Introduction
Does the public apply a “double standard” for human rights abuses based on the perpetrator’s alliance status? Research shows that the public is more supportive of military action against states that violate human rights – a preference rooted in the perception that violators are more threatening and the moral duty to intervene when informed about human rights violations (Tomz and Weeks Reference Tomz and Weeks2020). However, condemnations of human rights abuses are often politicized and contingent on strategic relations with the perpetrator (e.g., Terman and Byun Reference Terman and Byun2022).Footnote 1
The notion of a double standard on human rights reflects previous studies that show that states enforce norms only when it aligns with their interests (Krasner Reference Krasner1999; Mearsheimer Reference Mearsheimer1994). The application of supposedly absolute, universal principles can often be self-interest in disguise (Carr Reference Carr2001). Indeed, U.S. leaders are often criticized for employing double standards on human rights when dealing with strategically important partners, such as the Shah of Iran during the Carter administration (Goshko Reference Goshko1977) and Saudi Arabia during the Biden administration (Detrow Reference Detrow2022). Similarly, citizens exhibit a diminished propensity to support sanctions or reduce aid allocations in response to human rights violations when the perpetrators are strategic partners (Zarpli Reference Zarpli2024; Heinrich, Kobayashi, and Long Reference Heinrich, Kobayashi and Long2018; Esarey and DeMeritt Reference Esarey and DeMeritt2017). When contemplating backing military intervention against human rights violators who are aligned with the U.S., leaders and the public must weigh the moral obligation to protect human rights against the potential loss of a strategic partner.
Despite its importance, there is surprisingly limited research investigating how the human rights practices of foreign states interact with strategic relations to shape public support for war. If there is a double standard, the public will be less supportive of attacking human rights violators if the violator is a strategic partner. Put differently, the standard for tolerable behavior will be different for allies and non-allies. Understanding whether citizens in democracies apply this double standard holds paramount importance for our comprehension of public support for war.
To investigate the dynamics between strategic interests and human rights in shaping public support for war, we implemented a 2
$ \times $
2 factorial experiment in two separate U.S. nationwide surveys. In both surveys, we independently manipulated two key factors: (1) whether the foreign state violates human rights and (2) whether the foreign state is a U.S. ally. Both experiments yield two consistent findings. First, respondents are more inclined to support military action against states that violate human rights compared to those that respect human rights. Second, the effect of human rights violations on support for war is not contingent on the alliance status of the target state. In other words, we find no evidence of a double standard: domestic citizens are concerned about human rights violations regardless of the violator’s alliance status.
Our contributions are twofold. First, we successfully replicate the pioneering study by Tomz and Weeks (Reference Tomz and Weeks2020) on human rights and public support for war. Replication enables the accumulation, progression, and generalization of knowledge through the validation of findings and the identification of boundary conditions (Jasny et al. Reference Jasny, Chin, Chong and Vignieri2011; Maxwell, Lau, and Howard Reference Maxwell, Lau and Howard2015; Shadish, Cook, and Campbell Reference Shadish, Cook and Campbell2002). Our study reaffirms that human rights shape public support for war and that perceptions of threat and morality drive these preferences.
Second, we demonstrate that the effect of human rights violations on public support for war is not contingent on the violator’s alliance status. Contrasting with conventional notions of a double standard on human rights (e.g., Zarpli Reference Zarpli2024; Heinrich, Kobayashi, and Long Reference Heinrich, Kobayashi and Long2018), we find that the alliance status of human rights violators has a negligible effect on public support for war. This finding has important implications for U.S. foreign policy. While we often observe leaders employing double standards (Terman and Byun Reference Terman and Byun2022; Terman and Voeten Reference Terman and Voeten2018), the domestic public may not endorse such actions, which may prompt leaders to reconsider how they balance their image as human rights advocates while pursuing strategic interests. An important caveat is that our research focuses solely on democracies. The extent to which our findings generalize to a non-democratic context remains an open question that we leave for future investigation.
Public opinion, strategic interests, & human rights
Public opinion is an important consideration when leaders form foreign policy. Increasingly, experiments using elite samples show that leaders are more likely to pursue foreign policies that have public support (Tomz, Weeks, and Yarhi-Milo Reference Tomz, Weeks and Yarhi-Milo2020; Chu and Recchia Reference Chu and Recchia2022; Lin-Greenberg Reference Lin-Greenberg2021). This evidence suggests that democratic leaders consider public opinion when determining whether to initiate conflict. We examine two factors that could affect public support for war against a target state: its human rights record and alliance status.
