The Emotional Underpinnings of Attitudes toward Transitional Justice

Abstract

What explains citizens’ attitudes toward transitional justice? Studies that examined the support for transitional justice mechanisms identified three sets of factors: individual, socialization, and contextual. Building on the hot cognition theory, this article argues that the past political regime is an emotionally charged sociopolitical object encoded with its evaluative history with consequences in people’s opinion-formation process. Drawing on a specialized survey in Spain, the results first suggest that negative emotions, especially anger and fear, significantly influence the support for stronger transitional justice measures, even after adjusting for relevant confounders such as ideology, religiosity, or victimization. Second, the findings show that those who lack an emotional engagement toward the past regime, the so-called bystanders, hold attitudes toward transitional justice that are indistinguishable from those who report positive feelings (pride, patriotism, and nostalgia) toward the past regime. The effects of emotions are sizable relative to other important determinants, including ideology, religiosity, and family’s ideology.

Keywords

transitional justice democratization political attitudes emotions Spain

Transitional justice (TJ) is at the heart of the peace-building process of democratizing countries.¹ In many instances, negotiated transitions lead to the forgive-and-forget amnesty policy to criminals of the past regime. In this post-transitional scenario, the legitimacy of the newly created institutions might be vulnerable to the justice deficit that derives from an unsatisfied societal demand for justice. Much research on TJ has attempted to understand the effects of historical justice mechanisms on macro- and micro-political factors. Scholars have found that some of these mechanisms can reduce conflict, as well as consolidate democratic institutions (Sikkink and Walling, 2007), people’s perceived justice, fairness and acceptance (Gibson, 2002, 2004; Gibson and Gouws, 1999), societal reconciliation (Gibson, 2006), or psychological healing (Sonis et al., 2009). By contrast, skeptics have pointed out that digging into the past may, under some circumstance, reinvigorate the past conflict and undermine the supposed beneficial effects of TJ (Backer, 2010; Meernik, 2005; Olsen et al., 2010; Thoms et al., 2010).

While research on the consequences of historical justice abounds, fewer scholars have paid attention to the determinants of the demand for TJ mechanisms. International actors ranging from foreign countries to the UN have played active roles in determining the justice mechanisms to implement in transitional countries. Additionally, democratic theory suggests that as a country transitions from an authoritarian to a democratic regime, institutions should become more responsive to citizens’ preferences. The passage of time consolidates democratic institutions and reduces concerns about reprisals, as well as worries of return to the conflict. Not only does time attenuate people’s demands for justice it can also exacerbate them. As Skaar (1999) argues, government’s choice for transitional policy is largely a product of the relative strength of public’s demand for truth and justice. In sum, understanding the dynamics of TJ mechanisms around the world requires dissecting how citizens are willing to punish or forgive the past regime.

Mass surveys on attitudes toward TJ have shed some light on attitudes toward peace, justice, and reconciliation in a number of countries, including Hungary, Poland, and the Czech Republic (Nalepa, 2010); South Africa (Gibson, 2002, 2004); RD Congo (Vinck et al., 2008); Cambodia (Pham et al., 2004, 2009; Sonis et al., 2009); Liberia (Vinck et al., 2011), Northern Uganda (Pham and Vinck, 2010); Burundi (Samii, 2013); and Spain (Aguilar et al., 2011; Balcells, 2012). However, most of these surveys either have been reported in a purely descriptive manner (e.g. Pham et al., 2009; Pham and Vinck, 2010; Vinck et al., 2011) or have focused on the effect of TJ on people’s preferences (e.g. Gibson, 2002, 2004).² Those few studies that focused on attitudes identified three types of determinants of preferences for TJ: individual factors (e.g. educational level, ethnicity), socialization (victimization and early politicization), and contextual factors (e.g. conflict versus non-conflict areas). Even though the role of emotional engagement is assumed to underlie some of their findings, it has not been empirically explored. Consequently, because the literature does not formally and empirically incorporate the emotional engagement with the past, it remains incomplete.³

This article begins to fill this gap by bringing together an emerging literature on emotions in politics to push the boundaries of what we know about individual attitudes toward TJ policies. I argue that affective dimensions, constituted by positive versus negative emotions, strongly predict preferences toward TJ policies even after adjusting for relevant confounders. Consistent with the view of the role of passive agents or bystanders in human rights violations and war crimes (Barnett, 1999; Staub, 1993; Verdeja, 2012), I also contend that those who lack an emotional engagement with an authoritarian past regime implicitly conform with the status quo, which leads to a lower demand for further TJ mechanisms.

To test these relationships, I draw from a nationally representative survey with 2936 respondents conducted in Spain in 2008. This survey allows me to measure six emotions: three negative (anger, fear, and sadness) and three positive (pride, patriotism, and nostalgia), and an item tapping into individuals’ lack of emotional engagement. In addition, I use controls for individual-level determinants (basic demographics, education, religiosity, ideology, political knowledge, and engagement in politics), socialization factors (personal and familial victimization, family’s ideology, and early political socialization), and contextual factors (region).

The results strongly support the thesis that the past regime is an emotionally charged political object with consequences on people’s attitudes toward TJ. In particular, the results reveal that the six discrete emotions in the models—anger, fear, sadness, nostalgia, patriotism, and pride—significantly influence the willingness for TJ in the expected direction. First, anger, fear, and sadness significantly increase desire for more justice procedures toward the past authorities. Among these three negative emotions, anger and fear exert a particularly strong effect as compared to non-emotional determinants.

And, second, feelings of pride, patriotism, and nostalgia positively influence the desire for TJ, but their effect is not distinguishable from the effect among those who lack an emotional engagement with the past. Even, more importantly, the findings also show that those who report a lack of engagement with an authoritarian past regime are implicitly on the same side of their supporters; that is, they lower the demand for TJ mechanisms. This is so much the case that the effect of reporting no emotions is as strong as the effect of reporting a feeling of pride, patriotism, or nostalgia toward the past regime.

The remainder of the article proceeds as follows. In the next section, I provide an overview of some relevant prior literature on the demand for TJ mechanisms. Then, I describe emotions’ role in the formation of attitudes toward political objects and describe the affective approach of the determinants of the demand for historical justice. Next, I contextualize the reader with a brief description of Spain’s TJ. Following this, I present the survey data, the measurements of the variables, and some descriptive statistics. I then discuss the results of the analysis. The final section identifies some extensions to this work, its limitations, and broader implications of the findings.

Relevant Literature

TJ scholars have identified three types of determinants of preferences for TJ: individual factors, socialization, and contextual factors (see Aguilar et al. (2011) for an extended review of the theoretically relevant determinants of TJ).

Individual factors refer to those that vary depending on relatively exogenous personal conditions such as gender and age (Aguilar et al., 2011; Nalepa, 2010), education (Pham et al., 2004), religiosity—especially with regard to religious-based regimes (Montero et al., 2008)—and an ideological connection with the past regime (Aguilar et al., 2011).

Socialization factors are, perhaps, the most well-studied set of determinant of TJ mechanisms. They refer to factors that are exogenous to the individual, including personal experiences of victimization with the regime. Not only victims of the regime are thought to be those most benefited from TJ mechanisms but also those who have a greater willingness for revenge. Consequently, scholars have generally found that victims are more likely to support TJ mechanisms in general and, especially, of a more punitive kind (Aguilar et al., 2011; Balcells, 2012; Pham et al., 2004; Samii, 2013). In addition, scholars have also found other relevant socialization factors that influence political attitudes toward the past regime such as family’s ideology or degree of within-family politicization (Aguilar et al., 2011; Balcells, 2012).

Finally, contextual factors such as the existing political regime or whether citizens belong to previously repressed ethnic identities seem to play a fundamental role in predicting one’s attitudes toward TJ mechanisms (Balcells, 2012).

