Political Behavior

, Volume 40, Issue 2, pp 521–543 | Cite as

The Gag Reflex: Disgust Rhetoric and Gay Rights in American Politics

  • Shana Kushner Gadarian
  • Eric van der Vort
Original Paper


Political scientists have increasingly looked to the role that disgust plays in shaping public opinion and attitudes. This emotion plays an important role in building and reinforcing boundaries in the polity. It is particularly important in shaping attitudes toward gay rights. We analyze data from the 1993 American National Election Studies (ANES) data and two original studies. We find that disgust is a powerful but contingent rhetorical tool. It can powerfully shape public attitudes, especially on issues of sexual purity, but that efficacy must come with a strong caveat: our findings show that some members of the public will reject disgust rhetoric as an indignant reaction against the speaker.


Disgust Emotion Gay rights Rhetoric 


The politics of disgust is alive and well. In 2013, future Secretary of Housing and Urban Development Dr. Ben Carson compared homosexuality to pedophilia and bestiality, saying: “[marriage is] a well-established, fundamental pillar of society and no group, be they gays, be they NAMBLA, be they people who believe in bestiality—it doesn’t matter what they are, they don’t get to change the definition” (Ayres-Deets 2013). While Carson later apologized for his comments, the continued use of such language sometimes surprises modern ears, who are more accustomed to hearing arguments emphasizing the normalcy of lesbians and gays. These arguments helped create a more inclusive society (Flores and Barclay 2015). But for some portions of the population, it seems that disgusting language is a powerful emotional counterargument to pro-inclusive messages, one that lesbian and gay advocates have difficulty countering (Schilt and Westbrook 2015). How do those parts of the population react to language like Carson’s? How do messages that emphasize deviance and perversion manage to be so effective for some Americans and yet provoke an opposite reaction in others? We argue that disgust is a contingent political tool, the use of which comes with no small risk of provoking backlash.

Disgust may remain a powerful tool for shifting political attitudes, but not always in the way that elites intend (Clifford et al. 2015; Kam and Estes 2016). We argue that this contingency is rooted in how messaging and rhetoric have evolved in response to the public’s increasing acceptance of lesbians and gays (Harrison and Michelson 2012). Our politics has shifted enough that the mere existence of lesbians and gays no longer prompts condemnation from broad parts of the population. Lesbians and gays are now among the most effective advocates for their own rights, through personal contact and persuasion (Herek and Capitanio 1996; Broockman and Kalla 2016; Harrison and Michelson 2017). Contact theory argues that those who personally know lesbians or gays are more likely to hold inclusive attitudes. Yet research also suggests that lesbians and gay remain widely stigmatized, in part because of disgust (Herek and McLemore 2013).

Like other emotions, disgust is politically relevant. At its evolutionary core, it alerts individuals to the presence of toxicity and noxiousness in the world. It helps individuals to avoid toxic substances, thus maintaining cleanliness and relative safety (Cosmides and Tooby 2000). Its connection in maintaining physical purity is mirrored in the ways it helps to enforce and maintain various forms of social purity (Douglas 2003). The particular forms of purity enforced vary by culture, but it is a critical element in drawing and sustaining social and political barriers (Haidt et al. 1997). Disgust is a powerful predictor of moral and political judgments, particularly as they relate to lesbians and gays (Inbar et al. 2009; Inbar et al. 2012; Inbar and Pizarro 2014). Further, the increased openness of lesbians and gays may even increase the relevance of disgust for some Americans (Casey 2016). In their study of disgust, Kam and Estes (2016) quote writer Upton Sinclair, “I aimed at the public’s heart and by accident I hit it in the stomach.” We argue that, sometimes, elites may aim at the public’s stomach and find that they hit it in the heart or the head instead.

By this, we mean that there is a contingency to disgust’s ability to shape political attitudes and its perils and limits as a rhetorical strategy. Elites can attempt to use disgust to drive politics. Disgusting language and imagery are used to communicate stigma in the hopes of shaping public attitudes and policy (Smith 2007; Nussbaum 2010; Hatemi and McDermott 2012; Kam and Estes 2016). We argue that rhetoric like Carson’s represents an appeal to what Martha Nussbaum calls projective disgust. Projective disgust is a kind of “sympathetic magic” that involves “linking the allegedly disgusting group or person somehow with the primary objects of disgust” (Nussbaum 2010). “Practical connections” may exist between the object and the target, but usually “the extension works in more fantasy-laden ways, by imputing to… groups properties similar to those that are found disgusting in the primary objects: bad smell, ooziness, rottenness, germiness, decay. Typically, these projections have no basis in reality” (Nussbaum 2010). We call the political use of disgusting language, which invokes these projections, disgust rhetoric (Kam and Estes 2016).

In the United States, many groups have been portrayed as threatening the polity, including “sexually deviant” ones (Morone 2004). It has historically been easy to use disgust to depict sexual minorities in this light (Horberg et al. 2009; Herman 1996; Canaday 2009). This supposed deviance remains a theme in modern political discourse. Anita Bryant, a prominent anti-gay activist in the 1970s, asserted that “…if [children] are exposed to homosexuality, I might as well feed them garbage” (Eskridge 2008, Emphasis added). In 2014, Pat Robertson argued, “Nobody can ever produce a child through homosexual sex or lesbian sex—you cannot do it. This is for procreation and God has said that those who violate it, the land will vomit them out” (Wong 2015, Emphasis added). Robertson shows that disgusting rhetoric is not a relic of the past; it remains a tool used to stigmatize lesbians and gays.

Disgust influences political attitudes by increasing negative affect toward some groups and by lowering support for policies that benefit the targeted group. This emotion is powerful enough to transcend partisan cues that would normally drive political attitudes (Clifford and Wendell 2015). It tends to lead to disapproval of lesbians and gays and a preference for exclusive policies (Terrizzi et al. 2010). It can alter implicit and explicit support for lesbian and gay rights, whether through induction, social exposure, or overt disgust stimuli (Inbar et al. 2012; Cunningham et al. 2013; Casey 2016). There are clear connections between felt disgust and political attitudes. We know that language and rhetoric are critical in shaping perceptions of groups in the mass public, and that emotional language is part of this process (Schneider and Ingram 1993; Smith 2007; Albertson and Gadarian 2015). The ways in which disgusting rhetoric actually works in politics are less clear. In particular, we know little about the ability of political elites to effectively marshal disgust among the mass public or what its consequences might be.

Disgust plays a role in shaping attitudes and policies towards lesbians and gays. Public opinion toward lesbians and gays has shifted rapidly and substantially toward greater inclusion. In this context, should we expect that disgust rhetoric works as it has in the past? We argue that the answer to this question is no—but also yes. Disgust is simply no longer as relevant or as powerful as it once was. We also believe that disgust remains a powerful force among some portions of the population, making it an effective rhetorical strategy among that populace. There is also reason to believe it may work more broadly: some research suggests that public opposition to lesbian and gay rights may be higher than usually reported (Coffman et al. 2016). The ability to move even a small segment of an otherwise polarized population can make the use of disgust rhetoric an attractive option for elites (Hillygus and Shields 2009). We expand on existing research about disgust’s relevance for political attitudes by asking how disgust influences these attitudes in an era of increasing acceptance. In the following section, we describe our theory of how disgust rhetoric influences attitudes about gay rights.

Theory and Hypotheses

Highly passionate political debates provide elites the opportunity to use emotional language to appeal to the mass public (Holbrook and O’Shaughnessy 1984; Vakratsas and Ambler 1999). We focus on disgust rhetoric, defined as the political use of projective disgust in a message intended to communicate stigma and uncleanness in relation to social groups (Smith et al. 2011).1 It encourages social judgments intended to keep individuals and society safe and clean, which may lead to avoiding and sanctioning others seen to pose the threat of contamination, and to the endorsement of policies seen as protective. It stigmatizes populations deemed outside of a society’s conventional boundaries (Douglas 2003). It allows elites to signal that the “bounds of toleration are being reached” and may help convince citizens that a group’s threat is “weighty enough to deprive” said group of freedoms (Devlin 1965, p. 17).