The public is more supportive of military action against foreign states that violate human rights (Tomz and Weeks Reference Tomz and Weeks2020). A state’s willingness to use force to solve its domestic problems is likely to be reflected in the way it solves international disputes (Peterson and Graham Reference Peterson and Graham2011; Sobek, Abouharb, and Ingram Reference Sobek, Abouharb and Ingram2006; Caprioli and Trumbore Reference Caprioli and Trumbore2003, Reference Caprioli and Trumbore2006; Maoz and Russett Reference Maoz and Russett1993). Thus, states with peaceful internal conflict resolution practices are perceived as less threatening than those with violent practices (Tomz and Weeks Reference Tomz and Weeks2020). Public support for intervention is also driven by perceptions of morality (Tomz and Weeks Reference Tomz and Weeks2020; Kreps and Maxey Reference Kreps and Maxey2018; Hildebrandt et al. Reference Hildebrandt and Courtney Hillebrecht2013; Pomeroy and Rathbun Reference Pomeroy and Rathbun2023). The public may feel morally obligated to intervene against human rights abusers and have fewer misgivings toward initiating conflict against states that abuse their citizens (Tomz and Weeks Reference Tomz and Weeks2020).
However, the public’s preference to promote human rights may be in tension with other important goals. For example, in pursuit of security, economic, or political goals, states sharing common interests may form alliances (Bueno De Mesquita Reference Bueno De Mesquita1981). Reflective of these shared interests, alliances increase national security and autonomy (Morrow Reference Morrow1991; Snyder Reference Snyder1990). Since allies “burden share” security costs (Snyder Reference Snyder1990), allies face an opportunity cost for terminating their alliance (Maoz Reference Maoz2009). Thus, turning against an ally – should it violate human rights – risks severing the ties that provide these strategic advantages.
States may also suffer reputational damage for turning against an ally (Tomz and Weeks Reference Tomz and Weeks2021; Simmons Reference Simmons2010; Crescenzi et al. Reference Crescenzi, Kathman, Kleinberg and Wood2012). Tomz and Weeks (Reference Tomz and Weeks2021) find that the public’s preference to intervene militarily on behalf of an ally over a non-ally is driven, in part, by concerns that abandoning an ally could damage its reputation. Thus, should an ally violate human rights, the public’s preference to safeguard human rights vies with the reputational costs of turning against an ally.
Recent studies show that leaders and the public are not always willing to uphold human rights, revealing evidence of a double standard for perpetrators determined by their strategic value as partners (e.g., Terman and Byun Reference Terman and Byun2022). For example, regarding their willingness to name and shame, impose sanctions, or grant foreign aid, states tend to be more tolerant of abuses by their strategic partners (Terman and Byun Reference Terman and Byun2022; Terman and Voeten Reference Terman and Voeten2018; Peksen, Peterson, and Drury Reference Peksen, Peterson and Drury2014; Heinrich, Kobayashi, and Long Reference Heinrich, Kobayashi and Long2018; Sandlin Reference Sandlin2022; Nielsen Reference Nielsen2013). Further, citizens are less supportive of sanctioning human rights violations if the perpetrator is a strategic partner (Zarpli Reference Zarpli2024; Heinrich, Kobayashi, and Long Reference Heinrich, Kobayashi and Long2018).
In summary, prior research indicates that public support for war is likely shaped by considerations of safeguarding human rights and advancing other strategic goals. However, current studies fall short in exploring the nuanced interplay between these two variables. By using a factorial design, we examine the trade-off between the desire to protect human rights and maintain beneficial relationships from the public’s perspective. Following Tomz and Weeks (Reference Tomz and Weeks2020), we expect that a foreign state having a record of human rights violations will increase public support for war in contrast to a state not having a record (Hypothesis 1). We also expect that being a U.S. ally will decrease public support for war in contrast to not being a U.S. ally (Hypothesis 2). Finally, if a double standard on human rights exists, we should observe that the public is less supportive of initiating conflict against a U.S. ally that violates human rights, compared to a non-U.S. ally that violates human rights (Hypothesis 3).
