The many faces of the politics of shame in European policymaking

This paper analyzes shaming attempts in the European Parliament (EP) over a long period. Drawing on existing literature on shaming and stigmatization in International Relations, as well as on studies on blame avoidance (Public administration), this paper explores the extent to which (and how) shaming attempts were used in day-to-day European policymaking. The paper first shows how the word ‘shame’ has been employed by key policymakers in different policy areas. Data analyzed include EP speech acts (mainly debates) from 1994 to 2014. The second part of the paper consists of an interpretative and contextualizing qualitative analysis, exploring in-depth social and economic policy areas. This paper shows that, in these policy areas, shaming attempts have often served as an ideological tool, or have become entangled in turf wars between supranational institutions and Member States. The in-depth study also illustrates the circumstances under which shaming attempts have led to compliance, non-compliance or shame backlashes.


Introduction
Public shaming is common currency in European politics. Human rights campaigners have time and again shamed European Union (EU) institutions for their response to the 2015 refugee crisis (Sanchez Salgado, 2021a). The EU frequently engages in shaming attempts in their public declarations and in social media. In response to the Ukrainian war, the former European Council president Donald Tusk declared that EU governments had disgraced themselves by failing to impose the toughest possible sanctions ('Donald Tusk,' 2022). Following the Qatargate scandal, Home Affairs Commissioner Ylva Johansson said to those who had received money: 'Shame on you. Shame on you for violating trust.' 1 In the context of democratic backsliding in Hungary, and referring to a bill that makes it illegal to portray homosexuality, President von der Leyen tweeted 'The Hungarian bill is a shame.' 2 Shaming attempts have not only been used in relation to EU fundamental values or scandals: they have also been used for day-to-day topics such as the EU tax haven blacklist (Rusina, 2020). Despite all this, not much is known about the politics of shame in European politics. The frequent use of shaming attempts in policymaking raises interesting questions. How have EU policymakers used shaming attempts? Have shaming attempts served to defend and promote EU values and democratic principles, or are they just part of blame games in the context of policy controversies?
The main contribution of this paper is to develop our understanding of shaming attempts in European policymaking, bringing together for the first time advances in the study of shaming and stigmatization (prominent mainly in the study of International Relations (IR)) and literature on blame games and blame management (prominent in the study of public administration). To this purpose, this paper analyzes the EU's most open and representative institution: the European Parliament (EP). As a public and transparent body offering a wealth of information concerning the attitudes and opinions of politicians, the EP has been considered an ideal site for the study of the intersubjective character of emotions (Sanchez Salgado, 2023). While there has been much academic interest in the EP, only very recent studies have focused explicitly on the role of emotions there (Gürkan, 2021;Sanchez Salgado, 2021b).
To unpack the role of shaming attempts in the EP, this paper explores the following two questions: how has the word 'shame' been used by European officials and Members of the EP across policy areas? Under which circumstances have shaming attempts been more likely to lead to compliance? This paper provides a framework to analyze the role of shaming attempts in political institutions, considering not only existing IR literature on this topic, but also literature on blame games and blame management. This framework offers a more nuanced approach to shaming attempts, establishing a distinction between normative, disciplinary and policy-oriented shaming.
Based on a qualitative Atlas.ti content analysis of EU speech acts from 1994 to 2014, the empirical part of this paper shows that shaming attempts have been common in European politics. An in-depth study of EP debates on social and economic policy shows that shaming attempts have often been used to pursue ideologically based policy preferences. The empirical section of this paper also discusses the web of interrelated circumstances and factors that help understand why shaming attempts led to strong social pressure and compliance. This paper shows, for example, that in cases when ideological preferences have been presented as moral imperatives, when this interpretation received the support of a broad audience, and when it referred to relatively circumscribed policy issues, the pressures to comply were strong.

