CAN WE PRODUCTIVELY STUDY THE ROLE OF SPECIFIC EMOTIONS IN VIOLENT MOBILIZATION?

In arguing for the importance of emotions, it is helpful to first recognize and address several points made for downplaying the overall significance of emotions for political phenomena.

First, emotions are short-term. Unlike moods, emotions follow trajectories. Despite being a powerful influence at the time of onset, the intensity of an emotion may quickly decline. In laboratory settings, psychologists often measure the duration of emotions in terms of seconds and minutes. Michael Potegal, in the Handbook of Anger (2010), estimates that “a half hour is the modal duration of everyday anger.”

Second, emotions are an individual-level phenomenon while politics, especially in terms of issues like mobilization, is a matter of collectives. Even if many individuals are imbued with the same emotion, there is still the question of how that motivational force is aggregated to shape group-based outcomes.

Third, emotions are so diffuse and complex as to be analytically intractable. For emotion to be useful in explaining variation in an outcome, it is necessary to make predictions based on specific emotions and their specific action tendencies rather than to talk about emotion in general. But can we ever say that a population is primarily driven by just one emotion? Aren’t emotions so ubiquitous and diffuse that several are in play at any one given point?

Fourth, emotions are usually characterized in terms of changes in physical states (and, more recently, in changing neural states) rather than changes in beliefs and longer-term attitudes that are most relevant to politics and mobilization.

In sum, how can individual-level, short-term, physical and neural phenomena affect collective, longer-term, and attitudinal or mental processes and outcomes?

If emotions are the residue of experience, then we should start by considering types of experiences that generate emotions. In response to the second and third points above, can we identify types of experiences that create specific identifiable emotions that are widely shared across members of a collective? This question is an empirical one, and a brief consideration of times of violence brings up several positive examples related to violence and mobilization. Consider fear. During times of ethnic cleansing, collectives are imbued with fear and engage in the predicted behavior of flight. On the subject of anger, the al-Qaeda leader in Iraq, al-Zarqawi, clearly and correctly believed that the bombing of the golden dome in Samarra would imbue large numbers of the Shia community with a punishment-driven anger that would set off a desired spiraling violence between the sects. On humiliation, checkpoint systems in occupied areas create constant and common experiences of subordination that generate the emotion across whole identity categories. While emotions are an individual-level phenomenon, emotions help generate collective action in situations in which members of a group sense that almost all other members of the group are likely experiencing the same emotion. This sense of shared emotion solidifies the group and helps coordinate action.

Do these commonly held experiences really generate the same emotion? The test for the social science value of any concept is whether it can be used to generate hypotheses and specify the mechanisms behind those hypotheses. For both theoretical and policy reasons, it is useful to know whether people are violently mobilizing out of fear, anger, or some other emotion. I argue that we do possess the knowledge and ability to both operationalize specific emotions as variables and to study the sub-mechanisms relating to information collection and belief formation that define those emotions. Our ability to distinguish a given emotion is enhanced by our increasing knowledge about the specific components of that emotion. These components include not only cognitive antecedents and action tendency but also predicted effects on information collection and belief formation. Social psychologists employing an appraisal tendency framework have perhaps gone the farthest in establishing that specific emotions have identifiable cognitive and motivational sub-processes (Lerner et al. Reference Lerner, Li, Valdesolo and Kassam2015). By observing the cognitive antecedents, the action tendency, and these predicted effects on information collection and belief formation, we can effectively separate out which specific emotion is operating at a particular juncture.

For example, we can separate anger from fear—two emotions central to questions of violent mobilization. The cognitive antecedent of anger is that an individual or group has committed a bad action against one’s self or group. The action tendency is to heighten desire for punishment against a specific actor. Under the influence of anger, individuals become “intuitive prosecutors” (Goldberg, Lerner, and Tetlock Reference Goldberg, Lerner and Tetlock1999; Petersen and Zukerman Reference Petersen, Zukerman, Potegal, Stemmler and Spielberger2009). That is, individuals tend to specify a perpetrator and then seek retribution. On information collection, anger distorts information in predictable ways, producing attention funneling. As with other emotions, once under the influence of anger, individuals “perceive new events and objects in ways that are consistent with the original cognitive-appraisal dimensions of the emotion (Lerner and Keltner Reference Lerner and Keltner2000).” Following this point, individuals imbued with anger privilege information that helps specify a perpetrator as well as ways to punish that perpetrator. On belief formation, anger produces several predictable effects critical to violent mobilization. The angry person lowers the threshold for attributing harmful intent. Anger enhances the fundamental attribution error—angry people blame humans, not the situation (Keltner, Ellsworth, and Edwards Reference Keltner, Ellsworth and Edwards1993). Anger also tends to produce more stereotyping (Bodenhausen, Sheppard, and Kramer Reference Bodenhausen, Sheppard and Kramer1994). Perhaps most importantly for the study of violent politics, under the influence of anger, individuals lower their risk estimates and become more willing to engage in risky behavior (Gallagher and Clore Reference Gallagher and Clore1985; Lerner and Keltner Reference Lerner and Keltner2001; Lerner et al. Reference Lerner, Gonzalez, Small and Fischoff2003; Mano Reference Mano1994).