Even though emotions have been found to predict a great number of political attitudes and behaviors, such as intolerance (Gibson et al., 2017; Halperin et al., 2009), anti-immigration attitudes (Brader et al., 2008), violent behavior (Claassen, 2016), or political participation (Groenendyk and Banks, 2014; Valentino et al., 2009), the exploration of their relationship with the demand for TJ has remained unexplored. The next section presents a theoretical framework that motivates why we should expect emotional engagement to fundamentally shape people’s demand for TJ mechanisms.

Support For TJ: An Affective Approach

Debates on political behavior and public opinion formation, including preferences for TJ, conventionally revolve around the process of conscious deliberation in which citizens retrieve values, attitudes, and beliefs from their long-term memory. In this traditional view, surveys are aimed at pulling out fixed opinions from the respondents’ minds. Early scholars were largely skeptical about the ability of the masses to process information (Campbell et al., 1960; Converse, 2006; Delli Carpini and Keeter, 1996). In an attempt to close the gap between classic democratic theory and rational voters, some scholars looked at voters’ use of heuristics and information shortcuts as a tool to guide decision-making process (Lupia, 1994; Popkin, 1991). Heuristics, however, can lead citizens to citizens’ preferred decision as often as it leads them astray by false cues (Kuklinski and Quirk, 2000; Kuklinski et al., 2000; Lau and Redlawsk, 2001).

The role of emotions in the formation of public opinion aligns with the literature of heuristics. In this line of research, emotions can be conceptualized as triggers that may facilitate an efficient use of heuristics. The most outstanding contribution in the understanding on how emotions are linked to political behavior is the Affective Intelligence (AI) theory (Marcus et al., 2000). According to the AI, the emotions that people experience derive from a dual system. On the one hand, enthusiasm (as opposed to apathy) works through the disposition system, which is responsible for governing individuals’ habits and routines. On the other hand, anxiety responding to threat (as opposed to calm) works through the surveillance system and is responsible for redirecting individuals’ attention to concentrate in the new environmental situation that is the source of the anxiety. As a result of this theoretical framework, Marcus et al. (2000) predict that enthusiasm would cause people to rely on prior beliefs and habitual cues for voting and involvement in political campaigns (Marcus and MacKuen, 1993). In contrast, anxiety would motivate individuals to seek out new information, rely less on heuristics, such as ideology, and be more attentive to candidate’s policy stances in the political campaign (Brader, 2006; Marcus et al., 2000; Marcus and MacKuen, 1993).

Even though the AI is likely to be the best attempt to apply a theory of emotions in politics, some scholars have failed to replicate its main predictions. In a first attempt, Nadeau et al. (1995) applied the AI theory to the case of Quebec. They failed to find an effect of anxiety on learning through perceived importance of the issue at hand. In a more recent study, Ladd and Lenz (2008, 2011) attempted to replicate the predictions from the AI theory on candidate evaluations against simpler hypotheses of direct influence. While Ladd and Lenz find no evidence of the AI theory, their findings lend credence to a simpler valence model of emotions.

Emotional theories have contributed to our understanding of not only how people process information and efficiently use heuristics but also how people access information to form their opinions. The most basic conceptualization suggests that all emotional experiences can be classified based on their valence. Feelings are understood as a function of a single underlying positive to negative affect dimensions. In political science, the most prominent theory grounded on an emotional valence approach in the opinion formation process is the “hot cognition hypothesis” (Lodge and Taber, 2005; Morris et al., 2003).

Figure 1 presents a basic diagram of emotions’ role within the framework of the hot cognition hypothesis. A bedrock assumption of this theory is that all sociopolitical concepts that an individual has evaluated in the past are affectively charged sociopolitical objects with effects on beliefs, attitudes, and opinions today. Thus, political objects are stored in the long-term memory with affective tags. Then, when a person is called on to evaluate the stimulus, such as a question in a survey, the affective charge previously developed is unconsciously activated from the long-term memory and integrated with the new stimulus, leading to the report of a positive or negative discrete emotion. Thus, the cognitive process is colored by its unconsciously activated affective tag. In fact, the hot cognition hypothesis might also be seen as a theory of unconsciously emotion-driven motivated reasoning (Civettini and Redlawsk, 2009; Redlawsk, 2002; Redlawsk et al., 2010; Taber and Lodge, 2006). In line with the bipolar approach to emotions, some scholars have argued that most of the variation in emotional responses can be captured in a uni-dimensional scale of positive versus negative affect (Green et al., 1993).

Figure 1.

Emotions in the Hot Cognition Hypothesis.

Consistent with Lodge and Taber’s (2005) and Green et al.’s (1993, 1999) valence model, an important number of empirical studies have found that discrete negative emotions load onto a common factor and have similar effects on political attitudes and behavior (e.g. Marcus and MacKuen, 1993). Yet, the multi-dimensional scaling usually leads to two factors: negative versus positive valence and emotional engagement versus lack of emotional engagement.

Because of the lack of prior literature on the role of emotions in influencing the demand for TJ mechanisms, I directly draw empirical implication based on the hot cognition theory more generally. Just as with other political attitudes and behavior, I argue that valence emotional responses underlie attitudes toward TJ policies, which give rise to the following empirical expectations:

Hypothesis 1 (negative valence effect): Individuals for whom the past regime evokes negative emotions are more likely to support for further TJ mechanisms.

Hypothesis 2 (common negative affect): All discrete negative emotions are equally associated with the support for further TJ mechanisms.

Hypothesis 3 (positive valence effect): Individuals who report feeling positive emotions toward the past regime will be less likely to hold attitudes favoring stronger TJ mechanisms.

Hypothesis 4 (common positive affect): All discrete positive emotions are equally associated with the support for further TJ mechanisms.

From a theoretical point of view, the affective approach considers an individuals’ strength of emotional engagement—as opposed to the direction of the emotional valence alone—an important determinant for demanding TJ mechanisms. Scholars of war crimes, human rights violations, and atrocities committed by authoritarian regimes have also emphasized the fundamental role of those who become indifferent, those who lack an emotional engagement with the past. Also referred to as bystanders, these individuals are ostensibly passive agents who remain unwilling to intercede in matters related to justice, violence, and even mass killings. Apathetic citizens are those who actually may help the persistence of the status quo, including prosecutions and killings as reported by Barnett (1999), Staub (1993), Verdeja (2012), and others. In the extreme case, indifferent individuals would be equivalent to those with positive feelers in their request for a lack of action on the matter given that the status quo conforms with the regime’s supporters. This perspective rises to the following two empirical expectations:

Hypothesis 5 (bystanders’ hypothesis): Those respondents who report indifference toward the past regime would be less likely to demand more severe TJ mechanisms.

Hypothesis 6 (extreme bystanders’ hypothesis): Those respondents who report indifference toward the past regime would as likely to demand a set of forget-and-forgiveness set of policies as citizens who report positive feelings toward the past regime.

To empirically investigate these hypotheses, I use the case of contemporaneous Spain’s TJ. Interestingly, Spain is characterized by being a least-likely case of emotional engagement. The four decades of consolidated democratic institutions and the long time that has passed after the crimes were committed may have moderated individuals’ affective attachment to the past regime. Just like for many new democracies, Spain faced a dilemma between ethics and political constraints. The constraints led to the dominance of the forget-and-forgive policies, which explains why there might still today exist some degree of controversy around the mechanisms that Spain should pursue to reestablish justice for the victims (Grodsky, 2010; Zalaquett, 1992). This existing controversy allows that micro-level research on TJ mechanisms in Spain be contextually meaningful.

Case Context: Spain’s TJ

The military revolt supported by conservative and traditionalist ideological factions against the republican government of Spain led to the beginning of the Spanish Civil War (1936–1939)—the Nationalists, on the one side, and the Republicans, on the other side. The eventual victory of the Nationalists in 1939 led to the establishment of the Francoist regime (1939–1975). The triumphant Nationalists continued to use violence to suppress political dissidents, especially among the early years of the postwar period. Throughout the period of Francisco Franco’s rule, the regime executed tens of thousands of people and imprisoned some hundreds of thousands in concentration camps.⁴ Apart from executions, detentions and imprisonments were common repression practices.

