Hostname: page-component-cd9895bd7-dk4vv Total loading time: 0 Render date: 2024-12-27T13:40:35.481Z Has data issue: false hasContentIssue false
  • English
  • Français

Is there a religious bias? Attitudes towards military humanitarian intervention in Germany

Published online by Cambridge University Press:  02 April 2024

Benjamin Daßler Open the ORCID record for Benjamin Daßler [Opens in a new window] ,
Bernhard Zangl  and
Hilde van Meegdenburg
Benjamin Daßler*
Affiliation:
Geschwister-Scholl-Institute of Political Science, LMU Munich, Munich, Germany
Bernhard Zangl
Affiliation:
Geschwister-Scholl-Institute of Political Science, LMU Munich, Munich, Germany
Hilde van Meegdenburg
Affiliation:
Institute of Political Science, Leiden University, Leiden, The Netherlands
*
Corresponding author: Benjamin Daßler; Email: [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Is individual support for internationally agreed military humanitarian interventions (MHIs) subject to a religious bias? Conducting a vignette-based survey experiment, the paper provides micro-level evidence for such a bias within a highly unlikely sample: German university students. Participants in our survey experiment were more compassionate and indeed more supportive of an MHI when the victims of war-related violence were Christians rather than Muslims. The paper thus contributes to the literature on support for MHIs in two important ways: first, whereas the existing literature has a strong focus on the United States, this paper studies individuals’ support in another Western country that regularly contributes to MHIs, namely Germany. Second, while the existing literature has mainly examined how other social factors, such as the race or gender of the victims, affect individuals’ support for MHIs, drawing on social identity theory, this paper claims that religious identification also has an impact. Moreover, by showing that the religion of the victims of war-related violence shapes individuals’ attitudes towards MHIs through compassion, the paper also speaks to more recent literature that demonstrates that individuals’ attitudes towards refugees depend on – among other things – their religion. Against the background of a general rise of identitarian politics in many Western societies, our findings seem to be of particular relevance.

Keywords

military humanitarian interventionpublic opinionreligious identitysocial identity theory
Type
Research Article
Information
European Journal of International Security , Volume 9 , Issue 3 , August 2024 , pp. 434 - 448
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of The British International Studies Association.

Introduction

Western countries’ military humanitarian interventions (MHIs) are highly selective.Footnote 1 Western countries have found themselves accused of an interventionism ‘à la carte’ that has its root not only, but also in a religious bias within Western societies. For example, comparing the terrorist attacks in Paris in January 2015 and the killing of 2,000 people in Nigeria by Boko Haram, Mohamed ElBaradei – the former head of the International Atomic Energy Agency (IAEA) – stated in an interview: ‘The world goes crazy when French people are attacked. But when an ordinary Nigerian is killed, we don’t care at all.’Footnote 2 Western populations are allegedly biased. Among other things, they care more when Christians are the victims of war-related violence than when, for instance, Muslims are suffering from comparable war-related violence, and they are thus more likely to support MHIs to reduce the suffering of Christians than when victims are from another religion.Footnote 3

We define an MHI as the use of military force by a state (or group of states) with the intent of ending severe human suffering in another state which has not given permission for the intervention. MHIs are thus military interventions in another state that are justified as necessary to reduce human suffering in the target country and where this justification finds (some) acceptance in the general public of the intervening state(s).Footnote 4 For instance, if their humanitarian justification finds acceptance, peacekeeping and peace-enforcement operations, no matter whether authorised by the United Nations (UN) or not, can be regarded as MHIs. Examples of MHIs include those in Somalia, Bosnia, and Kosovo in the 1990s, as well as those in Libya and Mali in the 2000s. While some MHIs find widespread support in Western societies, others find little support.Footnote 5 Germany is no exception. While Germans are generally considered particularly reluctant to support MHIs,Footnote 6 Germany has, throughout the last decades, obviously abandoned its non-interventionist policy doctrine and participated in various MHIs, from Kosovo in the late 1990s to Mali in the early 2020s. However, public support for German participation varies from one MHI to the other.Footnote 7 Whereas the German participation in the United Nations Mission in South Sudan found support only among 30 per cent of the German population in 2022,Footnote 8 German participation in the International Security Assistance Force (ISAF – ‘Operation Resolute Support’, since 2014) in Afghanistan found support among 36 per cent of the German population in 2020, and in the same year Germany’s continued support for the Kosovo Force (KFOR) was approved by as many as 44 per cent of the German population.Footnote 9 The literature on MHIs – and the related responsibility to protect – has identified a variety of conditions shaping Western countries’ participation in military interventions that aim to help the victims of war-related violence.

A rationalist literature strand highlights that Western democracies will only engage in MHIs when their national self-interest is at stake.Footnote 10 Those countries that contribute to MHIs are more likely to do so either when war-related human suffering has adverse effects on their own countries that need to be alleviated or when an intervention has positive effects on their countries. It highlights, for instance, that the prospect of refugee flows into their countries can drive Western countries’ readiness to contribute to an MHI.Footnote 11 They also indicate that the availability of primary goods, such as oil in the target country, might prompt Western democracies to engage in humanitarian interventions.Footnote 12 A related literature strand also argues that the power of the target country has a decisive impact on whether Western democracies are willing to intervene in order to reduce war-related human suffering.Footnote 13

A constructivist literature strand argues, by contrast, that a norm of humanitarian intervention has emerged since the Cold War,Footnote 14 and that a sense of moral duty or obligation explains why Western democracies are willing to contribute to international MHIs.Footnote 15 Many contributions to this literature claim that, due to war-related human suffering, the governments of Western democracies come under ‘morally motivated pressure’ from their populations to help the victims.Footnote 16 This pressure is assumed to vary according to the severity of the war-related human suffering, thus explaining why Western democracies are more likely to engage in MHIs when the suffering is particularly severe. Moreover, a related literature strand has discussed whether the morally motivated pressure to act varies with the attention the media devotes to the war-related human suffering.Footnote 17

While these literature strands have greatly improved our understanding of the selectivity of Western countries’ MHI, they do not address whether Western MHIs are actually biased.Footnote 18 This is, however, the focus of a more recent literature that goes beyond the macro-level theories of the above literature strands, theorising the micro-level foundations of individuals’ support for humanitarian interventions. As this literature has shown, studying the micro-level of MHIs is crucial, as parliamentarians as well as government officials tend to orientate their decisions about military interventions strongly towards public opinion.Footnote 19 Within this growing micro-level literature on MHIs, a constructivist-leaning strand highlights the conditions that shape people’s identification with the victims of war-related violence and how this translates into their support for MHIs.Footnote 20 This literature focuses particularly on how social markers such as the victims’ gender or race impact individuals’ support of MHIs.Footnote 21 In this paper, we contribute to this literature on the effect of social co-identification on individuals’ support for MHIs in two important ways: (1) Our first contribution is theoretical. While the existing literature has its focus on how the victims’ race – and sometimes also their gender – shapes individuals’ support for MHIs,Footnote 22 we study how the victims’ religion impacts on individuals’ support for MHIs, thereby producing the alleged religious bias of Western MHIs. (2) Our second contribution is empirical. Whereas the existing literature has its focus on individuals’ support of MHIs in the United States, we study the conditions for individuals’ support in Germany, a country whose contribution to internationally agreed MHIs is on the rise. Moreover, by theorising and demonstrating how the religion of the victims of war-related violence shapes Germans’ support for MHI through a compassion mechanism, we also speak to the recent literature on how the religion of refugees shapes whether recipient countries’ citizens develop compassionate feelings for them and thus support the admission to their country.Footnote 23 We contribute to this literature by showing that religion is not only relevant for individuals’ support of refugees (i.e. ‘strangers in need that came to us’), but also for their support of MHIs (and thus ‘strangers in need that are abroad’). Against the background of a general rise of identitarian politics in many Western societies, these findings seem to be of particular relevance.Footnote 24