Anger can be distinguished from fear on each of these dimensions. Fear stems from cognition of a situation of danger with an action tendency towards defensive fight or flight. Under fear, individuals value self-preservation above all else. As with other emotions, once under the influence of fear, individuals will process information and form beliefs to confirm the cognitive-appraisal aspects of the emotion; that is, they will privilege information about danger (Gardner Reference Gardner2008; Lerner and Keltner Reference Lerner and Keltner2000). Crucially, both anger and fear affect beliefs about risk. However, anger and fear produce opposite effects on risk perception, with individuals under the sway of fear becoming more risk-averse. There are other established findings on belief formation; for instance, unlike anger, fearful subjects have been found to have low assessments of blame (Gallagher and Clore Reference Gallagher and Clore1985). Finally, we have ideas about the conditions that are most likely to trigger anger rather than fear. In experimental research, participants with a strong sense of in-group solidarity are more likely to feel anger against out-groups (Mackie, DeVos, and Smith Reference Mackie, DeVos and Smith2004). Likewise, fear levels can be manipulated by categorizing the victims as in-group or out-group (Dumont et al. Reference Dumont, Vincent, Yzerbyt, Daniel, Wigboldus and Gordijn2003). Emotions are complex phenomena. When the various aspects of a specific emotion line up and can be observed, we can assume that specific emotion is at work.

This discussion also helps us address the fourth point above concerning the relationship among emotions, beliefs, and longer-term attitudes. Previous conventional wisdom saw emotion and reason—thinking and feeling—as separate. Reason and thinking were seen as “correcting” emotion. Led by the work of Damasio and others, this conventional wisdom no longer holds. Belief formation is not a matter of reason, but of both reason and emotion. Accordingly, research on the impact of emotions on decision making has exploded in recent years. As Jennifer Lerner and colleagues (2015) state in the abstract of their summary piece in a recent volume of the Annual Review of Psychology: “A revolution in the science of emotion has emerged in recent decades, with the potential to create a paradigm shift in decision theories. The research reveals that emotions constitute potent, pervasive, predictable, sometimes harmful and sometimes beneficial drivers of decision making. Across different domains, important regularities appear in the mechanisms through which emotions influence judgments and choices.” The key point here is that emotions can help create regularities in judgments.

Coming back to the idea of emotions as the residue of lived experience helps clarify how emotions can change long-term beliefs and attitudes. Emotions can change cognitions and beliefs because the cycle of beliefs, emotions, preference changes, and information collection may keep turning over and over, especially in situations regarding identity and conflict. In many situations, members of Group X form beliefs about their unfair treatment at the hands of members of Group Y. For members of X, these cognitive antecedents create emotions such as humiliation or resentment. These emotions heighten desires to end Y’s perceived dominance. In turn, these emotions skew X’s information collection; members privilege sources and images that prove and illustrate Y’s dominance. The emotion then affects belief formation by shaping ideas about the out-group’s nature and motivations, the ability to change the situation without violence, and the amount of risks involved with confrontation. This cycle can turn again and again. Emotions are a powerful and constant force within this cycle for hardening long-term attitudes.

These points also bring us back to the first point above—the short-term durations of emotions. As mentioned above, laboratory experiments measure emotion in minutes. But real life is not a laboratory experiment. Although the actual intensity of emotion fades after onset, in many situations of everyday life individuals are constantly activating the same emotion. When the same experience triggers the same emotion in everyday interactions across an entire population over an extended period of time, then that emotion can have long-term collective effects.

THE FUNDAMENTAL ROLE OF EMOTIONS IN VIOLENT MOBILIZATION AND THE RELATIONSHIP OF EMOTIONS WITH IDEOLOGY

While some scholars see the role of emotion as important, they also see that role as limited. In their International Security article, for example, Stefano Costalli and Andrea Ruggeri (2015) posit a specific ordering of the effects of emotion and ideology. In their treatment, a shock produces an emotion which is a “push-factor” moving individuals off acceptance of the status quo. Once this emotion is in place, ideology comes along as a “pull-factor” which provides new strategies and new beliefs about possibilities for action. In that study, emotion produces a reshaping of the preference set, while ideology produces new beliefs about strategies.