The dictator’s death in 1975 triggered a number of rapid political developments that led to Spain’s transformation into a democracy. In the democratic period, TJ had been largely focused on distributive policies with scattered monetary compensations for Republican participants in the Civil War and citizens imprisoned during Franco’s era. A turning point in the process was the establishment of the Association for the Recovery of Historical Memory (ARMH) (Asociación Para la Recuperación de la Memoria Histórica) in 2000 with the aim to locate and exhumate mass graves as well as to build public support to investigate abuses that occurred during the Franco era. The social and political debates triggered by the ARMH put the historical justice on the political agenda for several years.

In this political climate, political parties positioned themselves in the debate. The Partido Socialista Obrero Español (PSOE), Spain’s major center-left party, and Izquierda Unida (IU), the former communist party, stood in favor of legislating to recover the dignity of the victims and broadened their rights. In contrast to this, the Partido Popular (PP), the major conservative party, positioned against “digging into the past.” The electoral victory of the PSOE in 2004 allowed Zapatero’s government to initiate the legislative works for the elaboration of a law for the historical memory. Spain officially condemned the Francoist regime, ordered the withdrawal of Francoist symbols from public spaces, declared judicial trials for political and ideological reasons during the Civil War, as well as the dictatorship, as illegitimate and legally invalid, and provisioned economic compensations to the victims of the Francoist regime and their descendants in 2007.⁵ The incumbent electoral defeat in 2011 meant the return of political power to the major conservative party. Despite the PP’s opposition throughout the legislative process before passing the law and their electoral promises of revoking the law, the Historical Memory Law has not been officially revoked, although it has been left inapplicable with no budget endowment since 2013.

In short, political constraints during Spain’s transition to democracy led to weak and limited distributive and restorative policies, and the absence of retributive policies. As suggested by Grodsky (2010), the lack of goods and services with regard to TJ to the constituents may have spurred a nascent demand for further TJ among some segments of the population. The long time span between today and the time when the crimes were committed, as well as the nascent demand of TJ, makes the Spanish case an ideal setting for this study.

Research Design

This section describes information about the Spanish data used in the analysis and the measurement of the dependent, independent, and control variables.

Data

The Spanish Parliament promulgated the Historical Memory Law in December 2007 with the support of the left-wing parties. The controversial law divided the left and the right and immersed public opinion into a dispute about its necessity after 30 years of a peaceful coexistence between supporters of the two political factions. In April 2008, only few months after the promulgation of the law, the Centro de Investigaciones Sociológicas (CIS) conducted a representative sample of the adult population with 2936 face-to-face interviews. The survey included a number of questions on the victimization of the respondent during the Francoist regime (only for respondents older than 45 years) and/or family members or individuals close to the respondent (for the entire sample), feelings and attitudes toward the Francoist period (1939–1975), and preferences toward different forms of TJ.

Dependent Variable: Desire for TJ

The components of TJ can be conceptualized in a “TJ spectrum” by mapping each justice mechanism into a one-dimensional continuum that captures the strength in which victims were restored and criminal authorities were prosecuted or reparation/punishment versus forgiveness (Grodsky, 2009). A country that starts with a clean slate by providing full amnesty for past authorities does not initiate any truth-telling mechanism and does not provide any monetary nor symbolic reparation for victims would lie at one extreme of the continuum. By contrast, a country that engages in investigations for truth-telling, prosecutes past authorities involved in some sort of criminal activities, and restores economically and symbolically their victims would lie at the other extreme.⁶

Based on this conceptual framework, I create a latent construct that locates individuals according to their willingness to seek reparations for the victims and their being tough against past criminals. This underlying desire for TJ is my dependent variable. To measure it, the CIS questionnaire asks respondents to what extent they agree about different sorts of justice mechanisms. These questions include justice toward the Civil War and the Francoist regime. The survey was distributed 69 years after the end of the Civil War and 33 years after the end of the Francoist regime. Since it is not feasible to implement full prosecution toward those who personally participated in the Civil War, the analysis exclusively refers to victims and authorities under the Francoist regime.

There are five relevant questions in the survey for this analysis: three questions on forms of restorative justice, a question halfway between restorative and procedural, and a question on retributive justice. Table A1 in the Online Appendix presents the univariate distributions of whether respondents would be in favor of nullifying political trials carried out during the dictatorship, attitudes toward the withdrawal of Francoist symbols from public spaces, and the creation of a monument for all victims of the regime. These questions capture a common component of TJ that aims at restoring the dignity of the victims.

Descriptive statistics reveal that the majority of the respondents approve making steps toward the recognition of the victims. In particular, about a majority of respondents would agree with nullifying the political lawsuits that occurred under the Franco regime and less than 20% of the sample would oppose it. Similarly, most respondents (56%) would approve the withdrawal of Francoist symbols, while only 23% oppose it. Finally, a wide majority of over 60% would agree with the creation of a monument in memory of the victims under the Franco regime. Generally, most Spanish agree with the need to recover symbolically the dignity of the victims and to officially condemn the past regime.

A step further in the TJ scale involves the creation of a Truth Commission to clarify human rights violations during the Francoist regime. Table A1 also indicates that citizens are split in two halves, with two-fifths of the respondents being in favor of setting up a Truth Commission, two-fifths being in favor of not doing anything in this respect, and one fifth of the sample reporting no opinion on this issue. Finally, Table A1 additionally presents the distribution of the respondents for the toughest component of TJ: prosecution of authorities who committed human rights violations in the past. Notice that there is more consensus about bringing criminals to be judged if they committed human rights violation than investigating whether there were violations or not in the first place through a Truth Commission. Since the transition, Spain’s priority has been political stability and the strength of the political institutions. In this line, creating a Truth Commission without enforcement to judge the criminals may be regarded as a policy to revive the past without solving any issue. Thus, some individuals may feel that the negative effects of stirring up the past would only be compensated if the outcome is the actual judicial accountability of the previous authorities, while this might be counter-productive if a Truth Commission does not involve the imposition of true penal consequences. To create a single scale of desire for TJ, I factor-analyzed the five relevant variables to create a scale on respondents’ approval of the TJ components: the Justice Score (JS).⁷

Independent Variables: Emotions toward the Francoist regime

The survey allows me to classify nine emotions into two categories: negative and positive (or non-negative) emotions. Table 1 presents their univariate distributions. The data reveal that negative emotions are those that best express respondents’ feelings. In particular, the most prevalent emotion toward the Francoist regime is anger with 37.1% of the respondents mentioning it as either the first or the second emotion. Anger is followed by incomprehension (28.7%), sadness (27.6%), fear (18.9%), and discomfort (18.0%). By contrast, a very limited number of people report holding positive emotions such as nostalgia (5.9%), patriotism (5.1%), or pride (3.1%). By contrast, a number of respondents show indifference toward the past (16.2%), a lack of emotional engagement.

Table 1.

Emotions toward the Francoist Regime.