Understanding shame and shaming in politics and policymaking
Shame has generally been interpreted as a negative emotion consisting of an 'intense and painful sensation that is bound up with how the self feels about itself' (Ahmed, 2014: 103). Shame is often considered to belong to an affect family of social self-conscious (or selfdirected) emotions, including emotions such as embarrassment, pride and guilt (Brader & Marcus, 2013;Douglas Creed et al., 2014). The differences between shame and kindred emotions are summarized in Table 1. When policymakers engage in shaming attempts, they do not specify if they target specific behaviors (guilt and remorse) or the whole self (shame). It is also generally not known if the shaming attempts lead to mild feelings in the target or to a substantial negative evaluation of the self. For these reasons, while focusing on shaming attempts, this paper also covers emotions such as guilt and embarrassment.
This diversity of manifestations of shame leads to different interpretations of its function. Shame has often been perceived as crucial for moral development (Ahmed, 2014;Koschut, 2022;Probyn, 2005). Shame is seen as a profound intra-subjective experience that provokes an emotional disturbance, eventually leading to a productive reflection on the self. Shame-closer here to guilt or remorse-encourages individuals to approximate to social ideals and prevents the betrayal of such ideals.
Shame is also related to the concept of stigma and stigma power (Link & Phelan, 2014). Stigma processes have served to achieve the aims of stigmatizers with respect to the exclusion, control or exploitation of others. Individuals socialized through shame and shaming are particularly susceptible to manipulation. They have lost their capacity of discernment and are more likely to accept abusive behavior from, for example, authority figures and hierarchical superiors. 3 Vulnerable individuals may feel shame without being directly responsible for acts that are considered shameful by any given standard (e.g., witnesses and victims of shameful acts).
Studies on shame have been carried out within a diversity of academic disciplines. Within the study of politics, most studies that explicitly discuss shaming attempts have been carried out within the field of IR. The main interest within this field has been on how states and/or policymakers can be shamed into compliance with international norms. Shaming can play a key role in processes of social conformity, since social norms are sustained by feelings of guilt, embarrassment and shame (Finnemore & Sikkink, 1998). 4 Following this logic, many IR scholars have studied shaming dynamics in relation to human rights and peace in world politics (Franklin, 2008;Hafner-Burton, 2008;Lebovic & Voeten, 2006;Peterson et al., 2018;Smidt et al., 2021;Terman & Voeten, 2018;Vadlamannati et al., 2018). Shaming is often presented as a tool for social justice, using strategies such as naming and shaming, sanctions or blacklists (Ilgit & Prakash, 2019;Sanchez Salgado, 2018). Many studies on shaming also focus on the conditions under which shaming attempts are effective (Adkins, 2019;Snyder, 2020). Existing research also shows that shaming processes in political institutions are inherently political and relational and are often mediated by strategic ties between shamer and target (Terman & Voeten, 2018). IR literature on naming and shaming usually focuses on policy outcomes and seldom centers on the emotional component of shame (Ilgit & Prakash, 2019;Snyder, 2020). Shame, in the act of naming and shaming, is assumed quite uncritically, without considering what emotions are really felt and to what extent. 5 Shaming attempts can also be related to broader dynamics of blame avoidance. Since Weaver (1986) introduced the concept of blame avoidance, both credit and blame-related activities have been considered to be key motivating forces for policymakers (Leong & Howlett, 2017). While the study of blame avoidance behavior has become increasingly popular in political science, especially in the domains of social policy and public administration (Hansson, 2018;Hinterleitner, 2017), this focus has left the blame generation side (the shamer and the shaming attempts) unexplored (Johannesson & Weinryb, 2021). From the blame avoidance perspective, blame generation is seen as the result of boundedly rational calculations made by politicians (Weaver, 2018). Shaming attempts could be interpreted as a form of negative messaging used strategically by policymakers to stay in office or to pursue their policy preferences. In the context of blame games, blame makers and blame takers might try to persuade audiences to side with their position (Hansson, 2018;Hinterleitner, 2020). Shaming attempts could, for example, be used within the framework of agency-based blame avoidance strategies such as buck passing, blame shifting and blame diffusing (Hood, 2010). Blame generation can produce multiple policy benefits at multiple stages of the policy process. It can keep issues or options off the agenda (or put them on), help defeat or modify specific alternatives during policy adoption, or alter the course of policy implementation (Weaver, 2018). While there is a clear connection between blame generation and shaming attempts, this literature has not considered how individual strategies could be connected to the broader dynamics of compliance and social conformity.
Shame has played a continuing role in explanations of social conformity within the sociological field of inquiry (Barbalet, 1998), but the emphasis has often been placed on the informal policing and maintenance of group norms by groups with wealth, power and high status (Tisseron, 2006). The literature on this topic, often inspired by Foucault, Bourdieu or Goffman, has investigated the oppression of a variety of groups, including women, minorities and people with disabilities or mental illnesses (Phelan, 2014;Scambler, 2018). Shame serves as a mechanism of surveillance and policing that perpetuates social injustices and inequalities (Shefer & Munt, 2019). Shaming and stigma processes are hidden, indirect and taken for granted. Having said this, emotion codes are not set in stone and, thus, many studies also explore how shaming and stigma processes are contested and reversed.

