suggest that the public costs of using violence against civilians are not insurmountable for secessionists. Indeed, secessionists can reduce or eliminate these costs through social service provision. Although international legitimacy does not guarantee recognition, foreign policy preferences may constrain and shape elite behavior (Levendusky and Horowitz, 2012; Fearon, 1994): a government could provide humanitarian, military, or political support to a secessionist group, or, alternatively, it could reduce support for the state the secessionist group is fighting. By elites, we mean individuals who have extensive information about foreign policy (e.g., policymakers) and are well positioned to influence the foreign policy attitudes of less informed individuals (Zaller 1992). For example, in the United States, members of the Democratic (Republican) Party increasingly sympathize with Palestinians (Israelis) in the long-standing conflict between these two states (Pew Research Center, 2018). These trends could increase pressure on elected officials to alter the extent or type of aid offered to both Israel and Palestine. While pro-Palestinian or pro-Israeli shifts in U.S. policy may not result in the formation of a new state, such shifts are profoundly consequential for Middle East policy. This article sheds light on how citizens are likely to form preferences about their government’s actions toward international secessionist conflicts.

Strategies of secession and international public opinion

Secessionist rebels must gain international recognition of sovereignty to achieve their goal of becoming a state (Bob, 2005; Fazal, 2013; Lasley and Thyne, 2015; Cunningham, 2014). To gain international recognition, some secessionist rebels attempt to cultivate legitimacy internationally (Heraclides, 1990; Fazal, 2013; Jo, 2015; Stewart, 2018; Seymour, 2017). One of the ways secessionist insurgents attempt to legitimate themselves is by engaging in certain behaviors that they believe will be looked upon favorably by the international community, while eschewing other behaviors that they believe will be looked upon unfavorably (Heraclides, 1990; Lasley and Thyne, 2015; Fortna, 2015; Jo, 2015; Huang, 2016a; Stewart, 2018; Fazal, 2018b). Two common legitimizing behaviors are complying with international expectations regarding wartime violence (Fazal, 2013; Fortna, 2015; Lasley and Thyne, 2015; Stanton, 2016) and the development of robust governing institutions (Grynkewich, 2008; Mampilly, 2011; Arjona, 2016; Stewart, 2018). While secessionist rebels also have domestic reasons for deploying these strategies (Grynkewich, 2008; Mampilly, 2011; Cammett, 2014; Arjona, 2016), these behaviors nonetheless have international repercussions that secessionist rebels actively propagate globally. We consider both of these strategies in turn.

Although civilian predation may yield immediate strategic benefits (Lapan and Sandler, 1993; Hoffman and McCormick, 2004; Kalyvas, 2008; Wood, 2010; Cohen, 2013; Fortna, 2015), violence against civilians entails substantial public costs and may undermine secessionists’ ability to cultivate international legitimacy. When rebels victimize civilians, public support for the perpetrators declines, even among people who may be likely to support the movement (Lyall, Blair and Imai, 2013). Because of these public costs, secessionists often avoid using terrorism and violence against civilians (Fazal, 2013; Stanton, 2016). Moreover, limiting civilian victimization adheres to international expectations and norms of warfare, thereby increasing the perceived legitimacy of the organization and its goals (Fazal, 2013; Lasley and Thyne, 2015; Jo, 2015; Stanton, 2016). Secessionists who refrain from using indiscriminate violence against civilians demonstrate a capacity to instill discipline and control, contributing to an overall sense of order.

Another common strategy of secession that may promote legitimacy is the establishment of governing institutions, e.g., schools, hospitals (Stewart, 2018; Arjona, 2016; Mampilly, 2011). The extent and inclusivity of rebel governance varies widely. Some secessionists offer services inclusively, to both supporters and those less likely to support the organization (Wilson, 1991, 94), while others offer services restrictively, only to supporters. When secessionists provide services inclusively, they clearly perform and mimic the role of the state (Mampilly, 2011; Stewart, 2018). For instance, the POLISARIO of the Western Sahara claims to govern their territory in order “to prove that they are ready for self-rule—a practice-run for statehood” (Organization for Statehood and Freedom, 2010). This sort of inclusive service provision may enhance secessionists’ domestic legitimacy relative to the incumbent state and could increase international support for the secessionist group’s long-term goals (Grynkewich, 2008; Mampilly, 2011; Beath, Christia and Enikolopov, 2012; Cammett, 2014).