We expect that disgust will be especially potent in negatively influencing attitudes about gays and lesbians. However, we do not expect disgust rhetoric to be universally effective. Changing attitudes have fundamentally altered the terms on which lesbians and gays are discussed. As issues have shifted over time (Brewer 2007), we believe that disgust is less effective for lesbians and gays as a broad category and disgust rhetoric simply may not work as intended. This is one reason that using emotional rhetoric in politics is a risky prospect. Emotional rhetoric may provoke the intended reaction—but it could also backfire. Elites using disgust rhetoric to achieve their political goals should be attentive to the threat of backlash as citizens hear and possibly reject their messages (Brehm 1966; Brehm and Brehm 1981; Mann 2010). This is particularly true since disgust objects are not fixed—old language can produce new results. The particular domains of disgust are stable, but how they apply varies significantly by culture (Haidt et al. 1997, Rozin and Haidt 2013). We contend that it is possible that disgust objects can vary over time within a culture. Social and moral disgust in particular are subject to social construction, which may change over time through experience or learning (Olatunji and Sawchuk 2005). Disgust is an emotion that signals the need for avoidance, but large numbers of Americans no longer consider lesbians and gays a group to be avoided.

Trends in disgust research support this assertion about the change in the emotional substrates of attitudes towards lesbians and gays over time. Herek (1984) found that disgust is one contributing factor toward anti-gay attitudes, consistent with later psychological research. A study of college students in 1989 asked them about attitudes toward homosexuality, including feelings of disgust when asked about public displays of affection between same-sex couples (Geller 1991). These students then went through several months of coursework on sexuality and completed a survey at the end. This study found no significant reduction in reports of feeling disgusted by the idea of public displays of affection. By comparison, a 2016 Harris poll shows that the numbers of Americans feeling such discomfort has greatly diminished over time. While asking different questions, the number of Americans who report being very or somewhat uncomfortable seeming same-sex couples holding hands in 2014 were 36 and 29% in 2015 (GLAAD 2016). This is a large reduction compared to the 1989 study.

These social changes have consequences for the politics of disgust (Hodson and Costello 2007). Efforts to break down those barriers and increase intergroup contact might reduce disgust’s efficacy not by lowering individual disgust sensitivity or making the emotion less prevalent but rather by changing disgust’s salience in evaluating particular groups. Personal contact, persuasion, and media effects have all combined to shift the emotional substrates of attitudes towards lesbians and gays in a way that we argue made disgust less relevant for at least some portions of the population (Herek 1990; Brewer 2007; Smith et al. 2009; Garretson 2015; Broockman and Kalla 2016; Harrison and Michelson 2017). Other portions may more actively seek to disguise their attitudes. The cumulative effects of these efforts may not simply make disgust less relevant and individual preferences more inclusive, but may also cause some Americans to react to disgust appeals with indignation.

Indignation can be expressed as a negative reaction directed against a speaker on behalf of the targeted population when people attempt to communicate disgust and moral stigma through rhetoric. Reactions against overt racism are sometimes expressed this way (Cowan and Hodge 1996; Uhlmann et al. 2014). We believe that the affective politics of sexuality have shifted in the face of cultural and social change, as the politics of race and racism did before them. This change involves a fundamental revision in the role of disgust in informing these attitudes. For some Americans, particularly those who endorse more traditional social arrangements, disgust will remain a powerful predictor of opposition to lesbian and gay rights. For others, more open to lesbian and gay rights, disgust may be directed toward elites who oppose those rights in the form of indignation. And some Americans may find themselves newly disgusted by the increased prevalence of lesbians and gays in American society (Casey 2016).2 Given the potential for anti-egalitarian backlash (Klarman 2004, 2012; Abrajano and Hajnal 2015; Bishin et al. 2016), we believe that pro-egalitarian backlash also exists. We argue that this pro-egalitarian backlash can take the form of indignation. This ‘new’ politics of disgust is what makes a powerful political emotion simultaneously a dangerous one: Americans may punish speakers for attempting to communicate stigma about lesbians and gays.

Indignation is an altruistic punishment fueled by anger (Sunstein and Kahneman 2007). Anger frequently co-occurs with disgust. Some respondents even seem to respond to disgust stimuli with no substantive expressions of disgust but instead with much stronger expressions of anger (Chapman and Anderson 2013). Appraisal theories of emotion argue that specific emotional reactions depend on how individuals understand and explain their experiences (Smith and Ellsworth 1985). While disgust is a prominent emotion associated with gay men in particular (Tapias et al. 2007), to the extent that respondents who support rights hear a message about gay rights that relies on disgust, (e.g. an anti-rights activist using disgusting language), they may see the messenger as blocking a desired goal and become angry. Anger can lead to an indignant altruistic punishment of offenders (Sunstein and Kahneman 2007), and supporting rights that are opposite of the ones advocated by anti-rights groups may be one way to “punish” in this situation. In the absence of barriers enforced by disgust, anger may become a viable alternative to disgust for those who would continue to enforce previous social barriers (Feldman and Huddy 2005; Nussbaum 2010; Banks and Valentino 2012; Banks 2014). In racial politics, old-fashioned racism is affectively linked to disgust and has become stigmatized. As a result, opponents of civil rights rearticulated their arguments in ways that centered on anger and outrage (Banks 2014). We do not believe this explanation applies to the politics of sexuality in the United States.

Rather, we suspect that disgust toward lesbians and gays has not yet been subsumed by other emotional substrates and translated into a fully differentiated rhetoric. This means that attempts to deploy disgust rhetoric are in a transitional state. As the examples we cite show, they rely on a set of tropes about lesbians and gays rejected by many but not all in the mass public. Disgust is a less relevant framing device, but not an irrelevant one. This means it is likely to be deployed but that, once deployed, may provoke anger among a mass public that is pushed too hard to stigmatize a group they have come to accept or no longer wish to stigmatize. Anita Bryant’s garbage rhetoric may be too strong a push for a more tolerant public and prompt backlash (Mann 2010). This backlash may even take the form of indignation and punish the anti-rights speaker by increasing support for lesbian and gay rights.

We test these ideas by using three studies to examine the following hypotheses:


Disgust rhetoric will produce feelings of disgust that lower support for rights for and decrease evaluations of lesbians and gays among some individuals.


In a more accepting time, disgust is no longer a reliable predictor of opposition to lesbian and gay rights.


To test our hypotheses, we analyze data from the 1993 American National Election Studies (ANES) and two original studies.3 The ANES data allow us to demonstrate the potential effect of disgust on political attitudes towards lesbians and gays in a less inclusive era. The first original study asks participants to describe their own emotions through thinking and writing about what made them feel that emotion towards lesbians and gays. The second study experimentally exposes participants to a news story about a same-sex marriage bill in Indiana with positive, neutral, and disgust frames. This design allows us to test the impact of elite use of emotional rhetoric as reported by the media. With this final study, we can test the causal impact of disgust rhetoric on attitudes without concern that those people most sensitive to disgust are simply more attentive and thus more reactive to this type of rhetoric. We believe there is a significant difference in how disgust informs the politics of lesbian and gay rights before the modern shift in attitudes. We first turn to the ANES 1993 Pilot test to establish a baseline of how disgust operated in the political world before major shifts in attitudes among the mass public on gay rights (Brewer 2007, 2014; Flores 2015). The ANES data demonstrates that disgust was a potent predictor of attitudes towards lesbians and gays in the early 1990s, right as significant debate over gay rights began. In contrast to the ANES data, our original studies show that the contemporary effects of disgust are less powerful and more contingent. While feeling disgust can dampen support for policies that benefit lesbians and gays, disgust must also compete with anger, an emotion that we find may offset the effects of disgust.

ANES Pilot 1993

To understand the extent to which disgust influenced support for gay rights in the past, we turn to the ANES 1993 Pilot test. The 1993 pilot interviewed 750 respondents originally interviewed in the 1992 election study. The 1993 study provides an ideal way to test the impact of disgust on attitudes because it includes an extensive battery measuring respondents’ views on homosexuality as well as multiple gay rights questions.4 The 1993 study asked three policy questions about lesbian and gay rights—whether lesbians and gays should be able to serve in the military, should be legally permitted to adopt children, and whether laws should protect gays and lesbians from job discrimination. These rights questions were also asked in the 1992 wave and we control for respondents’ prior attitudes on these policies in our models. Respondents also rated how warmly they felt toward gays and lesbians on a 101-point feeling thermometer.

Overall, attitudes toward gay rights differed significantly by policy area. A majority of respondents in both 1992 and 1993 favored allowing lesbians and gays to serve in the military and supported laws that prohibited job discrimination based on sexual orientation. In 1992, 61 percent of respondents supported allowing gays to serve in the military. By fall 1993, the month before the signing of “Don’t Ask, Don’t Tell”, 62 percent supported military service, with 10 percentage point more respondents in the “strong support” category than the previous year. Similarly, 64 percent of respondents in 1992 and 63 percent in 1993 support job discrimination protections. In contrast, there was strong opposition to allowing lesbians and gays to adopt children and evaluations of lesbians and gays themselves were quite cold. In both 1992 and 1993, 71 percent of respondents opposed adoption rights for same-sex couples. In 1993, respondents gave lesbians and gays an average feeling thermometer rating of 39, rating them on the “cold” end of the thermometer.