What is the main emotion that the Francoist regime evokes in you?And the second main emotion?	First emotion % (N)	Second emotion % (N)
Sadness	16.2 (477)	11.5 (336)
Discomfort	9.0 (264)	9.0 (265)
Patriotism	3.2 (93)	1.9 (55)
Anger–rage	23.5 (690)	13.7 (403)
Indifference	11.8 (345)	5.1 (150)
Pride	1.2 (34)	1.9 (56)
Fear	8.6 (254)	10.3 (303)
Nostalgia	2.5 (72)	3.5 (102)
Incomprehension	10.5 (308)	18.3 (538)
Another feeling	4.7 (137)	2.2 (64)
Don’t know/No answer	8.9 (260)	22.5 (663)
N	2935	2935

I begin to assess the relationship between discrete emotions and respondent’s willingness for justice by a bivariate analysis. This allows me to simplify the empirical exploration by including only those emotions that are more likely to be related to the outcome of interest. Table 2 presents the exploratory bivariate relationships between each emotion and the score of willingness for TJ. Columns 1 to 3 present the means in the TJ scale for those respondents who do not mention the emotion, those who mention it as the second feeling, and those who mention it as the first feeling. Columns 4 to 6 compare the means between the three groups of respondents by implementing F test, which tests the null hypothesis that columns 1 and 2, 1 and 3, and 2 and 3 are not significantly different, respectively.

Table 2.

Difference in Justice Score Means According to the Feelings Expressed in the Survey.

List of emotions	(1)	(2)	(3)	Col. 1 = 2	Col. 1 = 3	Col. 2 = 3	One-way ANOVA	η ²
List of emotions	Not mentioned	As second feeling	As first feeling	Prob. > F	Prob. > F	Prob. > F	Prob. > F	η ²
Sadness	2.63 (0.03)	2.99 (0.08)	2.83 (0.06)	F_(1,2629) = 17.2*** p < 0.001	F_(1,2629) = 8.4*** p < 0.001	F_(1,2629) = 2.3 p = 0.13	F_(2,2627) = 14.1 p < 0.001	0.011
Discomfort	2.72 (0.03)	2.70 (0.09)	2.62 (0.09)	F_(1,2629) = 0.04 p = 0.84	F_(1,2629) = 1.14 p = 0.28	F_(1,2629) = 0.44 p = 0.50	F_(2,2627) = 0.74 p = 0.48	0.001
Patriotism	2.76 (0.03)	2.11 (0.21)	1.69 (0.15)	F_(1,2629) = 9.7*** p < 0.002	F_(1,2629) = 47.1*** p < 0.001	F_(1,2629) = 2.59 p = 0.11	F_(2,2627) = 38.8 p < 0.001	0.029
Anger–rage	2.39 (0.04)	3.12 (0.07)	3.20 (0.05)	F_(1,2629) = 89.6*** p < 0.001	F_(1,2629) = 179.7*** p < 0.001	F_(1,2629) = 1.13 p = 0.29	F_(2,2627) = 133.6 p < 0.001	0.092
Indifference	2.84 (0.03)	2.39 (0.12)	1.98 (0.08)	F_(1,2629) = 13.4*** p < 0.001	F_(1,2629) = 100.2*** p < 0.001	F_(1,2629) = 8.19*** p < 0.005	F_(2,2627) = 72.3 p < 0.001	0.052
Pride	2.75 (0.03)	1.67 (0.18)	1.71 (0.24)	F_(1,2629) = 34.8*** p < 0.001	F_(1,2629) = 18.0*** p < 0.001	F_(1,2629) = 0.02 p = 0.87	F_(2,2627) = 30.8 p < 0.001	0.023
Fear	2.60 (0.03)	3.13 (0.07)	3.10 (0.08)	F_(1,2629) = 42.7*** p < 0.001	F_(1,2629) = 35.2*** p < 0.001	F_(1,2629) = 0.08 p = 0.77	F_(2,2627) = 37.4 p < 0.001	0.028
Nostalgia	2.76 (0.03)	2.23 (0.16)	1.75 (0.18)	F_(1,2629) = 10.4*** p < 0.002	F_(1,2629) = 31.4*** p < 0.001	F_(1,2629) = 4.0* p < 0.05	F_(2,2627) = 28.7 p < 0.001	0.021
Incomprehension	2.72 (0.04)	2.71 (0.06)	2.68 (0.08)	F_(1,2629) = 0.00 p = 0.97	F_(1,2629) = 0.22 p = 0.64	F_(1,2629) = 0.14 p = 0.71	F_(2,2627) = 0.14 p = 0.87	0.000

ANOVA: analysis of variance.

The bivariate analysis demonstrates that those respondents who mention a feeling first or second tend not to have an average JS statistically different from each other. Among the possible emotional responses, anger and indifference have the most substantial differences in means in the JS scale with impact sizes (η² = 0.092 and η² = 0.052, respectively). Furthermore, the five other emotions emerging with statistically significant effect sizes, ranging from small and medium-small (0.011 < η² < 0.029) according to Cohen’s (1988) rule of thumb for η² after the one-way analysis of variance (ANOVA), are sadness, patriotism, pride, fear, and nostalgia. Therefore, these emotions will be included in the multivariate analysis. By contrast, discomfort, incomprehension, and other feelings are all excluded from the analysis because they have negligible differences across their means in the JS scale.

Individual Controls

To adjust the relationships for plausible confounders, I control for a number of variables that are generally relevant in studies of the determinants of attitudes toward TJ in Spain, as well as in a comparative perspective. These are grouped in individual (generation, gender, education, religiosity, and individual’s ideology), socialization (victimization and/or family victimization, family’s ideology, and family’s politicization), and contextual factors (region of residence).

Generation

Respondents’ generational cohort is likely to be an important determinant for emotional engagement with the past, as well as the desire for TJ. The respondent’s age is a factor that might explain differences of attitudes toward the TJ of past events. Most obviously, a number of respondents have never had a first-hand experience of the regime by the time of the survey—those under the age of 30 years (19.4%). Thus, we might expect older people to have stronger feelings and be more emotionally engaged toward the Francoist regime because of their personal proximity to past events. However, the strength of the Francoist regime in the collective imaginary in the Spanish society may have spilled this emotional engagement to all generations. Then, we should expect a negative effect of age and the desire for historical justice. While the personal proximity might enhance individual’s emotional engagement toward the past, it might also make older individuals reluctant to support reparation policies. The mechanism that some scholars have associated with this is fear: fear of reprisal by those affected by TJ policies or a fear of reigniting the conflict and return to the authoritarian regime (Aguilar et al., 2011; Nalepa, 2010). However, the fear mechanism loses its strength in a consolidated democracy where the possibility of returning to the old regime is not credible. Furthermore, Samii’s (2013) analysis of a survey about attitudes toward TJ in Burundi indicates that the mechanism of fear/insecurity does not have a negative effect on the preference for punishment. Overall, I expect a negative effect of age on demand for more extensive and deeper justice, and I do not expect this to be channeled through emotional engagement.

Gender

The main rationale for gender differences in the impact of emotions lies in the socialization practices early in life which results in individual differences in children’s ability to express and regulate emotions, and socialization processes are dependent on the nature of parent–child interactions (e.g. Garside and Klimes-Dougan, 2002). In particular, parental contingent responses to expressed emotion in children have been found to facilitate (e.g. reward, magnify) or inhibit (e.g. override, neglect, punish) the expression of various discrete emotions such as sadness, anger, and fear. Research on early stages of children development consistently found that parents’ socialization of emotions is based on the gender of their child (e.g. Eisenberg et al., 1998). This is finally reflected in adults with women being more emotionally expressive than men (e.g. Fischer, 1993). Among boys, research suggests that with development, parental socialization inhibits the expression and attribution of most emotions, especially sadness and fear. By contrast, socially unacceptable emotions such as anger are inhibited in girls (e.g. Aldrich and Tenenbaum, 2006). Therefore, I expect gender to be associated with expressed emotions as well as TJ mechanisms, which justifies its inclusion as a control variable.

Education

This confounder might also affect emotional engagement with the past as well as attitudes toward TJ. In a Cambodian survey, Pham et al. (2004) found that educational attainment was a main determinant of attitudes toward TJ. As a consequence, I expect that those who report a higher level of educational attainment will support stronger measures against former authorities and in favor of the victims.