Secessionist rebels often highlight these strategic choices—including avoiding terrorism and providing services—to international publics uninvolved in the conflict and who will never benefit directly from secessionists’ efforts.² These public relations and propaganda campaigns are frequently aimed at citizens in foreign countries who can lobby their home governments and, in some cases, send material resources to support the insurgency (Bob, 2005; Coggins, 2015), such as Kurdish rebel attempts to appeal to the US public by highlighting their inclusive democratic institutions and avoidance of terrorism. Secessionists’ diplomatic efforts now include appeals on social media platforms such as Twitter, Facebook, and YouTube (Bodine-Baron et al., 2016; Jones and Mattiacci, 2017). Secessionists also rely on global networks of advocates and supporters in foreign countries. Sudanese secessionists, for example, hired public relations firms to help the organization craft a positive public image in the United States (Bob, 2005). Co-ethnic diaspora support is also important. For instance, Canadian Tamils and the American Irish have lobbied their host countries to support independence projects in their home countries. Politicians are often highly attuned to such pressures from ethnic communities (Saideman, 1997). For Eritreans abroad, details about rebel governing institutions during the Eritrean War of Independence “were the most desired and coveted”; “by documenting every aspect of the [Eritrean People’s Liberation Front]’s state-like structure, [Eritrean activists] not only felt a sense of belonging to the nation, but also solidified their commitment to ushering it into existence” (Hepner, 2009: 85).

Despite secessionists’ best efforts to engage in and highlight these legitimacy-seeking behaviors to international publics, international recognition of sovereignty remains elusive (Seymour, 2017): even secessionists that both avoid terrorism and provide inclusive governance often fail to achieve recognition. This pattern of events seems to suggest that secessionists’ attempts to propagate their behavioral choices internationally are unsuccessful. However, because existing research has largely failed to consider how international publics respond to information about secessionist behavior, it remains unclear whether sovereignty remains elusive because (1) secessionists’ appeals to foreign publics are ineffective, or (2) because secessionists’ appeals are effective, but other factors cause recognition or support to be withheld (e.g., elite preferences or great power politics; e.g., Coggins, 2011).

Consistent with the expectations of scholars and rebels themselves, we argue that secessionists can generate international legitimacy by avoiding terrorism and providing services. International audiences are generally sensitive to civilian victimization: historical public opinion data from the United States show that citizens generally “abhor” terrorism abroad (Downes-Le Guin and Hoffman, 1993: 16) and are skeptical of foreign military interventions that could involve high levels of civilian casualties (Eichenberg, 2005: 172; Burk, 1999: 56; Gelpi, Feaver and Reifler, 2009: 256). By contrast, when secessionists limit violence against civilians, they comply with international expectations regarding norms of warfare and demonstrate discipline within their ranks (Fazal, 2013; Jo, 2015; Stanton, 2016). This compliance with international expectations should generate legitimacy for the secessionist movement; conversely, engaging in violence against civilians should reduce international legitimacy.³ We thus expect that:

H1: International audiences will view secessionist insurgencies that avoid targeting civilians as more legitimate than secessionists that target civilians.

Similarly, governance represents a clear imitation of state behavior, especially when services are provided broadly. These demonstrations of state capacity, and attempts to highlight them to foreign publics, increase the likelihood that international audiences recognize insurgents and their goals as legitimate (Stewart, 2018; Mampilly, 2011, 2015). Moreover, the more extensive the services—and thus the more state-like the rebel group—the more legitimacy secessionist-seeking rebels will enjoy. Thus, governance—and especially more inclusive governance—will enhance international legitimacy:

H2: International audiences will view secessionists as more legitimate as social service provision becomes more extensive.

Of course, these two strategies of secession are not deployed in a vacuum. Information about both violent and non-violent behaviors shapes public opinion simultaneously. International audiences often process multiple pieces of information about rebel groups, and civilian victimization or service provision alone will not determine how these audiences form legitimacy evaluations. For instance, despite nearly universal public distaste for terrorism, almost half of the American public reports that terrorists could have legitimate grievances (Downes-Le Guin and Hoffman, 1993). Thus, even when rebels do victimize civilians, negative attitudes toward violence are not insurmountable, and can be mitigated by demonstrations of state-building or compliance with international expectations. We therefore predict that:

H3: Violent and non-violent tactics will interact to shape international perceptions of insurgent legitimacy, such that the negative effect of violence on legitimacy will be attenuated as service provision becomes more extensive.