Different approaches to shaming attempts
Given the divergences in the representation and understanding of shaming in current research, it is important to distinguish between different types of shaming for analytical purposes. Table 2 introduces three main types of shaming attempt: normative shaming, disciplinary shaming and policy shaming.
Normative shaming, based on general moral principles, is presented as apolitical and universal. Actors engaged in this type of shaming attempt present themselves as moral actors pursuing goals such as human dignity or social justice. From a disciplinary perspective, shaming attempts are based on authority and dependency. Social norms and moral authority stem from the perception of social rank and position and, for this reason, they would be primarily deployed by wealthy or high-status individuals. From a policy-centered perspective, loyalty, trust (or a sense of belonging) to your ideological or national group would take precedence over universal values and status to determine what is socially appropriate behavior.
While analyzing shaming attempts or blame generation, the literature has focused on effectiveness (Adkins, 2019) or on the cost of benefits from the perspective of the perpetrator (Weaver, 2018). Taking this literature into account, I argue that, to fully understand processes of social conformity activated by shaming attempts, it is important to consider the differences between types of shaming. By taking into account these nuances, I develop a framework showing the complex web of interrelated factors that can lead to compliance.
When shaming attempts are based on universal norms, the normative-affective pressures to comply will be stronger when the shamer, the shamed and the audience share the same ideas of what is right and wrong (Adkins, 2019). Both publicity and a favorable audience create pressure to conform to norms. From a policy perspective, shaming attempts are also used publicly to persuade audiences to support specific policy preferences (Hansson, 2018). In sharp contrast, disciplinary shaming is usually done in indirect and hidden forms. Rather than relying on publicity, it relies on subtle emotional commands from leaders or peers that are aimed at creating dependence (Tisseron, 2006). 6 Table 2 Approaches to the study of shaming attempts (elaborated by the author) While NGOs present themselves as moral actors, they can also engage in self-serving behavior (e.g., looking for funding to ensure survival, etc.). It is also common that political actors present themselves as moral actors In organizations, shame can be used, for example, to impose the power of leaders or patrons over their subordinates. In the context of the EP, this type of shaming could take place within national delegations or political groups to impose specific disciplinary practices. MEPs in vulnerable positions such as women or representatives from new Member States would be among the most likely targets of processes of stigmatization. The present analysis, focusing on the analysis of the keyword shame, is not the most adapted to identify this type of shaming that could be based on symbols, metaphors and nonverbal communication.
Under non-favorable conditions, shaming attempts lead to shame backlashes (Adkins, 2019;Snyder, 2020). Up until the time of writing, the study of shame backlashes has primarily concerned the normative type of shaming. In a shame backlash, shame is redirected back onto the person who is attempting to shame. Shame backlashes are frequent when the powerless seek political change against the powerful (Adkins, 2019). Shaming is also likely to mobilize backlash when it comes from outsiders and when it is aimed at cultural practices (Snyder, 2020). From a normative perspective, shaming attempts have also been found to fail because audiences are becoming increasingly saturated, manipulated and indifferent to the profit of certain evildoers, who know that they can often proceed with their evil acts without consequences (Keenan, 2004).
In the context of policy shaming and blame games, shaming attempts have been found to lead to alienation (a negative reaction from audience) and retribution (a target may withhold resources or cooperation) (Weaver, 2018). Despite non-favorable conditions from the perspective of compliance, shamers may engage in shaming attempts for a variety of reasons. Policymakers may use shaming attempts (both normative and policy-based) to avoid blame (Weaver, 2018), or to support friends and punish adversaries (Terman & Voeten, 2018). It has also been found that successful non-elite shaming at one point in time can lower the chances of successful shaming in the future (Johannesson & Weinryb, 2021;Terman & Voeten, 2018).
In IR, disciplinary shaming has been analyzed among states: the most powerful and wealthy states can dictate their norms (defined as normality) to the rest of the international community. These processes often involve stigmatization, but the targeted states do not always conform. Counter-stigmatization implies that the stigma is accepted by the shamer, but as an emblem of pride and virtue. In cases of stigma rejection, the stigmatized share the same values and norms of the group, but they deny being different from the norm abiders. Shaming or stigmatization might, however, be successful in cases of shame recognition, when the stigmatized, sharing the same values and norms as the stigmatizer, work to become normal and eventually succeed. Table 3 recapitulates the conditions applicable to the present study on shaming attempts at the EP. Since the present study only covers public emotional expressions, the normative and policy-based types of shaming attempt that take place in a public setting seem most appropriate.