The ambition of this article is to discuss theoretically and assess empirically whether the religion of the victims of war-related violence shapes individuals’ attitudes in Germany for the countries’ contribution to MHIs. Are individual attitudes for MHIs in Germany subject to a religious bias? The article is structured as follows. In the following section, we draw on social identity theory (and related socio-psychological research) to argue that individuals’ support for MHIs is shaped, inter alia, by the religious identification with the victims of war-related violence. Our religious identification hypothesis suggests that religious identity cues generally increase people’s identification with the victims of war-related violence, thereby driving not only their compassionate feelings, but also their individual support for a contribution of their country to an MHI. In the third section, we present the design of the survey experiment that we conducted with 449 German students to assess the effect of the religion of the victims of war-related violence on participants’ support for a German contribution to an MHI in a hypothetical conflict. In the fourth section, we show that support for such a German contribution is higher when the victims are said to be Christians rather than Muslims and that this greater support is partially explained by participants’ compassionate feelings. In the concluding section, we summarise our findings and highlight avenues for further research.

Theory: The religious identification hypothesis

Drawing on social identity theory,Footnote 25 we hold that individuals are more prone to support their countries’ contribution to MHIs when they share their religious identity with the victims of war-related violence.Footnote 26 We claim that religion is an important cue which, amongst many other social factors, shapes the degree of individuals’ identification with the victims of war-related violence.Footnote 27 The stronger the religious co-identification with the victims, the stronger are their compassionate feelings and the more they can be expected to support their countries’ participation in an MHI to reduce the victims’ suffering.

Following socio-psychological theories,Footnote 28 we argue that people with joint group memberships typically identify with each other and thus feel close to each other, while people who have no joint group memberships tend to see each other as remote and typically do not identify to the same degree with each other. If people share the same culture, religion, gender, race, class, profession, club, or family, they can relate to each other more easily, which in turn helps them to identify with each other. In other words, shared membership in social groups creates social bonds among people who can identify with each other.

Joint group memberships and the social identification that comes with them also shape, according to social identity theory, how individuals evaluate events that affect people that belong to the same group. The unexpected death of a family member is evaluated differently from the unexpected death of a celebrity one has hardly any joint group memberships with. As the example illustrates, even when an event has no ‘impact on the individual directly’,Footnote 29 it can unleash strong emotional reactions when the event affects people that belong to the same social group – i.e. when it affects people to whom they are socially proximate. The emotional reaction is much weaker – or even nullified – when the same event affects people that do not belong to the same social group.Footnote 30 The more people share common group memberships, the more likely it is that events that affect the ‘other’ will also affect the ‘self’, leading to emotional responses such as sadness for bad events or happiness for good events. Parkinson, Fischer, and Manstead explain this socio-emotional mechanism as follows:

We are all members of certain kinds of groups or collectivities, ranging from families, sports teams, and clubs to broader social categories relating to class, gender, social status, and so on. As general rule, what happens to members of our own groups also affects us personally.Footnote 31

In other words, social identity theory leads us to expect that joint memberships in social groups create not only the identification of group members with each other, but also emotional ties and thus empathy among the members. Therefore, any suffering of group members evokes compassionate feelings among the rest of the group, driving their readiness to provide help. Thus, the shared membership in social groups reinforces feelings of compassion for people suffering, whereas the lack of such joint memberships typically undercuts feelings of compassion with those who suffer from harm. Therefore, social identity theory leads us to expect that people that identify with each other due to their joint memberships in social groups are more willing to help each other than people that lack the emotional ties stimulated by joint group memberships.

However, as social identity theory also shows, not all social groups are equally relevant for how people identify with each other and thus whether they feel obliged to help each other if need be. Some social groups are more important for people’s identities and thus produce much stronger ties among their members than others.Footnote 32 Members of the same family typically identify with each other much more than members of the same soccer club. By the same token, members of the same nation, religion, church, ethnicity, or cult tend to have stronger ties and thus tend to identify much more with each other than people belonging to the same gender, class, profession, or gym. In general, the identity of human beings is much more shaped by their membership in cultural-religious groups than their belonging to the same social-structural groups. While the former can be seen as an expression of their members’ world views, moral attitudes, values, norms, and beliefs, the latter typically comes with common interests, preferences, concerns, or tastes.Footnote 33

As the identification of people with each other tends to be stronger in cultural-religious groups than in social-structural groups, we expect that individuals in Western democracies who are confronted with massive war-related human suffering in another country are more prepared to support a prospective MHI when the victims are Christians rather than from another religion and thus share with them this religious background. To be sure, to allow for the identification with the victims of war-related violence based on their shared religion, we do not consider it necessary that people actually practise their religion. We rather assume it to be sufficient when they share the same religious background and thus the same traditions and values. After all, recent research has shown that despite their secularisation, their religious background remains of utmost importance for people’s social identification in Western societies.Footnote 34 Therefore, we argue that people can already identify with the victims of war-related violence through their shared religious background, which drives their compassion towards them and their willingness to support MHIs to end the victims’ suffering. This identification driven by joint religious background with the victims can lead to compassionate feelings that can then be expressed through people supporting their countries’ contribution to MHIs.

Suggesting that a joint religious background with the victims of war-related violence helps people from the West to support their country’s contribution to MHIs does not imply that we consider Western societies to be homogeneous regarding religion. There are religious differences across Western countries, as there are religious differences within Western countries, and these differences should also impact on people’s support for MHIs to help the victims of war-related violence. Moreover, whether religious identities are shared is a matter of interpretation, an interpretation that is socially constructed in public discourse. In fact, religious identity is ‘imagined’ and can thus be shaped by how victims of war are framed in public discourse.Footnote 35 In public discourse, actors such as the government, the opposition, journalists, or humanitarian groups may highlight religious communalities, thereby creating social proximity and thus promoting people’s identification with the victims, or they may hint at religious differences, thereby creating social distance and thus preventing people’s identification with the victims. They may even use religious communalities or differences to justify their respective positions with regard to an MHI.Footnote 36 However, if they do, this only underscores our assumption that religious distance or proximity shapes peoples’ support for or against an MHI. Like any other narrative or discourse, they must be credible to be accepted by the target audience.Footnote 37

Research design: Survey experiment

To offer a first assessment of our religious identification hypothesis and the related socio-emotional mechanism, we conducted a vignette-based survey experiment with 449 German students from the authors’ university. In survey experiments using vignettes, researchers invent short stories to which participants are invited to respond.Footnote 38 As with all survey experiments, vignette-based survey experiments allow for the manipulation of the variable of interest ‘not easily manipulated in the real world’, while controlling for confounding variables.Footnote 39 They thus enable us to keep the social setting of war-related human suffering constant, while at the same time manipulating – through the hypothetical situation of a vignette-based treatment – the victims’ religion. Thereby, vignettes allow us to discern the impact of the victims’ religion on participants’ support for their country’s contribution to an MHI.

In our experiment, we confronted participants with vignettes that looked like real newspaper articles in the Neue Zürcher Zeitung, a Swiss-German newspaper, on war-related human suffering in Benin, West Africa (for the German vignettes used in the experiment, as well as an English translation, see Online Appendix A). The newspaper articles described exactly the same rapidly deteriorating humanitarian catastrophe with clashes, raids, torture, displacements, and unlawful killings involving government forces. However, the victims of war portrayed in the newspaper articles differed: one vignette portrays Christians as the main victims, while the other vignette depicts Muslims as suffering from the assaults of government forces. As participants in our sample are all from a Christian background, the ‘Muslim vignette’ recipients are considered the baseline group while the recipients of the ‘Christian vignette’ form the treatment group of the experiment.