To measure disgust, the survey includes two questions directly tapping respondents’ feeling of disgust about homosexuality, asking respondents (1) whether they felt disgusted about the “very idea of homosexuality” and (2) whether homosexuality made them uncomfortable. We use a measure that combined these questions into a 4-point scale from “not uncomfortable or disgusted” to “strongly disgusted” with higher values indicating more disgust. Most ANES respondents express some level of disgust or discomfort with homosexuality5—45 percent said they were neither disgusted nor uncomfortable with 42 percent expressing disgust and 13 percent expressing discomfort if not outright disgust. Men were more likely to be disgusted by homosexuality than women (X2 = 15.02, p < 0.01) as were respondents who endorsed more traditional views of morality (X2 = 158.52, p < 0.01),6 Republican respondents compared to Democrats and Independents (X2 = 50.00, p < 0.01) and conservative respondents compared to liberals and moderates (X2 = 40.81, p < 0.01). These associations may be related to the higher prevalence of disgust sensitivity among conservatives (Inbar et al. 2009) or to more consistent use of disgust rhetoric by conservative religious and political leaders (Herman 1996).

While baseline support varies across different types of policies, the effect of disgust on these attitudes is consistent across the four policy measures. Respondents who felt more disgusted about homosexuality were significantly less supportive of rights. These data cannot test the causal direction, whether disgust is a byproduct of political talk or the antecedent of political identification. However, there are clear connections between disgust about homosexuality and political attitudes. We use the 1993 data to model the effect of disgust on support for job protections, the right to serve in the military, adoption rights, and the feeling thermometer using OLS. We scaled the policy dependent variables to vary between 0 and 1, with higher values indicating more support for lesbian and gay rights. The feeling thermometer varies between 0 and 100, with higher values indicating more warmth toward lesbians and gays. The models also control for respondents’ partisanship and ideology (7 point scales), gender, moral traditionalism, and policy attitude measured in 1992, all rescaled to vary between 0 and 1 with higher values indicating Republicans, more conservative, and more morally traditional, and a dummy variable for gender with 1 equal to women.

As Fig. 1 demonstrates, disgust significantly decreases evaluations of and support for rights policies for lesbians and gays, even controlling for prior attitudes. Figure 1 displays the impact of moving across the disgust scale on attitudes. Disgust decreases support for job discrimination protections by.18 (SE = 0.05) or 19% of the scale, military service by 0.29 (0.05), and lowers already low support for adoption rights by 0.15 (0.03). Similarly, respondents who find homosexuality very disgusting rate lesbians and gays more negatively by 27 percentage points than respondents more comfortable with homosexuality. Disgust is the only variable that consistently predicts lesbian and gay rights attitudes, and is a significantly stronger predictor of feelings toward lesbians and gays than partisanship [F(1, 350) = 20.85, p < 0.01] and ideology [F(1, 350) = 9.63, p < 0.01]7 and a stronger predictor of attitudes about military service than moral traditionalism [F(1342) = 4.75, p < 0.05] and ideology [F(1342) = 4.40, p < 0.04]. Emotion substantially structured attitudes about lesbian and gay rights in 1993 over and above the effect of prior attitudes about these rights or political identities such as partisanship.
Fig. 1

Disgust decreases support for lesbian and gay rights 1993

Study 1: Bottom-up Disgust Manipulation

In the two decades since the 1993 ANES, support for gay rights has substantially increased.8 Yet, even with the overall increase, this support is not universal and we expect that disgust toward homosexuality may still undergird opposition to rights. We use a bottom-up induction task to ask respondents to conjure feelings of disgust about lesbians and gays. We then utilize their open-ended answers to describe the nature of that disgust—why or what about lesbians and gays disgusted them. This study demonstrates that there is significant variation in the causes of disgust about lesbians and gays, that disgust may not be limited only to prompting restrictive attitudes, and that respondents may feel not only disgust but also anger/indignation in the face of disgust rhetoric (Kam and Estes 2016). Overall, this analysis shows that disgust targeted at lesbians and gays can be evoked in the public, but that anger is present as well. When instructed to feel disgusted about lesbian and gays as a group, many respondents expressed disgust about traditional issues like physical purity. Yet, others responded to the same instruction by becoming indignant and disgusted at those who would discriminate against lesbians and gays.

We ran a study through Amazon’s Mechanical Turk (MT) workplace in summer 2014 that randomly assigned respondents to feel a discrete emotion (Berinsky et al. 2012). We recruited 317 MT respondents to take a survey of their views of “current political issues.”9 The MT sample is a convenience sample made up of respondents who have access to the Internet, and it skews toward young (average age = 34), educated (57% of respondents have a college degree or higher), and male (52% of respondents). The sample is diverse politically but does tilt liberal (ideological self-placement: 53% liberal, 24% moderate, 23% conservative) and Democratic (partisan identification: 57% Democrats, 19% Independents, 24% Republicans). Most importantly, though, we randomly assigned people to either describe disgust or not, and the randomization was successful, and thus, it is the experimental treatment rather than any underlying difference that should influence expressions of disgust (see Table A.1 in the appendix for the experimental balance check).

The study randomly assigned respondents to feel disgust or not feel any emotion.10 In order to manipulate emotions, we utilize a bottom up emotion induction task where respondents focused on a group, object, or event that causes them to feel a particular emotion and describe what makes them feel that emotion in detail.11 Through the exercise of describing what causes the emotion, respondents start to feel that emotion (Valentino et al. 2009, 2008; Gadarian and Albertson 2014; Lerner and Keltner 2001).

Respondents were asked to concentrate on what made them feel either disgusted toward gays and lesbians in the disgust condition or to explain what came to mind when thinking about lesbians and gays in the neutral condition.12 See the exact wording for the disgust condition in footnote 12. We compared respondents in this treatment to those respondents in the control condition where respondents wrote about what came to mind when they thought about gays and lesbians.13 Across conditions, we hold the target group constant and vary whether respondents are asked to feel a discrete emotion.

Writing about disgust increases respondents’ level of expressed disgust but also their anger over and above the effect of demographics, suggesting that in this current period, communications that direct citizens to feel disgusted may not always be successful. After the writing exercise, Study 1 respondents answered a series of questions about their emotional reactions. We asked how strongly they felt: disgusted, sickened, angry, furious, hopeful, relieved, enthused, and proud on a 9-point scale from “not at all” to “strongly”. We created three scales from these separate measures: Disgust (disgusted, sickened: Cronbach’s alpha = 0.91), Anger (angry, furious: Cronbach’s alpha = 0.88), and Enthusiastic (hopeful, relieved, enthused, and proud: Cronbach’s alpha = 0.87). The general emotion measures in Study 1 vary between 1 and 8, with higher values meaning stronger emotion.14 One thing to note is that these are measures of emotions themselves, not emotions attached to lesbians and gays particularly. For comparison, Table 1 also displays an OLS model of the impact of gender, partisanship, ideology, and moral traditionalism on disgust about homosexuality from the 1993 ANES data. In the ANES models, the dependent variable is the same disgust/discomfort with homosexuality discussed previously. The measure varies between 0 and 1 with higher values indicating more disgust about homosexuality.
Table 1

Determinants of disgust over time


ANES 1993

Experiment 1: neutral condition

Experiment 1: among all respondents








Disgust treatment


2.07*** (0.285)

1.17*** (0.280)

−1.34*** (0.242)


−0.18*** (0.045)

−0.26 (0.385)

−0.40 (0.382)


0.00 (0.307)

−0.08 (0.301)

−0.16 (0.261)



−1.49** (0.682)

−1.71** (0.676)

0.65 (0.662)

−0.60 (0.572)

−1.03* (0.561)

0.65 (0.486)


0.12 (0.081)

−0.77 (0.915)

−0.28 (0.907)

1.08 (0.888)

0.01 (0.705)

−0.25 (0.691)

0.44 (0.598)


0.22** (0.101)

0.40 (1.142)

0.12 (1.132)

−2.48** (1.109)

0.03 (0.904)

−0.28 (0.887)

−1.45* (0.767)


0.00 (0.001)

−0.01 (0.014)

−0.02 (0.014)

−0.00 (0.014)

−0.01 (0.011)

−0.02 (0.011)

−0.00 (0.010)

Morally tradition

0.56*** (0.108)

1.65* (0.946)

0.42 (0.937)

−1.21 (0.918)

1.56** (0.733)

0.70 (0.719)

−0.27 (0.622)

Disgust sensitivity


0.27 (0.993)

−0.24 (0.984)

1.22 (0.964)

0.02 (0.742)