Methodology
In this paper, the study of emotions is carried out through the analysis and interpretation of explicit emotion words present in discourses (Flam, 2015). By combining an interpretative and contextualizing approach, this paper contributes to the generation of knowledge about the emotional underpinnings and implications of discourses (Koschut, 2020). A first descriptive mapping was performed to determine how many times the keyword 'shame' was used by key players across policy areas. The keyword 'shame'-often used in the sense of 'guilt' by key players-was more appropriate for this analysis than other potential keywords such as 'remorse' or 'guilt.' 7 A manual coding of all occurrences of the keyword 'shame' was done using Atlas.ti (including policy area, political affiliation, gender and type of shame).
I expected that, when the word 'shame' was explicitly used in speeches, it would refer mainly to shaming attempts. I did not expect shame (remorse) as a process of self-reflection or recognition of wrongdoings to appear so explicitly in written discourse. The feeling of shame or guilt does not only arise from an evaluation of oneself; it is also possible to experience vicarious shame (Welten et al., 2012). Vicarious shame occurs when people experience shame on behalf of a group they belong to. It can also be felt when people take the perspective of others so they experience the same emotions as though they were in the situation themselves.
Secondly, an in-depth qualitative content analysis served to analyze how and under what circumstances shaming attempts lead to compliance in the EP. The in-depth analysis, combining an interpretative and contextualizing approach, covered policy areas in which the policy-based type of shaming was considered most likely to be used: budget, competition policy and employment and social affairs. 8 Through this choice, I aimed at generating knowledge about dynamics beyond the human rights-based naming and shaming (normative shaming) typically covered by IR studies. The purpose of the qualitative analysis was Table 4 Use of the keyword 'shame' compared to the use of other emotion keywords (elaborated by the author) The numbers indicate the number of speech events (e.g., debates) in which the keywords appeared, but not the number of times the keywords appeared in those speech events, which is usually higher 1994-1999 1999-2014 2004-2009 2009-2014 2014- not only to demonstrate the presence of themes; it was also to articulate the participants' understandings and uses of the keyword 'shame' in light of the literature presented above (Watts, 2014). The circumstances under which the keyword 'shame' was employed were carefully examined to determine how shaming was used by MEPs and EU officials, as well as the context and circumstances under which shaming attempts were conducted, including a discussion on the results in terms of policy choices. Data come from the EP website, including the transcripts from EP plenary sittings that took place from 1994 onward. The analysis included speech events that took place at the EP between 1994 and July 2014. Using the EP search function, it was possible to select all the transcriptions of paragraphs in which the keyword 'shame' was employed. This data collection technique permitted the reduction of a large amount of primary data to a manageable number of speeches and quotations that could be studied in-depth. Translations of EP speech events into English were only available up to 2012 9 (European Parliament, 2012).
While using large amounts of data, this paper is based exclusively on qualitative analysis methods and assumptions . Using the search function as a shortcut means that it is not possible to relate the results obtained to the total universe of cases. For example, the paragraphs collected show how many females used the keyword 'shame' from 1994 to 2014, but the data do not show the total number of females that spoke in the EP during this same period. It is thus not possible to know from these data if women engaged more frequently than men in a discourse in which the keyword 'shame' appeared. The data collected are, however, sufficient for the purpose of this article. To place my findings in a broader perspective, I also used data available from the EP Legislative Observatory. The findings can be compared with previous studies using similar data to analyze the use of generic emotion words in the EP (Sanchez Salgado 2022).

Shame and shaming in the EP
As expected, this first section shows that shaming attempts are much more prevalent in written discourses than other expressions of shame, and that they were mainly used by challengers to the status quo in policy areas such as foreign policy. Table 5 shows that, in most instances (81% out of the total number of times this keyword was used), 10 the keyword 'shame' was used in the context of a shaming attempt. While Commission representatives rarely used the word, a careful reading of the debates shows that the Commission has engaged in shaming indirectly. For example, during the suspension of the Food for Free program, an MEP affirmed that 'The Commissioner is right to shame those Council Members who are incapable of showing solidarity at this time' (Esther Herranz García, EPP, September 28, 2011). This supports existing research, 9 Since the analysis only includes English text, the keyword shame is expected to appear fewer times in the years 2013 and 2014. 10 The reader should consider that the coding of the usages of shame is a subjective process. In many occasions, the usages of the word shame were unclear (coded as other). If other coders had performed the analysis, the results would have probably differed. However, given the results, it is unlikely that inter-coding differences would have altered the main findings.

The prevalence of shaming attempts at the EP
affirming that the Commission tends to avoid explicit emotion-loaded language (Sanchez Salgado, 2021b;Schumacher et al., 2016) and that shaming by individuals with authority (high status) is often done in hidden and indirect ways.
Table 5 also shows that Renew Europe and the Greens used the keyword 'shame' more often than other emotion keywords, indicating that when these groups engage in emotional talk, they frequently use naming and shaming. A shaming strategy also seems to have been used by challengers to the current Parliament organization rules, since MEPs non-attached (NA) to any group frequently used the keyword 'shame,' especially on topics such as the organization of the Union.