As indicated above, the participants of our experiments were German students from the authors’ university attending selected classes in the social sciences or humanities.Footnote 40 Before the experiment, participants were comprehensively informed about the study, the authors, the procedure, the handling of their data, and their rights.Footnote 41 In each of the selected classes, we equally distributed the two vignettes. Participants were told that the survey was conducted in the context of a study that was interested in individuals’ attitudes towards MHI. Participants were asked to read the distributed fictitious ‘newspaper article’ on war-related violence in Benin and to answer subsequently the attached questionnaire asking them – among other things – about their support for a German military contribution to an UN-led humanitarian intervention. We asked participants for their support of an HMI formally backed by the UN, because we assumed that otherwise many respondents would disagree with the intervention – regardless of their identification with the victims of war – simply on the grounds that it was not authorised by the UN. After all, the German constitution is very restrictive with regard to German contributions to military interventions that are not authorised by the UN, and many Germans consider UN backing as absolutely essential for any German participation in such an intervention.Footnote 42 More specifically, the question of interest was: To what extent do you agree with the following statement? Germany should support an intervention by the UN in the conflict in Benin (by dispatching around 500 soldiers). Participants could select their extent of agreement on a four-point scale: (1) strongly agree; (2) somewhat agree; (3) somewhat disagree; (4) strongly disagree.Footnote 43

To isolate the religious identification effect, we removed all participants from our sample who did not indicate that they also had a Christian background.Footnote 44 Within the resulting cleaned sample, recipients of the different vignettes do not differ significantly with regard to fundamental social demographics such as gender, age, or party identification (see Table 1). In our sample, we thus control automatically for these potentially confounding variables; this facilitates the isolation of the effect our treatment – i.e. joint religion with the victims of war – has on participants’ support for MHI in our sample.

Table 1. Participant socio-demographics.

However, due to the social demographics of our participants, inferences beyond our sample of German students may seem to be difficult, and hence, the external validity limited.Footnote 45 This has two dimensions. First, wider inferences beyond Germany appear difficult because when it comes to MHIs, Germany does not seem representative of Western democracies. Second, our sample is based on students who are not representative of German society. Although we do not seek bold generalisations and acknowledge these considerations, we do believe they are mitigated by several factors.

In relation to the first, it is true that, compared to citizens from other Western countries, Germans were for a long time, on average, more sceptical towards military interventions in general and MHIs more specifically. This scepticism, which travelled beyond German unification, contributed to the German reluctance to participate in the MHIs of the early 1990s. However, over the course of the 1990s this German exceptionalism subsided.Footnote 46 Beginning with its contribution to the 1999 Kosovo intervention, Germany became a frequent participant in Western MHIs. As Stengel recently argued, the taboo in question eroded, and military interventions, once unthinkable, are now ‘considered a self-evident requirement of a post–Cold War world’ in German foreign policy circles.Footnote 47 Moreover, we also see that Germans’ attitudes towards MHIs have been ‘normalized’.Footnote 48 For example, despite the German government’s conspicuous abstention in the respective UN vote, only 39 per cent of the Germans opposed the NATO-led 2011 humanitarian intervention in Libya, a percentage that is much lower than, for instance, in Italy (49 per cent) and comparable to the United States (37 per cent).Footnote 49 Thus, with regard to MHIs, today’s Germany may well be considered a typical Western democracy.

Second, regarding the student-based sample, we acknowledge that our sample differs from the average German population in terms of age, gender, and level of education (see Table 1 for participant’s socio-demographics, and Online Appendix C6 for benchmark statistics comparing the sample with the average German population). Furthermore, our sample also differs from the German population in terms of party identification. Of the participants, 40.6% said they identified with the Green Party, which only scored 8.9% in Germany’s 2017 general election.Footnote 50 However, this leaves us with a sample in which the religion of the victims of war-related violence is ‘least likely’ to make a difference for participants’ support of an MHI. The sample thus works against our expectations: after all, younger people, women, the better educated, and people who vote for Green parties have, on average, a more secular worldview, according to which their support for MHIs is less likely to be influenced by the religion of the victims of war-related violence.Footnote 51 Thus, if the victims’ religion makes a difference within our sample, it should make a difference in the broader German population as well.

Empirical analysis: The religious identification effect

Our vignette-based experiment supports our theoretical expectations that the victims’ religion matters for the support of MHIs.

To begin with, participants in our experiment are generally supportive of a German military contribution to an UN-led humanitarian intervention to reduce the (hypothetical) war-related human suffering in Benin. In our survey experiment, 57.2% of the respondents ‘strongly agreed’ or ‘somewhat agreed’ to such a contribution. However, as expected, the baseline group of those participants who received the vignette with Christian victims agree more to an MHI than the treatment group of participants who received the Muslim vignette. While 10.8% and 49.8% of participants within the baseline group indicate that they ‘strongly agree’ or ‘somewhat agree’ with a German contribution to an MHI, these percentages drop to 8.4% and 45.6% within the treatment group (Figure 1). Thus, the overall agreement to an intervention drops by 6.6 percentage points. A standard T-Test indicates that the differences are significant at the 90% level (see Online Appendix C1).

Figure 1. Agreement to support a German military intervention.

To assess our religious identification hypothesis, we further estimated the effect strength of our religious treatment on participants’ support for military interventions by calculating ordinal logistic regression models using odds ratios (see Table 2).Footnote 52 Within these models, the independent variable of interest is victims’ religion, which varies with the vignettes. The dependent variable is the participants’ level of support for the suggested military intervention.Footnote 53 Our average treatment effect indicates that recipients of the ‘Christian vignette’ were 1.36 times more likely to agree more with an intervention as compared to recipients of the ‘Muslim vignette’ (Table 2). In addition, our data further support the religious identification hypothesis by indicating that the average treatment effect based on odds ratios is not driven or biased by a strong effect within a particular subgroup (see Table 2).Footnote 54 In our experiment, we find odds ratios of our treatment variable larger than 1.0 across almost all subgroups (Table 2).

Table 2. Agreement to military intervention: Odds ratios based on ordinal logistic regression models.

Standard errors in parentheses

*** p < 0.01, ** p < 0.05, *p < 0.1.

The evidence for our hypothesis is further strengthened by an analysis of the treatment effect among those participants who not only have a Christian background but also feel attached to this religious community.Footnote 55 In our experiment, we did not only ask participants whether they have a Christian background, but also whether they feel emotionally attached to the respective group. We did this because our religious social identification hypothesis leads us to expect that people who feel emotionally attached to their group will have more compassionate feelings for victims of the same group than people who have only an emotionally looser connection to that group. By implication, in our experiment the treatment effect among those people who feel attached to their respective groups should be greater than the general treatment effect. This expectation is largely borne out by our experiment. The (statistically significant) treatment effect based on odds ratios increases from 1.36 to 1.71 when participants not only report having a Christian background but also claim to have an emotional attachment with this religious community.

Finally, we do not only find evidence for our religious identification hypothesis, but also for the proposed underlying causal mechanism that shared religion shapes people’s support for MHIs because it affects the level of compassion they feel for the victims of war-related violence. This leads us to expect that the treatments in our experiments are associated with higher levels of support for MHIs because participants feel more compassion for the victims of war. To assess this mechanism, we asked participants in our survey experiment about the level of compassion they feel for the victims.Footnote 56 To analyse whether shared religion affects the support of military interventions through compassion, we performed a statistical mediation analysis.Footnote 57 Mediation analyses are ‘designed to estimate the role of causal mechanisms that transmit the effect of a treatment variable on an outcome’.Footnote 58 It thus allows us to quantify the effect of our religious treatment that operates via our proposed compassion mechanism. To this end, we constructed a structural equation model (SEM) with our variable indicating the degree of compassion with the depicted victims as the mediator, our religious treatment as the independent, and support for military intervention as the dependent variable (Figure 2).