Emotions, however, do far more than simply push people off the status quo. In fact, emotions do essential work in creating and reinforcing the identities that allow an ideology to take hold. Consider two major ideologies—nationalism and Islamic jihadism. In his account of the rise of nationalism, Ernest Gellner provides the illustrative story of Ruritania. In this story, multiethnic empires needed to become efficient states in order to compete military and economically in a changing world. To gain this efficiency, these reforming political entities needed to mobilize the population through education and military conscription. In turn, education and conscripted armies required a common language. In multilingual empires, the elevation of one language involved the relative lowering of the status of minority and regional languages. When peasants from a rural region entered the university or were conscripted into the military, they would now need to do all of their everyday business in the language of the center. No aspect of life is perhaps as pervasive as the language in which one must speak and think. In turn, this experience of becoming conscious of one’s own group as subordinate can breed an emotion of resentment. I have argued (Petersen Reference Petersen2002) that this commonly shared emotion of resentment is crucial in understanding why the rise of nationalist ideology was so pervasive across all of Eastern Europe in the twentieth century. In turn, this ideology, under some circumstances, helped lead to violent mobilization.

One does not have to agree with all of Gellner’s treatment of nationalism to understand that the processes of state-building he describes can produce common experiences which set off widespread, specific, shared emotions. Apartheid and caste systems, military occupations (especially ones with checkpoints), ethnically segmented labor markets, colonialism, and other situations can also create group-based hierarchies and produce similar everyday interactions. In the Jim Crow south of the US, the civil rights movement was as much about pushing back against the daily humiliation of African Americans and the resulting emotions of resentment and humiliation as it was about an ideology of equality. In these cases, there was no need for a shock and its emotional power to detach individuals from a status quo; the status quo was already unacceptable due to the lived experiences and emotions of daily life. The powerful emotional quality of daily life created the foundations for the ideologies that followed.

To take another case, for the past decades Western powers have been in a dominant position in relation to the Islamic World. The US and its allies made short work of the Iraqi Army in 1991 and in the 2003 invasion. Coalition troops occupied Iraq for several years and currently hold what seems to be a permanent presence in Afghanistan. Israelis control the West Bank and Jerusalem and maintain a checkpoint system in the occupied territories. Drones fly over the region and attack with impunity. In this situation, the constant images of Western dominance were arguably likely to mobilize some anti-Western violent response. Once again long-term experience, rather than shocks, produces a widespread emotion of humiliation. Dominique Moisi (Reference Moisi2010) goes so far as to claim that Arab Muslim culture has evolved into one dominated by the emotion of humiliation. Humiliation is a powerful emotion that is capable of generating mechanisms affecting information collection and belief formation. In turn, these forces help ideology form and take hold in the first place.

As argued above, in situations centered on identity and conflict, a cycle of lived experience, belief formation about that experience, triggered emotions, preference changes, and skewed information collection can turn over and over. The argument above places lived experience and emotions as the central forces driving this cycle. Admittedly, there are some cases in which ideology no doubt plays an important role in consolidating belief formation along the lines argued by Costalli and Ruggeri (Reference Costalli and Ruggeri2015). In some circumstances, ideology can play a coordinating role shaping how individuals imagine and articulate their struggles and also providing the belief that many other individuals share similar experiences.

Furthermore, in some cases ideology can provide a narrative of conflict that can shape which specific emotion is triggered in the above cycle. Consider this point in assessing the effects of a “shock.” To illustrate, recall the cognitive antecedents of anger. Anger results from a belief that an identifiable perpetrator has committed a blameworthy action against oneself or one’s group. Without an identifiable perpetrator, a shock or attack is more likely to create fear than anger. Ideologies often provide a narrative about who the perpetrator is. In these cases, the ideology has ready-made enemies to blame. With a set of cognitions already embedded within the words and imagery of an ideology, the resulting emotion from a terrorist shock, for example, is more likely to be anger than fear. Mobilization will likely be geared toward punishment more than protection. For groups lacking a coherent ideology, and thus without a clear perpetrator to automatically punish, anger is less likely to result and the intensity and the purpose of mobilization may be very different.

The relationship between emotions, ideology, and leadership is an empirical one. In some cases, the nature of lived experience is so similar, ubiquitous, and powerful among a certain collective that the resulting emotional force will naturally and predictably produce leaders and ideology to build on that force. In other cases, ideology and leadership may play dominant roles in shaping mobilization outcomes. In almost all cases, however, grievances alone do not automatically lead to predictable actions and outcomes, especially for the issue of violent mobilization.