Individual and vicarious shame
Key policymakers participating in EU debates used the keyword 'shame' to refer to their own shame (individual or vicarious) on around 90 occasions (4.77% out of the total number of times this keyword was used) (Table 5). This result is congruent with the little usage of the keyword 'remorse' in the EP (only 18 times in more than 20 years). Table 5 also shows that Members of the Commission or the Council, or the President of the EP, almost never talked about their own shame. Only once did Commissioner Emma Bonino (in a 1996 intervention full of passion) use the keyword 'shame' when referring to the shame that all of us should have been feeling due to the lack of EU response to the situation in Eastern Zaire. In this quotation, Commissioner Bonino admitted that she was not using 'diplomatic' language. It is also remarkable that her words were applauded. Talking about shame in a political context where emotional language is not considered appropriate may have been seen as an act of courage and honesty that was appreciated by the audience: Now, apparently, the question is no longer whether we shall have a multinational force, but when. When? This is the real essence of this mission, because if that force arrives in three weeks' time it will be kept very busy burying the bodies, but it will have very little else to do. (Applause) So I hope that, even if some of my language has not been very diplomatic, you will understand the anger and frustration and even shame -yes, the shame, that all of us … (Applause) (Emma Bonino, Commissioner, November 13, 1996. Situation in Eastern Zaire) The keyword 'shame' was mostly used in this individual or vicarious sense by MEPs. MEPs expressed their shame at being identified with EU institutions, Europe, their Member States or region, their political group or, more broadly, the western world or civilization. These MEPs realized that the broad group to which they felt attached was responsible for actions they saw as morally wrong, or which were considered not to be in congruence with their own standards. In most of these cases, vicarious shame was de facto used as a shaming attempt. The following quotation, for example, shows an MEP from the Greens expressing vicarious shame regarding the incapacity of her own national government to make the right decisions: Bearing this in mind, I must also point out that this Greek crisis -this crisis facing the euro -is actually a European crisis and that discussions have been going on for weeks and months without Europeans being able to garner themselves to make the necessary decisions; as far as I am concerned, it is simply shameful. As a German MEP -I hope you are listening, Mr. Langen -I am ashamed of my national government. (Rebecca Harms, Greens, March 24, 2010 Preparation for the European Council meeting debate) MEPs also employed the word 'shame' in cases when they were indirectly identified with the perpetrators of shameful acts, or when the degree of identification was unclear. A few MEPs also expressed their shame and frustration at not being able to do anything to improve specific situations.

Shame across policy areas: the prevalence of foreign policy
The keyword 'shame' was mostly used in connection with foreign policy (Table 6). Data showed that the word 'shame' was more frequently used in this policy area than might have been expected if only considering the level of parliamentary activity (see Table 8 in the appendix). In foreign policy, normative shame was frequent: policymakers frequently made a connection between morality and emotions. The EU was criticized for failing to act in many conflict situations during the time under study, such as in Zaire, Israel/Palestine and ex-Yugoslavia. Policymakers were expected to feel empathy and to take action in cases of disaster and conflict (Sanchez Salgado 2022). Following this same logic, the keyword 'shame' also appeared frequently in connection with home affairs and migration.
Particularly interesting is that, in some policy areas, the keyword 'shame' was more frequently used than other generic keywords referring to emotions (Table 6). This means that explicit talk about 'shame' was more frequent than discourse on emotions more generally. Some of these policy areas related to issues with strong ideological components, such as competition/the single market, and employment and social affairs. Other interesting policy areas in which discussion on 'shame' was more frequent than expected were budget and the organization of the Union.
Finally, in policy areas such as the environment, energy and food safety and public health, 'shame' did not appear very often when compared to general discourse on emotions. This can be explained by the fact that, in these policy areas, it is often argued that emotions should be left out of policymaking processes (Sanchez Salgado 2022). The existence of this so-called feeling rule (which artificially opposes reason/science and emotions) may have prevented key players from engaging in shaming strategies.

How shaming attempts develop in political institutions: the EP
This section proposes an in-depth analysis of shaming attempts in key social and economic policy areas, combining interpretative and contextualizing approaches. It shows, first, how shaming attempts in the EP have often been used to promote specific policy preferences. Second, this section uses illustrative examples to show the circumstances under which shaming attempts have led to pressures to comply, or to shame backlashes or to non-compliance.

How policy preferences shape shaming attempts
While moral standards such as human rights and human dignity are of great importance in foreign policy, data show that, in social and economic policy areas, ideology also plays a key role in determining what should be considered shameful (and why). Political groups engage in shaming dynamics on topics that are relevant in their ideological programs ( Table 7)  As the following quotations show, shaming has mainly been directed toward Member States as a tool to promote specific policy preferences. In these cases, the shaming process is often supported by a sense of belonging to a European community of destiny that is, necessarily, an 'ever closer union.' In the quotations below, when Member States decided not to harmonize certain rules or to maintain national restrictions, the use of shaming attempts allowed these MEPs rhetorically placing European integration at the same level (and with the same consequences in terms of social pressure) as a moral imperative: I am pleased that the Council has recognized that it has to speed through the issue of tractors, because the lack of a harmonized type-approval process for a very important sector is scandalous, and Member States should be ashamed of their foot-dragging on this. (Malcolm Harbour, EPP, December 14, 2005 Machinery) The Commission must be extremely stringent with Member States that choose to maintain transitory restrictions. These will have to be justified on the basis of sound economic data. The alleged vulnerability, or serious disturbances, of national labor markets will have to be proven on the basis of rational figures and will have to be scrutinized on the basis of statistics and facts. If Mr. Barroso wants to relaunch the single market, then it is about time we acted according to our words. It is about time to tear down these shameful walls of economic protectionism and nationalism.

Circumstances under which shaming attempts lead to compliance
When shaming is based on national interests and ideological differences, one might expect much controversy regarding what the object of shame should be. However, in some cases, shaming attempts exerted pressures to comply. In these cases, there was a circumscribed problem (a program to be restored), consensus about the targets of shame and broad support from the audience.
A first example refers to the suspension of the Food for Free program (2011) by a minority of Member States in the Council. This suspension was considered to be shameful, not only by the majority of MEPs, but also by the Presidency of the Council. MEPs and European officials agreed that the shaming should be directed to six selfish EU Member States. EU officials and MEPs shared the same values and views to such an extent that an MEP affirmed '… this is not a debate; it is a succession of monologues all along the same lines. We are all of the same opinion, though there are different views on the methods to be used' (Marc Tarabella, S&D, 2011). In this case, what began as a power dispute between supranational institutions and Member States was successfully re-framed as a normative question related to the values of the Union.
A similar example relates to a debate on the elimination of poverty in October 1996. During this debate, S&D and ALDE MEPs used shaming attempts in reference to the cancelation of the poverty program due to a dispute concerning its legal basis. Commission officials, MEPs and the Irish Presidency of the Council agreed that a legal basis to fight poverty should exist in the EU. Commissioner Flynn, while not using direct or explicit shaming attempts, presented the fight against poverty as more than an economic issue-as a moral imperative. The following is an example of the use of the word 'shame' referring to this program: Consequently, the fight against and the elimination of poverty should be the European Union's prime target and first priority. I fear that it is not. I wish it were, and we are striving to make it so, to rid the European Union of the shame of its refusal to promote programs to combat poverty. There is no place here for any legalistic sterility or bureaucratic or accounting small-mindedness. (Papakyriazis, S&D, October 23, 1996, Elimination of Poverty) In this case, a dispute regarding the attribution of competencies to the EU was also framed as a normative issue, which led to a consensus about the targets of shame: Germany and the UK, the two Member States blocking the program and which were reluctant to adopt a legal basis for the fight against poverty. The only German EPP MEP participating in this discussion was rather ambiguous: he seemed to agree on the need to fight poverty at the EU level and praised the suspended programs, but he also mentioned obstacles and reservations. In my interpretation, his initial position seems to have been nuanced by the existing normative pressures: I doubt whether the countries in question will vote for the proposal at the Intergovernmental Conference. Another factor is that definitions of the term poverty vary from country to country. Some countries also have reservations, considerable reservations in some cases, about the so-called alternatives or the second labor market. (…) The Council and Commission must reach an early agreement to enable them to provide rapid, unbureaucratic assistance in the future. It is irrelevant here whether the main focus is at national or European level, or whether Europe provides the framework and the Member States coordinate the individual initiatives (Shiedermeier, EPP, October 23, 1996, Elimination of Poverty).

How shaming attempts led to shame backlashes
Shame backlashes at the EP have taken different forms. First, shame backlashes present the same issue from a different angle or perspective. The new perspective indicates that the shamer should have been ashamed of having used a shaming attempt. EP debates on the Working Time Directive illustrate this use of shaming attempts and shame backlashes in relation to the controversy around the Working Time Directive, as a manifestation of the conflict between a neo-liberal vision and a center-left alternative ('Debates,' 2013). In their shaming attempts, S&D and left group MEPs consistently argued that forcing people to work more than 48 h per week represented a return to the nineteenth century in terms of workers' rights. In the following quotation, the MEP pictured her ideological preferences as a question of human rights and human dignity (moral appeals), presenting workers as people (not machines) and rejecting slavery: This is one of the most blatant examples of capitalist exploitation and threatens everything that has been said about reconciling work and family life. This proposal represents a backward step of nearly 100 years in the hard-won rights of workers, who are people, not machines. We therefore support the rejection of this shameful Council position and call on Members, in their voting, to listen to the protests of workers and prevent more serious social tensions, more backward steps and a return to a kind of slavery right now in the 21st century. In a backlash shaming attempt, an EPP MEP also engaged in a humanizing speech (focusing on families and children) based on different moral appeals, emphasizing individual freedom and the avoidance of restrictions: Let me tell you a story. My father was a bus driver and whenever we got an unexpected bill, whenever there was a school trip that he wanted to send me on, he would put in a few hours' extra overtime just so he could pay that extra bill or he could send me on a school trip. If the Working Time Directive had been around, none of this would have been possible. No one should be forced into working extra hours against their will. I think we all agree on that, no matter where we sit in this Parliament, but if you look at the impact of what we voted on today, this is a kick in the teeth for working men and women who wish to work a few extra hours to give their families a better life. Shame on the Socialists! (Syed Kamall, EPP, December 17, 2008 Explanations of Vote) Second, shame backlashes also consisted of bringing up a different topic in which the EU (or the shamer) was not up to the required standard, in a process of buck passing, blame diffusing or blame shifting (Hood, 2010). For example, the rapporteur for the organization of the raw tobacco market (UPE) claimed that economic support to tobacco growers should not be the object of shame or shaming. According to this MEP, the object of shame should instead be EU corruption and fraud: It is true, according to information provided by some of the Commissioner's emissaries and ayatollahs to the Committee on Budgets, that it costs 1,000 million ECU to support the tobacco growers. It is true! There is no need to conceal it or be ashamed! I am fed up with all the shame felt in Europe over spending money on people and the money not spent on fighting corruption. I do not see the same concern with fighting fraud or the billions of ECU of unpaid VAT, or the harmonization of tax systems wallowing in a criminal system in which violators are punished but where violations are paid for at bargain basement prices in today's Europe. (Rosado Fernandes, UPE, July 17, 1997, Common Organization of the Raw Tobacco Market Debate) Placing the emphasis on financial misconduct is seen as a type of scandal talk with the capacity to undermine trust in individuals (Hansson, 2018). This shame backlash could have undermined the audience's trust in the MEPs who were worried about the effects of tobacco on health. In the case of the organization of the raw tobacco market, MEPs using public health arguments did not use the word 'shame' or 'shaming,' but they engaged in shaming without using the word explicitly. They presented the growing of tobacco as a moral issue, pointing out what they saw as the hypocrisy of the EU in this domain: Mr. President, Parliament has made a decision to ban tobacco advertising. … We also give support to research to combat the harmful effects of tobacco on health. However, the biggest support we give is to the growing of tobacco, a crop which claims hundreds of thousands of lives every year in the Union. Throughout the world, many more people die from tobacco than from Aids. That does not make sense. The subsidies cannot be justified on the grounds of employment or on the grounds that someone else would have produced the tobacco. By the same logic we could just as easily subsidize the production of nuclear weapons, or of biological and chemical weapons. The fundamental issue is not an economic, but a moral one. The Union's citizens do not want to finance the growing of this harmful crop. (Wibe, S&D, Common Organization of the Market in Raw Tobacco, June 16, 1998) These examples show that shaming attempts and shame backlashes have been used within the framework of a blame game in which key players try to persuade audiences to side with their position (Hansson, 2018). Since there was much at stake in the cases referred to above policy opponents might have preferred to use strategically shaming attempts (based on alternative moral appeals) than to keep a low profile. Even if the chances of success are low, shaming attempts have also been used in cases where perpetrators do not perceive alternatives for achieving their objectives (Weaver, 2018). The existence of competing audiences, including active stakeholders and lobbyists, may also have helped introduce a variety of competing moral appeals for MEPs to use into public debate. In the absence of a moral consensus on a topic, EU decisions have been contradictory or ambiguous. The EU was both subsidizing the production of tobacco and including restrictions to avoid its consumption. Additionally, the Working Time Directive was not as strong as it could have been since it included op-outs to the 48-h limit and derogations. According to the European Trade Union Confederation (ETUC), the Directive included sufficient loopholes that it might not be properly transposed and enforced in some Member States ('Working Time' n.d.).

Circumstances under which shaming attempts lead to non-compliance
Even in the absence of a moral consensus and of shame backlashes, shaming attempts may not necessarily lead to compliance. One of the most interesting trends to emerge from the data analysis was the extensive number of quotations that referred to debates discussing the report of the European Court of Auditors (ECA). Part of the work of accountability institutions such as this Court involves shaming Member States when their behavior does not correspond to international standards. However, as shown in the example below, the ECA's shaming attempts to improve EU accounts were constrained by adverse circumstances.
The ECA regularly signs off the reliability of the EU accounts. As a professional audit body, account-holding is its core task. Following its instauration in 1995, the ECA refused to offer a judgment on EU finance until 2007. From 1995 to 2015, the ECA gave an adverse opinion regarding the existence of material errors in payments. 11 Most errors were found in budget items administrated by Members States, such as agricultural aid and structural funds. 12 For its shaming attempts to lead to compliance, the ECA would need a certain level of authority and trust (Adkins, 2019). In this case, the ECA was perceived (or pictured) as a political actor trying to promote EU integration, rather than as an independent institution aiming at improving the quality of EU accounts. There was controversy over what the role of the EU should be in controlling Member States' use of EU funds. Instead of inspiring trust, the ECA was suspected to act as a political entrepreneur, using the signaling of errors strategically (Sánchez-Barrueco, 2015). National authorities did not necessarily feel ashamed by the adverse opinions because they distrusted the audits made by the Commission and the ECA. According to Sánchez-Barrueco (2015), EU funds have also often been seen as a sort of manna that Member States should take advantage of, even if in doing so EU rules are not 100% respected.
The ECA's lack of visibility (Tidå, 2021) also helps explain its lack of success in its shaming attempts. The ECA regularly presented shameful numbers and conclusions using technical and obscure language detached from any emotion, which had no impact on a broader audience. MEPs admitted there was a problem with EU accounts but, while the problem remained circumscribed to the EU elite, no substantive policy changes were made. The ECA and mainstream MEPs were rather lenient with the Commission. This lenience was justified by the complexity of the task and the good disposition of Commission officials. From their perspective of supranational actors, it was more natural to direct the blame and shame toward Member States and the Council of Ministers, as in the following example: The Member States avoid responsibility, misappropriate funds and do nothing whatsoever to exercise control and, every time criticisms are made, a storm of criticism rains down upon those in Brussels, that is to say the Commission and others. I think the Member States should be ashamed of themselves. In 1995, a convention to protect the EU's economic interests was adopted. There is only one Member State which has implemented this in its legislation. (Blak, S&D Group, 2000) The situation only changed when Eurosceptic groups started to engage a broader audience. The changes that finally took place seem to have been motivated by fear of losing support in the context of this increasing Euroscepticism. In the early 2000s, populist parties and Eurosceptics used the reports of the ECA in their broader shaming strategies, exaggerating facts or citing them out of context to create a dramatization effect: Mr. President, I find it very difficult to take seriously a debate about voluminous documents that only a real euro-nerd would be able to read, let alone understand. Furthermore, what is the point of a debate about a 2006 budget which we all know will bear no relationship to the way the money will have been spent? The European Court of Auditors has, for good reason, refused to sign off the EU's accounts for the last 10 years and said, in its last report, that 95% of the EU's budget was open to fraud. Yet every year this House just shrugs its shoulders, sighs and looks the other way. That is a shameful indictment of this institution and demonstrates how pitifully useless it really is. (Jeffrey Titford, IND/DEM, 2005) As this example shows, while everyone agreed there were problems with EU accounts (and that this was morally unacceptable), the shaming attempts were aimed at harming political foes, while being lenient to political allies (Terman & Voeten, 2018). While mainstream MEPs blamed the Council and Member States, populist groups and Eurosceptics primarily shamed the Commission and supranational institutions, picturing them as failing to perform essential controls, wasting money and neglecting the real needs of the people.

Conclusion
This paper has shown that shaming attempts are common currency in European policymaking, contributing to emerging research on emotions in European politics. Combining studies in IR (naming and shaming) and in public administration (blame generation) has offered a more complete framework for analyzing the circumstances under which shaming attempts lead to compliance. From the perspective of blame avoidance, studies on the blame generation side would benefit from considering a sociological perspective in which normative pressures to comply play a relevant role. Regarding IR literature on naming and shaming, I have shown that political preferences and strategic reframing by key players are of utmost importance for understanding what is perceived as shameful. Within the EU context, this paper has shown that the sense of belonging at different levels of governance (Member States versus EU) can be crucial. This paper has also shown the circumstances under which shaming attempts have led to compliance, or to non-compliance or shame backlashes. In the illustrative examples studied, shaming attempts led to compliance in circumscribed cases where there was a large consensus regarding moral values and a favorable audience. In more controversial and politically relevant cases, shame backlashes were more frequent. Further analysis, including interviews, would be necessary to explore the reasons why politicians comply: politicians might feel ashamed and be persuaded by normative arguments, but they might also engage in strategic calculation, preferring to keep a low profile in adverse circumstances.
Even when MEPs agree on the existence of a problem, shaming attempts may not necessarily lead to compliance. The example of the ECA showed that, despite acknowledging shameful figures in EU accounts, there was little progress for many years. Since political actors tend to target their adversaries and spare their friends in their shaming attempts, nobody felt responsible for the situation. Action was postponed (or superficial) in the absence of critical audiences.
Further research could also pay more attention to how shamers emphasize or modify certain aspects of social reality through linguistically informed discourse analysis (Hansson, 2018;Koschut, 2018). Such analysis could uncover important aspects of politics and policymaking, such as the understanding of processes of institutional stability and change (Douglas Creed et al., 2014). Further research based on interviews or on observation might also focus more specifically on processes of stigmatization directed toward potentially vulnerable policymakers (Adler-Nissen, 2014;Scambler, 2018). In these cases, the circumstances leading to compliance may be very different from those discussed here. Shaming could also play a key role in the dissemination of rules that have been central in guaranteeing the stability of the EU institutions, such as the culture of consensus.
NA (Non-attached members). TDI (Technical Group of Independent). UEN (Union for the Europe of Nations). EDN (Independents for a Europe of Nations). ERA (European Radical Alliance).