Whose War is it Anyway? Explaining the Black-White Gap in Support for the Use of Force Abroad

Abstract

Building on long-standing work on a “gender gap” in war support, this article documents a recurring “race gap” in which Black Americans display less enthusiasm for war than their White counterparts. We compile time-series data on public opinion during the Iraq War collected from over fifty national polls and successive waves of the Cooperative Congressional Election Study to assess potential explanations for the gap. We show that concerns about casualties best explain lower levels of support for war among Black Americans. Feelings of political alienation and preferences for domestic spending serve as more salient contributors to Black disapproval of war during the George W. Bush years. Meanwhile, having a family member in the military does not explain lower Black support for war. Black antiwar rhetoric suggests that our casualty sensitivity and alienation findings stem from linked fate attitudes and concerns about fairness and “justness” of the war effort among Black Americans.

In April 1917, the United States officially entered into World War I. Members of the US Congress overwhelmingly supported the war declaration, with votes of 82–6 in the Senate and 373–50 in the House of Representatives. Miles away in the town of Friars Point, Mississippi, some members of the local Black American population were less excited. That same month, a circular in the town called out to local Black citizens with a word of caution:

Young negro men and boys what have we to fight for in this country? Nothing. Some of our well educated negros are touring the country urging our young race to be killed up like sheep, for nothing. If we fight in this war time we fight for nothing. Rather than fight I would rather commit self death.

Signed by a Negro Educator.

[P.S.] Stick to your bush and fight not for we will only be a breastwork or a shield for the white race. After war we get nothing.¹

Instances of Black concern about foreign war have continued, even to the present day. During World War II, US Black newspapers made the point that victory in the struggle against fascism abroad was equally as important as victory in the struggle against discrimination and racism at home,² calling for a “Double V” victory campaign.³ A generation later, Dr. Martin Luther King Jr. channeled rising opposition to the Vietnam War within the Black American community, condemning the hypocrisy of the United States in sending Black soldiers abroad to die for the liberties they could seldom enjoy at home by saying: “We were taking the black young men who had been crippled by our society and sending them eight thousand miles away to guarantee liberties in Southeast Asia which they had not found in southwest Georgia and East Harlem.”⁴ After Vietnam, race continued to be intertwined with antiwar sentiment. Protests against the 1991 Gulf War featured prominent Black leaders such as Coretta Scott King and Jesse Jackson Sr.⁵ More recently, in September 2013, public polls indicated limited support among African Americans for a potential military strike in Syria, and a large number of Black congressional representatives began to express skepticism. Congressional Black Caucus (CBC) chairwoman Marcia Fudge urged the caucus to “limit public comment” until after a briefing from the White House, worried that the reliably antiwar CBC bloc would hamstring President Barack Obama’s credibility.⁶

Surprisingly, this history has attracted little attention in international relations (IR) scholarship on public support for the use of military force. Whereas long-standing observations of a “gender gap” in support for war have given rise to an extensive research program⁷—and amid a growing acceptance of the role of mass attitudes in shaping foreign policy decision making⁸—examinations of race and ethnicity in IR have been largely confined to the study of ethnic conflict and civil war. This is striking. The results from decades of social science research on the use of force abroad show significantly different levels of support for war between Black and White respondents in the United States, with Black respondents expressing much less enthusiasm for US military extroversion.⁹ Across myriad conflicts, time periods, and geographical regions in the United States, the “race gap” in support for the use of force has surfaced repeatedly. Although Black support for war has fluctuated across conflicts, the gap persists.

However, the plethora of studies that have discovered a race gap were often not necessarily looking for one. In fact, remarkably little empirical research has taken Black-White differences in support for war as its main topic of focus.¹⁰ Instead, studies of attitudes toward particular conflicts or overall foreign policy have relegated racial identities to the background—frequently controlled for in individual-level analyses, seldom analyzed in their own right. At times, the race gap is only briefly mentioned in footnotes, or not at all. Furthermore, among the handful of studies that have taken racial differences in attitudes toward the use of force abroad as their main question, researchers have not tested a broad set of explanatory variables simultaneously. Though some scholars have proposed theories regarding why the race gap exists, there is a need for cross-temporal evidence that adjudicates among competing hypotheses to explain its persistence in modern public opinion.

Through this study, we aim to fill gaps in the literature about race and foreign policy attitudes by systematically considering available data. We make a conscious and intentional effort to extend insights specific to Black public opinion to the topic of the use of force. Throughout the study, we focus on how Black Americans’ experiences within US domestic politics and society shape attitudes toward US actions abroad. Our study asks several questions regarding Black preferences toward the use of force and Black-White differences in support for war. Do empirics confirm a continued hesitance about the use of force among Black Americans in the context of more recent US military action overseas? What variables explain Black preferences regarding war? Furthermore, how have preferences regarding the use of force among Black respondents changed between presidential administrations in recent years? To answer these questions, we examine support for the US war in Iraq across the George W. Bush and Obama administrations, employing data from over fifty national public opinion polls and repeated national Cooperative Congressional Election Study (CCES) polling.

In this article, we present the most comprehensive study in political science thus far on the Black-White race gap in support for war. We find that a higher level of casualty sensitivity among Black Americans best explains consistently lower levels of support for war. Meanwhile, political alienation plays a significant role in Black-White war support differences, but alienation is vastly more prevalent among Black respondents during the Bush years. Similarly, preferences for domestic spending contribute to Black disapproval of the war during the George W. Bush years, though the effect is considerably smaller. Our results show that who sits in the Oval Office can change the relationship between race and the use of force. Alternatively, our analyses do not lend support to the idea that being more likely to have a family member in the military accounts for lower Black support for war. We offer historical evidence to show how linked fate ideology, as well as concerns about fairness of treatment of Black Americans at home and justness of wars abroad, help explain our findings.

Toward a Research Agenda on the Race Gap in War Support

The relationship between race and politics in the United States has led to a significant divergence in the political views of Black Americans and other racial groups across a broad range of issue areas, as documented by a vibrant research stream within the Race and Ethnic Politics (REP) literature.¹¹ However, Melissa V. Harris-Lacewell singles out foreign policy, war, and international affairs as “glaring examples” of the limits to political science scholarship on Black American public opinion.¹² Although earlier work has outlined general trends in Black foreign policy attitudes from a historical, descriptive perspective,¹³ and some recent mainstream political science work has noted the distinctive effects of race on IR,¹⁴ much work remains. The time is ripe for a broader research agenda on African American preferences regarding certain foreign policies, particularly those entailing the use of force abroad.

We envision this research agenda as similar to the now-robust literature on the gender gap in support for war. The sub-subfield of IR scholarship on gender has made substantial headway in establishing gender differences in support for the use of force abroad¹⁵ and exploring the mechanisms that generate these differences.¹⁶ Contemporary political scientists can now be relatively secure in speaking of a “gender gap” in support for war, even if the specific causal mechanisms operating at the individual level remain up for debate.

By contrast, race and ethnicity are frequently relegated to mere control variables in analyses of mass attitudes toward war and conflict,¹⁷ save in the discussion of particular anomalous findings—which appear with some regularity. John Zaller, for example, pauses an empirical demonstration of his two-message reception-acceptance model of public opinion to puzzle over shifting Black attitudes on the Vietnam War,¹⁸ as initially high support for the conflict gives way to a greater inclination by Black survey respondents toward a liberal, antiwar message on Vietnam after 1966. Racial differences (White to non-White) are also associated with substantially greater casualty sensitivity in results reported in a 2006 article about Iraq War support by Christopher Gelpi and colleagues, but the authors do not spend much time discussing these differences.¹⁹

Scattered across the literature, however, a handful of studies examine the Black-White race gap in support for the use of force directly. Taken together, this work indicates that meaningful, persistent differences in racial attitudes toward military action do exist in the United States—and that they warrant further study. In an analysis of between-group differences in support for the Gulf War, Clyde Wilcox et al. find that Black Americans were much less likely to support military action throughout the period under consideration, particularly following President Bush’s veto of the Civil Rights Act of 1990.²⁰ Likewise, articles by Henry Richardson and Robert Smith and Richard Seltzer focus on the significantly less prowar attitudes of Black Americans during the course of the Gulf War.²¹ Coauthors Miroslav and Donna J. Nincic find both racial and gender differences in US attitudes toward the use of force in contexts such as Vietnam, Operation

Table 1. Notable studies identifying racial differences in US public support for the use of force abroad. Relevant page numbers for books cited in the table are included in the footnotes.

Publication	Outcome, Year(s) of Data Collection	Black-White Gap? Method?	Stat. Signif.: Bivariate?/ w. Controls?
Verba et al., “Public Opinion and the War in Vietnam”^22	Support for escalation in Vietnam War, 1966	Yes, diff-in-means	Yes/NA
Hamilton, “Research Note on the Mass Support for ‘Tough’ Military Initiatives”	Support for escalation in Korean War, 1952	Yes, poll results
	Support for a “strong stand” in Vietnam War, 1964	Yes, poll results
Robinson and Jacobson, “American Public Opinion about Vietnam”^23	Support for escalation in Vietnam War, 1966	Yes, lit review
	“Extreme dove” war orientation during Vietnam War, 1967	Yes, poll results
Hahn, “Correlates of Public Sentiments about War”^24	Referendum votes against Vietnam War, 1966–68	Yes, bivariate correlation	Yes (in 1/6 cities) / NA
Rosenberg et al., Vietnam and the Silent Majority^25	Support for Vietnam War, multiple years	Yes, poll results
Wright, “Working Class, Authoritarianism, and the War in Vietnam”	Support for escalation in Vietnam War, 1964	Yes, poll results
	Support for escalation in Vietnam War, 1968	Yes, poll results
Mueller, War, Presidents, and Public Opinion^26	Was Korean War a mistake, 1950–53	Yes, poll results
	Was Vietnam War a mistake, 1965–71	Yes, poll results
Wilcox et al., “Group Differences in Early Support for Military Action in the Gulf”^27	Support for military action in the Persian Gulf (Gulf War), 1990	Yes, OLS reg.	NA/Yes
Mueller, Policy and Opinion in the Gulf War^28	Support for Gulf War, 1990–91	Yes, poll results
Gartner and Segura, “Race, Casualties, and Opinion in the Vietnam War”^29	Support for Vietnam War, 1965–68	No, logit reg.	NA/No
	Support for Vietnam War, 1969–72	Yes, logit reg.	NA/Yes
Smith and Seltzer, Contemporary Controversies and the American Racial Divide^30	Support for Gulf War, 1990–91	Yes, poll results
Nincic and Nincic, “Race, Gender, and War”^31	Support for Korean War, 1952	Yes, Tau-b/probit	No/Yes^32
	Support for Vietnam War, 1966	Yes, Tau-b/probit	No/No
	Support for Vietnam War, 1972	Yes, Tau-b/probit	Yes/Yes
	Support for Desert Storm/Shield, 1990	Yes, Tau-b/probit	Yes/Yes
Harlow and Dundes, “‘United’ We Stand”^33	Support for attack on Afghanistan in response to 9/11, 2001	Yes, chi-square difference test	Yes/NA
Rohall and Ender, “Race, Gender, and Class”^34	Support for US invasion of Iraq, 2003	Yes, one-way	Yes/Yes
		ANOVA/logit
Baumgartner et al., “Clash of Civilizations?”^35	Support for US invasion of Iraq, 2005/2006	Yes, probit reg.	NA/No
Burris, “From Vietnam to Iraq”^36	Support for all major US conflicts between Vietnam and Iraq, multiple years	Yes, bivariate gamma coef.	Yes/NA
Holsti, Public Opinion and American Foreign Policy^37	Support for the Persian Gulf Conflict, 1991	Yes, poll results
	Support for Iraq War, 2003	Yes, poll results
Tate, What’s Going On?^38	Support for Iraq War, 2003	Yes, poll results
Allison, “Race, Gender, and Attitudes toward War in Chicago”^39	Support for US invasion of Iraq, 2008	Yes, logit reg.	Yes/No

Desert Shield, and Operation Desert Storm.⁴⁰ David E. Rohall and Morten G. Ender find diminished Black American support for the Iraq War at its outset even among US military officers and enlisted soldiers—a sample where we might otherwise expect self-selection to eliminate any divergence in views.⁴¹ Katherine Tate discusses the Black-White gap in support for the Afghanistan and Iraq Wars, and potential military strikes against Iran and North Korea, in the context of changing orientations toward isolationism, multilateralism, and US domestic politics.⁴² Meanwhile, historian Kimberley Phillips Boehm’s War! What Is It Good For? paints an intricate picture of the relationship between Black political activism, military involvement, and views toward war.⁴³

Yet, for the most part, explanations for findings on racial differences in support for war within the political science literature have tended toward the ad hoc or the contingent. Zaller explains shifting Black American attitudes toward the Vietnam War as a function of declining enthusiasm for Lyndon B. Johnson’s presidency, with Johnson’s championing of civil rights legislation fading into the past.⁴⁴ However accurate an observation, it seems incomplete to develop general theory to explain attitudes for the majority of the population and then explain away apparent group-based differences as a function of particular historical events—especially given the longstanding race gap in war attitudes.

Certainly, racial or ethnic gaps in support of the use of military force in particular conflicts have exhibited greater variance than gender-based differences. Scott Sigmund Gartner and Gary M. Segura find that Black support for the Vietnam War diverged in a statistically significant manner only during the second half of the war, and they further demonstrate that the number of casualties from a respondent’s hometown had a similar impact on war support for both Black and White respondents.⁴⁵ Val Burris documents a recurring race gap over time,⁴⁶ yet with a reduced or even reversed gap in support for Clinton-era engagements such as those in Somalia or Haiti.⁴⁷ Using an intersectional approach, Rachel Allison also finds a Black-White gap in attitudes toward the Iraq War among Chicago residents.⁴⁸ However, when gender is controlled for, she discovers that the race gap persists only for female respondents.⁴⁹ Additionally, other demographic differences might drive an apparent race gap in support for war, such as lower incomes among Black versus White Americans, or the overwhelming association of Black American voters with the more liberal (and frequently more dovish) Democratic Party.⁵⁰

Still, the race gap in war support has appeared and reappeared in evidence spanning decades of American foreign military engagements, even in studies that have controlled for relevant demographic variables (see Table 1). Thus, our baseline expectation regarding between-group differences is:

Hypothesis 1 (H1).

Black respondents are less supportive of war than White respondents, even when controlling for other demographic characteristics.

Casualties and War Effort

John E. Mueller’s 1973 book, which followed the collapse of public support for the Vietnam War, emphasized public aversion to mass casualties as a clear constraint on US presidential power, with citizens potentially punishing officials at the polls for imposing direct (and visible) costs on US society through the use of force abroad.⁵¹ Further work built upon these initial findings, indicating that factors such as citizen assessments of war costs might vary with a given conflict’s perceived aims. For example, work by Bruce W. Jentleson and Rebecca L. Britton has posited greater public acceptance for multilateral or humanitarian military action.⁵²

Here, we test the hypothesis that concerns about casualties, in particular, might account for much of the reason that Black Americans are wary of war. Following the Vietnam War, and amid other conflicts, several scholars expressed the perception that Black soldiers were more likely to die during war as a reason for Black Americans’ aversion to the use of military force.⁵³ Likewise, in 1967, Sidney Verba et al. speculated that the race gap in support for the use of force might stem from perceptions of a disproportionate war effort by Black Americans.⁵⁴ However, findings by Gartner and Segura brought this postulation under greater scrutiny.⁵⁵ Using Vietnam War data, Gartner and Segura investigated the effects of “proximate casualties” from the same racial group. Their findings used logistic regressions to demonstrate that both Black and White respondents surveyed across time during the Vietnam War reacted similarly to the weekly number of proximate casualties, defined as casualties of service members from the same geographic county as survey respondents. During the first part of the war, for both racial groups, a higher number of casualties in the previous week was associated with a similar drop in support for the war. Moreover, in their study, Black respondents reacted similarly to proximate Black and White casualties. Some scholars have interpreted these findings as evidence that Black Americans are not any more sensitive to casualties than White Americans.

The Gartner and Segura piece is based on a theory that assumes Bayesian updating: in short, that proximate deaths during combat in Vietnam should shape opinion of the war for Americans back home. Yet Black American deaths need not be geographically proximate for Black Americans to feel that the Black community as a whole will bear a disproportionate burden during wartime. If driven by this political belief, then Black concern about casualties might be aggravated by news of real-time battle deaths, but it should not solely depend upon it. See, for example, the results from a USA Today poll in December 1990 in Figure 1.⁵⁶ In the run-up to the Gulf War, Black respondents were more than twice as likely as White respondents to agree that: “If there is a war with Iraq, black Americans will bear an unfair burden in the fighting of the war.” Since this poll took place before fighting began, respondents could not have based their answers on any real-time data. Instead, they made their assessments based on preexisting beliefs about the war burden.

Figure 1. Percentage from a December 1990 USA Today poll in agreement with the statement that Black Americans would face an unfair burden in a war in the Persian Gulf.

The idea that war leads to an “unfair burden” on Black American communities is rooted in history. Interestingly, Verba et al. wrote as if surprised by a “high level of self-consciousness among the Negro population that allows the connections to be made between their social position and the war.”⁵⁷ Yet in the 1960s, when they were writing their article, many Black Americans were concerned about real racial inequities in the number of battlefield casualties during the early years of the Vietnam War. Black enlistment and deployment patterns during this period were high.⁵⁸ In fact, Black Americans comprised approximately 11% of the US population but almost 25% of US combatants killed in Vietnam in 1965. By 1967, the percentage of Black battle deaths had decreased to 12.7%, but perceptions within Black communities that Black Americans were more likely to die in the war persisted.⁵⁹

The belief that Black soldiers are more likely to die in US wars has been aided by the disproportionate representation of Black Americans in the contemporary, all-volunteer US military.⁶⁰ Though the racial proportions of war casualties are no longer as skewed as they were at the beginning of the Vietnam War, the percentage of active-duty military forces that was Black was 17% in 2015, compared with approximately 13% in the national population. Comparatively, non-Hispanic White Americans made up 60% of the active-duty military while representing approximately 62% of the national population.⁶¹ Hence, it is plausible that Black respondents might object to war because of the disproportionate number of Black soldiers that would be put in harm’s way.⁶²

Hypothesis 2 (H2).

Black respondents are less likely to support war due to being more likely to exhibit concern about the potential for war casualties than White respondents.

The overrepresentation of Black Americans in the US military also raises the possibility of a more direct explanation. Individuals either in the military or with family in the military might exhibit less support for war out of concern for the personal risks they or their family member(s) face.⁶³ Though this idea is similar to the proposition that disproportionate casualty numbers or war effort drive the race gap, it identifies an individual-level rather than group-level mechanism: direct concern for a member of one’s family. If this is the case, having a family member who has served in the US military may help explain the race gap in support for war.

Hypothesis 3 (H3).

Black respondents are less likely to support war due to being more likely than White respondents to report serving in, or having a family member in, the US military.

Guns-and-Butter Trade-Offs

Beyond the human costs of war, Black Americans might also be more concerned that spending on guns (defense) detracts from spending on butter (domestic welfare programs). Black Americans have historically exhibited greater race- than class-based solidarity in broad attitudes toward redistribution, although this has begun to shift in recent years.⁶⁴ Accordingly, Black Americans might oppose the use of force because they see government budget expenditures on foreign wars as direct threats to domestically focused social spending.⁶⁵ In his 1967 “Beyond Vietnam” speech, Martin Luther King Jr., referring to the antipoverty efforts of President Johnson’s Great Society, spoke of growing despair that “America would never invest the necessary funds or energies in rehabilitation of its poor so long as adventures like Vietnam continued to draw men and skills and money like some demonic, destructive suction tube.” He continued: “So I was increasingly compelled to see the war as an enemy of the poor and to attack it as such.”⁶⁶ Given this line of thought, empirical examination of the link between concern for domestic issues and support for foreign wars among different racial groups is warranted.

Hypothesis 4 (H4).

Black respondents are less likely to support war due to being more likely than White respondents to prioritize domestic spending over military spending.

Political Alienation

Upon discovering both a gender gap and a race gap in support for the Vietnam War, Verba et al. attempted to explain both by including a variable gauging how informed respondents were; however, when statistical analyses showed that the information variable could explain neither the gender nor the race gap, the authors remarked:

The racial patterns are particularly interesting. This does not appear to be a social class phenomenon in disguise, but a difference patterned by the racial distinction. The ad hoc explanation is obvious; at the present moment the American Negro is alienated from American society and unimpressed by arguments that our commitment is necessary to preserve freedom and justice for ourselves and others in Southeast Asia.⁶⁷

Several studies have echoed the idea of alienation that Verba et al. present without offering further analysis.⁶⁸ On the other hand, Miroslav and Donna Nincic operationalize alienation via two National Election Studies questions asking respondents to rate their agreement with the following statements: “I don’t think public officials care much about what people like me think,” and “People like me don’t have a say about the government.” Though they confirm via their research that Black respondents are more likely than White respondents to feel politically alienated, they find that the alienation variable does not fully absorb the race gap, and its effects actually dissipate in later conflicts. The authors note, “The decline of alienation’s significance during the Gulf operations, and the concomitant increase of race’s impact, is noteworthy, suggesting that race may be playing a more direct role, i.e., one less mediated by alienation.”⁶⁹ Roxanna Harlow and Lauren Dundes offer qualitative, focus group–based examples of alienation as expressed by Black students.⁷⁰ Somewhat related to the issue of alienation is a phenomenon that Burris explains. Burris argues that a weaker identification with national symbols and leaders among American minorities suppresses the “rally-’round-the-flag” effect among non-White groups.⁷¹

We presume that alienation will negatively affect Black support for war so long as Black Americans are more politically alienated than their White counterparts. During Barack Obama’s tenure as the first Black president of the United States, this was likely not the case. Previous work has shown that greater support for policies among Black Americans during the Obama administration was linked to increased political trust and decreased alienation.⁷² When Obama was president, race and racial attitudes predicted support for even nonracial policies, such as healthcare: Americans holding anti-Black stereotypes supported Obama’s policies less, and Black Americans, in particular, supported Obama’s policies more.⁷³ In fact, under Obama, Black political trust surpassed levels of political trust among White Americans.⁷⁴ Though we test for differences in relationships between the Obama and Bush years for multiple explanatory variables, we predict that Obama’s tenure was particularly likely to have reversed racial trends in political alienation.

Hypothesis 5 (H5).

When Black respondents are more politically alienated than White respondents, alienation leads the former to express less support for war.

Data and Methodology

Our article uses the best available polling and survey data to adjudicate between the four potential rationales for the race gap in war support proposed above for one of the most major wars in recent history: the US war in Iraq. Using Iraq War data for this study allows us to gauge the extent to which the race gap in support for war has persisted into recent years. Additionally, frequent polling during the conflict allows us to take advantage of a large number of regular polls with sufficient samples of both Black and White respondents—a data feature that was not as reliable during the time periods when other wars occurred. Data regarding opinions about this war have the added benefit of spanning both the George W. Bush administration (2001–2009) and the Barack Obama administration (2009–17).

In this article, we use an oversimplified racial coding system including only the designations of “Black” and “White.” We subset surveys to include only non-Hispanic Black and White respondents so as to make clear the Black-White gap in attitudes. This approach has limitations. First, there are numerous other racial groups in the United States equally worthy of study—a point we return to in the conclusion. Second, the terms “White” and “Black” group together people of myriad historical backgrounds, ethnicities, and nationalities. We acknowledge these limitations and proceed with the humble assertion that these simplistic categories still help us capture—however imperfectly—a better picture of the racial dynamics displayed within our data than would any other available variables.

To address the question of whether casualty sensitivity might explain part of a racial gap in support for war, we identify eighteen polls conducted by ABC between 2003 and 2008 (N = 17,299, 13.8% identifying as non-Hispanic Black).⁷⁵ Each of these polls asked respondents about their level of support for the Iraq War.⁷⁶ More importantly, each survey asked a repeated question that we use to indicate casualty sensitivity: “Do you think the United States should keep its military forces in Iraq until civil order is restored there, even if that means continued U.S. military casualties?”

We interpret opposition to keeping US military troops in Iraq as concern over the possibility of further US casualties, or casualty sensitivity. Though it is possible that responses to this question might simply reflect respondents’ direct views of the war, this is not strictly the case in the data: about 75.3% of respondents presented “consistent” responses to these questions—that is, the Iraq War was (not) worth fighting and the United States should (not) keep forces in Iraq. Meanwhile, around 6.5% supported the war but did not support keeping US troops in-country given the risk of casualties, and 18.2% did not support the war but believed the United States should keep troops in-country even if it meant more casualties. We present (consistent) results for our casualty sensitivity analyses in the online appendix using an alternative measure that focuses on whether there has been an “acceptable or unacceptable number of U.S. military casualties in Iraq.”

In the online appendix, we also identify one survey from March 2002 that asks about support for the Afghanistan War, as well as concern with casualty levels, which allows us to ensure that our findings are not solely specific to the Iraq War.⁷⁷ There are no comparable polls in the Roper polling database from the era of the Obama administration that ask about casualty levels. However, we do not expect that the president in office would change levels of concern about casualties during war.

To determine whether alienation contributes to Black aversion to war, we gather a total of 39 surveys administered by several different polling companies during the Bush administration (N = 33,629, 8.7% Black) and 11 administered during the Obama administration (N = 3,503, 7.9% Black).⁷⁸ Bush administration polls for this section were administered from 2002 to 2007, whereas Obama administration polls occurred between 2009 and 2011. We use polls from different companies because each company asked a similar question about whether the current president “‘cares about’/’understands’ people like you.”⁷⁹ Here, respondents indicating that the president cared about or understood them were coded as not alienated; respondents indicating that the president did not care about or understand them were coded as being alienated. While views on the president are relevant to foreign policy attitudes, given the presidency’s outsized role in foreign policymaking, they reflect a narrower measure of political alienation than more traditional measures of “trust in government.”⁸⁰ Views on particular presidents may diverge from broader views on the US political system. Still, as only questions related to views on the president were regularly available in polling from this time period, we rely on this measure in our present analysis while accounting for its potential shortcomings.

To determine the effects of preferences for domestic spending and the effects of having a family member in the military on Black-White war support, we use five waves of CCES data. The 2006 and 2008 CCES waves took place during the Bush administration (N = 43,295, 6.7% Black), and the 2010, 2012, and 2014 waves took place during the Obama administration (N = 143,277, 14.1% Black). All five waves asked about military family members, support for domestic and military spending, and support for the Iraq War.⁸¹

Our third explanatory variable is constructed from survey questions about whether respondents value domestic spending over military spending when asked about where they would cut funding to balance the budget. In our view, respondents who said they would cut the defense budget before cutting domestic spending could be said to prioritize domestic spending. Finally, we construct an indicator of whether a respondent is in a military family based on whether they or a family member currently serves in the US military.

Across all the surveys used for this project, we code demographic control variables into standardized categories of gender, age, education level, income level, and partisanship. Coding of the demographics is explained in further detail in the online appendix.

For our analyses, we first perform ordinary least squares (OLS) regressions, including poll fixed effects and demographic controls.⁸² In the majority of our regressions, we use support for war as the outcome variable. However, a series of regressions also use the explanatory variables (casualty sensitivity, alienation, family in the military, or spending preferences) as the outcome variable to assess whether Black respondents differ from White respondents in their orientations to these variables. Given that different political surveys in our analysis asked different questions, the data only allow for examining the four potential explanatory variables in separate regressions.⁸³

We also run mediation analyses to determine the extent to which each of the explanatory variables mediates the relationship between race and support for war in each sample. Mediation analysis helps researchers go beyond assessing the correlation between an independent variable and an outcome variable by determining “the process in which the treatment causally affects the outcome.”⁸⁴ In our mediation analyses, we take the binary race variable—in which “treatment” is indicated by being African American—and examine the extent to which each of our explanatory variables acts as a mediating variable that accounts for the effect of race on support for war.

Other scholars, such as Maya Sen and Omar Wasow, have raised concerns that indicators such as race cannot serve as causal or treatment variables.⁸⁵ These scholars argue that researchers should not take regression results showing significant “effects” of a race variable as causal explanations because race indicates a multiplicity of underlying factors and assignment occurs at birth. They encourage scholars to consider which factors might be associated with race, and thus to explore the underlying factors that motivate racial differences. This is very much the spirit of the current analysis, which aims to move beyond demonstrating a racial gap in support for war by explaining it. Nevertheless, in line with Sen and Wasow’s suggestions, we also take the extra steps of performing within-sample analyses of only Black and only White respondents so as to examine whether the relationship between the explanatory variables and support for war is different within these two groups separately.

Results

The Race Gap in Support for War

Across all of the analyses performed in this study, Black respondents expressed less support for war than did White respondents, regardless of the sample of surveys used or the president in office (see Figure 2).⁸⁶ This gap is persistent across time, although it narrows during the later years of the Obama administration. Table A4 in the online appendix reports the results of linear probability models that estimate support for the Iraq War for each sample, first using only race and poll fixed effects, then including other demographic controls. Regression models show a consistently negative and significant coefficient on the “Black” variable, indicating that Black respondents were less likely to support the Iraq War (see Figure 3). Analyses in the online appendix show the same persistent race gap in support for the war in Afghanistan (Figure D1, Table D1). These results lend support to H1 in finding a consistent race gap even when controlling for relevant demographic variables.

Figure 2. Support for the Iraq War by race, 2002–14.

Figure 3. Coefficient plot, race and support for the Iraq War.

These results also indicate, however, that a substantial portion of the between-group gap in support for war can be explained by various demographic controls, particularly under the Obama administration (Table A4 in the online appendix). While full controls reduce the coefficient on “Black” by approximately 50%–70% for polls conducted during the Bush administration, controls reduce the effect size even further during the Obama years. The largest effect sizes across our regressions are those of party identification.⁸⁷ Still, the association between race and support for the Iraq War is both consistently signed and one of the largest effect sizes across our full samples. Though we likewise observe a consistent “gender gap” across regressions, for all except the CCES-Obama sample the magnitude is far smaller than that of the “race gap.” Furthermore, in most samples, the magnitude on “Black” is comparable to the effect size of having a postgraduate education (relative to high school education or less). More education is generally associated with a lower likelihood of supporting the Iraq War. By contrast, higher income brackets are consistently associated with a greater likelihood of support, albeit typically with smaller effect sizes than that of the race gap. Age categories are not consistently associated with support for the Iraq War.

Casualty Sensitivity

OLS regressions provide evidence in support of H2 (casualty sensitivity decreases Black support) but no evidence for H3 (family member in the military decreases Black support). Black respondents were consistently more likely to express concern about potential casualties resulting from US troop presence in Iraq by around 16.4%–20.6%, even when controlling for demographic variables.⁸⁸ Only changes in partisanship affect casualty sensitivity to a similar or greater degree (Table A5 in the online appendix, Model 1). Casualty sensitivity is in turn negatively associated with support for the war among both Black and White respondents.⁸⁹ Figure 4 and subsequent corresponding figures show coefficients on the independent variables (IVs) of interest for the three regressions used to provide these results; moreover, the figures also list the dependent variable (DV) and sample specification used in each regression.⁹⁰

Figure 4. Coefficient plot, race and casualty sensitivity.

Mediation analysis provides evidence that a substantial amount of the association between race and support for war is mediated by the casualty sensitivity variable. Figure A1 in the online appendix plots the effect sizes for the average causal mediation effect (ACME, defined here as the effect of being Black on support for war that works through the casualty sensitivity variable) and the average direct effect (ADE, the effect of being Black on support for war that operates through anything other than casualty sensitivity) as compared to the total effect of being Black on support for the war in Iraq.⁹¹ The mediated effect constitutes between 38% and 50% of the total effect at 95% confidence intervals (proportion mediated not pictured), with both the ADE and ACME as negative.⁹²

Black respondents were slightly more likely to report that they or a family member were serving in the US military in CCES samples during the Bush (≈ 2.3%–5.2%) and Obama (≈ 2.8%–3.8%) administrations (Figure A2 in the online appendix), despite controlling for demographic variables.⁹³ However, we find that members of both Black and White military families were, if anything, more supportive of the Iraq War.⁹⁴ Accordingly, the mediated effects (Figure A1 in the online appendix) are substantively nonexistent.

Spending Preferences

We observe partial evidence in support of H4 (guns-vs.-butter trade-off). CCES responses regarding domestic spending during the Bush administration help explain the race gap in war support: Black respondents prioritized domestic spending (over defense spending) slightly more than White respondents (by ≈ 0.5%–4.5%), and Black and White respondents who prioritized domestic spending were markedly less likely to support the Iraq War (see Figure 5).⁹⁵ Mediation analysis (Figure A3 in the online appendix) suggests that around 4%–18% of the total effect is driven by the association between race and views on domestic spending. However, during the Obama administration, Black respondents were slightly less likely than White respondents to prioritize domestic spending (by ≈ 3.3%–4.9%) after controlling for demographic variables.

Figure 5. Coefficient plot, race and spending preferences.

Other variables are consistently signed across the two samples. Respondents who were male, identified as independent or Republican, and those aged thirty or over were all less likely to prioritize domestic spending over defense spending. As we might expect, those in the lowest income bracket (average annual household income of $30,000 or less) were more likely to prioritize domestic spending. Those with a college education (associate degree or higher) were also consistently more likely to prioritize domestic spending (Table A6 in the online appendix).

Political Alienation

We find partial support for H5 (political alienation). Our findings suggest that alienation can play a role in explaining the race gap in support for war within the United States, albeit one that is highly dependent on political context and leadership. As expected, feelings of political alienation moved in different directions for Black respondents during the Bush and Obama years. While an average of 81.0% of Black respondents indicated feelings of political alienation in surveys given during the Bush administration, this figure dropped to just 7.4% for surveys given during the Obama administration: a lower percentage than among White Democrats.⁹⁶

Under the Bush administration, Black respondents were 11.1%–14.7% more likely to be politically alienated from US leadership (see Figure 6), and “alienated” Black and White respondents were each less likely to support the Iraq War.⁹⁷ Political alienation predicted opposition to the Iraq War for both White and Black respondents more than demographic variables. The effect size of alienation on war support for White Americans was greater in magnitude (more negative) than the corresponding effect size for Black respondents. However, under President Bush, 81.0% of Black respondents expressed feelings of alienation, compared to only 46.7% of White respondents. Thus, the larger negative effect size indicates that for the smaller number of White Americans who were alienated, these feelings played a larger role in predicting war support. Overall, mediation analysis suggests that during the Bush years, 27%–36% of the Black-White difference in support for the Iraq War could be explained by the mediating effect of alienation on respondents’ views.

Figure 6. Coefficient plot, race and political alienation.

Political alienation plays a more complicated role in data collected during the Obama administration. Only 7.4% of Black respondents expressed feelings of alienation (compared to 56.1% of White respondents), a considerable shift from during the Bush administration. Controlling for demographics, Black respondents were 13.8%–22.6% less likely than White respondents to be politically alienated under Obama (see Figure 6). Both Black and White respondents who were alienated from President Obama appeared to support the Iraq War more. Still, alienation operated as a negative and statistically significant mediator of race during both the Bush and Obama administrations (Figure A4 in the online appendix). Taken together, our results indicate that alienation was a more relevant explanatory variable under President Bush, when the vast majority of Black Americans felt alienated.⁹⁸

Discussion

In sum, we find evidence of distinctive Black American war attitudes that cannot be explained away by individual-level demographic differences alone. Several group-level explanations find support through our regression and mediation analyses.⁹⁹

Casualty Sensitivity and War Support

Our strongest evidence shows that casualty sensitivity is a predictor of Black war aversion. Black Americans consistently express greater objection to casualties during the war, with no evidence that this result is driven by individuals’ direct concern for themselves or family members. Though we find higher reported rates of military participation among Black Americans’ families, this difference does not contribute to a between-group gap in support for war.

There may be several different rationales for heightened casualty sensitivity in Black communities, but high adherence to “linked fate” ideology—demonstrated through antiwar rhetoric across several different contexts—indicates that Black casualty sensitivity stems from concerns about Black compatriots who will be put in harm’s way due to war. Previous work on US foreign policy attitudes has emphasized the role of “sociotropic” perceptions—namely, concern with how policy affects the nation as a whole over the direct effect of policy on an individual—in public opinion.¹⁰⁰ Sociotropic perceptions operate with respect to a wide range of reference communities. Black Americans have been shown to exhibit a strong sense of “linked fate,” an individual-level concern for the improvement of social and economic conditions for all members of the race.¹⁰¹ Linked fate, long credited as a key source of Black Americans’ relatively unified political and policy views, might be seen as a representation of sociotropic concern not for the entire nation, but for the entire Black subpopulation within it.¹⁰² Linked fate helps predict disparate political phenomena from political participation to congressional representation on foreign policy issues, and it exists for other marginalized groups within the United States as well.¹⁰³ During a research study performed just before the US invasion of Iraq, Black Americans who scored high in linked fate were significantly less likely to support military action in Iraq after reading a news story that emphasized the high likelihood of Black and Hispanic casualties if war were to occur.¹⁰⁴

Vastly disproportionate rates of Black casualties at the beginning of the Vietnam War offended many Black Americans, and they did not soon forget this unequal distribution of harm. After Vietnam, though casualty rates became less imbalanced, Black overrepresentation in the US military stood out as a source of concern. To illustrate this point, despite the fact that Black service personnel experienced relatively low casualty rates during the 1991 Gulf War,¹⁰⁵ the idea of disproportionate risk continued to feature in Black congressional leaders’ opposition to war, as well as in the positions of other Black elites. At an antiwar rally in January 1991, Rev. Jesse Jackson Sr. lamented:

30% of the ground troops in the Persian Gulf are African-American, 45% black and brown, 75% black, brown and poor … And because of lack of racial justice, if the war breaks out, the black and the brown and the poor will burn first in the Gulf.¹⁰⁶

At another anti-Gulf War rally in Washington, DC’s Lafayette Park, racial themes also emerged, as expressed by one sign reading: “Rich Man’s War, Black Man’s Fight.”¹⁰⁷ Further reflecting these concerns, in 2005, Democratic members of Congress requested a Government Accountability Office report on the demographics of US casualties during Operation Iraqi Freedom to address worries “that certain subgroups of the U.S. population are disproportionately represented among those fighting and dying in support of the war on terrorism.”¹⁰⁸

Imbalanced casualty numbers at the beginning of the Vietnam War exacerbated linked fate concerns about casualties among Black Americans, but these sociotropic concerns also existed before Vietnam, even in contexts where Black and White casualties were more evenly distributed. For instance, the rates of war deaths between Black and White soldiers were comparable both before and after desegregation of US forces in the Korean War.¹⁰⁹ Nevertheless, because of troop segregation at the beginning of that war, the perception of Black American activists at the time was that Black soldiers faced greater risk of death in Korea due to White officers’ racism. Prominent performer and international activist Paul Robeson was quite vocal against the Korean War, and his words often made explicit how linked fate affected his war attitudes. The February 1951 issue of his monthly newspaper, Freedom, connected themes of domestic oppression, “group responsibility” within Black America, and condemnation of the war in Korea. On the front page, Robeson wrote: “No people feels more than we that what one Negro does affects the whole people … until we all, every one of our little boys and girls all over this land—have full equality in every phase of our social life, Negro history cannot rest.”¹¹⁰ He continued by critiquing the war in Korea, saying that Black American soldiers had to “face white-supremacy frameups in a jimcrow army.” Later in the issue, Black lawyer and civil rights activist Charles P. Howard expounded on the “frameups” mentioned by Robeson, saying, “One of the things I dislike most about the American system is the abominable segregation in the Army … They drive us into purely suicidal missions so as to create an opportunity to declare us cowards … Those conditions not only existed in World Wars I and II, but they exist in the present Korean conflict.”¹¹¹

Intertwined with the themes of sociotropic concern for other Black Americans in the Freedom issue is a palpable feeling of the unfairness of the US military effort in Korea. Throughout the publication, the authors derided the idea that Black people should be asked to fight and die for a state that mistreated and sometimes even killed them. A front-page editorial expressed outrage jointly at the court-ordered execution of the “Martinsville Seven” in Virginia and at US militarism in Korea,¹¹² ominously warning that “Anglo-American ‘superiority’” threatened to “drown the lives and liberties of Negro Americans in a sea of blood.”¹¹³ Two decades later, Black Panther Party leader Eldridge Cleaver quipped that while Black Americans “are asked to die for the system in Vietnam, in Watts they are killed by it.”¹¹⁴

In more recent years, some Black community leaders have linked concerns about Black military casualties abroad to the loss of Black life domestically through deadly violence sanctioned by the American state. As activists for Black Lives Matter and the Movement for Black Lives decry the militarization of American police forces, they reinforce the ugliness of loss of life in America’s foreign conflicts as well.¹¹⁵ Black Americans experience greater rates of violent force perpetrated by US institutions at home. The many examples of this inequality fuel perceptions of unfairness. Black civilians are subjugated to higher rates of police violence than White counterparts.¹¹⁶ Black Americans make up one-third of the US prison population, though they make up only 12% of the country’s total population.¹¹⁷ When facing criminal charges, Black criminals who have White victims are disproportionately more likely to receive the death penalty compared to White criminals who have Black victims.¹¹⁸ Instances of police violence, death penalty sentences, and pardoned vigilante killings that lead to Black deaths domestically perpetuate the view that the state itself is not a fair actor, and thus it will not adequately protect Black lives overseas. While previous scholarship has demonstrated that concerns about the unfairness of state-level burden-sharing can affect support for international agreements,¹¹⁹ these examples illustrate how perceptions of unfairness in the application of state violence can affect domestic stakeholders’ tolerance for foreign war costs.¹²⁰

Related to but distinct from the issue of fairness in treatment of Black stakeholders during wartime is the idea of the justness of the war effort. Historically, Black discourse on war casualties has also included references to justness, arguing that unjust wars cheapen the loss of soldiers’ lives. From previous scholarship, we know that Americans make judgments about how many casualties to tolerate based on assessments about the importance of the mission of a given war.¹²¹ Reasons that Black Americans have seen wars as unjust have varied. Black journalist William Worthy took issue with US Asia policy during the Korean War, calling it “wrong A to Z.” As a result, he said, Black soldiers in Korea had died “utterly in vain.”¹²² On the other hand, when a former classmate died in the Vietnam War in 1965, civil rights activists in McComb, Mississippi, released a public antiwar statement in which they objected to the war because it would only guarantee “the White Man’s freedom.”¹²³

Prominent arguments within some schools of liberal and radical Black political thought have expressed discontent about the unjustness of US wars. These narratives depict American militarism and interventionism overseas as imperialist or neocolonial in nature.¹²⁴ In her 1948 book, The United States and Armaments, Black IR scholar Merze Tate—a prominent female representative of the Howard School of International Relations¹²⁵—presented an early, critical history of US attempts to promote a peace agenda through disarmament, faulting these efforts for their failure to address deeper threats to peace embedded in colonial modes of production and imperial domination. In her conclusion, she wrote:

What was needed in the twenties and thirties was not a series of disarmament conferences to palaver over such terms and questions as war potential … the states should have established a system of collective security against aggression, and then proceeded to the … reexamination and revision of imperialism.¹²⁶

It would be useful to better capture sentiments surrounding lack of fairness in the application of state violence and unjustness of war efforts empirically in future studies—as well as their relationship to Black Americans’ casualty sensitivity.

Political Alienation and War Aversion

We also demonstrate that racial differences in political alienation are associated with differences in Black and White respondents’ support for war under both the Bush and Obama administrations—albeit with notable differences between the two time periods. During the Bush administration, greater Black alienation from the US presidency accounted for a substantial portion of racial differences in support for the use of force (more so than spending preferences). Under the Obama administration, however, Black Americans were substantially less likely to report feelings of alienation from Obama, and political alienation was associated with a greater likelihood of supporting the invasion of Iraq.

We consider the trends in Black alienation under President Obama to be largely related to conceptions of identity,¹²⁷ feelings of political efficacy,¹²⁸ and positions on domestic policy (which tend to be of greater concern to Americans than foreign policy, as noted by scholars as far back as the 1960s).¹²⁹ Linked fate also played a role:

Turning to the theory of inverted Black linked fate, Black Americans, in particular, can feel a greater sense of trust in the American political system and its political institutions during the Obama presidency because, as they trust in a figurehead with a shared racial background and Democratic partisanship, government becomes a symbolic manifestation of their political desires and an amplification of the Black political presence in the highest office of the land.¹³⁰

Still, different relationships between alienation and support for the Iraq War across the two administrations likely also reflect the fact that our measure for this variable asked respondents about particular presidents rather than their feelings about the US political system as a whole. The association of political alienation with prowar views under the Obama administration is probably also driven in part by candidate Obama’s outspoken opposition to the Iraq War in both the 2007 primary and the 2008 general election, having voiced his opposition to the war as “a dumb war … a rash war” as early as 2002.¹³¹ Accordingly, respondents might have felt that President Obama “cared about people like them” in part because he channeled their opposition to the war, which centered on issues of justness as discussed above. Hence, respondents’ views on the war might be constitutive of “alienation” as measured by available survey questions rather than caused by a deeper alienation from the US political system. It would be worthwhile for future research to explore conceptualizations of alienation from the state or political institutions rather than political leadership. We expect that broader measures of the kind of deep-seated, long-term mistrust of US political institutions that some studies have associated with a “black racial consciousness” would show a more consistent relationship between racial differences in alienation and racial differences in support for war across both administrations.¹³²

We also readily admit that some of the concerns about fairness described in the previous section might be interpreted as “alienation,” though the sense of distrust of state-sanctioned violence related to casualty sensitivity described above is more specific than a general sense of political alienation from leaders or institutions. Regardless, casualty sensitivity and political alienation are likely mutually causal to some extent.

Trade-Offs between Domestic and Defense Spending

Finally, our analysis suggests that greater Black American concern over the impact of military spending on domestic welfare programs explains a small part of between-group differences in support for the Iraq War—albeit only during the Bush administration, which clearly prioritized defense spending over domestic welfare programs. We suspect that this result derives from the fact that spending on “defense” may not always be clearly distinct from “domestic” spending for Black Americans, given both their overrepresentation in the ranks of the US armed forces and the financial benefits of military service as a source of employment. Notwithstanding the numerous forms of discrimination that have persisted within the US armed forces,¹³³ and Black critiques of the US military as sustaining racial injustice abroad,¹³⁴ military service has historically provided numerous Black Americans with a path to attaining middle-class socioeconomic status.¹³⁵ Accordingly, some Black respondents might be hesitant to support budget cuts to an institution that provides an important source of income to numerous Black American individuals and households—particularly under a Democratic administration that was more likely to cut defense spending relative to domestic spending. Our results point to a need for more nuanced research on US public perceptions of who benefits from military spending.

Taking Race and Ethnicity Seriously in Research on Foreign Policy Attitudes

Our article sheds light on the race gap in war support, and it represents the most comprehensive quantitative study of differences in White and Black support for the use of force abroad to date. Our findings point to the need for a broader research agenda focused on the drivers of public opinion on national security policy within different racial and ethnic groups. In future studies, surveys that include questions about linked fate, perceptions of representation in the military, justness, fairness, our four explanatory variables, and war support all together could help further clarify the relationship between these variables. More generally, it will be important for scholars to incorporate concepts and measures from REP scholarship into research on IR. The desire for more research in this area is more than a complaint that the discipline is “missing a stamp from its collection”; it indicates instead the opportunity for new, substantive explorations of public opinion toward foreign policy.

Though we primarily consider the US invasion of Iraq, racial and ethnic differences in American views on use of force might depend on the particular countries where force will be used.¹³⁶ Numerous historical works and some historically oriented political science studies have emphasized the importance of transnational ties between Black Americans and Black communities abroad in shaping aspects of US foreign policy.¹³⁷ Some scholars have also indicated that Black Americans exhibit solidarity with internationally based Black groups.¹³⁸ When an urgent threat confronts Black people in other parts of the world, Black respondents might demonstrate a kind of transnational linked fate inclusive of people of color overseas. Burris, for example, finds significantly greater Black than White support for the use of force in the 1994 operation to restore Black, democratically elected Bertrand Aristide to the Haitian presidency after he was deposed by military action.¹³⁹

A number of further analyses could extend the scope of the research agenda related to the Black-White race gap. Previous work has shown that elites influence both the public and decision makers on the use of force.¹⁴⁰ Further research should work to understand the role of Black elites in influencing Black public opinion on the use of force. Moreover, it is unclear how and whether support from international institutions affects war attitudes for Black and other racial minority groups.¹⁴¹

In the future, we hope more scholars will take up the mantle of researching racial attitudes about foreign policy. After all, we do not live in a world of only Black and White—there are a multitude of other racial groups and ethnic backgrounds in the United States and elsewhere whose unique foreign policy preferences and predispositions are worthy of greater exposition. We limit our study to only Black and White Americans here because we feel that attempting to explain the war preferences of more racial groups in one article would do a disservice to them. There are complex factors that explain the preferences of different racial groups in the United States,¹⁴² as other important research in the field has shown.¹⁴³ Scholars who take on this research agenda should take a cue from the REP literature and be attentive to intragroup mechanisms.¹⁴⁴ To this point, future work might distinguish between war attitudes of Black American descendants of slavery and more recent Black immigrants. And beyond studying what motivates war hesitance within racial groups, scholars should also investigate what factors within different racial communities spur increased support for conflicts abroad.¹⁴⁵

There are two compelling reasons to pursue this broader research agenda. First, public opinion can play an important role in the trajectory of America’s military actions abroad. Examination of gendered attitudes toward conflict is typically justified in terms of the potential for such “gender polarization” to tip the balance in electoral competition or send a meaningful signal to political elites. Observable differences in opinion toward war and conflict across racial and ethnic groups are equally if not more likely to affect partisan calculations and political activity given the greater partisan polarization of many racial and ethnic groups in the United States and elsewhere. Black Americans, for example, overwhelmingly affiliate with and vote for Democratic Party candidates—by margins of 87%–95% in the past five presidential elections.¹⁴⁶ In different ways, the political preferences and policy priorities of other racial groups, such as Hispanics, Asians, and Native Americans, are also distinct and unique. To the extent that different racial groups hold distinctive views on foreign policy, those views may constrain political parties’ ability to pursue particular foreign policy stances or lead them to select for candidates who hold more dovish views.

Second, investigating these mechanisms in the context of the United States, whose foreign policy experiences and populations tend to dominate our theorizing on and empirical testing of the dynamics of international relations, provides a firm academic starting point from which to explore these dynamics in other political contexts, where ethnic or sectarian differences may play a prominent role in foreign policy decision making. Scholarship on the various racial and ethnic trends in other countries’ foreign policy attitudes would add to the already burgeoning literature on comparative foreign policy preferences.¹⁴⁷ For these reasons, we would be pleased to see a wellspring of future work on the relationship between race and foreign policy preferences in the coming years. We hope that future scholarship will build upon our initial findings to generate a more nuanced research agenda around race, ethnicity, and foreign policy attitudes.

Acknowledgments

We would like to extend our sincere thanks to the scholars who took time to provide feedback on this project, especially: Joshua Kertzer, Alex Mierke-Zatwarnicki, Katy Powers, Robert Trager, Neta Crawford, Pete Cuppernull, Richard Eichenberg, Scott Sagan, and Steve Ansolabehere. Thanks also to Jessica Weeks, Mike Tomz, Joseph Grieco, Christopher Gelpi, Jason Reifler, and Peter Feaver for providing data and insights from their previous work during the preparation of this article.

Data Availability Statement

The data and materials that support the findings of this study are available in the Security Studies Dataverse at https://doi.org/10.7910/DVN/PFAGEE.

Additional information

Funding

This study was funded by Institute for Quantitative Social Science, Harvard University; and Center for American Political Studies, Harvard University.

Notes on contributors

Naima Green-Riley

Naima Green-Riley is an assistant professor in the Department of Politics and at the School of Public and International Affairs at Princeton University.

Andrew Leber

Andrew Leber is an assistant professor in the Department of Political Science and the Middle East & North African Studies program at Tulane University.

Notes

1 Mark Ellis, “W. E. B Du Bois and the Formation of Black Opinion in World War I: A Commentary on ‘The Damnable Dilemma,’” Journal of American History 81, no. 4 (March 1995): 1584–85.

2 Lee Finkle, “The Conservative Aims of Militant Rhetoric: Black Protest during World War II,” Journal of American History 60, no. 3 (December 1973): 692–713.

3 Christopher S. Parker, Fighting for Democracy: Black Veterans and the Struggle against White Supremacy in the Postwar South (Princeton, NJ: Princeton University Press, 2009), 42.

4 Martin Luther King Jr., “Beyond Vietnam: A Time to Break Silence,” speech at Riverside Church, New York, NY, 4 April 1967, American Rhetoric Online Speech Bank, http://www.americanrhetoric.com/speeches/mlkatimetobreaksilence.htm.

5 Susan Dodge, “Thousands of College Students Protest Persian Gulf War in Rallies and Sit-Ins; Others Support Military Action,” Chronicle of Higher Education, 30 January 1991, https://www.chronicle.com/article/thousands-of-college-students-protest-persian-gulf-war-in-rallies-and-sit-ins-others-support-military-action/; Lynne Duke, “Emerging Black Anti-War Movement Rooted in Domestic Issues,” Washington Post, 8 February 1991; Robert C. Smith and Richard Seltzer, Contemporary Controversies and the American Racial Divide (Lanham, MD: Rowman & Littlefield, 2000).

6 John Hudson, “Congressional Black Caucus Instructed to Hold Tongue on Syria,” Foreign Policy, 5 September 2013, https://foreignpolicy.com/2013/09/05/congressional-black-caucus-instructed-to-hold-tongue-on-syria/; Perry Bacon Jr., “Why the Congressional Black Caucus Could Determine if the US Strikes Syria,” The Grio, 5 September 2013, https://thegrio.com/2013/09/05/why-the-congressional-black-caucus-could-determine-if-the-u-s-strikes-syria/.

7 Pamela Johnston Conover and Virginia Sapiro, “Gender, Feminist Consciousness, and War,” American Journal of Political Science 37, no. 4 (November 1993): 1079–99; Nancy Gallagher, “The Gender Gap in Popular Attitudes toward the Use of Force,” in Women and the Use of Military Force, ed. Ruth H. Howes and Michael R. Stevenson (Boulder, CO: Lynne Rienner, 1993), 23–38; Richard C. Eichenberg, “Gender Differences in Public Attitudes toward the Use of Force by the United States, 1990–2003,” International Security 28, no. 1 (Summer 2003): 110–41; Deborah Jordan Brooks and Benjamin A. Valentino, “A War of One’s Own: Understanding the Gender Gap in Support for War,” Public Opinion Quarterly 75, no. 2 (Summer 2011): 270–86; Joslyn N. Barnhart et al., “The Suffragist Peace,” International Organization 74, no. 4 (Fall 2020): 633–70.

8 John H. Aldrich et al., “Foreign Policy and the Electoral Connection,” Annual Review of Political Science 9 (2006): 477–502; Joshua D. Kertzer and Thomas Zeitzoff, “A Bottom-Up Theory of Public Opinion about Foreign Policy,” American Journal of Political Science 61, no. 3 (July 2017): 543–58; Michael Tomz, Jessica L. P. Weeks, and Keren Yarhi-Milo, “Public Opinion and Decisions about Military Force in Democracies,” International Organization 74, no. 1 (Winter 2020): 119–43; Erik Lin-Greenberg, “Soldiers, Pollsters, and International Crises: Public Opinion and the Military’s Advice on the Use of Force,” Foreign Policy Analysis 17, no. 3 (July 2021): 1–12.

9 Sidney Verba et al., “Public Opinion and the War in Vietnam,” American Political Science Review 61, no. 2 (June 1967): 317–33; Richard F. Hamilton, “A Research Note on the Mass Support for ‘Tough’ Military Initiatives,” American Sociological Review 33, no. 3 (June 1968): 439–45; John P. Robinson and Solomon G. Jacobson, “American Public Opinion about Vietnam,” in Vietnam: Some Basic Issues and Alternatives, ed. Walter Isard (Cambridge, MA: Schenkman, 1969), 63–79; Harlan Hahn, “Correlates of Public Sentiments about War: Local Referenda on the Vietnam Issue,” American Political Science Review 64, no. 4 (December 1970): 1186–98; Milton J. Rosenberg, Sidney Verba, and Philip E. Converse, Vietnam and the Silent Majority: The Dove’s Guide (New York: Harper & Row, 1970); James D. Wright, “The Working Class, Authoritarianism, and the War in Vietnam,” Social Problems 20, no. 2 (Autumn 1972): 133–50; John E. Mueller, War, Presidents, and Public Opinion (New York: John Wiley & Sons, 1973); Henry J. Richardson III, “The Gulf Crisis and African-American Interests under International Law,” American Journal of International Law 87, no. 1 (January 1993): 42–82; Clyde Wilcox, Joseph Ferrara, and Dee Allsop, “Group Differences in Early Support for Military Action in the Gulf: The Effects of Gender, Generation, and Ethnicity,” American Politics Quarterly 21, no. 3 (July 1993): 343–59; John E. Mueller, Policy and Opinion in the Gulf War (Chicago: University of Chicago Press, 1994); Scott Sigmund Gartner and Gary M. Segura, “Race, Casualties, and Opinion in the Vietnam War,” Journal of Politics 62, no. 1 (February 2000): 115–46; Smith and Seltzer, Contemporary Controversies and the American Racial Divide; Miroslav Nincic and Donna J. Nincic, “Race, Gender, and War,” Journal of Peace Research 39, no. 5 (September 2002): 547–68; Roxanna Harlow and Lauren Dundes, “‘United’ We Stand: Responses to the September 11 Attacks in Black and White,” Sociological Perspectives 47, no. 4 (Winter 2004): 439–64; David E. Rohall and Morten G. Ender, “Race, Gender, and Class: Attitudes toward the War in Iraq and President Bush among Military Personnel,” Race, Gender & Class 14, nos. 3–4 (2007): 99–116; Jody C. Baumgartner, Peter L. Francia, and Jonathan S. Morris, “A Clash of Civilizations? The Influence of Religion on Public Opinion of U.S. Foreign Policy in the Middle East,” Political Research Quarterly 61, no. 2 (June 2008): 171–79; Val Burris, “From Vietnam to Iraq: Continuity and Change in Between-Group Differences in Support for Military Action,” Social Problems 55, no. 4 (November 2008): 443–79; Ole R. Holsti, Public Opinion and American Foreign Policy, revised ed. (Ann Arbor: University of Michigan Press, 2009); Katherine Tate, What’s Going On? Political Incorporation and the Transformation of Black Public Opinion (Washington, DC: Georgetown University Press, 2010); Rachel Allison, “Race, Gender, and Attitudes toward War in Chicago: An Intersectional Analysis,” Sociological Forum 26, no. 3 (September 2011): 668–91.

10 This dearth of research is particularly noticeable within the mainstream American IR literature.

11 Michael C. Dawson, Behind the Mule: Race and Class in African-American Politics (Princeton, NJ: Princeton University Press, 1994); Donald R. Kinder and Nicholas Winter, “Exploring the Racial Divide: Blacks, Whites, and Opinion on National Policy,” American Journal of Political Science 45, no. 2 (April 2001): 439–56; Claudine Gay, “Putting Race in Context: Identifying the Environmental Determinants of Black Racial Attitudes,” American Political Science Review 98, no. 4 (November 2004): 547–62; Vincent L. Hutchings and Nicholas A. Valentino, “The Centrality of Race in American Politics,” Annual Review of Political Science 7 (2004): 383–408; Ismail K. White, “When Race Matters and When It Doesn’t: Racial Group Differences in Response to Racial Cues,” American Political Science Review 101, no. 2 (May 2007): 339–54; Jane Junn, Tali Mendelberg, and Erica Czaja, “Race and the Group Bases of Public Opinion,” in New Directions in Public Opinion, ed. Adam J. Berinsky (Abingdon, UK: Routledge, 2011), 119–38.

12 Melissa V. Harris-Lacewell, “Political Science and the Study of African American Public Opinion,” in African American Perspectives on Political Science, ed. Wilbur C. Rich (Philadelphia: Temple University Press, 2007), 107–29.

13 Alfred O. Hero Jr., “American Negroes and U.S. Foreign Policy: 1937–1967,” Journal of Conflict Resolution 13, no. 2 (June 1969): 220–51; Jake C. Miller, The Black Presence in American Foreign Affairs (Washington, DC: University Press of America, 1978); Milfred C. Fierce, “Black and White American Opinions towards South Africa,” Journal of Modern African Studies 20, no. 4 (December 1982): 669–87; Smith and Seltzer, Contemporary Controversies and the American Racial Divide.

14 Adam J. Berinsky, In Time of War: Understanding American Public Opinion from World War II to Iraq (Chicago: University of Chicago Press, 2009); Zoltán I. Búzás, “The Color of Threat: Race, Threat Perception, and the Demise of the Anglo-Japanese Alliance (1902–1923),” Security Studies 22, no. 4 (October–December 2013): 573–606; Zoltán I. Búzás, “Racism and Antiracism in the Liberal International Order,” International Organization 75, no. 2 (Spring 2021): 440–63; Errol A. Henderson, “Hidden in Plain Sight: Racism in International Relations Theory,” in Race and Racism in International Relations: Confronting the Global Colour Line, ed. Alexander Anievas, Nivi Manchanda, and Robbie Shilliam (London: Routledge, 2014), 19–43; Robert Vitalis, White World Order, Black Power Politics: The Birth of American International Relations (Ithaca, NY: Cornell University Press, 2015); Adom Getachew, Worldmaking after Empire: The Rise and Fall of Self-Determination (Princeton, NJ: Princeton University Press, 2019); Kelebogile Zvobgo and Meredith Loken, “Why Race Matters in International Relations,” Foreign Policy 237 (2020): 11–13; Bianca Freeman, D. G. Kim, and David A. Lake, “Race in International Relations: Beyond the ‘Norm against Noticing,’” Annual Review of Political Science 25 (2022): 175–96.

15 Barbara Bardes and Robert Oldendick, “Beyond Internationalism: A Case for Multiple Dimensions in the Structure of Foreign Policy Attitudes,” Social Science Quarterly 59, no. 3 (December 1978): 496–508, esp. 505; Ole R. Holsti and James N. Rosenau, “Does Where You Stand Depend on When You Were Born? The Impact of Generation on Post-Vietnam Foreign Policy Beliefs,” Public Opinion Quarterly 44, no. 1 (Spring 1980): 1–22, esp. 7; Eichenberg, “Gender Differences in Public Attitudes toward the Use of Force by the United States.”

16 Brooks and Valentino, “War of One’s Own”; Richard C. Eichenberg, “Gender Difference in American Public Opinion on the Use of Military Force, 1982–2013,” International Studies Quarterly 60, no. 1 (March 2016): 138–48; Barnhart et al., “Suffragist Peace.”

17 Peter Hays Gries, The Politics of American Foreign Policy: How Ideology Divides Liberals and Conservatives over Foreign Affairs (Stanford, CA: Stanford University Press, 2014), 277.

18 John Zaller, “Information, Values, and Opinion,” American Political Science Review 85, no. 4 (December 1991): 1228–29.

19 Christopher Gelpi, Peter D. Feaver, and Jason Reifler, “Success Matters: Casualty Sensitivity and the War in Iraq,” International Security 30, no. 3 (Winter 2005/06): 33, 37.

20 Wilcox, Ferrara, and Allsop, “Group Differences in Early Support for Military Action in the Gulf.”

21 Richardson, “Gulf Crisis and African-American Interests under International Law”; Smith and Seltzer, Contemporary Controversies and the American Racial Divide.

22 Non-White vs. White racial categories (appears most of the non-White sample is Black).

23 Chapter is a review of other surveys about support for Vietnam; it includes results from Verba et al., “Public Opinion and the War in Vietnam,” and a 1967 Newsweek poll; Robinson and Jacobson, “American Public Opinion about Vietnam,” 65–66.

24 Measured at the census-tract level; study measures the percentage of non-White individuals in each area; gap appears in many cities, but it is statistically significant in San Francisco only.

25 Chapter is a summary of other survey work on war support; Rosenberg et al., Vietnam and the Silent Majority, 74–75.

26 Gap exists for both Black/White men and Black/White women across most poll dates; Mueller, War, Presidents, and Public Opinion, 142–43, 147–48.

27 Controls for party, education, income, information, sex, ethnicity, and age.

28 Mueller, Policy and Opinion in the Gulf War, 43.

29 California sample; controls for party, income, region, religion, sex, ethnicity.

30 Smith and Seltzer, Contemporary Controversies and the American Racial Divide, 39–60.

31 Controls for sex, “alienation,” party, age, income; Tau-b coefficients show bivariate race differences in approval, while probit coefficients show multivariate race differences.

32 The race gap in the regression with controls in Nincic and Nincic, “Race, Gender, and War,” for Korean War support (1952) has a reported p value of 0.06. Though authors report that it is not significant, it is significant at the alpha < 0.1 level.

33 Undergraduate sample, private liberal arts college.

34 US armed forces service member sample; logit controls for sex, rank, military affiliation, partisan ideology, job satisfaction.

35 Gap is not significant at the alpha < 0.1 level; coefficient is for “black Protestants,” so it should be noted that this study isolates just a subset of the Black population; controls for several religious denominations, party, sex, education, income.

36 Non-White vs. White respondents; multiple analyses with statistically significant bivariate gamma coefficients; gap reverses for interventions in El Salvador, Somalia, Bosnia and Herzegovina, and Haiti.

37 Multiple Gallup polls; Holsti, Public Opinion and American Foreign Policy, 227–31.

38 Various Gallup poll questions, Tate, What’s Going On?, chap. 8.

39 Race variable is significant in the bivariate regression, but not with the full set of controls.

40 Nincic and Nincic, “Race, Gender, and War.”

41 Rohall and Ender, “Race, Gender, and Class.”

42 Tate, What’s Going On?, chap. 8.

43 Kimberley L. Phillips, War! What Is It Good For? Black Freedom Struggles and the U.S. Military from World War II to Iraq (Chapel Hill: University of North Carolina Press, 2012).

44 Zaller, “Information, Values, and Opinion,” 1229.

45 Gartner and Segura, “Race, Casualties, and Opinion in the Vietnam War.”

46 Burris, “From Vietnam to Iraq,” 462–66.

47 Though Burris uses a White/non-White racial categorization scheme, much of his discussion focuses on Black Americans. Black Americans were, until recently, the largest non-White population in the United States.

48 Allison, “Race, Gender, and Attitudes toward War in Chicago.”

49 Allison compares levels of Black and White support within genders. Her small, geographically limited sample may explain the null results in her male-only analysis. Nevertheless, the intersection of race and gender has often shown dual effects during past crises, with White men being the largest proponents of war and Black women being the biggest opponents. See Mueller, War, Presidents, and Public Opinion, 142–43. Also see Smith and Seltzer, Contemporary Controversies and the American Racial Divide, 47, 50.

50 Less wealthy individuals are less likely to support the use of military force. See: Hamilton, “Research Note on the Mass Support for ‘Tough’ Military Initiatives”; Hahn, “Correlates of Public Sentiments about War”; Wright, “Working Class, Authoritarianism, and the War in Vietnam”; Burris, “From Vietnam to Iraq”; Douglas L. Kriner and Francis X. Shen, “Conscription, Inequality, and Partisan Support for War,” Journal of Conflict Resolution 60, no. 8 (December 2016): 1419–45. Moreover, Rohall and Ender suggest that Black Americans’ preference for the Democratic Party may make them more likely to hold antiwar positions, though they do not investigate this question empirically. See Rohall and Ender, “Race, Gender, and Class,” 102. Additionally, Tate shows trends over time of convergence between the views of Black Americans and Democrats on foreign policy issues. See Tate, What’s Going On?

51 Mueller, War, Presidents, and Public Opinion.

52 Bruce W. Jentleson, “The Pretty Prudent Public: Post Post-Vietnam American Opinion on the Use of Military Force,” International Studies Quarterly 36, no. 1 (March 1992): 49–74; Bruce W. Jentleson and Rebecca L. Britton, “Still Pretty Prudent: Post-Cold War American Public Opinion on the Use of Military Force,” Journal of Conflict Resolution 42, no. 4 (August 1998): 395–417.

53 Wilcox, Ferrara, and Allsop, “Group Differences in Early Support for Military Action in the Gulf”; Smith and Seltzer, Contemporary Controversies and the American Racial Divide; Rohall and Ender, “Race, Gender, and Class.”

54 Verba et al., “Public Opinion and the War in Vietnam.”

55 Gartner and Segura, “Race, Casualties, and Opinion in the Vietnam War.”

56 The survey had a total of 741 White respondents and 206 Black respondents.

57 Verba et al., “Public Opinion and the War in Vietnam,” 331.

58 David Coffey, “African Americans in the U.S. Military,” in Encyclopedia of the Vietnam War: A Political, Social, and Military History, ed. Spencer C. Tucker (Oxford: Oxford University Press, 1998), 9.

59 Gerald F. Goodwin, “Black and White in Vietnam,” New York Times, 18 July 2017, https://www.nytimes.com/2017/07/18/opinion/racism-vietnam-war.html.

60 Kim Parker, Anthony Cilluffo, and Renee Stepler, “6 Facts about the U.S. Military and Its Changing Demographics,” Pew Research Center, 13 April 2017, http://www.pewresearch.org/fact-tank/2017/04/13/6-facts-about-the-u-s-military-and-its-changing-demographics/; Robert Burns and Lolita C. Baldor, “Floyd Death Pushes Military to Face ‘Own Demons’ on Race,” AP News, 5 June 2020, https://apnews.com/article/american-protests-race-and-ethnicity-politics-b3540ebbf461cbf59e75c7a4aa7c6c21/; Richard V. Reeves and Sarah Nzau, “Black Americans Are Much More Likely to Serve the Nation, in Military and Civilian Roles,” Brookings Institution, 27 August 2020, https://www.brookings.edu/blog/up-front/2020/08/27/black-americans-are-much-more-likely-to-serve-the-nation-in-military-and-civilian-roles/.

61 Military statistics come from Parker, Cilluffo, and Stepler, “6 Facts about the U.S. Military and Its Changing Demographics”; population statistics come from the Census Bureau’s 2015 American Community Survey data.

62 The “burden” of war includes not only loss of life but also other types of harm, such as injuries/wounds, which are often overlooked in quantitative examinations of war damage. See: Tanisha M. Fazal, “Dead Wrong? Battle Deaths, Military Medicine, and Exaggerated Reports of War’s Demise,” International Security 39, no. 1 (Summer 2014): 95–125; Tanisha M. Fazal, “Life and Limb: New Estimates of Casualty Aversion in the United States,” International Studies Quarterly 65, no. 1 (March 2021): 160–72.

63 Rohall and Ender, “Race, Gender, and Class.”

64 Jennifer Hochschild and Vesla Weaver, “Is the Significance of Race Declining in the Political Arena? Yes, and No,” Ethnic and Racial Studies 38, no. 8 (June 2015): 1250–57.

65 Richardson, “Gulf Crisis and African-American Interests under International Law,” 53; Wilcox, Ferrara, and Allsop, “Group Differences in Early Support for Military Action in the Gulf,” 346.

66 King, “Beyond Vietnam.”

67 Verba et al., “Public Opinion and the War in Vietnam,” 331.

68 Robinson and Jacobson, “American Public Opinion about Vietnam,” 65; Holsti, Public Opinion and American Foreign Policy, 231.

69 Nincic and Nincic, “Race, Gender, and War,” 562.

70 Harlow and Dundes, “‘United’ We Stand,” 450.

71 Burris, “From Vietnam to Iraq,” 464–65.

72 William M. Mason, James S. House, and Steven S. Martin, “On the Dimensions of Political Alienation in America,” Sociological Methodology 15 (1985): 111–51; Shayla C. Nunnally, Trust in Black America: Race, Discrimination, and Politics (New York: New York University Press, 2012).

73 Michael Tesler and David O. Sears, Obama’s Race: The 2008 Election and the Dream of a Post-Racial America (Chicago: University of Chicago Press, 2010); Michael Tesler, “The Spillover of Racialization into Health Care: How President Obama Polarized Public Opinion by Racial Attitudes and Race,” American Journal of Political Science 56, no. 3 (July 2012): 690–704.

74 Shayla C. Nunnally, “Race, Trust, and the American Presidency: Black-White Confidence in the Executive Branch in the Obama Era and Beyond,” in After Obama: African American Politics in a Post-Obama Era, ed. Todd C. Shaw, Robert A. Brown, and Joseph P. McCormick II (New York: New York University Press, 2021), 45–71.

75 One dataset lacks measures for income, though we find that our results are robust to the inclusion or exclusion of an income variable.

76 The wording of the question was: “All in all, considering the costs to the United States versus the benefits to the United States, do you think the war with Iraq was worth fighting, or not?”

77 This poll included 873 respondents, of whom 73 (8.4%) identified as Black. We measure support for the war in Afghanistan using a question that asks respondents whether they “support or oppose” US military action in the conflict. We measure casualty sensitivity based on whether respondents believed the US government was not “doing all it reasonably [could] do to try to avoid U.S. military casualties in Afghanistan.”

78 Four of these surveys did not include an income variable, so we replicate our analyses without a control for income in the online appendix.

79 See the online appendix for all questions we included in the analysis to measure alienation and support for war.

80 Jack Citrin, “Comment: The Political Relevance of Trust in Government,” American Political Science Review 68, no. 3 (September 1974): 973–88.

81 We measure support for the Iraq War based on questions where respondents indicated the invasion was “the right thing” or was not “a mistake.” The 2010 and 2012 waves of the CCES also ask about support for the Afghanistan War, which we analyze in the online appendix.

82 We use HC₁ heteroskedasticity-robust standard errors across all regressions.

83 Since several of our proposed explanatory variables may explain the race gap, there is undoubtedly some omitted variable bias in each of our sets of regression results. Granted, this means that all previous studies that only explored one of the explanations presented here suffered from omitted variable bias as well.

84 Dustin Tingley et al., “Mediation: R Package for Causal Mediation Analysis,” Journal of Statistical Software 59, no. 5 (August 2014): 1.

85 Maya Sen and Omar Wasow, “Race as a Bundle of Sticks: Designs That Estimate Effects of Seemingly Immutable Characteristics,” Annual Review of Political Science 19 (2016): 499–522.

86 Error bars represent 95% confidence intervals throughout. The vertical line divides the Bush and Obama presidencies.

87 Subsetting to only self-identified Democrats, we continue to observe a negative Black-White gap in polling only during the Bush administration, and the gap actually reverses in the CCES-Obama sample (Table C1, Models 9–10 in the online appendix).

88 Predicted likelihoods are calculated by subtracting and adding 1.96 standard error measures from the coefficients in the regressions.

89 The effect size for “casualty sensitivity” is larger for White respondents; however, Black respondents were significantly more likely to express concern about casualties in the first place, as indicated in Figure 4. For the smaller proportion of White respondents who expressed concern over casualties, this factor appeared to hold greater weight in determining their war attitudes.

90 Table A5 in the online appendix shows full regression results, which include demographic controls (Models 1–3).

91 For this and all subsequent mediation analyses, we have used list-wise deletion to address missingness in the outcome variable.

92 For our alternative measure of casualty sensitivity, across fewer surveys, the mediated effect constitutes a smaller but still substantial portion of the total effect: between 0.18 and 0.31 (95% confidence intervals). Responses are reported in the online appendix.

93 Table A5 in the online appendix shows full regression results, which include demographic controls (Models 4–9).

94 The effect size for Black respondents is smaller and not significantly distinguishable from a null relationship at the α < 0.1 level.

95 Regression table A6 in the online appendix shows full regression results, which include demographic controls (Models 1–6).

96 Across the four polls, an average of 18.1% of White Democrats felt that President Obama did not care about or understand people like them.

97 Regression Table A7 in the online appendix shows full regression results, which include demographic controls (Models 1–6).

98 The results analyzing the relationship between race, political alienation, and support for the Afghanistan War across these two administrations (in the online appendix) displayed nearly identical trends with one exception: the coefficient on alienation for the subsample of Black respondents during the Obama administration (Table D5, Model 5) was negative and not statistically significant. In other words, although Black Americans who were alienated from President Obama were more likely to support the Iraq War, they were not any more or less likely to support the Afghanistan War.

99 In observational studies such as this one, it can be difficult to tell whether the sequential ignorability assumption required for mediation holds. Sensitivity analyses developed by Kosuke Imai et al. can determine how robust results are to potential violations of this assumption. Sensitivity tests show that in our mediations for each of the explanatory variables correlated with lower Black support for war, confounders would have to have relatively large effect sizes to overturn the conclusions that we reach here; for the casualty sensitivity, domestic spending, and Bush alienation findings, the mediation effects reach zero when ρ = −0.4. In mediations without control variables, ρ would have to be at least −0.5 to alter the results. For analytical details, see Kosuke Imai, Luke Keele, and Dustin Tingley, “A General Approach to Causal Mediation Analysis,” Psychological Methods 15, no. 4 (December 2010): 309–34.

100 Sociotropic policy concerns are more commonly discussed in terms of economic voting issues. See: Donald R. Kinder and D. Roderick Kiewiet, “Sociotropic Politics: The American Case,” British Journal of Political Science 11, no. 2 (April 1981): 129–61; Edward D. Mansfield and Diana C. Mutz, “Support for Free Trade: Self-Interest, Sociotropic Politics, and Out-Group Anxiety,” International Organization 63, no. 3 (July 2009): 425–57.

101 Dawson, Behind the Mule.

102 Michael J. Donnelly, “Material Interests, Identity and Linked Fate in Three Countries,” British Journal of Political Science 51, no. 3 (July 2021): 1119–37.

103 Todd C. Shaw, Kirk A. Foster, and Barbara Harris Combs, “Race and Poverty Matters: Black and Latino Linked Fate, Neighborhood Effects, and Political Participation,” Politics, Groups, and Identities 7, no. 3 (2019): 663–72; Menna-Weiwot A. Demessie and Errol A. Henderson, “Race and Ethnicity in Black Congressional Representation: The Case of US Foreign Policy towards Africa,” Congress & the Presidency 48, no. 2 (2021): 147–74; Claudine Gay, Jennifer Hochschild, and Ariel White, “Americans’ Belief in Linked Fate: Does the Measure Capture the Concept?” Journal of Race, Ethnicity, and Politics 1, no. 1 (March 2016): 117–44.

104 White, “When Race Matters and When It Doesn’t,” 344.

105 Lynne Duke, “U.S. War Toll Defies Racial Predictions,” Washington Post, 21 March 1991, https://www.washingtonpost.com/archive/politics/1991/03/21/us-war-toll-defies-racial-predictions/5ae579c3-fb2f-4d60-b2ee-e8198798ff72/.

106 Jesse Jackson Sr., “Rev. Jesse Jackson at MIT Peace Rally against Gulf War,” 14 January 1991, MIT Black History, https://www.blackhistory.mit.edu/archive/rev-jesse-jackson-mit-peace-rally-against-gulf-war-1991.

107 Dodge, “Thousands of College Students Protest Persian Gulf War in Rallies and Sit-Ins; Others Support Military Action.”

108 US Government Accountability Office, Military Personnel: Reporting Additional Servicemember Demographics Could Enhance Congressional Oversight, Report to Congressional Requesters GAO-05-952 (Washington, DC: US Government Accountability Office, 2005), 2.

109 Connor Huff and Robert Schub, “Segregation, Integration, and Death: Evidence from the Korean War,” International Organization 75, no. 3 (Summer 2021): 858–79.

110 Paul Robeson, “Here’s My Story,” Freedom 1, no. 2 (February 1951): 1, https://mc.dlib.nyu.edu/files/books/tamwag_fdm000003/tamwag_fdm000003_hi.pdf.

111 “‘I Love All Humanity’ Howard Says; Midwest Leader Defends Position,” Freedom 1, no. 2 (February 1951): 5, https://mc.dlib.nyu.edu/files/books/tamwag_fdm000003/tamwag_fdm000003_hi.pdf.

112 The Martinsville Seven were a group of seven Black men from Virginia who were executed by the state after being accused of raping a White woman. In August 2021, their convictions were posthumously pardoned by the governor of Virginia.

113 “The Men of Martinsville,” unsigned editorial, Freedom 1, no. 2 (February 1951): 1, https://mc.dlib.nyu.edu/files/books/tamwag_fdm000003/tamwag_fdm000003_hi.pdf.

114 Eldridge Cleaver, Soul on Ice (New York: Dell, 1968), 165.

115 Patrick Deer, “Black Lives Matter in Wartime,” RSAJournal 29 (2018): 65–91; Adaugo Pamela Nwakanma, “From Black Lives Matter to EndSARS: Women’s Socio-Political Power and the Transnational Movement for Black Lives,” Perspectives on Politics 20, no. 4 (December 2022): 1246–59.

116 Dean Knox, Will Lowe, and Jonathan Mummolo, “Administrative Records Mask Racially Biased Policing,” American Political Science Review 114, no. 3 (August 2020): 619–37.

117 John Gramlich, “Black Imprisonment Rate in the U.S. Has Fallen by a Third since 2006,” Pew Research Center, 6 May 2020, https://www.pewresearch.org/fact-tank/2020/05/06/share-of-black-white-hispanic-americans-in-prison-2018-vs-2006/.

118 David C. Baldus, Charles Pulaski, and George Woodworth, “Comparative Review of Death Sentences: An Empirical Study of the Georgia Experience,” Journal of Criminal Law and Criminology 74, no. 3 (Fall 1983): 661–753; Frank R. Baumgartner et al., Deadly Justice: A Statistical Portrait of the Death Penalty (Oxford: Oxford University Press, 2017).

119 Kathleen E. Powers et al., “What’s Fair in International Politics? Equity, Equality, and Foreign Policy Attitudes,” Journal of Conflict Resolution 66, no. 2 (2022): 217–45.

120 Similarly, Jason Lyall’s argument about ethnic marginalization indicates there is a depressive effect of mistreatment of ethnic groups on the sense of common purpose in wartime and armies’ battlefield performance. See Lyall, Divided Armies: Inequality and Battlefield Performance in Modern War (Princeton, NJ: Princeton University Press, 2020).

121 Peter D. Feaver and Christopher Gelpi, Choosing Your Battles: American Civil-Military Relations and the Use of Force (Princeton, NJ: Princeton University Press, 2011); Gelpi, Feaver, and Reifler, “Success Matters.”

122 Quoted in Phillips, War! What Is It Good For?, 154–55.

123 “McComb Project Comes Out against the Vietnam War,” SNCC Digital Gateway, SNCC Legacy Project and Duke University, 2017, https://snccdigital.org/events/mccomb-project-comes-vietnam-war/.

124 Penny M. Von Eschen, Race against Empire: Black Americans and Anticolonialism, 1937–1957 (Ithaca, NY: Cornell University Press, 1997); Getachew, Worldmaking after Empire; Anthony Monteiro, “Race and Empire: W.E.B. Du Bois and the US State,” Black Scholar 37, no. 2 (Summer 2007): 35–52; Nada Elia, “Violent Women: Surging into Forbidden Quarters,” in Fanon: A Critical Reader, ed. Lewis R. Gordon, T. Denean Sharpley-Whiting, and Renée T. White (Oxford: Blackwell, 1996), 163–69.

125 Vitalis, White World Order, Black Power Politics.

126 Merze Tate, The United States and Armaments (Cambridge, MA: Harvard University Press, 1948), 204.

127 Tesler and Sears, Obama’s Race; Thomas Craemer et al., “‘Race Still Matters, However . . . ’: Implicit Identification with Blacks, Pro-Black Policy Support and the Obama Candidacy,” Ethnic and Racial Studies 36, no. 6 (2013): 1047–69.

128 Emily A. West, “Descriptive Representation and Political Efficacy: Evidence from Obama and Clinton,” Journal of Politics 79, no. 1 (January 2017): 351–55.

129 See, for example, Warren E. Miller and Donald E. Stokes, “Constituency Influence in Congress,” American Political Science Review 57, no. 1 (March 1963): 45–56.

130 Nunnally, “Race, Trust, and the American Presidency,” 53.

131 Obama did not politicize the war in Afghanistan to the same extent, which may help to explain the slight differences in results for the effects of alienation on support for the Iraq and Afghanistan Wars. Quote from: “Transcript: Obama’s Speech against the Iraq War,” National Public Radio, 20 January 2009, https://www.npr.org/templates/story/story.php?storyId=99591469.

132 James M. Avery, “The Sources and Consequences of Political Mistrust among African Americans,” American Politics Research 34, no. 5 (September 2006): 653–82; James M. Avery, “Political Mistrust among African Americans and Support for the Political System,” Political Research Quarterly 62, no. 1 (March 2009): 132–45.

133 James Burk and Evelyn Espinoza, “Race Relations within the US Military,” Annual Review of Sociology 38 (2012): 401–22.

134 Manning Marable, “The Military, Black People, and the Racist State: A History of Coercion,” Black Scholar 12, no. 1 (January–February 1981): 6–17.

135 John J. Hisnanick, “Military Service as a Factor in the Economic Progress of African American Men: A Post-Draft Era Analysis,” Journal of African American Men 5, no. 4 (March 2001): 65–79; W. Bradford Wilcox, Wendy R. Wang, and Ronald B. Mincy, Black Men Making It In America: The Engines of Economic Success for Black Men in America (Washington, DC: American Enterprise Institute, 2018), https://www.aei.org/wp-content/uploads/2018/06/BlackMenMakingItInAmerica-Final_062218.pdf?x91208.

136 Stephen M. Utych, “Human or Not? Political Rhetoric and Foreign Policy Attitudes,” Political Science Research and Methods 10, no. 3 (July 2022): 642–50; Scott D. Sagan and Benjamin A. Valentino, “Weighing Lives in War: How National Identity Influences American Public Opinion about Foreign Civilian and Compatriot Fatalities,” Journal of Global Security Studies 5, no. 1 (January 2020): 25–43; Jonathan Chu and Carrie Lee, “Race, Religion, and American Support for Humanitarian Intervention” (preprint, last revised 18 April 2022), https://doi.org/10.2139/ssrn.4060474.

137 Elliott P. Skinner, African Americans and U.S. Policy toward Africa, 1850–1924: In Defense of Black Nationality (Washington, DC: Howard University Press, 1992); Von Eschen, Race against Empire; Carol Anderson, Bourgeois Radicals: The NAACP and the Struggle for Colonial Liberation, 1941–1960 (Cambridge: Cambridge University Press, 2015); Alvin B. Tillery Jr., Between Homeland and Motherland: Africa, U.S. Foreign Policy, and Black Leadership in America (Ithaca, NY: Cornell University Press, 2011).

138 Richardson, “Gulf Crisis and African-American Interests under International Law”; Smith and Seltzer, Contemporary Controversies and the American Racial Divide.

139 Burris, “From Vietnam to Iraq.”

140 Elizabeth N. Saunders, “War and the Inner Circle: Democratic Elites and the Politics of Using Force,” Security Studies 24, no. 3 (July–September 2015): 466–501.

141 Joseph M. Grieco et al., “Let’s Get a Second Opinion: International Institutions and American Public Support for War,” International Studies Quarterly 55, no. 2 (June 2011): 563–83.

142 See, for example: Janelle S. Wong et al., Asian American Political Participation: Emerging Constituents and Their Political Identities (New York: Russell Sage Foundation, 2011); Taeku Lee, S. Karthick Ramakrishnan, and Ricardo Ramírez, eds., Transforming Politics, Transforming America: The Political and Civic Incorporation of Immigrants in the United States (Charlottesville: University of Virginia Press, 2007); Jeré Franco, “Empowering the World War II Native American Veteran: Postwar Civil Rights,” Wicazo Sa Review 9, no. 1 (Spring 1993): 32–37.

143 For some groups of Native Americans, it may even be the case that pre-US security dynamics in North America have affected contemporary views of the use of force. See Neta C. Crawford, “A Security Regime among Democracies: Cooperation among Iroquois Nations,” International Organization 48, no. 3 (Summer 1994): 345–85.

144 For example: Ali Adam Valenzuela, “Tending the Flock: Latino Religious Commitments and Political Preferences,” Political Research Quarterly 67, no. 4 (December 2014): 930–42; Jeron Fenton and LaFleur Stephens-Dougan, “Are Black State Legislators More Responsive to Emails Associated with the NAACP versus BLM? A Field Experiment on Black Intragroup Politics,” Journal of Race, Ethnicity, and Politics 7, no. 2 (July 2022): 203–18.

145 See, for example, Jon Green, “Does Race Stop at the Water’s Edge? Elites, the Public, and Support for Foreign Intervention among White U.S. Citizens over Time,” Political Science Quarterly 136, no. 2 (Summer 2021): 339–61.

146 “How Groups Voted,” Roper Center for Public Opinion Research, accessed 29 November 2022, https://ropercenter.cornell.edu/how_groups_voted.

147 Thomas Risse-Kappen, “Public Opinion, Domestic Structure, and Foreign Policy in Liberal Democracies,” World Politics 43, no. 4 (July 1991): 479–512; Eric Shiraev, “Toward a Comparative Analysis of the Public Opinion–Foreign Policy Connection,” in Decisionmaking in a Glass House: Mass Media, Public Opinion, and American and European Foreign Policy in the 21st Century, ed. Brigitte L. Nacos, Robert Y. Shapiro, and Pierangelo Isernia (Lanham, MD: Rowman & Littlefield, 2000), 297–304; Timothy B. Gravelle, Jason Reifler, and Thomas J. Scotto, “The Structure of Foreign Policy Attitudes in Transatlantic Perspective: Comparing the United States, United Kingdom, France and Germany,” European Journal of Political Research 56, no. 4 (November 2017): 757–76; Timothy B. Gravelle, Jason Reifler, and Thomas J. Scotto, “Personality Traits and Foreign Policy Attitudes: A Cross-National Exploratory Study,” Personality and Individual Differences 153 (20 January 2020): 109607; Clara H. Suong, Scott Desposato, and Erik Gartzke, “How ‘Democratic’ Is the Democratic Peace? A Survey Experiment of Foreign Policy Preferences in Brazil and China,” Brazilian Political Science Review 14, no. 1 (2020): 1–38, https://doi.org/10.1590/1981-3821202000010002.