Religiosity

This is among the most important factors that are likely to drive emotional engagement as well as TJ. The effect of religion might be strong for historical reasons in Spain. The religious division was significant in the escalation of violence preceding the Civil War, and the Catholic Church sided the Francoist regime during the dictatorship. In fact, Franco’s regime was partly established on religious grounds with Catholicism as the official religion of the country (Seco, 2002). Even though the effect of religion in Spain has receded in recent decades, it remains a strong predictor of vote choice (Montero et al., 2008). Therefore, I expect that more religious individuals will be more likely to report positive feelings and less likely to report negative feelings toward the Francoist regime, as well as less religious individuals to be more likely to support TJ mechanisms. I construct a scale of religiosity that ranges between 0, those who define themselves as non-believers or atheists, and 4, those who report attending religious services every Sunday and holidays (13.4%), as well as several times per week (1.9%) to take the value of 4 (see the Online Appendix B for further information on the construction of this scale).

Ideology

The perpetrators’ right-wing ideological leaning is likely to lead those individuals more congruent with the past regime’s ideology to reject more strongly TJ mechanisms. At the same time, leftist individuals may report holding more negative emotions toward the regime than rightist respondents. In the survey, respondents are asked to locate themselves in a scale between 1 (extreme left) and 10 (extreme right). However, there are a substantial number of respondents who failed to provide an answer to the question. In particular, 8.5% of the respondents admit that they do not know, and 8.3% of the respondents do not want to answer the question. My strategy to avoid losing observations is to recode the missing values from those respondents who do not know and do not answer to the mean of ideology for the sample that do provide an answer. However, since missing values in this answer are likely to be due to a non-random process, I include in my analysis a dummy variable that identifies those respondents who do not answer the question. As a result, the slope for these individuals will be adjusted by their non-response in this variable. Following Aguilar et al. (2011), I also include an individual-level confounder that captures the attentiveness in the debate about the promulgation of the law in Spain in the time of the survey: interest in politics.

This variable is measured by asking individuals whether they are very (6%), quite (26%), little (36.9%), or not at all (30.5%) generally interested in politics. The 20 individuals (0.7%) who do not provide an answer to this question are recoded as not at all. Also, I include a measure of general knowledge about politics that captures how much individuals know about country’ past, which might affect their emotional sensitivity and their attitudes toward TJ. The knowledge scale indicates the number of questions correctly answered by a respondent to three factual questions (see Online Appendix for further details).

Socialization Controls

Victimization

It refers to those individuals who consider themselves direct victims of the past regime or indirectly through a close individual. Table B in the Online Appendix presents the distribution of victimization in the Francoist regime. Even though victimization could be regarded as an individual-level process and not a product of socialization, I include them as part of the socialization process because an extremely limited number of respondents in the survey report having been directly a victim of the regime (see Online Appendix B for further information on the distribution of this variable). Therefore, I use a dummy that takes a value of 1 if the respondent has been a victim of the Francoist regime either personally or through a close acquaintance, and 0 otherwise. Following most literature, I expect victimization to influence the emotional engagement toward the past regime and the willingness for TJ (Aguilar et al., 2011; Balcells, 2012; Pham et al., 2004; Samii, 2013).

Family’s Ideology

The primary socialization process of the respondent is generally linked to later attitudes and beliefs. The direction of the effects is expected to be similar as those of the respondent’s own ideology, although their influence is expected to be weaker. Fortunately, respondents were asked about the ideology of their fathers and mothers. Using mothers’ and fathers’ ideologies separately in the analysis would preclude the effect of family ideology. What might be conceptually understood as a mixture of several elements: direction of the ideology (mainly left or right), ideological strength (strongly or weakly left or right), and within-family ideological consistency (congruence of father’s and mother’s ideology) weighted by the perceived authority of the source (in Spain’s male-dominant society, father’s ideology has a stronger influence in the family’s ideology). I construct an index of family’s ideology that combines these four features. Overall, this measure ranges from 0 to 9 by taking the weighted average of a respondent’s mother and/or father ideologies (see the Online Appendix B for further information on the construction of this variable).

Family Talked About Politics Politicization

This variable captures how important politics was during the primary socialization of the respondent, ranging from 0 to 4. This might have led to stronger feelings toward past sociopolitical objects, as well as stronger attitudes toward policies about the past (see the Online Appendix B for further information on the construction of this variable).

Contextual Confounders

Region-specific fixed effects are used to control for respondents’ context. These are justified given both the relevance of context for explaining people’s national identity in Spain (Barceló, 2014) and the particular repression of the Francoist regime against those ethnic identities that could challenge Spain’s national identity (Balcells, 2012). Citizens of some regions may have generated collective histories during the dictatorship different from those in other regions (Balcells, 2012). Consequently, models include fixed effects for Catalonia, Basque Country, Galicia, and Andalucía. Table C presents the main descriptive statistics of the three blocks of confounders after scaling them from 0 to 1.

Multivariate Analysis

This section first evaluates the hypothesis that a positive (negative) emotional engagement with the past regime influences a respondent’s support for TJ mechanisms independent of common confounders in the literature. I test this hypothesis by conducting a multivariate analysis predicting respondents’ willingness to pursue stronger TJ measures. This analysis adjusts for theoretically relevant confounders that are expected to correlate with both emotions and TJ. Table 3 reports the results of six ordinary least squares (OLS) regression models exploring the effects of six discrete emotions (anger, fear, sadness, nostalgia, patriotism, and pride), and indifference, after adjusting for all the above-mentioned controls, across different model specifications. To illustrate the size of the effects, I take the most conservative estimates (model 6 in Table 3) to directly compare the expected values of the dependent variable given the changes of similar magnitude in the distribution of the predictors (Figure 2). Changes in the outcome are based on a shift from 0 to 1 for dummy variables, or from the mean to 1 standard deviation above the mean for non-dummy variables, with all other covariates kept either at their reference category or at their mean. The solid and dashed lines indicate the baseline situation where all covariates are kept constant at their reference categories or at their mean.

Table 3.

Regression Effects of Discrete Emotions, Individual, Socialization, and Contextual Factors on the Desire for Transitional Justice.