Figure 2. Structural equation model: Religious identification, compassion, and agreement with intervention.

The analysis we ran based on the above model shows that the statistically significant average treatment effect that we found in the experiment is indeed indirect and operates via compassion. While there is still a direct effect of our treatment on participants’ support for a MHI, this effect does not only decrease (OR 1.36 to 1.30) when controlling for compassion, but it is no longer statistically significant. At the same time, the effect of our treatment on compassion (OR 1.46) as well as the effect of compassion on support for a humanitarian intervention (OR 1.36) are statistically significant at the 95% level. This provides important evidence for the causal mechanism underlying our religious identification theory: religious identity significantly increases our participants’ compassion for the victims of war-related violence, which, in turn, leads to an increase of their support for their government’s contribution to a military intervention to help the victims of war.

To be sure, the overall effect of the victims’ religion on German students’ support of MHIs is rather small; but given the small margins with which Western societies agree or disagree with MHIs to help victims of war-related violence, it is likely to be of political relevance.Footnote 59 In Germany, representative polls have shown that the approval or disapproval rates for Bundeswehr missions are often just slightly above or slightly below 50 per cent.Footnote 60 Therefore, even if relatively small, the effect of the religious identity of the victims can be consequential for political decisions on individual MHIs.

Conclusion

In this article, we argue that in cases of massive war-related violence, people’s attitudes towards MHIs are not only motivated by rational considerations but also depend on social identification with the victims. We argue that the ‘morally motivated pressure to act’ that emerges in cases of war-related human suffering does not simply reflect the severity of the suffering. Identification with the victims matters, too. And this identification depends on – among other things – whether the victims of war-related violence belong to the same religion. Conducting a vignette-based experiment, we show that people in Germany are more likely to feel compassion, and to be prepared to support an MHI, when the victims are Christians. By contrast, when the victims do not share the Christian religion, people feel less compassion and are thus less willing to support their country’s contribution to a MHI that seeks to help them.

Whereas the dominant rationalist and constructivist macro-level theories make us believe that Western MHIs are merely selective,Footnote 61 our micro-level study demonstrates that individuals’ support for MHIs in Germany is subject to a religious bias. It thus contributes to existing micro-level studies on MHIs, which have so far mainly shown that other social markers such as the victims’ race – and sometimes also their gender – shape individuals’ support for MHIs and thus produce similar biases. Moreover, our findings of a religious bias suggest that it might be worthwhile to study the effect of religious identification on individuals’ support for a wider range of (foreign) policies that aim to help ‘strangers in need’. It seems worthwhile to study the effect of religious identification not only with regard to Western migration policies and military humanitarian interventions, but also with regard to Western policies of distributing humanitarian aid, providing development finance, engaging in disaster relief, or supporting global health efforts. Like MHIs, these policies may be subject to religious bias at the micro level.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/eis.2024.12.

Acknowledgements

We thank the German Foundation for Peace Research (DSF) for generously funding our research. We are also very thankful for the great research support by Nadia El Ghali. We are very grateful for all the valuable comments we received from colleagues at various conferences. We also acknowledge our colleagues and participants of the GSI IR colloquium Berthold Rittberger, Felix Biermann, Lisa Kriegmair, Tim Heinkelmann-Wild, Andreas Kruck, Raphaela Hobbach, Rainer Hülsse, Sebastian Schindler, Moritz Weiß, and Laura Seelkopf for their great support and feedback during the work on this paper.

Funding statement

This research has been funded as a pilot study by the German Foundation for Peace Research (DSF) (Grant Number FP07/17 – PS03/05).

Competing interests

The authors declare no competing interests.

Benjamin Daßler is a postdoctoral researcher at the Ludwig-Maximilians-Universität München (LMU). His research focuses on global governance and international regime complexes. He is the author of the book The Institutional Topology of International Regime Complexes, published by Oxford University Press, and of peer-reviewed articles in the European Journal of International Relations, Global Constitutionalism, and The Journal of International Relations and Development, among others.

Bernhard Zangl is Professor of Global Governance and Public Policy at the Ludwig-Maximilians-Universität München (LMU). Before moving to Munich in 2009, he was a Professor of International Relations at the University of Bremen. He also held visiting positions at Harvard University, the European University Institute, Northwestern University, and the Berlin Social Science Center. His work has been published in, inter alia, International Studies Quarterly, the British Journal of Political Science, the Annual Review of Political Science, and the Review of International Organizations.

Hilde van Meegdenburg is Assistant Professor of International Relations at the Institute of Political Science, Leiden University, the Netherlands. Her research explores the socio-emotional and narrative bases of state foreign policymaking and world politics. She has a particular interest in the rationalisation and organisation of foreign and military interventions. Previous work has been published in Contemporary Security Policy, Cooperation and Conflict, and the Journal of European Public Policy, among others.

References

1 Patrick M. Regan, ‘Choosing to intervene: Outside interventions in internal conflicts’, The Journal of Politics, 60:3 (1998), pp. 754–79; Martin Binder, The United Nations and the Politics of Selective Humanitarian Intervention (Cham: Springer International Publishing Palgrave Macmillan, 2017); Dieter Senghaas, ‘Der Grenzfall: Weltrechtsordnung vs. Rowdiestaaten’, Sicherheit und Frieden (S + F)/Security and Peace, 17:3 (1999), pp. 134–38 (p. 136); James Pattison, ‘The ethics of humanitarian intervention in Libya’, Ethics & International Affairs, 25:3 (2011), pp. 271–7.

2 Christian Ultsch, ‘ElBaradei: “Viele Muslime fühlen sich vom Westen wie Dreck behandelt”’, Die Presse (25 January 2015), available at: {https://www.diepresse.com/4646594/elbaradei-viele-muslime-fuehlen-sich-vom-westen-wie-dreck-behandelt}.

3 Sidita Kushi, ‘Selective humanitarians: How region and conflict perception drive military interventions in intrastate crises’, International Relations, 1 (2022), pp. 1–40; Andrew Wallis, Silent Accomplice: The Untold Story of France’s Role in the Rwandan Genocide (London: I.B. Tauris, 2014); Stephan Maninger, ‘Heart of darkness: Western policy of non-interventionism in Africa’, African Security Review, 8:6 (1999), pp. 25–36; Pattison, ‘The ethics of humanitarian intervention’.

4 For a similar definition, see Sarah Kreps and Sarah Maxey, ‘Mechanisms of morality: Sources of support for humanitarian intervention’, Journal of Conflict Resolution, 62:8 (2018), pp. 1814–42 (p. 1816). For a more nuanced discussion of MHIs and its key distinction from other humanitarian interventions, see Jennifer Szende, ‘Humanitarian military intervention’, in Deen K. Chatterjee (ed.), Encyclopedia of Global Justice (Dordrecht: Springer, 2011), pp. 516–19. For discussions of the broader concept of humanitarian intervention which also includes non-military means, see Nicholas J. Wheeler, Saving Strangers: Humanitarian Intervention in International Society (Oxford: OUP, 2000); Jeff L. Holzgrefe and Robert O. Keohane (eds), Humanitarian Intervention: Ethical, Legal and Political Dilemmas (Cambridge University Press, 2003); Jennifer M. Welsh (ed), Humanitarian Intervention and International Relations (Oxford: Oxford University Press, 2004), Martha Finnemore, The Purpose of Intervention: Changing Beliefs About the Use of Force (Cornell University Press, 2019).