Research design
To investigate our hypotheses, we conducted two preregistered survey experiments that employed identical vignettes.Footnote 2 The first survey employs a two-wave approach, conducted by YouGov between November and December 2022, and includes 1,000 U.S. adults.Footnote 3 The second survey, executed by PureSpectrum, used a national sample of 4,002 U.S. adults in September 2023.Footnote 4
We first told participants that they would read about a hypothetical situation that the U.S. could face in the future. We then presented respondents with a description of the scenario based on Tomz and Weeks (Reference Tomz and Weeks2020): “A country is developing nuclear weapons and will have its first nuclear bomb within six months. The country could then use its missiles to launch nuclear attacks against any country in the world.” We hold several characteristics of the proliferating state constant: its level of trade with the U.S. (low), regime type (democracy), and military strength (half as strong as the U.S.). We elected to hold regime type constant because democracy, or the holding of elections, is conceptually different from human rights practices (Tomz and Weeks Reference Tomz and Weeks2020). Indeed, it is common practice for states – regardless of regime type – to engage in torture (Cingranelli and Richards, Reference Cingranelli and Richards2004, cf. Conrad and Moore, Reference Conrad and Moore2010). Further, the U.S. has historically partnered with states with elected leaders who have allegedly violated human rights, including Turkey and Israel (Amnesty International 2024). Thus, our experiment can help shed light on the conditional effect of alliance status on differences between electoral democracies that violate human rights versus those that protect them.
Our first treatment describes the foreign state’s human rights practices following Tomz and Weeks (Reference Tomz and Weeks2020). Half of the respondents were informed that: “The country does not violate human rights; it does not imprison or torture its citizens because of their beliefs,” and the other half were told that: “The country violates human rights; it imprisons or tortures some of its citizens because of their beliefs.” Our second treatment captures the foreign state’s alliance status. Half of respondents were told that: “The country has not signed a military alliance with the United States,” and the other half were told that: “The country has signed a military alliance with the United States.”
The scenario concluded with several points that were the same for all respondents. Participants were informed that “The country’s motives remain unclear, but if it builds nuclear weapons, it will have the power to blackmail or destroy other countries.” Finally, the scenario concluded that “by attacking the country’s nuclear development sites now,” the U.S. could “prevent the country from making any nuclear weapons.” After presenting this information, we asked whether respondents would favor or oppose using U.S. armed forces to attack the nuclear development sites on a 5-point Likert scale ranging from “Oppose Strongly” to “Favor Strongly.” We also asked questions about perceptions of threat, morality, costs, and success to investigate the mechanisms behind the treatment effects.Footnote 5
Several features of the experimental design are worth emphasizing. First, by independently randomizing Human Rights and Alliance, we can estimate how information about a state’s human rights record influences support for war, not only on average but also based on whether the target is a U.S. ally. This interaction allows us to examine whether public support for war against a state implicated in human rights abuses varies based on the extent of its shared strategic interests with the U.S. Second, as our experiment is based on Tomz and Weeks’ (Reference Tomz and Weeks2020) design, we can affirm whether the effect of human rights practices on public support for war is replicable.Footnote 6 Finally, in certain conditions, our vignette reflects the high-stakes dispute between the U.S. and Israel in the 1960s. Recently declassified archives reveal that despite the U.S.’ disapproval of Israel’s nuclear proliferation, neither country wanted to abandon their partnership. Thus, Washington exhibited leniency toward Israel due to its strategic significance (Burr and Cohen Reference Burr and Cohen2019; Cohen and Burr Reference Cohen and Burr2019).
Experimental findings
To assess the interaction between the human rights record and alliance status of the foreign state, we randomized two treatments. Human Rights is a binary variable equal to 1 if the state violates human rights, and 0 if it does not. Our second treatment, Alliance, is a binary variable set to 1 if the state has a military alliance with the U.S., and 0 otherwise. Following Tomz and Weeks (Reference Tomz and Weeks2020), we converted our dependent variable, support for war, into a binary variable, which is coded as 100 if the respondent thought that the U.S. military should attack, and 0 otherwise. This coding strategy enables us to interpret the average treatment effect as the percentage point change in support for war. We use the following OLS regression model to estimate the coefficients:
$${Y_i} = {\beta _0} + {\beta _1}Human\;Right{s_i} + {\beta _2}Allianc{e_i} + {\beta _3}Human\;Right{s_i} \times Allianc{e_i},$$
where
${Y_i}$
is an indicator of support for war, and
$HumanRight{s_i}$
and
$Allianc{e_i}$
are treatment indicators for human rights violations and alliance status, respectively. The outcomes of the PureSpectrum survey are presented in the main text, whereas the results of the first wave of the YouGov survey are reported in Appendix E.Footnote
7
We also pooled both surveys and reproduced the analyses in Appendix F. The results are consistent with those in the main text.