	Dependent variable: desire for transitional justice
	(1)	(2)	(3)	(4)	(5)	(6)
Discrete emotions
Anger		0.61*** (0.05)	0.41*** (0.05)	0.37*** (0.05)	0.38*** (0.05)	0.34*** (0.05)
Fear		0.44*** (0.06)	0.41*** (0.06)	0.39*** (0.06)	0.41*** (0.06)	0.37*** (0.06)
Sadness		0.17*** (0.05)	0.16*** (0.05)	0.18*** (0.05)	0.19*** (0.05)	0.17*** (0.05)
Indifference		−0.51*** (0.08)	−0.37*** (0.08)	−0.37*** (0.08)	−0.34*** (0.08)	−0.32*** (0.08)
Nostalgia		−0.59*** (0.13)	−0.30** (0.13)	−0.30** (0.13)	−0.26** (0.13)	−0.25* (0.13)
Patriotism		−0.62*** (0.12)	−0.27** (0.12)	−0.26** (0.12)	−0.30** (0.12)	−0.29** (0.12)
Pride		−0.54*** (0.13)	−0.25* (0.13)	−0.26** (0.13)	−0.22* (0.14)	−0.23* (0.13)
Individual factors
Gender	0.06 (0.05)		0.01 (0.05)	0.01 (0.05)	0.003 (0.05)	0.003 (0.05)
Age [0–1]	−0.56*** (0.11)		−0.40*** (0.11)	−0.47*** (0.11)	−0.49*** (0.12)	−0.54*** (0.11)
Education [0–1]	−0.33*** (0.09)		−0.33*** (0.09)	−0.33*** (0.09)	−0.33*** (0.09)	−0.36*** (0.09)
Municipality size [0–1]	0.09 (0.08)		0.01 (0.08)	0.02 (0.08)	0.03 (0.08)	0.07 (0.08)
Religiosity [0–1]	−0.48*** (0.08)		−0.43*** (0.08)	−0.42*** (0.08)	−0.41*** (0.09)	−0.37*** (0.08)
Ideology [0–1]	−1.94*** (0.15)		−1.68*** (0.14)	−1.65*** (0.14)	−1.57*** (0.16)	−1.51*** (0.16)
Ideology DK	0.01 (0.09)		0.09 (0.09)	0.12 (0.09)	0.12 (0.09)	0.06 (0.09)
Interested in politics [0–1]	0.12 (0.09)		0.10 (0.09)	0.07 (0.08)	0.08 (0.09)	0.10 (0.09)
Knowledge scale [0–1]	0.11 (0.07)		0.06 (0.07)	0.03 (0.07)	0.04 (0.07)	0.08 (0.07)
Socialization factors
Victimization
Arrested	0.10 (0.09)			0.14 (0.09)	0.09 (0.09)	0.08 (0.09)
Imprisoned	0.14 (0.09)			0.10 (0.09)	0.12 (0.09)	0.12 (0.09)
Sacked	−0.05 (0.12)			−0.02 (0.12)	−0.02 (0.12)	−0.01 (0.12)
Fined	0.30** (0.12)			0.29** (0.12)	0.29** (0.12)	0.27** (0.12)
Exiled	−0.05 (0.11)			−0.08 (0.11)	−0.09 (0.11)	−0.08 (0.11)
Executed	0.28*** (0.10)			0.33*** (0.10)	0.30*** (0.10)	0.25** (0.10)
Politicization
Talked about politics [0–1]	0.13 (0.09)				0.09 (0.09)	0.09 (0.09)
Family ideology [0–1]	−0.25* (0.13)				−0.18 (0.13)	−0.14 (0.13)
Contextual factors
Catalonia	0.44*** (0.07)					0.38*** (0.07)
Basque Country	0.70*** (0.11)					0.58*** (0.10)
Galicia	0.43*** (0.09)					0.41*** (0.10)
Andalucía	0.39*** (0.06)					0.31*** (0.06)
Constant	3.62*** (0.10)	2.47*** (0.05)	3.53*** (0.11)	3.51*** (0.11)	3.51*** (0.12)	3.31*** (0.12)
Observations	2428	2630	2378	2378	2275	2275
R ²	0.25	0.17	0.26	0.28	0.28	0.30
Adjusted R²	0.24	0.17	0.26	0.27	0.27	0.29
Residual std. error	1.11	1.14	1.09	1.08	1.07	1.06

The symbol [0–1] indicates that variables are rescaled from 0 to 1.

p < 0.1; **p < 0.05; ***p < 0.01.

Figure 2.

Expected Values of the Desire for Justice Scale at Changes in the Predictors (95% CI).

Negative Emotions

Models 2 to 6 in Table 3 establish the relationship between discrete emotions and TJ with a different set of adjustments in the models.⁸ The relationships provide support for the main hypotheses. Negative feelings toward the past regime are associated with a stronger desire for restoring the victims and prosecuting past authorities (Hypothesis 1). In particular, reporting to be angry, fearful, or sad toward the Francoist regime increases the expected value in the TJ scale by 0.61, 0.44, and 0.17, respectively. These figures can be conceptualized as the total effects of emotions. However, much of the variation in the TJ scale that is explained by this set of negative discrete emotions should actually be attributed to variation in confounders that are related to both emotions and the justice scale. For instance, once model 2 is controlled for individual-level confounders, then the coefficient of anger decreases from 0.61 to 0.41, although it remains statistically significant at 99.9% confidence level. This suggests that the previously identified effect of anger on the outcome is partially channeling the indirect effects of age, education, religiosity, and/or ideology on the justice scale. Similarly, the effect of fear is also slightly reduced, especially, after the introduction of the regional dummies, although it remains significantly positive.

While the size of the effects of anger and fear is statistically distinguishable between each other, both yield a much stronger effect on TJ attitudes than the effect of another negative emotion, sadness. Therefore, there is mixed support for the “hot cognition hypothesis” with regard to the joint effect of negative feelings (Hypothesis 2) because sadness seems to increase the demand for TJ mechanisms but not as much as anger and fear.⁹ Altogether, there is strong evidence that negative emotions, especially anger and fear, importantly shape the demand for stronger TJ mechanisms.

To assess the size of the effects of emotions, Figure 2 shows that a respondent who reports feeling anger or fear toward the former regime has an expected value of the TJ scale of 2.66 and 2.69, respectively, keeping all other variables constant. In both cases, the fitted values are significantly different from the baseline prediction of 2.32 (95% confidence interval: 2.20–2.43). In relative terms, the effects of anger and fear are much stronger than the effects of gender, municipality size, interest in politics, knowledge, and most forms socialization factors: about three times stronger than a shift in 1 standard deviation of education; 2.5 stronger than age, education, religiosity; and an equivalent effect of having a relative fined or executed under Franco’s rule as well as 1 standard deviation change in one’s ideology. Therefore, the effect of anger and fear is substantively important.

Positive Emotions

The number of people who report holding positive emotions toward the past regime is lower than negative emotions. As a consequence, the estimates for positive emotions tend to be less precise. Notwithstanding this fact, models 2 to 6 suggest that positive emotions toward Franco’s regime are significantly less likely to support TJ policies (Hypothesis 3). The coefficients for positive emotional engagement through nostalgia, patriotism, and pride range between −0.51 and −0.62 and are significant at 99.9%. Once I adjust for individual-level factors, these coefficients are shrunken to about half of their initial effect, ranging from −0.25 to −0.37 in model 3, being two of them significant at 95% confidence level. After I include all covariates in the model, they are further reduced, which leads the effect of nostalgia and pride to be only marginally significant, and patriotism significant at 95% level, yet none of them is distinguishable from one another (Hypotheses 3 and 4).¹⁰ Taking the last column to calculate expected values, Figure 2 illustrated that the size of their effects are not as strong as negative emotions but are still remarkably large. Overall, there is strong support for the importance of positive emotions in determining the lack of demand for TJ.

Lack of Emotional Engagement

Although the predictions for negative and positive emotions may be regarded as unsurprising, a noteworthy finding of these analyses is the role of indifference; that is, those who provide no emotional evaluation with regard to the past regime are indifferent to the past regime’s symbolic and political actions. Table 3 and Figure 2 reveal that those who report being indifferent to the past regime are more likely to support avoiding any sort of TJ. Thus, the effect of indifference is negative and statistically significant from zero (Hypothesis 5), but, in addition to this, it is statistically indistinguishable from any other positive emotion (Hypothesis 6).¹¹

Taking into account that the most common non-negative category was indifference, this finding supports the idea that bystanders implicitly are on the side of active supporters of the regime—those who feel pride, patriotism, or nostalgia—rather than with the outraged and the victims—those who feel anger, fear, or sadness toward the past regime.

Other Variables

Models 1 to 6 present the total and partial effect of non-emotional factors on the desire for TJ. Individual factors seem to play an important role in explaining attitudes toward justice. The most salient predictors are ideological self-placement and family victimization. The strength of the effect of ideology is expected due to the ideological leaning of the past regime, as well as the contemporaneous ideological divide that characterizes the debate in the public opinion about transitional policies. As for the effects of family victimization, we need to differentiate across types of victimization. For instance, only those who report having a close member who was fined or executed under the regime are more likely to support stronger TJ measures. By contrast, other forms of victimizations, such as having a family member exiled or sacked from work, have no effect on TJ attitudes.

In a second stage of relevance, age is significantly negative across all models, which is consistent with prior literature (e.g. Aguilar et al., 2011; Nalepa, 2010). Similarly, the effects of educational attainment and religiosity are also significant. Those who are more educated and religious are more forgiving toward the Francoist authorities and less likely to support TJ measures. The rest of control variables are less consistently associated with the support for TJ.