−0.48 (0.727)

1.10* (0.630)


0.10 (0.085)

3.83*** (0.853)

5.36*** (0.845)

3.70*** (0.828)

2.88*** (0.694)

4.61*** (0.681)

3.63*** (0.589)

















p < .1; ** p < .05; *** p < .01

To compare the impact of demographics on the strength of emotion, we can compare respondents Study 1′s neutral condition to the ANES respondents since lesbians and gays were salient for both sets of respondents. In Study 1, respondents in the neutral condition expressed low levels of disgust (Mdisgust = 2.69, SD = 2.31), anger (Manger = 2.86, SD = 2.29), and enthusiasm (Menthus = 4.03, SD = 2.28) after thought listing about lesbians and gays, indicating that these emotions are not necessarily close to the surface. Comparing across the samples, one major take away is that demographics are a much stronger predictor of disgust in the 1993 ANES than in the 2014 MT sample. In 1993, demographics such as gender, partisanship and ideology consistently determined disgust attached to lesbians and gays, but by 2014, only moral traditionalism remained a clear determinant of such disgust. Unlike respondents in the ANES, in Study 1, neither partisanship nor ideology was a significant determinant of disgust and ideology only mattered for enthusiasm. As respondents in the neutral condition went from liberal to conservatives, the less enthusiastic they felt by 2.48 (SE = 1.109) or almost a third of the 9-point scale.

The second major take-away from Table 1 is that the disgust treatment not only significantly increased respondents’ expressed levels of disgust but also increased anger significantly and depressed enthusiasm levels. Respondents who wrote about what made them disgusted about lesbians and gays expressed more disgust by 2.07 on the 9-point scale (or about 23% of the scale) compared to those in the neutral condition but also increased their anger by 1.17 on average and dampened their enthusiasm by a similar amount (b = −1.34, SE = 0.24). These models demonstrate that even when targeting one emotion, communications may trigger multiple emotions that could affect policy attitudes in multiple ways. Because respondents in Study 1 wrote down their thoughts as part of the exercise, we have qualitative evidence about what considerations they brought to mind in each condition.

Respondents articulated a wide variety of associations in response to the open-ended prompts. Table 2 provides several qualitative examples of responses. We analyzed a random sample of 30 open-ended responses for content and tone, 14 from the neutral condition and 16 from the disgust condition. The neutral group’s responses overwhelmingly presented a procedural equality frame, centering generally on the provision of equal rights to lesbians and gays. The disgust group’s responses were varied, falling into several categories. These include (1) pro-gay respondents who were disgusted by the ways that lesbians and gays are treated by the law and by their opponents, (2) pro-gay respondents who did not express disgust but rather anger directed at unequal treatment and the opposition, and (3) anti-gay respondents who expressed disgust based on sex acts and moral violations. This latter category reflects theoretical expectations about how disgust should act in the political world, connecting disgusting phenomenon (e.g. sexual deviance) with marginalized groups (e.g. those groups most likely to engage in deviant behavior). Though we directed respondents to feel a single, discrete emotion in each treatment, some respondents showed reticence to evoke those emotions, and buffered against these feelings by doing things like expressing indignation at the treatment of lesbians and gays rather than the behavior of lesbians and gays themselves. This demonstrates that while disgust may be increased through a relatively simple induction, the causes behind and targets of the disgust can vary significantly—and in ways that theory typically does not predict.
Table 2

Examples of statements in the conditions

No Emotion


I think that they should have the same rights as everyone else in the country. I do not personally want to be gay or anything, but I really don’t want anyone prying into my private life

I think it is disgusting because life should be about a man and a woman and they should be able to reproduce offspring

About the only thing that comes to mind is that gay people want to be treated like everyone else, and I can respect that. They have a different lifestyle than mine, but that’s okay

I felt disgusted… during the NFL draft a openly gay candidate was chosen and his lover embraced him with a huge kiss

Common stereotypes would say that these people are child molesters and murderers and are tainting society when in reality those things have nothing to do with what sex someone is interested in

I felt disgusted by the lack of compassion some people choose to show for gays and lesbians because of their hatred for those who are different from them

The qualitative coding suggests that disgust rhetoric may provoke disgust directed at the speaker and not the intended target. In our second study, we specifically test the efficacy and implications of disgust rhetoric.

Study 2: Disgust as Political Rhetoric

In Study 2, we use a newspaper story to evoke disgust to get closer to the ways that political elites using disgust rhetoric may influence policy attitudes. This allows us to increase external validity. Study 2 randomly assigned respondents to read one of three versions (positive, neutral, disgust) of a story about an actual proposed constitutional amendment to ban same-sex marriage in Indiana. All respondents read about the proposed amendment itself and the groups that rallied either in support or opposition to the amendment. We varied the emotion of the language that the groups used in their advocacy.

Table 3 below outlines the differences between the conditions. In the neutral condition, the quoted speaker used non-emotional language to advocate that people contact their legislators. In the positive emotion condition, the quoted speaker focused on the similarities between same-sex families and their neighbors. In the disgust condition, the quoted speaker advocated that people support the bill and evoked disgust by emphasizing the threats posed to the polity by deviants. In addition to the speaker’s language, the stories varied the language that appeared on a protest sign being held by a man in an accompanying photograph. In the neutral condition, the sign read “Call your representatives”. In the positive condition, the sign read “Love, Family, Commitment”, and in the disgust condition, the sign read “The Anus is a Grave”. We included a positive vignette as to test whether positive rhetoric similar to that used by the lesbian and gay movement could inspire positive feelings in our respondents and increased support (Harrison and Michelson 2012). We used a generally positive framing in order to mimic actual messages from advocates as closely as possible, similar to our efforts to mimic actual disgust rhetoric. See the appendix for the full transcript of the treatment.
Table 3

Experimental treatments

Our disgust treatment closely mirrors the kinds of disgust rhetoric deployed by anti-gay elites. For example, language in the disgust treatment states, “We have a moral duty to defend the institution of traditional marriage as being between one man and one woman,” Price said. “Two men will never be able to reproduce. Anal penetration can never result in the creation of new life.” This language is similar to a statement made by Pat Robertson on his show the 700 Club (Post 2014). Similarly, our positive condition mimics messaging that lesbian and gay rights advocates have used to increase support in marriage debates (Fisher 2009; Solomon 2014).

We used MT to recruit a (separate) sample of 636 respondents to participate in a study on public opinion in 2014. Study 2′s MT sample is diverse on several dimensions such as partisanship (39% Democrats, 17% Republicans) and ideology (53% liberals, 22% conservative), although it skews young (average age = 33) and male (66%). As in our first experiment, there is balance across the conditions, meaning that the randomization was successful and it is the experimental treatment rather than any underlying difference that should influence attitudes.

After reading the treatment stories, respondents were asked how disgusted, worried, angry, hopeful, and sickened they felt on a scale from 1 (not at all) to 8 (very strongly), which we rescaled to vary between 0 and 1 with higher values being stronger emotions.15 The disgust treatment successfully increased respondents’ negative emotional reactions over the neutral control condition and decreased the positive emotion of hope. As expected, the disgust treatment significantly raised respondents’ level of disgust by 0.15 on the 0–1 scale or 50% over the control condition (Mcontrol = 0.30, Mtreat = 0.45, t = 4.52, p < 0.01) but did not raise respondents’ worry (Mcontrol = 0. 31, Mdisgust = 0.32).

As Study 1 demonstrated, respondents often counter disgust, and persuasive language about disgust may increase the probability that respondents react against the language itself (Brehm and Brehm 1981). The disgust story not only raised feelings of disgust, but also affected anger and enthusiasm. The disgust treatment significantly increased anger by a similar magnitude of 0.13 (Mcontrol = 0.37, Mdisgust = 0.50, t = 4.22, p < 0.01) and also decreased enthusiasm by 0.21 or a 58% decrease over the control condition (Mcontrol = 0.39, Mdisgust = 0.18, t = 7.98, p < 0.01). In contrast, subjects reacted emotionally to the positive story in largely identical ways to the neutral, emotion-less control condition. Table 4 shows an OLS model of the effect of the treatments on feelings of disgust and anger that controls for respondents’ disgust sensitivity16 as well their gender, religious identity, partisanship, ideology, and education. Even controlling for the effects of psychological and political predispositions, the disgust treatment significantly increased both disgust and anger. When modeling the effect of the treatment on rights attitudes, we consider how both anger and disgust affect opinions.
Table 4

Disgust treatment significantly increased disgust and anger




Disgust sensitivity scale

0.34 (0.081)

0.37 (0.083)

0.06 (0.091)

0.09 (0.090)

Disgust treatment

0.14 (0.030)

0.14 (0.029)

0.14 (0.033)

0.14 (0.031)

Positive treatment

−0.02 (0.030)

−0.01 (0.029)

0.05 (0.033)

0.06 (0.032)



−0.01 (0.026)


0.05 (0.028)



−0.01 (0.009)


−0.01 (0.010)

Born again Christian


0.03 (0.035)


−0.04 (0.038)



−0.12 (0.067)


−0.07 (0.072)



−0.15 (0.070)


−0.33 (0.076)


0.15 (0.044)

0.28 (0.060)

0.34 (0.050)

0.49 (0.065)











Source: 2014 MT Rights Experiment. Model: OLS. Bold indicates coefficient is significant at p < 0.05. Dependent variable is a 9-point scale of how strongly respondent felt the emotion, rescaled from 0 to 1 with higher values indicating stronger emotion

Respondents answered a series of questions that asked how supportive they were of a variety of lesbian and gay rights issues, some of which are already law and some that are still being negotiated at the state and federal level. These rights included allowing gays and lesbians to adopt children, extending hate crime and job discrimination protections based on sexual orientation, providing health insurance to same-sex partners, and allowing same-sex partners to immigrate to the United States. Respondents also answered two questions specifically about marriage—whether they generally support same-sex marriage and, whether as an Indiana voter, they would vote for an amendment to ban same-sex marriage (−1), keep the status quo of a ban by statute (0), or legalize same-sex marriage (1).17 Respondents also evaluated whether they would vote for a lesbian or gay candidate for president and whether federal funding for AIDS/HIV should be increased, decreased, or kept the same. Except for the Indiana marriage question, each of the other questions is a 5-point scale that varies between 0 and 1 with higher values indicating higher levels of support for rights or more support for funding.

In Table 5, we use a mediation model that accounts for the level of disgust each respondent expressed to model the effect of disgust on rights attitudes. Because the disgust treatment also increased anger, we also include respondents’ level of anger in the mediation models and check anger’s effect on rights opinions. By using these measures, we can test whether disgust is mediating the relationship between the experimental manipulation and support or opposition to lesbian and gay rights. Consistent with our hypothesis, the disgust treatment led respondents to significantly decrease their support for a variety of rights and does so through increasing feelings of disgust. Table 5 shows the effect of being in the treatment condition and feelings of disgust on the rights questions including pretreatment variables (ideology and partisanship).18 The models in Table 5 show that feelings of disgust significantly mediate the effect of the disgust condition for 7 of the 9 rights questions. In other words, the disgust treatment decreases support for a variety of rights, including marriage, immigration rights for same-sex partners, and protection from job discrimination, through increasing respondents’ level of disgust. Feelings of disgust dampen support for rights in a substantively large way, varying between 10% of the scale (job discrimination protection) to 28% of the scale (marriage). Even in a liberal-leaning sample, invoking disgust decreased their support for policies that are already legal in many states, like allowing same-sex couples to adopt children. Disgust did not significantly affect opinions about AIDS spending or hate crime protections, suggesting that disgust may not expand to all issues easily or at least not when the disgusting language is geared toward a specific policy like marriage.
Table 5

Disgust decreases support for gay rights




Gay president

Hate crime


AIDS spending

Job discrim.


Indiana SSM

Disgust treatment

0.03 (0.02)

0.02 (0.02)

0.03 (0.03)

0.02 (0.03)

0.01 (0.02)

0.01 (0.02)

0.03 (0.02)

0.03 (0.02)

0.13 (0.05)

Positive treatment

0.02 (0.02)

0.03 (0.02)

0.03 (0.02)

0.01 (0.03)

0.01 (0.02)

0.00 (0.02)

0.01 (0.02)

0.02 (0.02)

0.11 (0.05)


0.21 (0.04)

0.27 (0.04)

0.23 (0.04)

−0.08 (0.05)

−0.16 (0.04)

0.06 (0.03)

−0.10 (0.04)

−0.28 (0.04)

−0.55 (0.09)


0.29 (0.04)

0.36 (0.04)

0.36 (0.04)

0.26 (0.04)

0.29 (0.04)

0.08 (0.03)

0.22 (0.03)

0.39 (0.04)

0.60 (0.09)


0.08 (0.05)

0.00 (0.06)

0.02 (0.06)

−0.02 (0.06)

−0.01 (0.05)

−.03 (0.04)

0.07 (0.05)

0.04 (0.05)

0.01 (0.11)


−0.51 (0.05)

−0.46 (0.06)

−0.55 (0.06)

−0.30 (0.06)

−0.45 (0.06)

−0.31 (0.05)

−0.37 (0.05)

−0.57 (0.06)

−0.99 (0.12)


0.87 (0.03)

0.88 (0.03)

0.85 (0.03)

0.78 (0.03)

0.84 (0.03)

0.68 (0.02)

0.85 (0.02)

0.89 (0.03)

0.92 (0.06)





















Mediation by disgust

Mediation effect










95% CI

[−0.05, −0.01]


[−0.05, −0.02]

[−0.03, 0.00]

[−0.04, −0.01]

[−0.00, 0.02]

[−0.03, −0.00]

[−0.06, −0.02]

[−0.12, −0.04]

Direct Effect

0.03 [−0.01, 0.07]

0.02 [−0.02, 0.07]

0.03 [−0.02, 0.08]

0.02 [−.03, 0.07]

0.01 [−0.03, 0.06]

0.01 [−.03, 0.05]

0.03 [−0.01, 0.07]

0.03 [−.01, 0.08]

0.13 [0.03, 0.23]

Total Effect

0.00 [−0.03, 0.03]

−0.02 [−0.05, 0.02]

−0.01 [−0.04, 0.03]

0.01 [−0.04,0.06]

−0.01 [−0.05, 0.03]

0.02 [−0.02, 0.06]

0.02 [−0.01, 0.06]

−0.01 [−0.04, 0.03]

0.05 [−0.02, 0.13]

Mediation by anger

Mediation effect










95% CI

[.02, .07]








[.05, .14]

Direct Effect

0.03 [−0.01, 0.07]

0.02 [−0.02, 0.07]

0.03 [−0.02, 0.08]

0.02 [−0.03, 0.07]

0.01 [−0.03, 0.06]

0.01 [−0.03, 0.05]

0.03 [−0.01, 0.07]

0.03 [−0.01, 0.08]

0.13 [0.03, 0.23]

Total Effect

0.07 [0.01, 0.13]

0.07 [−0.04, 0.03]

0.08 [0.01, 0.15]

0.05 [−0.01,0.13]

0.06 [0.01, 0.12]

0.05 [−0.01, 0.12]

0.02 [−0.02, 0.07]

0.09 [0.02, 0.16]

0.21 [0.08, 0.36]

Bold indicates coefficient is significant at p < 0.05

Across the rights models in Table 5, anger significantly increases support for rights across 8 of the 9 models in magnitudes almost identical to the effect of disgust. The mediation coefficients also show that the disgust treatment affected attitudes through increasing anger, not simply that pre-occurring anger influenced opinion.19 Anger strongly increases support for lesbian and gay rights across the spectrum, and we can speculate that this indignation is fueled by a reaction to the anti-rights rhetoric present in our disgust treatment.20 Combined with the findings about the different objects of disgust in the open-ended disgust measures in Study 1, disgust rhetoric may be a powerful tool but it does not come without a price. This price may be found in a form of emotional backlash among segments of the population who are supportive of rights and who react with indignation and in ways that anti-gay elites would not want them to.

Conclusion: From Disgust to Rights

Disgust shapes our politics. It is a powerful but contingent rhetorical tool that can powerfully shape public attitudes, especially on issues of sexual purity, but that efficacy must come with a strong caveat: our findings show that some members of the public will reject disgust rhetoric as an indignant reaction against the speaker. If they dislike the message they hear, they may become angry. And sometimes, that anger offsets any elicited disgust and respondents may relate disgust to gay rights opponents rather than its intended targets. Our argument about the changing politics of disgust may be partially rooted in changing norms that make it more socially acceptable to disguise anti-gay attitudes, but we argue that social desirability is an insufficient explanation for the indignant backlash that we observe in our results. If our young and liberal respondents were simply acceding to more inclusive norms, we would be unlikely to observe an indignant backlash—it is more likely that there would be no effect, with something that is not deeply felt.

We believe that disgust is simply less relevant in a world where 75% of Americans report personally knowing lesbians and gays. It is harder to discriminate against someone or find them disgusting when you are personally connected to them. This depth of personal contact likely explains the pro-gay anger and disgust toward gay rights opponents expressed by respondents in Study 1. But reduced relevance is not irrelevancy. Disgust proved effective in dampening support for a wide variety of rights in Study 2. Public acceptance of lesbians and gays has increased, but not universally. Some Americans remain opposed, and others may be flexible in their support. This presents anti-gay rights elites with an opportunity that we explore in Study 2: using disgust language and imagery to influence attitudes towards lesbians and gays. The most effective use of this tool appears to be not only to remind people that lesbians and gays are “disgusting” but to also draw the connection between this disgust and policies that elites want to oppose. The findings in Study 2 suggest that, when prompted by political elites to consider both disgust and relevant policies, individuals make a connection between the emotion and the policies in question. To this extent, disgust rhetoric still works with some people, shifting attitudes on policy in a manner similar to the 1993 ANES data. This may come at the cost of inadvertently provoking anger and indignation among other people.

In both of our studies, considering disgust aimed at lesbians and gays produced significant emotional backlash. In Study 1, this backlash takes the form of disgust on behalf of lesbians and gays, directed at their opponents. Some respondents also responded with anger or indignation, stating that lesbians and gays are treated unfairly. The disgust treatment in Study 2 also provoked anger, and in turn, anger increased support for rights. Respondents in Study 2 did not provide reasons for their anger, but given the open-ended responses from Study 1, we believe that anger occurring in response to disgust rhetoric and increasing support for rights observed in Study 2 comports with our content analysis of the open-ended responses and represents a pro-gay backlash.

While our analysis is limited to the gay rights context, our findings have implications beyond these issues. Our results suggest two major directions for future research. First, what kinds of rhetoric are most effective at prompting disgust in a political environment? As we note, many disgust studies use direct treatments such as exposure to disgusting stimuli to prompt participants to actually experience the emotion. These studies can demonstrate powerful effects, but they do not go far enough in approximating how politics actually works. A better understanding of the types and styles of disgust rhetoric is itself a useful topic, but we also suggest a practical result for advocacy groups—in better understanding disgust rhetoric, we can also gain more leverage in countering such language. Similarly, we can find whether disgust rhetoric targeted at other populations is equally effective. While disgust is uniquely predictive on issues of sexual purity, populations that are negatively targeted through such language (e.g. the implication that immigrants bring disease) may be subject to some of the same effects. The dynamics of backlash that we document in the gay rights context may not appear in areas where opinion has been more static.

Second, we wonder whether the nexus of negative emotions that often appears in rights-restricting discourses is necessary for them to be effective. These emotional messages may be finely tuned enough to be considered a form of affective microtargeting, but our results showing emotional backlash make us skeptical of the real efficacy of such microtargeting. Emotional rhetoric is a powerful but not particularly subtle tool. Negative emotions often co-occur with one another, may not be easily separable, and may not work as intended when provoked by political discourse. Further study of these clusters of negative emotion in politics and their possible unintended consequences would do much to link disgust research to other emotions.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The original research in this article were reviewed by the Syracuse University Institutional Review Board as study 13-159.


  1. 1.

    We focus on emotional appeals in the form of disgusting political rhetoric because, in the political world, discourse must be used to shape attitudes. Lab studies that create disgust through unclean rooms and fart spray show a strong connection between disgust and attitudes, but the political world does not easily mimic the laboratory. Elites rarely use physically disgusting stimuli to influence attitudes, instead relying on language and imagery. A speaker creatively employing physically disgusting stimuli as part of their message is unlikely to find an attentive or receptive audience.

  2. 2.

    Our data do not speak to this last conjecture, but recent research suggests that this is an important factor to consider in future research on the politics of disgust (Casey 2016).

  3. 3.

    Data and replication code for the analyses is available at

  4. 4.

    This is the only wave of the ANES that includes both gay rights questions and disgust questions that are unconnected from candidate evaluations. Disgust originates from a need to protect against toxic substances and groups, and we expect that the political reaction is to support policies that will most effectively protect the polity from the perceived source of disgust. People who are disgusted by Bill Clinton and people disgusted by homosexuality should support very different types of policies to avoid being “contaminated” by the objects of disgust. Political elites who draw on disgust imagery to oppose gay rights policies use homosexuality as their target, as the ANES questions do.

  5. 5.

    Disgust is measured in the 1993 wave of the ANES.

  6. 6.

    We combine four questions asked in the 1992 wave of the ANES into a “moral traditionalism” index. People were asked: 1. whether people should be more tolerant of those who live by their own moral standards, 2. if the country would have fewer problems if there was more emphasis on traditional family ties, 3. if newer lifestyles contribute to the breakdown of society, and 4. if it world is always changing and we should adjust our view of morality to changing times. We combine these into an index (Cronbach’s alpha = 0.65) rescaled from 0 to 1 so that higher values indicate more support for traditional morality.

  7. 7.

    These tests were calculated using the postestimation “test” command in Stata.

  8. 8.

    For example, in the 2012 ANES, 61 percent of respondents supported adoption rights for gays and lesbians, almost doubling support over the 1993 wave. Support for allowing gays to serve in the military similarly increased from 55 percent support in 1992 to 86 percent in 2012.

  9. 9.

    This part of the study was part of a larger study of emotions that included an additional two conditions that are not analyzed here. The larger sample has 659 respondents.

  10. 10.

    There were two additional conditions where respondents were asked to feel empathy or anger designed for a different purpose that will not be analyzed here.

  11. 11.

    Disgust research typically involves two approaches to eliciting disgust in respondents. One method is to expose respondents to a variety of physically disgusting stimuli. Examples of these kinds of treatments include exposure to bad smells such as ‘fart gas’, asking respondents to complete tasks in unclean environments, displaying disgusting videos, and asking respondents to recall physically disgusting experiences (Schnall et al. 2008). Variations on this use a combination of photographs of disgusting behaviors and implicit moral violations to elicit disgust responses, such as a photograph of a man eating a handful of worms (Smith et al. 2011). Alternatively, disgust researchers use techniques such as asking respondents to express judgments on a variety of moral transgressions, ranging from the trivial to the severe, usually presented as vignettes (Chapman and Anderson 2013). Study 1 varies this second technique by simply asking respondents to focus on what makes them feel disgusted.

  12. 12.

    The disgust prompt read: We’re interested in how people react to different groups. There’s been a great deal of attention lately to gays and lesbians. Please describe something about gays and lesbians that made you feel DISGUSTED. Please describe how you felt as vividly and in as much detail as possible. Think about the way the issues are talked about, recent court cases, and real world events. Examples of things that have made some people feel DISGUST are statements made on the media, things said during political debates and campaigns, or how everyday people discuss gays and lesbians. It is okay if you don’t remember all the details, just be specific about what exactly it was that made you feel DISGUST and what it felt like to be DISGUSTED. Take a few minutes to write out your answer.

  13. 13.

    The control prompt read: We’re interested in how people react to different groups. There’s been a great deal of attention lately to gays and lesbians. Please describe something that comes to mind when you think about gays and lesbians. Think about the way the issues are talked about, recent court cases, and real world events. Examples of things that may come to mind are statements made on the media, things said during political debates and campaigns, or how everyday people discuss gays and lesbians. It is okay if you don’t remember all the details, just be specific about what exactly it was that comes to mind. Take a few minutes to write out your answer.

  14. 14.

    Across all models, demographics are scaled to vary between 0 and 1 with higher values indicating Republican (a 7-point scale), conservative (a 7-point scale), more traditional morality (see question wordings in the ANES section) (16-point scale) and disgust sensitivity (16-point scale). Age is measured in years and varies between 18 and 76 in Study 1. Gender and sexual orientation are dummy variables with 1 equal to identifying as female and heterosexual. Disgust treatment is the effect of being assigned to the disgust treatment (coded as 1) compared to the neutral condition (coded as 0).

  15. 15.

    We combine “disgusted” and “sickened” together into an index (Cronbach’s alpha = 0.88).

  16. 16.

    We measure respondents’ sensitivity to disgust prior to exposure to the treatment using the DS-R scale, which includes multiple measures of comfort with objects like dead bodies or maggots (Olatunji et al. 2007).

  17. 17.

    These questions were asked prior to the Supreme Court’s decision legalizing same-sex marriage nationwide in 2015.

  18. 18.

    We include anger only in the second stage model that predicts opinion as a function of emotions and the treatment. Including anger as a predictor of disgust would imply that anger was a cause of disgust and would thus be a post-treatment confounder, which would violate the sequential ignorability assumption that underlies these mediation models.

  19. 19.

    It is not the case that it is different people become angry and disgusted; these emotions correlated at 0.67 (p < 0.01). Yet, the effects of these emotions are countervailing.

  20. 20.

    One alternative mechanism is that anger is an indicator of empathy here, but we do not have the requisite measure of empathy to test this mechanism.



The authors are listed in alphabetical order. We would like to thank Aaron Hoffman, Seth Jolly, Dan McDowell, Spencer Piston, Josh Thompson, participants at Purdue University, the Moynihan Research Workshop, and the Midwest Political Science Association 2014 annual meeting for feedback on earlier versions of this paper. We also owe thanks to the editor and the anonymous reviewers for their feedback and subsequent improvements to the paper. Finally, we thank the Department of Political Science and the Maxwell School at Syracuse University for their support of this project.

Supplementary material

11109_2017_9412_MOESM1_ESM.docx (26 kb)
Supplementary material 1 (DOCX 25 kb)


  1. Abrajano, M., & Hajnal, Z. L. (2015). White Backlash: Immigration, Race, and American Politics. Princeton: Princeton University Press.CrossRefGoogle Scholar
  2. Albertson, B., & Gadarian, S. K. (2015). Anxious Politics: Democratic citizenship in a threatening world. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  3. Ayres-Deets, A. (2013). Ben Carson compares homosexuality to bestiality and pedophilia. Mic. Retrieved from
  4. Banks, A. J. (2014). Anger and racial politics: The emotional foundation of racial attitudes in America. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  5. Banks, A. J., & Valentino, N. A. (2012). Emotional substrates of white racial attitudes. American Journal of Political Science, 56(2), 286–297. doi: 10.1111/j.1540-5907.2011.00561.x.CrossRefGoogle Scholar
  6. Berinsky, A., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research:’s Mechanical Turk. Political Analysis. doi: 10.1093/pan/mpr057.Google Scholar
  7. Bishin, B. G., Hayes, T. J., Incantalupo, M. B., & Smith, C. A. (2016). Opinion backlash and public attitudes: Are political advances in gay rights counterproductive? American Journal of Political Science, 60(3), 625–648. doi: 10.1111/ajps.12181.CrossRefGoogle Scholar
  8. Brehm, J. W. (1966). A theory of psychological reactance. New York: Academic Press.Google Scholar
  9. Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. San Diego: Academic Press.Google Scholar
  10. Brewer, P. R. (2007). Value war: Public opinion and the politics of gay rights. Lanham: Rowman & Littlefield Publishers.Google Scholar
  11. Brewer, P. R. (2014). Public opinion about gay rights and gay marriage. International Journal of Public Opinion Research, 26(3), 279–282. doi: 10.1093/ijpor/edu029.CrossRefGoogle Scholar
  12. Broockman, D., & Kalla, J. (2016). Durably reducing transphobia: A field experiment on door-to-door canvassing. Science, 352(6282), 220–224. doi: 10.1126/science.aad9713.CrossRefGoogle Scholar
  13. Canaday, M. (2009). The straight state: Sexuality and citizenship in twentieth-century America. Princeton: Princeton University Press.Google Scholar
  14. Casey, L. S. (2016). The politics of disgust: Public opinion toward LGBTQ people and policies (Unpublished doctoral dissertation). Ann Arbor, MI: University of Michigan.Google Scholar
  15. Chapman, H. A., & Anderson, A. K. (2013). Things rank and gross in nature: A review and synthesis of moral disgust. Psychological Bulletin, 139(2), 300–327. doi: 10.1037/a0030964.CrossRefGoogle Scholar
  16. Clifford, S., & Wendell, D. G. (2015). How disgust influences health purity attitudes. Political Behavior, 38(1), 155–178. doi: 10.1007/s11109-015-9310-z.CrossRefGoogle Scholar
  17. Clifford, S., Jerit, J., Rainey, C., & Motyl, M. (2015). Moral concerns and policy attitudes: Investigating the influence of elite rhetoric. Political Communication, 32(2), 229–248. doi: 10.1080/10584609.2014.944320.CrossRefGoogle Scholar
  18. Coffman, K. B., Coffman, L. C., & Ericson, K. M. M. (2016). The size of the LGBT population and the magnitude of anti-gay sentiment are substantially underestimated. Management Science. doi: 10.1287/mnsc.2016.2503.Google Scholar
  19. Cosmides, L., & Tooby, J. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. M. Haviland-Jones (Eds.), Handbook of Emotions (2nd ed.). Guilford Press.Google Scholar
  20. Cowan, G., & Hodge, C. (1996). Judgments of hate speech: The effects of target group, publicness, and behavioral responses of the target. Journal of Applied Social Psychology, 26(4), 355–374. doi: 10.1111/j.1559-1816.1996.tb01854.x.CrossRefGoogle Scholar
  21. Cunningham, E., Forestell, C. A., & Dickter, C. L. (2013). Induced disgust affects implicit and explicit responses toward gay men and lesbians. European Journal of Social Psychology, 43(5), 362–369. doi: 10.1002/ejsp.1945.CrossRefGoogle Scholar
  22. Devlin, P. (1965). The enforcement of morals. Oxford University Press.Google Scholar
  23. Douglas, M. (2003). Purity and danger: An analysis of concepts of pollution and taboo. New York: Psychology Press.Google Scholar
  24. Eskridge, W. N. (2008). Dishonorable passions: Sodomy laws in America, 1861–2003. Penguin.Google Scholar
  25. Feldman, S., & Huddy, L. (2005). Racial resentment and white opposition to race-conscious programs: Principles or prejudice? American Journal of Political Science, 49(1), 168–183. doi: 10.1111/j.0092-5853.2005.00117.x.CrossRefGoogle Scholar
  26. Fisher, S. (2009). It takes (at least) two to tango: Fighting with words in the conflict over same-sex marriage. In S. Barclay, M. Bernstein, & A. M. Marshall (Eds.), Queer mobilizations: LGBT activists confront the law. New York: NYU Press.Google Scholar
  27. Flores, A. R. (2015). Examining variation in surveying attitudes on same-sex marriage: A meta-analysis. Public Opinion Quarterly, 79(2), 580–593. doi: 10.1093/poq/nfv017.CrossRefGoogle Scholar
  28. Flores, A. R., & Barclay, S. (2015). Backlash, consensus, legitimacy, or polarization the effect of same-sex marriage policy on mass attitudes. Political Research Quarterly. doi: 10.1177/1065912915621175.Google Scholar
  29. Gadarian, S. K., & Albertson, B. (2014). Anxiety, immigration, and the search for information. Political Psychology, 35(2), 133–164.CrossRefGoogle Scholar
  30. Garretson, J. J. (2015). Does change in minority and women’s representation on television matter?: A 30-year study of television portrayals and social tolerance. Politics, Groups, and Identities, 3(4), 615–632. doi: 10.1080/21565503.2015.1050405.CrossRefGoogle Scholar
  31. Geller, W. W. (1991). Attitudes toward gays and lesbians: A longitudinal study (p. 340970). ED: Eric Documents.Google Scholar
  32. GLAAD. (2016) Accelerating Acceptance 2016. GLAAD. Retrieved from
  33. Haidt, J., Rozin, P., Mccauley, C., & Imada, S. (1997). Body, psyche, and culture: The relationship between disgust and morality. Psychology & Developing Societies, 9(1), 107–131. doi: 10.1177/097133369700900105.CrossRefGoogle Scholar
  34. Harrison, B. F., & Michelson, M. R. (2012). Not that there’s anything wrong with that: The effect of personalized appeals on marriage equality campaigns. Political Behavior, 34(2), 325–344. doi: 10.1007/s11109-011-9169-6.CrossRefGoogle Scholar
  35. Harrison, B. F., & Michelson, M. R. (2017). Listen, we need to talk: How to change attitudes toward LGBT rights. New York: Oxford University Press.CrossRefGoogle Scholar
  36. Hatemi, P., & McDermott, R. (2012). Policing the perimeter: Disgust and purity in democratic debate. PS. Political Science & Politics, 45(04), 675–687. doi: 10.1017/S1049096512000686.CrossRefGoogle Scholar
  37. Herek, G. M. (1984). Attitudes toward lesbians and gay men: A factory-analytic study. Journal of Homosexuality, 10(1/2), 39–51. doi: 10.1300/J082v10n01_03.CrossRefGoogle Scholar
  38. Herek, G. M. (1990). The context of anti-gay violence: Notes on cultural and psychological heterosexism. Journal of Interpersonal Violence, 5(3), 316–333. doi: 10.1177/088626090005003006.CrossRefGoogle Scholar
  39. Herek, G. M., & Capitanio, J. P. (1996). “Some of My Best Friends”: Intergroup Contact, Concealable Stigma, and Heterosexuals’ Attitudes toward Gay Men and Lesbians. Personality and Social Psychology Bulletin, 22(4), 412–424. doi: 10.1177/0146167296224007.CrossRefGoogle Scholar
  40. Herek, G. M., & McLemore, K. A. (2013). Sexual prejudice. Annual Review of Psychology, 64, 309–333. doi: 10.1146/annurev-psych-113011-143826.CrossRefGoogle Scholar
  41. Herman, D. (1996). (Il)legitimate Minorities: The American Christian Right’s Anti-Gay-Rights Discourse. Journal of Law and Society, 23(3), 346–363. doi: 10.2307/1410716.CrossRefGoogle Scholar
  42. Hillygus, D. S., & Shields, T. G. (2009). The persuadable voter: Wedge issues in presidential campaigns. Princeton: Princeton University Press.CrossRefGoogle Scholar
  43. Hodson, G., & Costello, K. (2007). Interpersonal disgust, ideological orientations, and dehumanization as predictors of intergroup attitudes. Psychological Science, 18(8), 691–698.CrossRefGoogle Scholar
  44. Holbrook, M. B., & O’Shaughnessy, J. (1984). The role of emotion in advertising. Psychology and Marketing, 1(2), 45–64. doi: 10.1002/mar.4220010206.CrossRefGoogle Scholar
  45. Horberg, E. J., Oveis, C., Keltner, D., & Cohen, A. B. (2009). Disgust and the moralization of purity. Journal of Personality and Social Psychology, 97(6), 963–976. doi: 10.1037/a0017423.CrossRefGoogle Scholar
  46. Inbar, Y., & Pizarro, D. A. (2014). Pollution and Purity in moral and political judgment. In H. Sarkissian & J. C. Wright (Eds.), Advances in experimental psychology. New York: Bloomsbury Academic.Google Scholar
  47. Inbar, Y., Pizarro, D. A., & Bloom, P. (2009a). Conservatives are more easily disgusted than liberals. Cognition and Emotion, 23(4), 714–725.CrossRefGoogle Scholar
  48. Inbar, Y., Pizarro, D. A., Knobe, J., & Bloom, P. (2009b). Disgust sensitivity predicts intuitive disapproval of gays. Emotion, 9(3), 435–439. doi: 10.1037/a0015960.CrossRefGoogle Scholar
  49. Inbar, Y., Pizarro, D. A., & Bloom, P. (2012). Disgusting smells cause decreased liking of gay men. Emotion, 12(1), 23–27. doi: 10.1037/a0023984.CrossRefGoogle Scholar
  50. Kam, C. D., & Estes, B. A. (2016). Disgust sensitivity and public demand for protection. The Journal of Politics, 78(2), 481–496. doi: 10.1086/684611.CrossRefGoogle Scholar
  51. Klarman, M. J. (2004). From Jim Crow to civil rights: The Supreme Court and the struggle for racial equality. Oxford: Oxford University Press.Google Scholar
  52. Klarman, M. J. (2012). From the closet to the altar: Courts, backlash, and the struggle for same-sex marriage. Oxford: Oxford University Press.Google Scholar
  53. Lerner, J. S., & Keltner, D. (2001). Personality processes and individual differences: Fear, anger, and risk. Journal of Personality and Social Psychology, 81(1), 146. doi: 10.1037/0022-3514.81.1.146.CrossRefGoogle Scholar
  54. Mann, C. B. (2010). Is there backlash to social pressure? A large-scale field experiment on voter mobilization. Political Behavior, 32(3), 387–407. doi: 10.1007/s11109-010-9124-y.CrossRefGoogle Scholar
  55. Morone, J. A. (2004). Hellfire nation: The politics of sin in American history. New Haven: Yale University Press.Google Scholar
  56. Nussbaum, M. C. (2010). From disgust to humanity: Sexual orientation and constitutional law. Oxford: Oxford University Press.Google Scholar
  57. Olatunji, B. O., & Sawchuk, C. N. (2005). Disgust: Characteristic features, social manifestations, and clinical implications. Journal of Clinical and Social Psychology, 24(7), 932–962. doi: 10.1521/jscp.2005.24.7.932.CrossRefGoogle Scholar
  58. Olatunji, B. O., Williams, N. L., Tolin, D. F., Abramowitz, J. S., Sawchuk, C. N., Lohr, J. M., et al. (2007). The Disgust Scale: Item analysis, factor structure, and suggestions for refinement. Psychological Assessment, 19(3), 281–297. doi: 10.1037/1040-3590.19.3.281.CrossRefGoogle Scholar
  59. Rozin, P., & Haidt, J. (2013). The domains of disgust and their origins: contrasting biological and cultural evolutionary accounts. Trends in Cognitive Sciences, 17(8), 367–368. doi: 10.1016/j.tics.2013.06.001.CrossRefGoogle Scholar
  60. Schilt, K., & Westbrook, L. (2015). Bathroom battlegrounds and penis panics. Contexts, 14(3), 26–31. doi: 10.1177/1536504215596943.CrossRefGoogle Scholar
  61. Schnall, S., Haidt, J., Clore, G. L., & Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34(8), 1096–1109. doi: 10.1177/0146167208317771.CrossRefGoogle Scholar
  62. Schneider, A., & Ingram, H. (1993). Social construction of target populations: Implications for politics and policy. American Political Science Review, 87(02), 334–347. doi: 10.2307/2939044.CrossRefGoogle Scholar
  63. Smith, R. A. (2007). Language of the lost: An explication of stigma communication. Communication Theory, 17(4), 462–485. doi: 10.1111/j.1468-2885.2007.00307.x.CrossRefGoogle Scholar
  64. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 48(4), 813–838. doi: 10.1037/0022-3514.48.4.813.CrossRefGoogle Scholar
  65. Smith, S. J., Axelton, A. M., & Saucier, D. A. (2009). The effect of contact on sexual prejudice: A meta-analysis. Sex Roles, 61(3), 178–191. doi: 10.1007/s11199-009-9627-3.CrossRefGoogle Scholar
  66. Smith, K. B., Oxley, D., Hibbing, M. V., Alford, J. R., & Hibbing, J. R. (2011). Disgust sensitivity and the neurophysiology of left-right political orientations. PLoS ONE, 6(10), e25552. doi: 10.1371/journal.pone.0025552.CrossRefGoogle Scholar
  67. Solomon, M. (2014). Winning marriage: The inside story of how same-sex couples took on the politicians and pundits-and won. New York: ForeEdge.Google Scholar
  68. Sunstein, C. R., & Kahneman, D. (2007). Indignation Psychology, Politics, Law. Working paper. Coase-Sandor Institute for Law and Economics.
  69. Tapias, M. P., Glaser, J., Keltner, D., Vasquez, K., & Wickens, T. (2007). Emotion and prejudice: Specific emotions toward outgroups. Group Processes & Intergroup Relations, 10(1), 27–39. doi: 10.1177/1368430207071338.CrossRefGoogle Scholar
  70. Terrizzi, J. A., Jr., Shook, N. J., & Ventis, W. L. (2010). Disgust: A predictor of social conservatism and prejudicial attitudes toward homosexuals. Personality and Individual Differences, 49(6), 587–592. doi: 10.1016/j.paid.2010.05.024.CrossRefGoogle Scholar
  71. Uhlmann, E. L., Zhu, L., & Diermeier, D. (2014). When actions speak volumes: The role of inferences about moral character in outrage over racial bigotry. European Journal of Social Psychology, 44(1), 23–29. doi: 10.1002/ejsp.1987.CrossRefGoogle Scholar
  72. Vakratsas, D., & Ambler, T. (1999). How advertising works: What do we really know? Journal of Marketing, 63(1), 26–43. doi: 10.2307/1251999.CrossRefGoogle Scholar
  73. Valentino, N. A., Hutchings, V. L., Banks, A. J., & Davis, A. K. (2008). Is a worried citizen a good citizen? Emotions, political information seeking, and learning via the internet. Political Psychology, 29(2), 247–273. doi: 10.1111/j.1467-9221.2008.00625.x.CrossRefGoogle Scholar
  74. Valentino, N. A., Banks, A., Hutchings, V., & Davis, A. K. (2009). Selective exposure in the internet age: The interaction between anxiety and information utility. Political Psychology, 30(4), 591–613. doi: 10.1111/j.1467-9221.2009.00716.x.CrossRefGoogle Scholar
  75. Wong, C. M. (2015). Pat Robertson: Gay Marriage Is “Unnatural” Because No One Has Conceived A Child Through Anal Sex. Huffington Post. Retrieved from Scholar

Copyright information

© Springer Science+Business Media, LLC 2017

Authors and Affiliations

  1. 1.Department of Political Science, Maxwell SchoolSyracuse UniversitySyracuseUSA

Personalised recommendations