Table 1 presents the estimates from three linear regression models. All models consistently demonstrate that Human Rights significantly affect respondents’ support for military action. Model 1 indicates that if the state violated human rights, respondents, on average, were 10.6 percentage points more likely to support attacking it. This equates to a shift in support from 37.3% when the state respected human rights to 47.9% when the state violated human rights. This result supports our first hypothesis and successfully replicates Tomz and Weeks’ (Reference Tomz and Weeks2020) major finding. We also replicate Tomz and Weeks’ (Reference Tomz and Weeks2020) finding that human rights exert influence on public support for war by altering perceptions of threat and morality (Appendix D.2).
Table 1. Regression estimates of support for war (PureSpectrum survey)
Note: The table shows the estimates of the coefficients and standard errors from OLS regression. The dependent variable is coded as a binary variable that takes a value of 100 if the public approves of attacking the target country and 0 otherwise.
***p < 0.01.
**p < 0.05.
*p < 0.1.
Model 1 reveals a reduction in public support for war due to alliance partnerships, but this difference does not achieve statistical significance (p = 0.11). However, upon employing an alternative, preregistered measure of public support for war (5-point Likert scale), Alliance does indeed significantly diminish public support for war (p = 0.006) (Table S5). Overall, we find some preliminary evidence suggesting that participants responded to our treatment in the expected direction.
Our primary focus is on Model 2 in which we interact Human Rights with Alliance to examine whether the effect of human rights violations is contingent on alliance status. Contrary to our expectations, the estimate of the interaction term (
${\beta _3}$
) in Model 2 is positive but insignificant. This result shows that the effect of human rights violations on support for war does not change based on whether the perpetrator was a U.S. ally. Our finding remains robust after accounting for pre-treatment covariates, as stipulated in our preregistration (Model 3).
Figure 1a shows average public support for war under each condition. Average support for war under our baseline condition, a non-allied state engaging in human rights abuses, stands highest at 46.9%. Figure 1b demonstrates the differences in means between each treatment and the baseline condition (Non-Ally, Violates Human Rights). Compared to the baseline, support for war decreases by 8.6 percentage points when the state is not a U.S. ally but protects human rights (p < 0.0001) and decreases by 13.1 percentage points when the state is a U.S. ally and protects human rights (p < 0.0001). However, when participants were told that the state is a U.S. ally but violates human rights, the proportion of support for war decreased only marginally by 0.5 percentage points (p = 0.81), compared to the baseline condition. Therefore, we find little evidence that the U.S. public applies a double standard for human rights violations based on the U.S.’ strategic relationship with the perpetrator.
Figure 1. Impact of treatments on support for war (95% confidence intervals). (a) Plots the means of public support for war in percentage points. (b) Plots the percentage point difference in support for war between each treatment and the baseline condition (Non-Ally, Violates Human Rights). Error bars represent 95% confidence intervals.
Given the unexpected finding, we further investigate the lack of a double standard on human rights. The most compelling evidence of the lack of a double standard is that the results of the PureSpectrum survey are consistent with the findings from our initial YouGov survey. Both surveys reveal that respondents do not apply a double standard on human rights (Table S9). The result remains robust when we utilize a 5-point Likert scale to measure public support for war (Table S5) or restrict the analysis to participants who successfully passed the “mock vignette check” (MVC) (Kane, Velez, and Barabas Reference Kane, Velez and Barabas2023).Footnote 8 We investigate several alternative explanations for the null effect in Appendix D.4 but maintain our conclusion that the alliance status of human rights violators has a negligible effect on support for war.
While this finding is contrary to the expectations of our third hypothesis, it joins scholars who do not find evidence of a double standard on human rights in foreign policy. For instance, Koliev (Reference Koliev2020) finds that democracies are just as likely to name and shame fellow democracies as non-democracies for labor rights violations. Indeed, for some forms of human rights violations, states are more likely to reprimand their strategic partners (Terman and Byun Reference Terman and Byun2022).Footnote 9 Further, Allendoerfer (Reference Allendoerfer2017) shows that public support for punishing foreign aid recipients who violate human rights is the same regardless of the recipient’s strategic importance.Footnote 10
Conclusion
If an ally violates human rights, supporting policies to uphold human rights may come at the cost of pursuing strategic interests. This research article asks: does the public apply a double standard on human rights abuses based on the perpetrator’s alliance status? To answer this question, we utilized two experiments on the U.S. public to explore the trade-offs between protecting human rights and pursuing national interests when determining support for conflict.
We find that individuals are more likely to support conflict against human rights violators and in some specifications, are less likely to support conflict against allies. However, we find little evidence of a double standard: the effect of violating human rights on support for conflict is consistent for allies and non-allies. In other words, individuals weigh the importance of a foreign state’s human rights record equally for allies and non-allies.
Our paper contributes to our understanding of double standards for human rights in several important ways. By replicating and expanding Tomz and Weeks’ (Reference Tomz and Weeks2020) experiment, we add to the expanding literature on human rights and public support for war. Our results show an important pattern: approximately one-third of respondents support war against states that prioritize human rights protection, but support surges to a near majority against states that violate human rights. Since leaders prefer foreign policies with public support (e.g., Tomz, Weeks, and Yarhi-Milo Reference Tomz, Weeks and Yarhi-Milo2020; Chu and Recchia Reference Chu and Recchia2022; Lin-Greenberg Reference Lin-Greenberg2021), these differences could exert substantive influence on elites’ decisions to engage in conflict.
Further, by demonstrating that the U.S. public does not exhibit double standards on human rights, this paper contributes to our understanding of the U.S. public’s commitment to safeguarding human rights irrespective of strategic ties (Koliev Reference Koliev2020; Allendoerfer Reference Allendoerfer2017). For example, Americans, particularly Democrats, have become increasingly concerned about Israel’s alleged human rights violations in Gaza (El Baz and Smeltz Reference El Baz and Smeltz2023, Reference El Baz and Smeltz2024; Data for Progress 2023), and protests on college campuses have sparked nationwide (Thompson and Beck Reference Thompson and Beck2024). The influence of public opinion could elucidate the Biden administration’s condemnation of Israel’s human rights violations (U.S. Department of State 2024) and its decision to temporarily halt some weapons shipments to Israel due to humanitarian concerns in May 2024 (Hudson Reference Hudson2024).
We acknowledge several important caveats regarding our results. First, our experimental design, which focuses only on proliferating democracies, does not allow us to investigate whether the lack of a double standard applies to non-democracies. Our results should be treated with caution regarding how well they generalize to a non-democratic context. Future experiments should investigate whether this lack of a double standard can be sustained in an autocratic context. In addition, there are multiple conceptualizations of double standards of human rights. Future research may investigate how the public reacts to diverse manifestations of double standards, encompassing not only those related to alliance partners but also disparities based on race and ethnicity.
Third, we recognize that when an ally pursues nuclear proliferation without U.S. consent, it may raise concerns about the ally’s reliability. Future experiments could examine the benefits that allies bring to the U.S. For example, researchers could investigate whether the public supports criticizing allies with poor human rights records, even when such actions jeopardize strategic interests, such as stationing troops or obtaining overflight permissions. Finally, our vignette investigates the double standard in a security-driven scenario where a state is proliferating nuclear weapons; therefore, caution should be applied regarding how generalizable our findings are for scenarios where the public might want to punish human rights violations. Future experiments could explore whether the U.S. public applies double standards for humanitarian intervention when an ally is abusing human rights.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/XPS.2024.19
Data availability
The data code and any additional materials required to replicate all analyses in this article are available at the Journal of Experimental Political Science Dataverse within the Harvard Dataverse Network, at: https://doi.org/10.7910/DVN/6GMO9C.
Acknowledgments
The authors thank Mark Souva, Kai Quek, Carlisle Rainey, Brad T. Gomez, Holger Kern, and Quintin Beazer for their feedback. The authors also extend their gratitude toward participants at Florida State University’s international relations workshop and panelists at the International Studies Association conference as their comments improved the manuscript. The authors especially thank Eddy Yeung for generously sharing his survey time. Finally, we thank our editor and three anonymous reviewers for their valuable comments.
Funding
This research was funded in part by the Political Science Department of Florida State University as part of its 2022 Cooperative Election Study Team Module.
Competing interests
No potential conflict of interest was reported by the authors.
Ethics statement
The research adheres to American Political Science Association’s (APSA) Principles and Guidance for Human Subjects Research. Approval was obtained from the Institutional Review Board at Florida State University (Protocols: STUDY00004425 and STUDY00003350). Informed consent was obtained from all participants included in the study. The authors did not compensate participants directly. Participants were compensated for their participation by the survey panel provider. The studies were preregistered prior to being conducted at Open Science Framework. The pre-registrations can be viewed here:
PureSpectrum: https://osf.io/wm8gy?view_only=cfaa9a4fac1d4c68a56a89c331cf470e
YouGov: https://osf.io/9wczj?view_only=3e3a079c04e84d898a65dad4fd879525
Open access