Conclusion

Emotional engagement with the past significantly influences the support for TJ. The discrete emotions of anger and fear have a large effect on the desire for justice, stronger than most other predictors and comparable to those considered the main drivers of attitudes such as ideology and some types of victimization. Additionally, the effect of sadness significantly influences attitudes with a remarkable strength relative to most other confounders, although less strongly than fear and anger. Similarly, even though positive emotions halve their effects once confounders are included in the analysis, and their presence significantly reduces support for TJ.

Importantly, this study has analyzed the relative impact of indifference. A long-standing thought in political science has assigned a great share of the blame for past atrocities to passive agents who implicitly supported authoritarian regimes with their lack of political engagement. The results presented here provide further ground for this reasoning but with regard to the demand for TJ. In particular, I find that those individuals who report holding a feeling of indifference toward the regime are not neutral in their demand for TJ mechanisms, but they tend to support a lack of justice mechanisms. Importantly, the effect of indifference on the reduction in the demand for justice mechanisms is as strong as the effect of feeling pride, patriotism, or nostalgia.

The results of this study underscore the central role that emotions play in people’s attitudes toward TJ. Certainly, this and other research document that TJ attitudes are a product of multiple factors. I confirm that some individual, socialization, and contextual factors stand out as salient determinants of attitudes. However, I also provide evidence that supports the incompleteness of a model of attitudes toward historical justice when emotions are not included. In line with scholarly research on the role of emotions in the opinion formation process, positive and negative emotions play a significant and distinct role on opinion formation even after adjusting for the most relevant covariates that have been found in the literature of TJ attitudes.

Importantly, I also find evidence supporting that a simple valence assessment of circumstances (or stimuli) as either punishing (negative emotions) or rewarding (positive emotions) is sufficient to capture the influences of emotional engagements. Even though I used a finer-grained set of emotions to accommodate a more complex array of relationships, a simpler valence assessment captures most variation between the variables. This is consistent with the theoretical model, based on the hot cognition hypothesis, that I used to build my empirical expectations. The fundamental role of emotions in the attitudinal model presented in this article has direct implications for policy action. Emotions override rationality, and as such, when considering TJ policy options, the importance of emotions needs to be considered. Results suggest that citizens will not necessarily become in internal peace with the past as their material opportunities improve, but policies should mainly aim at healing the psychological conditionings derived from the past that elicit a negative emotional reaction.

Readers will recognize some limitations to this analysis. Indeed, I rely on a survey conducted in Spain in 2008, which means 30 years after the death of the dictator and the transition toward democracy. Future research will have to address whether the salient role of emotions is even stronger in forming opinions toward justice mechanisms in cases where the transitional process is more recent or, in contrast, emotions’ role recedes in favor of a more conscious evaluation of their opinions. An obvious question that remains unresolved is whether the similar functioning of negative emotions, especially anger and fear, and positive emotions found in this research disentangles in more meaningful distinct effects in the aftermath of transitional process.

Additionally, the survey used for this analysis was not designed to capture the emotional underpinnings of respondents and, as a consequence, some limitations arise from the questions used itself. For instance, providing mutually exclusive responses for emotions, the lack of evaluation of their strength and only of their existence, or, perhaps, not tapping into the discrete emotions more relevant for the respondents are potential drawbacks that I am not able to address in this analysis. Additionally, this research has not experimentally manipulated emotional engagement with the past and, as any non-experimental research design, it is subject to potential omitted variable bias. Nevertheless, I have drawn on previous literature from my case study to identify those more likely confounders and I have adjusted for them. Despite the research limitations inherent to this research design, this is, to my knowledge, the first attempt to systematically capture the independent influence of affect toward a salient and an emotionally charged sociopolitical object such as a past dictatorship to understand people’s attitudes toward TJ mechanisms.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplementary Information

Additional supplementary information may be found with the online version of this article.

Appendix A: Construction of the Transitional Justice Scale

Figure A: Distribution of respondents in the Desire for Justice scale.

Table A1: Indicators of desire for transitional justice.

Table A2: Correlation Matrix of items and scale of desire for justice.

Appendix B: Construction of Controls

Table B1: Victimization of the Francoist regime.

Appendix C: Descriptive Statistics

Table C: Descriptive statistics of respondents’ characteristics.

Notes

Author biography

Joan Barceló is PhD Candidate of Political Science at the Washington University in St. Louis. His research focuses on comparative political behavior and institutions. He is particularly interested in the consequences conflict, political regimes and political events on people’s preferences and their institutions. His articles have appeared or are forthcoming in journals that include the British Journal of Political Science, Party Politics, Social Science Quarterly, Nations and Nationalism, Journal of Cross-Cultural Psychology and Personality and Individual Differences. His work has been cited by several media outlets, including the BBC.

References

Aguilar

Balcells

Cebolla-Boado

(2011) Determinants of Attitudes toward Transitional Justice: An Empirical Analysis of the Spanish Case. Comparative Political Studies 44 (10): 1397–1430.

Aldrich

Tenenbaum

(2006) Sadness, Anger, and Frustration: Gendered Patterns in Early Adolescents’ and Their Parents’ Emotion Talk. Sex Roles 55 (11–12): 775–785.

Backer

(2010) Watching a Bargain Unravel? A Panel Study of Victims’ Attitudes about Transitional Justice in Cape Town, South Africa. International Journal of Transitional Justice 4 (3): 443–456.

Balcells

(2012) The Consequences of Victimization on Political Identities Evidence from Spain. Politics & Society 40 (3): 311–347.

Barceló

(2014) Contextual Effects on Subjective National Identity. Nations and Nationalism 20 (4): 701–720.

Barnett

(1999) Bystanders: Conscience and Complicity during the Holocaust. Westport, CT: Greenwood Press.

Brader

(2006) Campaigning for Hearts and Minds: How Emotional Appeals in Political Ads Work. Chicago, IL: University of Chicago Press.

Brader

Valentino

Suhay

(2008) What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat. American Journal of Political Science 52 (4): 959–978.

Campbell

Converse

Miller

et al . (1960) The American Voter. Hoboken, NJ: John Wiley & Sons.

10.

Civettini

AJW

Redlawsk

(2009) Voters, Emotions, and Memory. Political Psychology 30 (1): 125–151.

11.

Claassen

(2016) Group Entitlement, Anger and Participation in Intergroup Violence. British Journal of Political Science 46 (1): 127–148.

12.

Cohen

(1988) Statistical Power Analysis for the Behavioral Sciences. Mahwah, NJ: Lawrence Erlbaum.

13.

Converse

. (2006) The Nature of Belief Systems in Mass Publics (1964). Critical Review 18 (1–3): 1–7.

14.

Delli Carpini

Keeter

(1996) What Americans Know about Politics and Why It Matters. New Haven, CT: Yale University Press.

15.

Eisenberg

Cumberland

Spinrad

(1998) Parental Socialization of Emotion. Psychological Inquiry 9 (4): 241–273.

16.

Elster

(2004) Closing the Books: Transitional Justice in Historical Perspective. Cambridge: Cambridge University Press.

17.

Elster

(2006) Retribution and Reparation in the Transition to Democracy. Cambridge: Cambridge University Press.

18.

Fischer

(1993) Sex Differences in Emotionality: Fact or Stereotype? Feminism & Psychology 3 (3): 303–318.

19.

Garside

Klimes-Dougan

(2002) Socialization of Discrete Negative Emotions: Gender Differences and Links with Psychological Sistress. Sex Roles 47 (3): 115–128.

20.

Gibson

(2002) Truth, Justice, and Reconciliation: Judging the Fairness of Amnesty in South Africa. American Journal of Political Science 46 (3): 540–556.

21.

Gibson

(2004) Does Truth Lead to Reconciliation? Testing the Causal Assumptions of the South African Truth and Reconciliation Process. American Journal of Political Science 48 (2): 201–217.

22.