Importantly, although we argue that foreign publics do respond to information about secessionist rebel groups, we certainly do not argue that international publics’ perceptions guarantee recognition. Rather, such information shapes foreign policy preferences of international publics (Kertzer and Zeitzoff, 2017). This may in turn constrain elite actions (Fearon, 1994; Levendusky and Horowitz, 2012), shifting how elites allocate political support short of independence (such as United Nations (UN) Observer Status or sponsoring rebel leaders to the UN), or how elites allocate humanitarian and military aid to either a foreign government or the secessionist rebel group fighting that government. Furthermore, while we cannot measure how elites respond to such information in this study, recent research suggests that elites’ and publics’ foreign policy preferences frequently overlap (Renshon et al., 2016). Thus, our theoretical framework encompassing both state-building and war-making behaviors may nevertheless be instructive for understanding how other populations—elites and diasporas—develop perceptions of the legitimacy of secessionist rebel groups.

Experimental design

We tested our hypotheses with survey experiments that randomize the information provided about secessionists’ records of civilian victimization and service provision before measuring multiple indicators of perceived legitimacy (see Huff and Kruszewska, 2016 for a similar approach).⁴ This allowed us to isolate the effects of violence and service provision—both independently and jointly—on the perceived legitimacy of secessionist insurgencies.

Study 1 asked a sample of respondents in the United States (N = 605) to evaluate the Karen National Union (KNU) in Myanmar, and Study 2 asked a sample of respondents in the United Kingdom (N = 617) to evaluate the Oromo Liberation Front (OLF; Oromo Liberation Front, 2017). in Ethiopia.⁵ Both the KNU and OLF are currently engaged in conflict, both are organized around an ethnic cleavage, and both are secessionists.

We selected the KNU and OLF to balance concerns of external validity endemic to experiments, while still maximizing internal validity. These groups are real, so our treatments are more realistic samples of information respondents might encounter. Yet, these groups are also low salience, thereby minimizing the risks of pre-treatment bias (Huff and Kruszewska, 2016: 1777). Both experiments used a 3 × 2 between-subjects design.⁶ All respondents were provided with background information about the insurgency, which covered its history, geographic location, and strategic objectives. Because all respondents received this information, the analyses below consider how public opinion responds to variation in insurgent behavior holding information constant. The vignettes used in both studies are highly similar in length and style (see Online Appendix A). Respondents were randomly assigned to one of six experimental conditions, displayed in Table 1.⁷ Following our theory, we manipulated two factors hypothesized to affect perceptions of insurgent legitimacy: violence against civilians and social service provision. We discuss each experimental factor in turn.

Table 1.

Experimental design.

	No indiscriminate violence against civilians	Indiscriminate violence against civilians
No mention of social services	Condition 1	Condition 2
Restrictive social service provision	Condition 3	Condition 4
Inclusive social service provision	Condition 5	Condition 6

Violence

While there are countless possible ways to operationalize violence, we focus on perhaps the most widely employed and discussed violent tactic: killing civilians (Gelpi et al., 2009; Eichenberg, 2005; Downes-Le Guin and Hoffman, 1993). In Table 1, conditions 1, 3, and 5 read that the insurgency actively avoids harming civilians during its operations, while respondents in groups 2, 4, and 6 read that the group targets civilians.⁸ The full text of all experimental stimuli is provided in Online Appendix A.

Social service provision

There are also many ways to measure rebel social service provision. In addition to health care and education provision, rebels build judicial systems (Mampilly, 2011), engage in taxation (Revkin, 2016), diplomacy (Coggins, 2015; Huang, 2016a) and policing (Huang, 2016b). Rebels even hold elections (Huang, 2016b). We focused on social services like education and health care because these are relatively value-neutral services (in comparison to taxation or policing, for instance) that allowed us to measure variation in access. The inclusivity of insurgent social service provision varies widely: from 1945–2003, about 25% of all insurgencies that controlled territory provided inclusive services, while the remaining 75% of rebels provided no services, or limited provision to active supporters only (Stewart, 2018). We therefore manipulated not only whether the group provides services, but also the inclusivity of the group’s governance. Conditions 1 and 2 in Table 1 were not provided with any information about service provision, conditions 3 and 4 read that the group provides services restrictively (i.e., to supporters only), and conditions 5 and 6 read that the group provides services inclusively (i.e., beyond supporters).

After reading the experimental treatment, participants responded to three outcome measures that captured perceived legitimacy: the extent to which the group is a legitimate alternative to the existing government (rating scale: 1–7), the extent to which the group is capable of governing an independent state (1–5), and support for an independent state (1–7).⁹ Outcome measures were coded such that higher values correspond to more favorable evaluations (e.g., higher perceived legitimacy). Below, we present results using all three outcome measures but focus our in-text discussion on the first measure (legitimate alternative). This question asked respondents to indicate whether they agreed or disagreed with the statement “The (KNU (Study 1)/OLF (Study 2)) is a legitimate alternative to the existing government in (Myanmar (Study 1)/Ethiopia (Study 2)).” Our results remain substantively similar across all outcome measures. (Question wordings for all dependent variables are provided in Online Appendix A.)

Our experimental design reflected the range of insurgencies currently operating across the world, as there are real-world analogues for all six possible combinations of violence and service provision treatments (see Online Appendix F). Moreover, our treatments resembled the types of information often contained in news coverage of international insurgencies (e.g., Hammer, 2016; Fuller, 2012; Wax, 2012; Gray, 2007) and in direct appeals by insurgents themselves. As a result, our experiments offer insight into how citizens evaluate a wide range of real-life insurgencies (see Online Appendix F for more information).

Results

In both studies, we tested our hypotheses by estimating an ordinary least squares (OLS) regression model. The models regressed respondents’ perceptions of the legitimacy of the KNU (Study 1) or OLF (Study 2) on binary indicators for violence, restrictive service provision, and inclusive service provision. (The omitted baseline categories were no violence and no service provision, respectively.) We also included interaction terms between violence and the two levels of service provision.¹⁰ As we discuss below, our tests of Hypothesis 3 relied on the interaction between violence and service provision. Therefore, following Pepinsky (2018), we present regression results in table form, which allows readers to easily discern the statistical significance of the interaction, and also present a marginal effects plot, which illustrates the magnitude of interaction effects.¹¹

We discuss the evidence for each hypothesis (in both studies) in turn, beginning with Hypothesis 1, which posits that violence will decrease the perceived legitimacy of international secessionists. We tested this hypothesis by examining the effect of violence on legitimacy evaluations when no information about social services is provided. (Below we consider how this effect changes when information about service provision is provided alongside information about violence.) The top row of Table 2 shows strong support for Hypothesis 1 across both studies. Compared with the group that does not commit violence against civilians, the violent group is viewed as less of a legitimate alternative to the existing state in both Study 1 ( $β = - 0.64$ , $p < . 01$ ) and Study 2 ( $β = - 0.42$ , $p < . 05$ ).¹² These results suggest that committing violence against civilians has consistent negative consequences for perceived legitimacy.

Table 2.

OLS regression models predicting legitimacy evaluations (studies 1 and 2).

Coefficients	DV = Legitimate alternative (1–7)
Coefficients	Study 1(United States)	Study 2(United Kingdom)
Violent	−0.64 ***	−0.42 **
	(0.19)	(0.18)
Restrictive Services	0.31	0.41 **
	(0.20)	(0.20)
Inclusive Services	0.11	0.40 **
	(0.19)	(0.19)
Violent × Restrictive Services	0.39	−0.04
	(0.28)	(0.27)
Violent × Inclusive Services	0.67 **	0.20
	(0.27)	(0.26)
Constant	4.07 ***	3.91 ***
Constant	(0.13)	(0.13)
Observations	605	611

Note: All independent variables are binary. The dependent variable (DV) is perceived legitimacy of the KNU (Study 1) or OLF (Study 2). See Online Appendix A for question wording.

***

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

We now turn to Hypothesis 2, which suggests that perceived legitimacy will increase as service provision becomes more inclusive. Here we consider the effects of service provision among groups that are described as non-violent (we discuss the joint effects of service provision and violence below). As discussed, the simplest test of Hypothesis 2 is whether groups that provide restrictive or inclusive services are viewed as more legitimate than groups that do not provide services. As shown in the second row of Table 2, restrictive service provision has a positive effect on perceived legitimacy in both studies, though this effect is statistically significant in Study 2 ( $β = 0.41$ , $p < . 05$ ) and insignificant in Study 1 ( $β = 0.31$ , $p = . 13$ ). Similarly, the effects of inclusive service provision are positive in both studies, but again significant in Study 2 ( $β = 0.40$ , $p < . 05$ ) but not Study 1 ( $β = 0.11$ , $p = . 58$ ).

Following our discussion above, we conducted an exploratory analysis examining whether inclusive services have a larger positive effect on perceived legitimacy than restrictive services (i.e., whether $β_{3} > β_{2}$ ). In both models in Table 2, we cannot reject the null hypothesis of no difference between these service provision coefficients (Study 1: $p = 0.33$ ; Study 2: $p = 0.96$ ). Collectively, these results suggest that service provision has a consistent positive effect on perceived legitimacy; however, citizens do not appear to distinguish between groups that provide services restrictively and inclusively.

Finally, we considered our key hypothesis (H3), which posits that the negative effects of violence on legitimacy will be attenuated as service provision becomes more inclusive. We present the estimated effect of violence conditional on various levels of service provision and 95% confidence intervals (in both studies) in Figure 1. In both studies, the statistical significance of the violence/service provision interaction is indicated by the (in)significance of the relevant interaction term in row 4 or 5 of Table 2 (Pepinsky 2018).

Figure 1.

Effect of killing civilians on perceived legitimacy, by level of social service provision (studies 1 and 2).

Our tests of Hypothesis 3 in Study 1 are displayed in the left panel of Figure 1. As shown in the left-most bar in this panel, the estimated effect of violence on perceived legitimacy among groups that do not provide services is negative and highly significant (95% CI: $[- 1.00, - 0.27]$ ). By contrast, the estimated effect of violence is indistinguishable from zero among groups that provide any governance. As shown in the middle bar in Figure 1, among groups that provide restrictive services, the 95% confidence interval around the estimated violence effect is $[- 0.14, 0.64]$ . Among groups that provide inclusive services, this interval is $[- 0.45, 0.37]$ . Consistent with Hypothesis 3, these patterns indicate that the public cost of killing civilians decreases as service provision becomes more inclusive. Indeed, the significant interaction between violence and inclusive service provision in the left column of Table 2 confirms that service provision does significantly reduce the effect of violence on perceived legitimacy.

Turning to the results from Study 2, which are presented in the right panel of Figure 1, we see that the negative effect of violence on legitimacy is more consistent here than in the previous study. In particular, the negative effect of violence on legitimacy is not significantly attenuated as service provision becomes more inclusive. Consider the magnitude of the negative violence effect under each of the three levels of service provision. As shown in the left-most bar in this panel, secessionists that provide no services incur a significant penalty for committing violence against civilians (95% CI: $[- 0.78, - 0.06]$ ). As shown in the middle bar in the same panel, groups that provide restrictive services pay a similar penalty (95% CI: $[- 0.84, - 0.08]$ ). Finally, groups that provide inclusive social services appear to pay a slightly smaller penalty for killing civilians (95% CI: $[- 0.58, 0.15]$ ). Regression results in the right column of Table 2 confirm the lack of a significant interaction between violence and service provision in this study. We therefore found support for Hypothesis 3 in Study 1 (the United States) but not in Study 2 (the United Kingdom).

As a whole, Hypothesis 1 was in the predicted direction and was statistically significant across both studies. While the results of both studies were consistent with directional expectations of hypotheses 2 and 3, the statistical significance of these effects were mixed across both studies.

Conclusions

In this article, we examined how violent and non-violent strategies of secession affect the opinions of important international audiences who are frequent targets of secessionist propaganda. We focused on secessionist insurgencies because they are fundamentally reliant on the international system for political support and material resources. Using survey experiments in the United States and the United Kingdom, we demonstrated that social service provision can allow rebels to decrease—and, in some cases, eliminate—the public costs of killing civilians. Inclusive service provision—a special case of rebel governance in which services are provided to both supporters and non-supporters of the insurgency—appears especially powerful, as it allowed both groups in our studies to completely eliminate the negative effect of killing civilians on legitimacy.¹³

As discussed above, we found that the magnitude of the negative effect of violence on legitimacy was more responsive to changes in secessionist tactics in Study 1 (United States) compared with Study 2 (United Kingdom). Put differently, service provision helped a rebel group reduce the costs of violence in the eyes of American subjects (study 1), but not among subjects in the United Kingdom (study 2). Although we do not have the data necessary to explain these divergent patterns across countries, we view this as an important topic for future research. In particular, scholars should consider the contexts in which rebel groups can successfully use service provision to reduce the public costs of violence (e.g., types of services provided, type of violence employed, country-level factors, etc.).

Broadly, our findings underscore the effectiveness of service provision for shaping international public opinion (Coggins, 2015; Huang, 2016a) and improve our understanding of the incentives facing rebels when deciding among several strategic behaviors. While our results suggest that state-building and non-violence cultivate international legitimacy, the relationship between legitimacy and recognition remains mixed (Coggins, 2011, 2014). As a result, secessionists who provide services or engage in non-violence may enjoy increased legitimacy but remain unrecognized by dominant states. Secessionists thus receive mixed messages from the international system that have important implications for policymakers responding to ongoing conflicts: if secessionists recognize that state-building and non-violence do not always result in international recognition, they may abandon these behaviors, which could result in more violent civil wars with weaker governance apparatuses to mitigate the humanitarian fallout that frequently accompanies domestic conflict (Fazal, 2018a, 2018b).

Of course, the experiments reported here are not without limitations, many of which may be addressed in future research. First, although our experiments focused on two commonly used strategies of secession, rebels often use other forms of violence (e.g., rape, torture) and governance (e.g., compliance with international law, taxation, elections) that may affect international perceptions of legitimacy. Future research should consider the extent to which our findings generalize across other tactics that secessionists frequently employ. Second, certain populations that are highly invested in foreign conflicts, such as diasporas or elite policymakers, may react differently to information about secessionist tactics. Future research should assess how these populations evaluate parties in civil wars. Third, non-secessionist rebels engage in similar governance behaviors as the ones we have studied here. Future research should investigate whether the effects of social services and violence we found are similar for non-secessionist groups. Finally, building on the multi-country data presented here, we again urge future researchers to evaluate the contexts in which rebel governance affects the opinions of important international audiences.

Supplemental Material

FlynnStewart_appendix – Supplemental material for Secessionist social services reduce the public costs of civilian killings: Experimental evidence from the United States and the United Kingdom

Supplemental material, FlynnStewart_appendix for Secessionist social services reduce the public costs of civilian killings: Experimental evidence from the United States and the United Kingdom by D.J. Flynn and Megan A. Stewart in Research & Politics

Footnotes

Acknowledgements

Authors are listed alphabetically and both contributed equally.

We thank Ethan Busby, Jamie Druckman, Jin Woo Kim, Rana B. Khoury, Thomas Leeper, Julia Macdonald, Brendan Nyhan, Jason Reifler, Jake Rothschild, Jackie Schneider, Richard Shafranek, Salma Al-Shami, and participants at the 2016 meeting of the American Political Science Association and the 2017 meeting of the International Studies Association for helpful comments

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: Study 1 was supported by a Qualtrics Behavioral Research Grant.

ORCID iD

Megan A. Stewart

Supplemental materials

The supplemental files are available at .

The replication files are available at

Notes

Carnegie Corporation of New York Grant

This publication was made possible (in part) by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.

References

Arjona

(2016) Rebelocracy: Social Order in the Colombian Civil War. New York: Cambridge University Press.

Beath

Christia

Enikolopov

(2012) Winning hearts and minds through development: Evidence from a field experiment in Afghanistan. Policy Research Working Paper, World Bank. Available at https://openknowledge.worldbank.org/handle/10986/11950 (accessed 20 September 2018).

Bob

(2005) The Marketing of Rebellion: Insurgents, Media, and International Activism. New York: Cambridge University Press.

Bodine-Baron

Helmus

Magnuson

, et al. (2016) Examining ISIS support and opposition networks on Twitter. Research Report, RAND Corporation. Available at http://www.rand.org/pubs/research_reports/RR1328.html (accessed 25 February 2017).

Burk

(1999) Public support for peacekeeping in Lebanon and Somalia: Assessing the casualties hypothesis. Political Science Quarterly 114(1): 53–78.

Cammett

(2014) Compassionate Communalism: Welfare and Sectarianism in Lebanon. Ithaca, NY: Cornell University Press.

Coggins

(2011) Friends in high places: International politics and the emergence of states from secessionism. International Organization 65(3): 433–467.

Coggins

(2014) Power Politics and State Formation in the Twentieth Century: The Dynamics of Recognition. New York: Cambridge University Press.

Coggins

(2015) Rebel diplomacy: Theorizing violent non-state actors’ strategic use of talk. In: Arjona

Kasfir

Mampilly

(eds) Rebel Governance in Civil War. New York: Cambridge University Press, pp. 98–118.

10.

Cohen

(2013) Explaining rape during civil war: Cross-national evidence (1980–2009). American Political Science Review 107(3): 461–477.

11.

Cunningham

(2014) Inside the Politics of Self-Determination. New York: Oxford University Press.

12.

Derik

(2018) We fought for our democracy: Now Turkey wants to destroy it. New York Times, 29 January. Available at https://www.nytimes.com/2018/01/29/opinion/turkey-erdogan-syria-kurds.html (accessed 20 September 2018).

13.

Downes-Le Guin

Hoffman

(1993) The impact of terrorism on public opinion, 1988 to 1989. Research Report, RAND Corporation. Available at https://www.rand.org/pubs/monograph_reports/MR225.html (accessed 20 September 2018).

14.

Druckman

Leeper

(2012) Learning more from political communication experiments: Pretreatment and its effects. American Journal of Political Science 56(4): 875–896.

15.

Eichenberg

(2005) Victory has many friends: U.S. public opinion and the use of military force, 1981–2005. International Security 30(1): 140–177.

16.

Fazal

(2013) Secessionism and civilian targeting. Paper presented at the Annual Meeting of the American Political Science Association, Chicago. Available at https://papers.ssrn.com/sol3/papers2.cfm?abstract_id=2300126 (accessed 20 September 2018).

17.

Fazal

(2018a) Go your own way: Why rising separatism might lead to more conflict. Foreign Affairs. Available at https://www.foreignaffairs.com/articles/2018–06–14/go-your-own-way (accessed 20 September 2018).

18.

Fazal

(2018b) Wars of Law: Unintended Consequences in the Regulation of Armed Conflict. Ithaca, NY: Cornell University Press.

19.

Fearon

(1994) Domestic political audiences and the escalation of international disputes. American Political Science Review 88(3): 577–592.

20.

Fortna

(2015).Do terrorists win? Rebels’ use of terrorism and civil war outcomes. International Organization 69(3): 519–556.

21.

Fuller

(2012) In Myanmar, Karen rebels deny signing a cease-fire. New York Times, 3 February. Available at http://www.nytimes.com/2012/02/04/world/asia/inmyanmar-karen-rebels-deny-signing-a-cease-fire.html (accessed 25 February 2017).

22.

Gaines

Kuklinski

Quirk

(2007) The logic of the survey experiment reexamined. Political Analysis 15(1): 1–20.

23.

Gelpi

Feaver

Reifler

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

24.

Goren

Schoen

Reifler

, et al. (2016) A unified theory of value-based reasoning and U.S. public opinion. Political Behavior 38(4): 977–997.

25.

Gray

(2007) Myanmar military accused of raping women. Washington Post, 12 February. Available at http://www.washingtonpost.com/wpdyn/content/article/2007/02/12/AR2007021200260.html (accessed 25 February 2017).

26.

Grynkewich

(2008) Welfare as warfare: How violent non-state groups use social services to attack the state. Studies in Conflict & Terrorism 31(4): 350–370.

27.

Hammer

(2016) Once a Bucknell professor, now the commander of an Ethiopian rebel army. New York Times, 31 August. Available at https://www.nytimes.com/2016/09/04/magazine/once-a-bucknell-professornow-the-commander-of-an-ethiopian-rebel-army.html (accessed 25 February 2017).

28.

Hepner

TMR

(2009) Soldiers, Martyrs, Traitors, and Exiles: Political Conflict in Eritrea and the Diaspora. Philadelphia: University of Pennsylvania Press.

29.

Heraclides

(1990) Secessionist minorities and external involvement. International Organization 44(3): 341–378.

30.

Herrmann

Fetlock

Visser

(1999) Mass public decisions to go to war: A cognitive-interactionist framework. American Political Science Review 93(3): 553–573.

31.

Hoffman

McCormick

(2004) Terrorism, signaling, and suicide attack. Studies in Conflict & Terrorism 27(4): 243–281.

32.

Huang

(2016a) Rebel diplomacy in civil war. International Security 40(4): 89–126.

33.

Huang

(2016b) The Wartime Origins of Democratization: Civil War, Rebel Governance, and Political Regimes. New York: Cambridge University Press.

34.

Huff

Kruszewska

(2016) Banners, barricades, and bombs: The tactical choices of social movements and public opinion. Comparative Political Studies 49(13): 1774–1808.

35.

(2015) Compliant Rebels. New York: Cambridge University Press.

36.

Jones

Mattiacci

(2017) A manifesto, in 140 characters or fewer: Social media as a tool of rebel diplomacy. British Journal of Political Science: 1–23.

37.

Kalyvas

(2006) The Logic of Violence in Civil War. New York: Cambridge University Press.

38.

Kalyvas

(2008) Promises and pitfalls of an emerging research program: The microdynamics of civil war. In: Kalyvas

Shapiro

Masoud

(eds) The Microdynamics of Civil War. New York: Cambridge University Press, pp. 397–421.

39.

Kam

Trussler

(2017) At the nexus of observational and experimental research: Theory, specification, and analysis of experiments with heterogeneous treatment effects. Political Behavior 39(4): 789–815.

40.

Kertzer

Powers

Rathbun

, et al. (2014) Moral support: How moral values shape foreign policy attitudes. Journal of Politics 76(3): 825–840.

41.

Kertzer

Zeitzoff

(2017) A bottom-up theory of public opinion about foreign policy. American Journal of Political Science 61(3): 543–558.

42.

Lapan

Sandler

(1993) Terrorism and signaling. European Journal of Political Economy 9(3): 383–397.

43.

Lasley

Thyne

(2015) Secession, legitimacy and the use of child soldiers. Conflict Management and Peace Science 32(3): 298–308.

44.

Levendusky

Horowitz

(2012) When backing down is the right decision: Partisanship, new information, and audience costs. Journal of Politics 74(2): 323–338.

45.

Lyall

Blair

Imai

(2013) Explaining support for combatants during wartime: A survey experiment in Afghanistan. American Political Science Review 107(4): 679–705.

46.

Mampilly

(2011) Rebel Rulers: Insurgent Governance and Civilian Life During War. Ithaca, NY: Cornell University Press.

47.

Mampilly

(2015) Performing the nation state: Rebel governance and symbolic processes. In: Arjona

Kasfir

Mampilly

(eds) Rebel Governance in Civil War. New York: Cambridge University Press, pp. 74–97.

48.

Marquez

(2016) The irrelevance of legitimacy. Political Studies 64(1): 19–34.

49.

Organization for Statehood and Freedom (2010) The Sahrawi struggle. Available at http://www.statehoodandfreedom.org/en/the-saharawi-stuggle (accessed 20 September 2018).

50.

Oromo Liberation Front (2017) OLF policies. Available at http://oromoliberationfront.org/en/olf-policies/ (accessed 26 February 2017).

51.

Page

Shapiro

(1992) The Rational Public: Fifty Years of Trends in Americans’ Policy Preferences. Chicago: University of Chicago Press.

52.

Page

Xie

(2010) Living with the Dragon: How the American Public Views the Rise of China. Chicago: University of Chicago Press.

53.

Palan

Schitter

(2018) Prolific.ac — a subject pool for online experiments. Journal of Behavioral and Experimental Finance 17: 22–27.

54.

Peer

Brandimarte

Samat

, et al. (2017). Beyond the Turk: Alternative platforms for crowdsourcing behavioral research. Journal of Experimental Social Psychology 70: 153–163.

55.

Pepinsky

(2018) Visual heuristics for marginal effects plots. Research & Politics 5(1). doi: 10.1177/2053168018756668.

56.

Pew Research Center (2018) Republicans and Democrats grow even further apart in views of Israel, Palestinians. Report, Pew Research Center: U.S. Politics and Policy. Available at http://www.people-press.org/2018/01/23/republicans-and-democratsgrow-even-further-apart-in-views-of-israel-palestinians/ (accessed 20 September 2018).

57.

Powell

(2011) Framing Islam: An analysis of U.S. media coverage of terrorism since 9/11. Communication Studies 62(1): 90–112.

58.

Rathbun

Kertzer

Reifler

, et al. (2016) Taking foreign policy personally: Personal values and foreign policy attitudes. International Studies Quarterly 60(1): 124–137.

59.

Reifler

Scotto

Clarke

(2011) Foreign policy beliefs in contemporary Britain: Structure and relevance. International Studies Quarterly 55(1): 245–266.

60.

Renshon

Yarhi-Milo

Kertzer

(2016) Democratic leaders, crises and war: Paired experiments on the Israeli Knesset and public. Unpublished manuscript. Available at https://people.fas.harvard.edu/~jkertzer/Research_files/DemocraciesWarCriseswebsite. pdf (accessed 20 September 2018).

61.

Revkin

(2016) The non-economic functions of insurgent taxation: Evidence from the Islamic State in Syria. Unpublished paper. Available at https://digitalcommons.law.yale.edu/ylsspps_papers/122/ (accessed 20 September 2018).

62.

Saideman

(1997) Explaining the international relations of secessionist conflicts: Vulnerability versus ethnic ties. International Organization 51(4): 721–753.

63.

Seymour

(2017) Legitimacy and the politics of recognition in Kosovo. Small Wars & Insurgencies 28(4–5): 817–838.

64.

Stanton

(2016) Violence and Restraint in Civil War: Civilian Targeting in the Shadow of International Law. New York: Cambridge University Press.

65.

Stewart

(2018) Civil war as state-making: Strategic governance in civil war. International Organization 72(1): 205–226.

66.

Wax

(2012) As separatists in Ethiopia disarm, a new chapter for D.C.’s Oromo community. Washington Post, 1 April. Available at https://www.washingtonpost.com/lifestyle/style/asseparatists-in-ethiopia-disarm-a-new-chapter-for-dcs-oromocommunity/2012/04/01/gIQArnUtpS_story.html?utm_term=.b98fb584719a (accessed 20 September 2018).

67.

Wilson

(1991) The Challenge Road: Women and the Eritrean Revolution. London: The Red Sea Press.

68.

Wood

(2010) Rebel capability and strategic violence against civilians. Journal of Peace Research 47(5): 601–614.

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.

0.00 MB

0.50 MB