5 See, e.g., Ben Clements, ‘Public opinion and military intervention: Afghanistan, Iraq and Libya’, The Political Quarterly, 84:1 (2013), pp. 119–31; Ben Clements, British Public Opinion on Foreign and Defence Policy: 1945–2017 (London: Routledge, 2019); Richard C. Eichenberg, ‘Victory has many friends: U.S. public opinion and the use of military force, 1981–2005’, International Security, 30:1 (2005), pp. 140–77 (p. 157); Gordon D. Cumming, Roel van der Velde, and Tony Chafer, ‘Understanding the public response: A strategic narrative perspective on France’s Sahelian operations’, European Security, 31:4 (2022), pp. 617–38.

6 Hans Kundnani, ‘Germany as a geo-economic power’, The Washington Quarterly, 34:3 (2011), pp. 31–45; Olivier Schmitt, ‘Strategic users of culture: German decisions for military action’, Contemporary Security Policy, 33:1 (2012), pp. 59–81; Regina Karp, ‘Identity and anxiety: Germany’s struggle to lead’, European Security, 27:1 (2018), pp. 58–81.

7 Fabrizio Coticchia and Francesco N. Moro, ‘Peaceful legislatures? Parliaments and military interventions after the Cold War: Insights from Germany and Italy’, International Relations, 34:4 (2020), pp. 482–503.

8 Timo Graf, ‘Zeitenwende im sicherheits- und verteidigungspolitischen Meinungsbild: Ergebnisse der ZMSBw-Bevölkerungsbefragung 2022’, Forschungsbericht des Zentrums für Militärgeschichte und Sozialwissenschaften der Bundeswehr, 133 (2022), pp. 1–27 (p. 9), available at: {https://www.bundeswehr.de/resource/blob/5510318/27bc160b1e7b392547290d059049bd20/download-bevoelkerungsbefragung-data.pdf}.

9 Timo Graf, Trendradar 2021: Die öffentliche Meinung zur Sicherheits- und Verteidigungspolitik in der Bundesrepublik Deutschland 2010–2020 (Potsdam: Zentrum für Militärgeschichte und Sozialwissenschaften der Bundeswehr, 2021), p. 28.

10 Karen A. Feste, Expanding the Frontiers: Superpower Intervention in the Cold War (New York: Praeger, 1992); Daniel Fiott, ‘Realist thought and humanitarian intervention’, The International History Review, 35:4 (2013), pp. 766–82.

11 Alan Dowty and Gil Loescher, ‘Refugee flows as grounds for international action’, International Security, 21:1 (1996), pp. 43–71; Idean Salehyan and Kristian Skrede Gledisch, ‘Refugees and the spread of civil war’, International Organization, 60:2 (2006), pp. 335–66; Bernhard Zangl, ‘Humanitäre interventionen’, in Mir Ferdowsi (ed), Internationale Politik im 21. Jahrhundert (Munich: Fink, 2002), pp. 105–22.

12 Michael J. Gilligan and Stephen J. Stedman, ‘Where do the peacekeepers go?’, International Studies Review, 5:4 (2003), pp. 37–54; Michael G. Findley and Josiah F. Marineau, ‘Lootable resources and third-party intervention into civil wars’, Conflict Management and Peace Science, 32:5 (2015), pp. 465–86.

13 Michael W. Doyle and Nicholas Sambanis, Making War and Building Peace: United Nations Peace Operations (Princeton, NJ: Princeton University Press, 2006); Mark J. Mullenbach, ‘Deciding to keep peace: An analysis of international influences on the establishment of third-party peacekeeping missions’, International Studies Quarterly, 49:3 (2005), pp. 529–55.

14 Francis K. Abiew, The Evolution of the Doctrine and Practice of Humanitarian Intervention (The Hague: Kluwer Law International, 1999); Wheeler, ‘Saving strangers’; Finnemore, ‘The purpose of intervention’; Alex J. Bellamy, ‘The Responsibility to Protect: Five years on’, Ethics & International Affairs, 24:2 (2010), pp. 143–69; Michael R. Tomz and Jessica L. Weeks, ‘Human rights and public support for war’, The Journal of Politics, 82:1 (2020), pp. 182–94.

15 Michael R. Tomz and Jessica L.P. Weeks, ‘Human rights and public support for war’, The Journal of Politics, 82:1 (2020), pp. 182–94; Carla Bagnoli, ‘Humanitarian intervention as a perfect duty: A Kantian argument’, in Terry Nardin and Melissa S. Williams (eds), Humanitarian Intervention (New York: New York University Press, 2006), pp. 117–40; Michael Tomz and Jessica L. Weeks, ‘Military alliances and public support for war’, International Studies Quarterly, 65:3 (2021), pp. 811–24; Martha Finnemore, ‘Constructing norms of humanitarian intervention’, in Richard K. Betts (ed), Conflict after the Cold War: Arguments on Causes of War and Peace (London and New York: Routledge, 2012), pp. 262–79; Martha Finnemore, The Purpose of Intervention: Changing Beliefs about the Use of Force (Ithaca, NY: Cornell University Press, 2019); Peter V. Jakobsen, ‘National interest, humanitarianism or CNN: What triggers UN peace enforcement after the Cold War?’, Journal of Peace Research, 33:2 (1996), pp. 205–15; Andreas Hasenclever, Die Macht der Moral in der internationalen Politik: Militärische Interventionen westlicher Staaten in Somalia, Ruanda und Bosnien-Herzegowina (Frankfurt a.M.: Campus-Verlag, 2001); Zangl, ‘Humanitäre interventionen’; Thomas G. Weiss, ‘The humanitarian impulse’, in David M. Malone (ed), The UN Security Council: From the Cold War to the 21st Century (Boulder, CO: Lynne Rienner, 2004), pp. 37–54; Michael R. Tomz and Jessica L. Weeks, ‘Public opinion and the democratic peace’, American Political Science Review, 107:4 (2013), pp. 849–65; Garrett Wallace Brown and Alexandra Bohm, ‘Introducing jus ante bellum as a cosmopolitan approach to humanitarian intervention’, European Journal of International Relations, 22:4 (2016), pp. 897–919; Seung-Whan Choi and Patrick James, ‘Why does the United States intervene abroad? Democracy, human rights violations, and terrorism’, Journal of Conflict Resolution, 60:5 (2016), pp. 899–926; Binder, The United Nations; Kreps and Maxey, ‘Mechanisms of morality’.

16 Binder, The United Nations.

17 Warren P. Strobel, Late Breaking Foreign Policy: The News Media's Influence on Peace Operations (Washington, DC: US Institute of Peace, 1997); Michael N. Barnett, ‘The limits of peacekeeping, spheres of influence, and the future of the United Nations’, in Joseph Lepgold and Thomas G. Weiss (eds), Collective Conflict Management and Changing World Politics (Albany: State University of New York Press, 1998), pp. 83–103; Piers Robinson, ‘The CNN-effect: Can the news media drive foreign policy’, Review of International Studies, 25:2 (1999), pp. 301–9; Piers Robinson, ‘The policy-media interaction model: Measuring media power during humanitarian crisis’, Journal of Peace Research, 37:5 (2000), pp. 613–33; David M. Malone (ed), The UN Security Council; Eytan Gilboa, ‘The CNN effect: The search for a communication theory of international relations’, Political Communication, 22:1 (2005), pp. 27–44; Benjamin Daßler, Bernhard Zangl, and Hilde van Meegdenburg, ‘Mitleids- und Hilfsmüdigkeit bei humanitären Krisen: Zum Effekt übermäßigen Medienkonsums’, ZIB Zeitschrift für Internationale Beziehungen, 28:2 (2021), pp. 64–82, available at: {https://www.nomos-elibrary.de/10.5771/0946-7165-2021-2-64/mitleids-und-hilfsmuedigkeit-bei-humanitaeren-krisen-zum-effekt-uebermaessigen-medienkonsums-jahrgang-28-2021-heft-2?page=1}; Hilde van Meegdenburg, Bernhard Zangl, and Benjamin Daßler, Humanitarian Interventions: Saving Close and Distant Strangers (Osnabrück: Deutsche Stiftung Friedensforschung, 2020).

18 With ‘bias’, we imply here the existence of an (probably largely subconscious) inclination in people to be more attentive to and affected by human suffering when those suffering have the same religion, gender, or race. For a similar understanding, see Richard Hanania and Robert Trager, ‘The prejudice first model and foreign policy values: Racial and religious bias among conservatives and liberals’, European Journal of International Relations, 27:1 (2021), pp. 204–31.

19 Michael Tomz, Jessica L. Weeks, and Keren Yarhi-Milo, ‘Public opinion and decisions about military force in democracies’, International Organization, 74:1 (2020), pp. 119–43; Jonathan A. Chu and Stefano Recchia, ‘Does public opinion affect the preferences of foreign policy leaders? Experimental evidence from the UK parliament’, The Journal of Politics, 84:3 (2022), pp. 1874–77; Clements, ‘Public opinion and military intervention’.

20 Jonathan Chu and Carrie Lee, ‘Race, religion, and American support for humanitarian intervention’ (2022), available at SSRN 4060474; Kreps and Maxey, ‘Mechanisms of morality’; Donghyun D. Choi, Mathias Poertner, and Nicholas Sambanis, Native Bias: Overcoming Discrimination against Immigrants (Princeton, NJ: Princeton University Press, 2022); Michael C. Grillo and Juris Pupcenoks, ‘Let’s intervene! But only if they’re like us: The effects of group dynamics and emotion on the willingness to support humanitarian intervention’, International Interactions, 43:2 (2017), pp. 349–74; Mattias Agerberg and Anne-Kathrin Kreft, ‘Sexual violence, gendered protection and support for intervention’, Journal of Peace Research, 60:5 (2023), pp. 853–67; Jonathan Chu, ‘Social cues by international organizations: NATO, the Security Council, and public support for humanitarian intervention’ (2019), available at SSRN 3977910; Kathleen E. Powers, Joshua D. Kertzer, Deborah J. Brooks, and Stephen G. Brooks, ‘What’s fair in international politics? Equity, equality, and foreign policy attitudes’, Journal of Conflict Resolution, 66:2 (2022), pp. 217–45.

21 Chu and Lee, ‘Race, religion, and American support’; Anne-Kathrin Kreft and Mattias Agerberg, ‘Imperfect victims? Civilian men, vulnerability, and policy preferences’, American Political Science Review, 118:1 (2024), pp. 274–90; Mary-Kate Lizotte, ‘Investigating the origins of the gender gap in support for war’, Political Studies Review, 17:2 (2019), pp. 124–35; Timothy M. Gill, ‘The Civilizing mission persists: Racism and justification for US intervention into socialist Venezuela’, Du Bois Review: Social Science Research on Race, 19:2 (2022), pp. 309–28, available at: {https://www.cambridge.org/core/journals/du-bois-review-social-science-research-on-race/article/abs/civilizing-mission-persists/6C30F8EC652BF9D3C3664B8EE0515673}.

22 For an important exception, see Chu and Lee, ‘Race, religion, and American support’.

23 Choi, Poertner, and Sambanis, Native Bias; Robert Shaffer, Lauren E. Pinson, Jonathan A. Chu, and Beth A. Simmons, ‘Local elected officials’ receptivity to refugee resettlement in the United States’, Proceedings of the National Academy of Sciences of the United States of America, 117:50 (2020), pp. 31722–28; Isabel W. Skinner, ‘How characterizations of refugees shape attitudes toward refugee restrictions: A study of Christian and Muslim Americans’, International Journal of Public Opinion Research, 34:3 (2022), p. edac022; Kirk Bansak, Jens Hainmueller, and Dominik Hangartner, ‘How economic, humanitarian, and religious concerns shape European attitudes toward asylum seekers’, Science, 354:6309 (2016), pp. 217–22.

24 Tobias Cremer, ‘Defenders of the faith? How shifting social cleavages and the rise of identity politics are reshaping right-wing populists’ attitudes towards religion in the West’, Religion, State and Society, 50:5 (2022), pp. 532–52; Nicholas Morieson, Religion and the Populist Radical Right: Secular Christianism and Populism in Western Europe (Wilmington, DE: Vernon Press, 2021); Jakob Schwörer and Belén Fernández-García, ‘Religion on the rise again? A longitudinal analysis of religious dimensions in election manifestos of Western European parties’, Party Politics, 27:6 (2021), pp. 1160–71.

25 James E. Cameron, ‘A three-factor model of social identity’, Self and Identity, 3:3 (2004), pp. 239–62; Toon Kuppens and Vincent Y. Yzerbyt, ‘Group-based emotions: The impact of social identity on appraisals, emotions, and behaviors’, Basic and Applied Social Psychology, 34:1 (2012), pp. 20–33; Mark Levine, Amy Prosser, David Evans, and Stephen Reicher, ‘Identity and emergency intervention: How social group membership and inclusiveness of group boundaries shape helping behavior’, Personality & Social Psychology Bulletin, 31:4 (2005), pp. 443–53; Pazit Ben-Nun Bloom, Gizem Arikan, and Marie Courtemanche, ‘Religious social identity, religious belief, and anti-immigration sentiment’, American Political Science Review, 109:2 (2015), pp. 203–21; Mark Tarrant, Nyla R. Branscombe, Ruth H. Warner, and Dale Weston, ‘Social identity and perceptions of torture: It’s moral when we do it’, Journal of Experimental Social Psychology, 48:2 (2012), 513–18, available at: {https://www.sciencedirect.com/science/article/pii/s0022103111002605}.

26 Muriel Dumont, Vincent Yzerbyt, Daniël Wigboldus, and Ernestine H. Gordijn, ‘Social categorization and fear reactions to the September 11th terrorist attacks’, Personality & Social Psychology Bulletin, 29:12 (2003), pp. 1509–20; Alexandra D. Blackman, ‘Religion and foreign aid’, Politics and Religion, 11:3 (2018), pp. 522–52; Brian Parkinson, Agneta H. Fischer, and Antony S. R. Manstead, Emotion in Social Relations: Cultural, Group, and Interpersonal Perspectives (London: Psychology Press, 2005), available at: {https://www.taylorfrancis.com/books/mono/10.4324/9780203644966/emotion-social-relations-brian-parkinson-agneta-fischer-antony-manstead}.

27 Joannie Tremblay‐Boire and Aseem Prakash, ‘Biased altruism: Islamophobia and donor support for global humanitarian organizations’, Public Administration Review, 79:1 (2019), pp. 113–24; Blackman, ‘Religion and foreign aid’; Jeffrey R. Seul, ‘“Ours is the way of God”: Religion, identity, and intergroup conflict’, Journal of Peace Research, 36:5 (1999), pp. 553–69; Michael A. Hogg, Janice R. Adelman, and Robert D. Blagg, ‘Religion in the face of uncertainty: An uncertainty-identity theory account of religiousness’, Personality and Social Psychology Review: An Official Journal of the Society for Personality and Social Psychology, Inc, 14:1 (2010), pp. 72–83; Kerem O. Kalkan, Geoffrey C. Layman, and Eric M. Uslaner, ‘“Bands of others”? Attitudes toward Muslims in contemporary American society’, The Journal of Politics, 71:3 (2009), pp. 847–62.

28 Kentaro Fujita, Marlone D. Henderson, Juliana Eng, Yaacov Trope, and Nira Liberman, ‘Spatial distance and mental construal of social events’, Psychological Science, 17:4 (2006), pp. 278–82; Yaacov Trope, Nira Liberman, and Cheryl Wakslak, ‘Construal levels and psychological distance: Effects on representation, prediction, evaluation, and behavior’, Journal of Consumer Psychology, 17:2 (2007), pp. 83–95, available at: {https://www.sciencedirect.com/science/article/pii/s105774080770013x}; Ido Liviatan, Yaacov Trope, and Nira Liberman, ‘Interpersonal similarity as a social distance dimension: Implications for perception of others’ actions’, Journal of Experimental Social Psychology, 44:5 (2008), pp. 1256–69, available at: {https://www.sciencedirect.com/science/article/pii/s0022103108000723}; Nira Liberman and Yaacov Trope, ‘The psychology of transcending the here and now’, Science, 322:5905 (2008), pp. 1201–5.

29 Eliot R. Smith and Diane M. Mackie, ‘Dynamics of group-based emotions: Insights from intergroup emotions theory’, Emotion Review, 7:4 (2015), pp. 349–54 (p. 350).

30 Dumont et al., ‘Social categorization’; Ernestine H. Gordijn, Vincent Yzerbyt, Daniel Wigboldus, and Muriel Dumont, ‘Emotional reactions to harmful intergroup behavior’, European Journal of Social Psychology, 36:1 (2006), pp. 15–30; Parkinson, Fischer, and Manstead, Emotion in Social Relations (2005); Kuppens and Yzerbyt, ‘Group-based emotions’; Smith and Mackie, ‘Dynamics of group-based emotions’.

31 Parkinson, Fischer, and Manstead, Emotion in Social Relations, p. 11.

32 Cameron, ‘Three-factor model’; Jeffrey Lyons and Stephen M. Utych, ‘You’re not from here! The consequences of urban and rural identities’, Political Behavior, 45:1 (2023), pp. 75–101, available at: {https://link.springer.com/article/10.1007/s11109-021-09680-3}; Danielle Jacobson and Nida Mustafa, ‘Social identity map: A reflexivity tool for practicing explicit positionality in critical qualitative research’, International Journal of Qualitative Methods, 18 (2019), p. 1609406919870075.

33 James Tilley, ‘“We don’t do God”? Religion and party choice in Britain’, British Journal of Political Science, 45:4 (2015), pp. 907–27; Renate Ysseldyk, Kimberly Matheson, and Hymie Anisman, ‘Religiosity as identity: Toward an understanding of religion from a social identity perspective’, Personality and Social Psychology Review: An Official Journal of the Society for Personality and Social Psychology, Inc, 14:1 (2010), pp. 60–71; James K. Wellman, Jr and Kyoko Tokuno, ‘Is religious violence inevitable?’, Journal for the Scientific Study of Religion, 43:3 (2004), pp. 291–6; Pazit Ben-Nun Bloom, Gizem Arikan, and Marie Courtemanche, ‘Religious social identity, religious belief, and anti-immigration sentiment’, American Political Science Review, 109:2 (2015), pp. 203–21; Jesse Graham and Jonathan Haidt, ‘Beyond beliefs: Religions bind individuals into moral communities’, Personality and Social Psychology Review: An Official Journal of the Society for Personality and Social Psychology, Inc, 14:1 (2010), pp. 140–50.

34 Egbert Ribberink, Peter Achterberg, and Dick Houtman, ‘Religious polarization: Contesting religion in secularized Western European countries’, Journal of Contemporary Religion, 33:2 (2018), pp. 209–27; Bloom, Arikan, and Courtemanche, ‘Religious social identity, religious belief, and anti-immigration sentiment’; Muniba Saleem and Srividya Ramasubramanian, ‘Muslim Americans’ responses to social identity threats: Effects of media representations and experiences of discrimination’, Media Psychology, 22:3 (2019), pp. 373–93; Skinner, ‘How characterizations of refugees shape attitudes’; Thomas Luckmann, The Invisible Religion: The Problem of Religion in Modern Society (Abingdon: Routledge, 2023); Kalkan, Layman, and Uslaner, ‘“Bands of others”?’; Samuel L. Perry, ‘American religion in the era of increasing polarization’, Annual Review of Sociology, 48 (2022), pp. 87–107; Steven L. Neuberg, Carolyn M. Warner, Stephen A. Mistler et al., ‘Religion and intergroup conflict: Findings from the Global Group Relations Project’, Psychological Science, 25:1 (2014), pp. 198–206; Bethany Lacina and Charlotte Lee, ‘Culture clash or democratic peace? Results of a survey experiment on the effect of religious culture and regime type on foreign policy opinion formation’, Foreign Policy Analysis, 9:2 (2013), pp. 143–70; Daniel Stegmueller, ‘Religion and redistributive voting in Western Europe’, The Journal of Politics, 75:4 (2013), pp. 1064–76.

35 Saleem and Ramasubramanian, ‘Muslim Americans’ responses’; Skinner, ‘How characterizations of refugees shape attitudes’; Rita Nassar, ‘Framing refugees: The impact of religious frames on U.S. partisans and consumers of cable news media’, Political Communication, 37:5 (2020), pp. 593–611; Christoph Ramm, ‘The Muslim makers’, Interventions, 12:2 (2010), pp. 183–97; Paul A. Djupe and Brian R. Calfano, ‘Divine intervention? The influence of religious value communication on U.S. intervention policy’, Political Behavior, 35:4 (2013), pp. 643–63, available at: {https://link.springer.com/article/10.1007/s11109-012-9211-3}; Laura P. B. Partain and Andrew J. Weaver, ‘(Un)veiling our biases: Activating religious, emotional, and contextual cues in news media representations of Syrian refugees’, International Journal of Communication, 16 (2022), pp. 2410–30, available at: {https://ijoc.org/index.php/ijoc/article/view/17215}.

36 Kevin R. den Dulk and Mark J. Rozell, ‘George W. Bush, religion, and foreign policy: Personal, global, and domestic contexts’, The Review of Faith & International Affairs, 9:4 (2011), pp. 71–82; Paul Froese and F. C. Mencken, ‘A U.S. holy war? The effects of religion on Iraq War policy attitudes’, Social Science Quarterly, 90:1 (2009), pp. 103–16; Rebecca A. Glazier, ‘Divine direction: How providential religious beliefs shape foreign policy attitudes’, Foreign Policy Analysis, 9:2 (2013), pp. 127–42.

37 Erik Ringmar, Identity, Interest and Action: A Cultural Explanation of Sweden’s Intervention in the Thirty Years War (Cambridge: Cambridge University Press, 2007); Patrick T. Jackson, Civilizing the Enemy: German Reconstruction and the Invention of the West (Ann Arbor: University of Michigan Press, 2006).

38 Janet Finch, ‘The vignette technique in survey research’, Sociology, 21:1 (1987), pp. 105–14 (p. 105).

39 Matthew S. Winters and Rebecca Weitz-Shapiro, Lacking Information or Condoning Corruption: When Do Voters Support Corrupt Politicians? (New York: City University of New York, 2013), p. 423, available at: {https://www.ingentaconnect.com/content/cuny/cp/2013/00000045/00000004/art00004}.

40 We selected classes in Communications Science, Political Science, and Law. The experiment was conducted within three days from 29–31 October 2018. None of the courses selected was taught by one of the authors.

41 For a detailed description of the study’s research ethics, see Online Appendix C7, which also provides further information on the data collection process and ethical integrity measures. Participants were not paid. They were instructed not to talk to each other and to refrain from using electronic devises during the experiment.

42 See especially Stefano Recchia and Jonathan Chu, ‘Validating threat: IO approval and public support for joining military counterterrorism coalitions’, International Studies Quarterly, 65:4 (2021), pp. 919–28, who demonstrate the generally positive effect of IO involvement on public support for joining military coalitions.

43 The original question in German read: ‘Inwieweit stimmen Sie folgender Aussage zu? Deutschland sollte eine militärische Intervention der UNO (durch die Entsendung von etwa 500 Bundeswehrsoldaten) in den Benin-Konflikt unterstützen.’ Answer options: (1) stimme voll zu; (2) stimme eher zu; (3) stimme eher nicht zu; (4) stimme überhaupt nicht zu. For the entire survey instrument in both the original German version and an English translation, see Online Appendix B.

44 More precisely, all participants in the cleaned-up sample of this experiment answered the question ‘In which cultural-religious context did you grow up?’ (‘In welchem religiös-kulturellen Kontext sind Sie aufgewachsen?’), also with ‘Christian’ (see Online Appendix B).

45 This is particularly true for the dimension of external validity Egami and Hartmann (REF) refer to as “X-Vailidity”, i.e. validity concerns related to differences in the composition of the sample and the target population. Naoki Egami and Erin Hartmann, ‘Elements of external validity: Framework, design, and analysis’, American Political Science Review, 117:3 (2023), pp. 1070–88.

46 Maja Zehfuss, ‘Constructivism and identity’, in Stefano Guzzini and Anna Leander (eds), Constructivism and International Relations: Alexander Wendt and His Critics (London and New York: Routledge, 2005), pp. 93–117; Frank A. Stengel, The Politics of Military Force: Antimilitarism, Ideational Change, and Post-Cold War German Security Discourse (Ann Arbor: University of Michigan Press, 2020).

47 Stengel, Politics of Military Force, pp. 2–3.

48 Matthias Mader, ‘Citizens’ perceptions of policy objectives and support for military action’, Journal of Conflict Resolution, 61:6 (2017), pp. 1290–314; Rainer Baumann and Gunther Hellmann, ‘Germany and the use of military force: “total war”, the “culture of restraint” and the quest for normality’, German Politics, 10:1 (2001), pp. 61–82; Thomas Risse, ‘Kontinuität durch Wandel: Eine “neue” deutsche Außenpolitik’, Aus Politik und Zeitgeschichte, 11:2004 (2004), pp. 24–31.

49 Clements, ‘Public opinion and military intervention’, p. 122.

50 The outcome of the last general elections was, in fact, as follows and overall quite different from the distribution of party identification of the participants in our sample: CDU/CSU: 32.9%; SPD: 20.5%; AfD: 12.6%; FDP: 10.8%; Die Linke: 9.2%; B’90/ Die Grünen: 8.9%; other: 5%.

51 Regarding gender and party identification, this is clearly underscored by our data collected in the context of our own survey experiment (see Table 2): the effect of our social distance treatment is very small and insignificant among women and even adversarial but insignificant among supporters of the Green Party. Regarding younger and better-educated individuals see, e.g., Ian Woodward, Zlatko Skrbis, and Clive Bean, ‘Attitudes towards globalization and cosmopolitanism: Cultural diversity, personal consumption and the national economy’, The British Journal of Sociology, 59:2 (2008), pp. 207–26 (p. 221); Pippa Norris, ‘Global governance and cosmopolitan citizens’, in Joseph S. Nye and John D. Donahue (eds), Governance in a Globalizing World (Cambridge, MA: Visions of Governance for the 21st Century; Brookings Institution Press, 2000); Jeannie Haubert and Elizabeth Fussell, ‘Explaining pro-immigrant sentiment in the U.S.: Social class, cosmopolitanism, and perceptions of immigrants’, International Migration Review, 40:3 (2006), pp. 489–507; Steffen Mau, Jan Mewes, and Ann Zimmermann, ‘Cosmopolitan attitudes through transnational social practices?’, Global Networks, 8:1 (2008), pp. 1–24; Florian Pichler, ‘“Down-to-earth” cosmopolitanism’, Current Sociology, 57:5 (2009), pp. 704–32; Timothy Phillips and Philip Smith, ‘Cosmopolitan beliefs and cosmopolitan practices’, Journal of Sociology, 44:4 (2008), pp. 391–9.

52 Table 2 reports the odds ratios of our treatment variable for the full sample (Average Treatment Effect) and within important subgroups of our sample (Subgroup Effects). For the full models underlying the reported subgroup effects, including the corresponding model statistics, see Online Appendix C.3.

53 Note that the dependent variable, i.e. agreement with a military intervention, is coded in the following way: 1 = strongly disagree; 2 = somewhat disagree; 3 = somewhat agree; 4 = strongly agree. Thus, odds ratios larger than 1 indicate higher probabilities of agreeing more with a military intervention (as compared to the baseline), while odds ratios smaller than 1 indicate lower odds of agreeing more, ergo, higher probabilities of disagreeing more with a military intervention as compared to the baseline. Odds ratios for each subgroup of the sample were calculated by means of ordinal regression models estimating agreement to military intervention (as DV) and the interaction terms for the treatment and the respective subgroup (as IV). Regarding the interpretation of odds-ratio based coefficients in logistic regression models including interaction terms, see also Maarten L. Buis, ‘Stata Tip 87: Interpretation of interactions in nonlinear models’, The Stata Journal, 10:2 (2010), pp. 305–8.

54 For a useful explanation including examples of the interpretation of odds-ratio based coefficients in logistic regression models including interaction terms, see Buis, ‘Stata Tip 87’.

55 We identified this group by asking the participants of our experiment whether they not only grew up in a Christian context, but whether they also feel themselves attached to this respective cultural-religious group. More precisely, we asked in German: ‘To which of these cultural-religious groups do you also feel connected?’ (see also Online Appendix B).

56 More specifically, we asked in German: ‘Did the newspaper report touch or move you emotionally?’. We coded the variable as follows: (1) no, not at all; (2) no, rather not; (3) yes, rather; (4) yes, very (see also Online Appendix B).

57 Doug Gunzler, W. Tang, N. Lu, P. Wu, and X. M. Tu, ‘A class of distribution-free models for longitudinal mediation analysis’, Psychometrika, 79:4 (2014), pp. 543–68, available at: {https://link.springer.com/article/10.1007/s11336-013-9355-z}; Bianca L. De Stavola, Rhian M. Daniel, George B. Ploubidis, and Nadia Micali ‘Mediation analysis with intermediate confounding: Structural equation modeling viewed through the causal inference lens’, American Journal of Epidemiology, 181:1 (2015), pp. 64–80.

58 Raymond Hicks and Dustin Tingley, ‘Causal mediation analysis’, The Stata Journal, 11:4 (2011), pp. 605–19 (p. 605).

59 For instance, public opinion polls on Germany’s participation in the UN Mission MINUSMA in Mali revealed a stalemate between opponents and supporters of the German participation with 32 percent agreement and 32 percent disagreement (and 30 percent considering themselves neutral); see Graf, ‘Zeitenwende im sicherheits- und verteidigungspolitischen Meinungsbild’, p. 9.

60 Ibid.; Graf, Trendradar 2021.

61 Martin Binder, ‘Humanitarian crises and the international politics of selectivity’, Human Rights Review, 10:3 (2009), pp. 327–48, available at: {https://link.springer.com/article/10.1007/s12142-009-0121-7}; Kushi, ‘Selective humanitarians’.

Figure 0

Table 1. Participant socio-demographics.

Figure 1

Figure 1. Agreement to support a German military intervention.

Figure 2

Table 2. Agreement to military intervention: Odds ratios based on ordinal logistic regression models.

Figure 3

Figure 2. Structural equation model: Religious identification, compassion, and agreement with intervention.

Supplementary material: File

Daßler et al. supplementary material

Daßler et al. supplementary material
Download Daßler et al. supplementary material(File)
File 232.8 KB