Gibson

(2006) The Contributions of Truth to Reconciliation Lessons From South Africa. Journal of Conflict Resolution 50 (3): 409–432.

23.

Gibson

Gouws

(1999) Truth and Reconciliation in South Africa: Attributions of Blame and the Struggle over Apartheid. American Political Science Review 93 (3): 501–517.

24.

Gibson

Claassen

Barceló

(2017) Is Political Intolerance Grounded in Emotions? Available at: http://www.chrisclaassen.com/emotion_intolerance.pdf.

25.

Green

Goldman

Salovey

(1993) Measurement Error Masks Bipolarity in Affect Ratings. Journal of Personality and Social Psychology 64 (6): 1029–1041.

26.

Green

Salovey

Truax

(1999) Static, Dynamic, and Causative Bipolarity of Affect. Journal of Personality and Social Psychology 76 (5): 856–867.

27.

Grodsky

(2009) Re-Ordering Justice: Towards A New Methodological Approach to Studying Transitional Justice. Journal of Peace Research 46 (6): 819–837.

28.

Grodsky

(2010) The Costs of Justice. Notre Dame, IN: University of Notre Dame Press.

29.

Groenendyk

Banks

(2014) Emotional Rescue: How Affect Helps Partisans Overcome Collective Action Problems. Political Psychology 35 (3): 359–378.

30.

Halperin

Canetti-Nisim

Hirsch-Hoefler

(2009) The Central Role of Group-Based Hatred as an Emotional Antecedent of Political Intolerance: Evidence from Israel. Political Psychology 30 (1): 93–123.

31.

Juliá Díaz

(1999) Víctimas de la Guerra Civil. Madrid: Temas de Hoy.

32.

Kuklinski

Quirk

(2000) Reconsidering the Rational Public: Cognition, Heuristics, and Mass Opinion. In: Lupia

McCubbins

Popkin

(eds) Elements of Reason: Cognition, Choice, and the Bounds of Rationality. Cambridge: Cambridge University Press, pp.153–82.

33.

Kuklinski

Quirk

Jerit

et al . (2000) Misinformation and the Currency of Democratic Citizenship. Journal of Politics 62 (3): 790–816.

34.

Ladd

Lenz

(2011.) Does Anxiety Improve Voters’ Decision Making? Political Psychology 32 (2): 347–361.

35.

Ladd

Lenz

(2008) Reassessing the Role of Anxiety in Vote Choice. Political Psychology 29 (2): 275–296.

36.

Lau

Redlawsk

(2001) Advantages and Disadvantages of Cognitive Heuristics in Political Decision Making. American Journal of Political Science 45 (4): 951–971.

37.

Lodge

Taber

(2005) The Automaticity of Affect for Political Leaders, Groups, and Issues: An Experimental Test of the Hot Cognition Hypothesis. Political Psychology 26 (3): 455–482.

38.

Lupia

(1994) Shortcuts versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections. American Political Science Review 88 (1): 63–76.

39.

Marcus

MacKuen

(1993) Anxiety, Enthusiasm, and the Vote: The Emotional Underpinnings of Learning and Involvement during Presidential Campaigns. American Political Science Review 87 (3): 672–685.

40.

Marcus

Neuman

MacKuen

(2000) Affective Intelligence and Political Judgment. Chicago, IL: University of Chicago Press.

41.

Meernik

(2005) Justice and Peace? How the International Criminal Tribunal Affects Societal Peace in Bosnia. Journal of Peace Research 42 (3): 271–289.

42.

Montero

Calvo

Martínez

(2008) El voto religioso en España y Portugal [Religious voting in Spain and Portugal]. Revista Internacional de Sociología 66: 19–54.

43.

Morris

Squires

Taber

et al . (2003) Activation of Political Attitudes: A Psychophysiological Examination of the Hot Cognition Hypothesis. Political Psychology 24 (4): 727–745.

44.

Nadeau

Niemi

Amato

(1995) Emotions, Issue Importance, and Political Learning. American Journal of Political Science 39 (3): 558–574.

45.

Nalepa

(2010) Skeletons in the Closet: Transitional Justice in Post-Communist Europe. Cambridge: Cambridge University Press.

46.

Olsen

Payne

Reiter

et al . (2010) When Truth Commissions Improve Human Rights. International Journal of Transitional Justice 4 (3): 457–476.

47.

Pham

Vinck

(2010) Transitioning to Peace: A Population-Based Survey on Attitudes about Social Reconstruction and Justice in Northern Uganda. Berkeley, CA: Human Rights Center.

48.

Pham

Vinck

Balthazard

et al . (2009) So We Will Never Forget: A Population-based Survey on Attitudes about Social Reconstruction and the Extraordinary Chambers in the Courts of Cambodia. Berkeley, CA: Human Rights Center.

49.

Pham

Weinstein

Longman

(2004) Trauma and PTSD Symptoms in Rwanda: Implications for Attitudes toward Justice and Reconciliation. JAMA 292 (5): 602–612.

50.

Popkin

(1991) The Reasoning Voter: Communication and Persuasion in Political Campaigns. Chicago, IL: University of Chicago Press.

51.

Redlawsk

(2002) Hot Cognition or Cool Consideration? Testing the Effects of Motivated Reasoning on Political Decision Making. Journal of Politics 64 (4): 1021–1044.

52.

Redlawsk

Civettini

AJW

Emmerson

(2010) The Affective Tipping Point: Do Motivated Reasoners Ever “Get It”? Political Psychology 31 (4): 563–593.

53.

Salas Larrazábal

(1977) Victimas De La Guerra Civil. Barcelona: Planeta Group.

54.

Samii

(2013) Who Wants to Forgive and Forget? Transitional Justice Preferences in Postwar Burundi. Journal of Peace Research 50 (2): 219–233.

55.

Sikkink

Walling

(2007) The Impact of Human Rights Trials in Latin America. Journal of Peace Research 44 (4): 427–445.

56.

Skaar

(1999) Truth Commissions, Trials-or Nothing? Policy Options in Democratic Transitions. Third World Quarterly 20 (6): 1109–1128.

57.

Sonis

Gibson

de Jong

et al . (2009) Probable Posttraumatic Stress Disorder and Disability in Cambodia: Associations with Perceived Justice, Desire for Revenge, and Attitudes toward the Khmer Rouge Trials. Journal of the American Medical Association 302 (5): 527–536.

58.

Staub

(1993) The Psychology of Bystanders, Perpetrators, and Heroic Helpers. International Journal of Intercultural Relations 17 (3): 315–341.

59.

Taber

Milton

(2006) Motivated Skepticism in the Evaluation of Political Beliefs. American Journal of Political Science 50 (3): 755–769.

60.

Thoms

ONT

Ron

Paris

(2010) State-Level Effects of Transitional Justice: What Do We Know? International Journal of Transitional Justice 4 (3): 329–354.

61.

Valentino

Gregorowicz

Groenendyk

(2009) Efficacy, Emotions and the Habit of Participation. Political Behavior 31 (3): 307–331.

62.

Verdeja

(2012) Moral Bystanders and Mass Violence. In: Jones

(ed.) New Directions in Genocide Research. New York: Routledge, pp.153–168.

63.

Vinck

Phuong

Tino

(2011) Talking Peace: A Population-Based Survey on Attitudes about Security, Dispute Resolution, and Post-Conflict Reconstruction in Liberia. Berkeley, CA: Human Rights Center.

64.

Vinck

Phuong

Suliman

et al . (2008) Living with Fear: A Population-based Survey on Attitudes about Peace, Justice, and Social Reconstruction in Eastern Democratic Republic of the Congo. Berkeley, CA: Human Rights Center.

65.

Zalaquett

(1992) Balancing Ethical Imperatives and Political Constraints: The Dilemma of New Democracies Confronting Past Human Rights Violations. Hastings Law Journal 43 (6): 1425–1438.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB