Ostracism and the EU’s contradictory approach to sanctions at home and abroad

ABSTRACT The European Union (EU) is one of the world’s most active imposers of foreign policy sanctions. By contrast, the EU has never used formal political sanctions against a member. Why does the EU favour sanctions as an instrument to deal with norm violations abroad, but not at home? The article argues that an understanding of sanctions as ostracism helps illuminate this discrepancy. Far from technical instruments that can be made to ‘work’ through improved design, sanctions are social instruments that operate through selective exclusion. An in-depth scrutiny of public justifications from the European institutions and individual politicians shows that both at home and abroad sanctions are strongly associated with ostracising attributes. However, whereas practical and symbolic distance-taking from the target is core for foreign policy sanctions, at home, the same ostracising properties run against the EU’s traditional insistence on resolving disagreements through rational dialogue.


Introduction
European integration has transformed sovereignty. National parliaments are occupied with turning EU-directives into national legislation, and member states are taken to court if they violate EU law. The members of the monetary union have delegated fiscal sovereignty, and the austerity paradigm of the euro crisis is based on invasive negative conditionality. Meanwhile, the EU struggles to establish itself as an international actor, and is often accused of doing too little, too late, and with lacking unity.
Yet, when it comes to sanctions against violations of fundamental norms, the EU is extraordinarily active abroad and reluctant at home. With measures in place against 33 different countries (see author's table in appendix 2), the EU is among the world's most active imposers of foreign policy sanctions (Portela, 2016). By contrast, despite evidence of value backsliding in member states, the EU has never used its internal sanctions provision, the Article 7:3 of the Treaty on European Union (TEU) (European Union, 2007/ 2009a). This pattern is unique among regional organisations (ROs). Other ROs have occasionally used sanctions outside of the region, but only the EU uses foreign policy Timmermanns, Sövndal, and Tuomioja (2013) emphasised that '[t]he credibility of the European project depends on us living up to the standards we have given ourselves'. In his 2013 State of the Union address, Barroso (2013) himself declared: 'There is no doubt about it. Our internal coherence and international relevance are inextricably linked'. His successor, Jean-Claude Juncker (2014), likewise expressed that 'we are credible to the outside world if we demand high standards of ourselves when it comes to fundamental values'. As a recurrent theme, CFSP High Representative Federica Mogherini speaks of '[a] world where internal and external policies are more intimately linked than ever' (2015e).
Yet, scholars have so far treated internal and external sanctions as different creatures (see, though, Börzel & van Hüllen, 2015). The literature on foreign policy sanctions has mainly been interested in issues of selection, impact, and effectiveness. Central contributions focus on why sanctions are used in some cases but not in others (e.g. Boogaerts, 2016;Hazelzet, 2001), on why sanctions became the EU's common response to the Russia-Ukraine crises (Sjursen & Rosén, 2017), on the structural environment in which the common sanctions policy emerged (Jones, 2007), on conditions for effectiveness (Portela, 2010), on coercion, constraint and signalling as means of influencing the target (Giumelli, 2011), on the transformation of sanctions through 'targeting' (Eriksson, 2011), and on human rights consequences for targets (e.g. Heupel & Zürn, 2017). Whereas this literature has advanced our understanding of the causes and consequences of the EU's turn to sanctions, until now few have engaged with the fundamental legitimacy andby extensionappropriateness of the EU's sanctions policy at large. This is surprising considering that, in contrast to other types of conditionality or countermeasures, foreign policy sanctions lack contractual basis, have unclear support in international law (see Doxey, 1980;White & Abass, 2010) and face growing opposition from a clear majority of the world's countries (see OHCHR, 2017 and linked UN resolutions).
That the EU has not imposed sanctions against Hungary and Poland has been much debated, first in blogs (e.g. Verfassungsblog) and eventually in academic publications (e.g. Closa & Kochenov, 2016;Kelemen & Blauberger, 2017;Kochenov, Magen, & Pech, 2016). We have now excellent insights into the Polish and Hungarian situations, as well as into the party-politics of sanctions at the European Parliament (Sedelmeier, 2014(Sedelmeier, , 2017. However, the reluctance to use sanctions at home has mostly been taken as a given premise rather than as something to be explained. Largely accepting policymakers' description that institutional requirements make article 7 sanctions unusable (e.g. Kelemen, 2017, p. 230;Scheppele, 2016a, p. 106), scholars have instead drafted proposals for alternative procedural strategies (see contributors in Closa & Kochenov, 2016;Müller, 2013).
The thresholds for imposing internal sanctions are, indeed, demanding. The European Council has full discretion in unanimously deciding when 'a serious and persistent breach' of Article 2 norms has occurred (Article 7.2 TEU). This step may be initiated by the Commission or one third of member states, and the ensuing decision requires consent by the European Parliament. Qualified majority in the European Council is thereafter needed to 'suspend certain of the rights deriving from the application of the Treaties to the Member State in question, including the voting rights of the representative of the government of that Member State in the Council' (7.3). 1 However, the institutional factor cannot explain the discrepancy in the EU's approach to sanctions at home and abroad. The threshold for CFSP sanctions to be imposed is also high, with first framework decisions requiring unanimity in the Council (Article 31:1 TEU). 2 Council Decisions contain all measures in a sanctions regime, and are sufficient for travel bans and arms embargoes (which are implemented directly by member states). Supplementary implementing legislation through Council Regulations (qualified majority, upon proposal by High Representative and the Commission) is needed for trade and financial measures (Art 215 TFEU, European Union, 2007/2009b. The crucial commonality is that for sanctions to emerge all member states must agree. In the case of internal sanctions, the potentially targeted country is excluded (Article 354 TFEU). The currently most discussed potential targets Poland and Hungary are in a 'loyalty pact' which would bloc sanctions against either. However, it has been proposed that this 'fellow-traveller veto' could be overcome by acting simultaneously against both countries (Scheppele, 2016b). Moreover, such pacts could be envisaged also for CFSP sanctions, where interests often diverge between member states.
Complementing the focus of the established literatures on effectiveness (abroad) and procedures (at home), this study is interested in the intersubjective meaning of sanctions. It speaks in at least three ways to scholarship on sanctions as well as on European integration and the EU as an international actor. First, the argument about ostracism shows the limits of rationalistic models for grasping the meaning of sanctions in international and regional relations. The basic understanding of sanctions as primarily social instruments contrasts with the tendency to treat them as technical instruments that can be made to 'work' through more sophisticated design and strict implementation. Second, the study of public justifications offers an unusual engagement with reasonable grounds for using sanctions in different contexts. The way that the EU assumes a right to use foreign policy sanctions contrasts with the centrality of legal and social legitimacy for arguments about the Article 7. Third, the article makes an empirical contribution to the concurrently topical and classical debate of how to ensure normative discipline within the Union, and how to extend normative loyalty into 'the world outside' (Diez, 2005, p. 627). The discursive framing of the Article 7 as a 'nuclear option' for insiders, next to CFSP sanctions' dialectical play with normality and punishment for outsiders, show how basic categories of belongingrather than effectiveness considerations or institutional prerequisitesdetermine policy.

Ostracism and sanctions
This article argues that if we are to understand the EU's contradictory approach to sanctions at home and abroad, we need to acknowledge that sanctions are primarily social instruments which work through selective exclusion from community. The ostracising properties of sanctions are appealing to the EU as an international actor claiming a high normative posture, while it does not fit with its internal self-understanding as a group of (normative) equals where disagreements are resolved through rational dialogue.
A bundle of concrete measures, from selective economic embargoes to suspension of membership (or other institutionalised) rights, to travel bans and asset freezes against individuals, organisations, and companies, fit under the umbrella term of 'sanctions'. The common denominator of these measures is that they combine an explicit message of disapproval with an actual removal of something considered valuable to the target. It is this characteristic combination of symbolic and material attributes within an institutional framework that distinguishes sanctions from other tools of influence (e.g. diplomatic demarches, 'naming and shaming', litigation), making them outstanding carriers of ostracism.
Ostracism -'the social death penalty' (Ouwerkerk, Kerr, Gallucci, & Van Lange, 2005)is about rejection from a community. At the extreme, sanctioning 'embodies judgment and seeks to associate the target with all that is reprehensible in the human character and human history', thereby excluding it 'from the circle of civilized society' (Schulz, 2015, p. 39). Not only does ostracism influence relations between countries and other collective actors, through the symbolic establishment of who is an insider and outsider in the international or a regional community; ostracism through sanctions is often literal. The expulsion of diplomats, a classical, often reciprocal, sanction in foreign policy disputes, amounts to the physical rejection of individuals from the territory. Similarly, travel bans prohibit individuals (and legal entities) from entering the sender's territory. The 'lifeline' of an ostracised individual is 'severed' (Case & Williams, 2004, p. 337), in the case of targeted sanctions by depriving access to social benefits and rendering economic participation in modern society impossible (De Goede, 2011). Hence, De Goede (2011, p. 500) sees blacklisting 'in terms of its symbolic function of banishment and exclusion, which simultaneously redraws the boundaries around normal, valued, ways of life'.
Within regional organisations, suspension of membership or rights associated with membership is the most used formal sanction. 3 Excluding a member from meetings, cancelling voting rights, cutting funds or in other ways putting a halt to political conversation or normal modes of conduct means to cut the member of a community away from its social belonging and privileges associated with it (see Hellquist, 2015, p. 327). If ostracism in the case of foreign policy sanctions is about determining the boundaries of international community in an abstract sense, and about access to Europe in a practical sense, in the case of internal EU sanctions it is about exclusion from the regional community and from practices of European cooperation. Beyond the reputational damage, a member state under sanctions could face considerable material costs (e.g. market reaction, decline in tourism and investments). Testimonies from other regions tell us that also in the case of sanctions against members, ostracism is 'not only' symbolic but amounts to physical exclusion. To give one example, in December 2016 Venezuelan foreign minister Delcy Rodriguez was physically hindered to enter Mercosur's summit meeting in Buenos Aires since her country's membership had been suspended (BBC, 2016).
Sanctions represent a rupture with 'normal' (or 'customary', Hufbauer, Schott, & Elliott, 1990, p. 2) affairs (see Baldwin, 1985, p. 59). It is through this rupture that the target is told that it has crossed a line. One way of thinking of this process is that the sender draws its 'limits of toleration' by using sanctions (Rawls, 1993;see Adler-Nissen, 2014, pp. 170-171;Hoffmann, 1967, p. 144). Foreign policy sanctions reply to appalling wrongdoings which are intimately connected to broader international norms (see Doxey, 1980, p. 9). Sanctions articulate opposition to these wrongdoings and thereby become formalised counterimages to the sender's norm-obedience (see Hellquist, 2016).
That the EU has not employed sanctions under Article 7:3 does not imply that there are no limits to its toleration at home. The death penalty is a declared red line -'incompatible with the membership of the European Union'which would justify sanctions (Timmermans, 2015b). Moreover, in 2000, 14 EU members unanimously judged unacceptable that the Freedom Party of Austria (FPÖ), a populist-right party with Nazi-roots (the EU's historical other -'past as other ' -Diez, 2005, p. 634)entered government in Austria. At the time, there was no Treaty provision covering situations of not yet materialised risks of value breaches. Instead, unique bilateral sanctions from all co-members followed, which shut Austria off Council cooperation.
Why then does the EU not use sanctions against Poland or Hungary, despite clearly documented breaches of common values and the availability of a designated Treaty clause, if it did so against Austria for letting a democratically elected party into government? The answer is twofold. First, as suggested above, the limits of toleration have changed. Far-right parties have become normalised, making it into government in all Visegrad countries (Czech Republic, Hungary, Poland and Slovakia), as well as in Italy, Greece, Bulgaria, Slovenia, Latvia, Lithuania, Finland, andagain -Austria, and acting as official support parties in the Netherlands and Denmark. None of these situations have produced a reaction similar to the 2000 sanctions against Austria. Although the populist far-right parties remain a headache for the European political establishment, they are nowadays largely tolerated.
Second, the Austrian sanctions case illustrates the difficulties of using ostracism within the European Union. Pushed for by the French and Belgian governments, the bilateral sanctions were likely motivated by a wish to contain far-right populist parties in their own countries (Merlingen, Mudde, & Sedelmeier, 2001). However, the warning example backfired, catering a convenient victim-role to the FPÖ and exposing the shaky legitimacy basis of the EU's action. According to Wodak and Pelinka (2002, p. 204), sanctions were 'nothing other than a sleight of hand clearly intended to convey the impression of EU legitimacy where none exists'. '[T]he Austrian government never accepted being labeled as transgressive', instead it shifted the debate to 'how normative discipline should be upheld in Europe' (Adler-Nissen, 2014, p. 160). As we will see in the following sections, this debate has been reopened and intensified through the recent rule of law crises among EU members. In foreign policy, by contrast, sanctions have become a near-routinised reaction against the worst normative breaches.

Justifying sanctions at home and abroad
Following the Weberian legacy of verstehen, and Habermas (1996) insistence on reflexive justification as a building-block of legitimate order, paying attention to public justifications 'as a form of intersubjective reasoning' is 'a key piece in the explanatory puzzle of our socio-political universe' (Abulof & Kornprobst, 2017, p. 4). Of interest here are public justifications -'explicit communicative exchanges about good reasons upon which to act' (Kornprobst, 2014, p. 192), which respond to '"the basic legitimation demand" (Williams, 2005, pp Abulof & Kornprobst, 2017, p. 3). Political statements are influenced by the form of the communication, and we can never know if they truthfully represent inner convictions. However, whether 'reasonable' (Boltanski & Thévenot, 2000, p. 228) and sincere or not, public justifications are relevant as political acts in their own capacity (Hurd, 1999, p. 391;Kratochwil, 1989).
In the absence of explicit justifications of the dual approach to sanctions, the following compares justifications of foreign policy sanctions with justifications of the non-use of Article 7.3 sanctions. Two main types of documents have been consulted: official documents from institutions and speeches and debate appearances by individual policymakers.
Official documents represent 'convergences of reasons' (Kornprobst, 2014, p. 197); the institutional acceptance of a dominant argument about a political standpoint or action. Justifications articulated by individual policymakers in front of an audience, or in engagement with other policymakers, let us into the 'communicative process of legitimacymaking (and unmaking) in the public sphere' (Abulof & Kornprobst, 2017, p. 4). Hence, to study justifications is to study the use of language for political purposes. For the CFSP, the empirical analysis of this article demonstrates how a space for political action is created through the use of attributive words presenting sanctions as necessary, and a causal argumentation in which they are the logical consequence of the target's wrongdoing. For the article 7, the study shows how the nuclear metaphor ('nuclear option') effectively closes the prospective space for political action through sanctions. Instead, the EU's approach to the value crises becomes defined in contradistinction to sanctions, emphasising attributes of dialogue, objectivity, and rationality.
For foreign policy sanctions, all Council decisions, conclusions, and press releases about the 31 sanctions cases in force as of January 2017 4 (European Commission, 2017a) have been analysed. These documents are particularly important since they constitute, or deal directly with, actual decisions on sanctions. Core strategic documents on sanctions and the CFSP provide a picture of how the policy, in general, is justified. Moreover, public appearances by the High Representative of the CFSP Federica Mogherini (2015Mogherini ( -2017, offer insights into the reasoning of a key player in defining the sanctions policy. European Parliament debates mentioning foreign policy sanctions between 1999 and 2012 (statements n = 1748) bring an additional comprehensive picture of contestation over justifications (translated data is only available until December 2012).
Since the Article 7 is a 'negative case' of a not materialised policy, there is no available data equivalent to the Council communications analysed for CFSP sanctions. Instead, justifications have been identified in milestone communications from the Commission, the Council, and the European Parliament on the Article 7 in general, as well as on the rule of law crisis in Poland in particular. Moreover, all speeches and debate appearances (2014-2017) by Vice-president of the Commission Frans Timmermans have been analysed. Timmermans has been the main voice of the Commission's engagement strategies for dealing with the situation in Poland. In May 2017, the Polish Foreign Minister even boiled down the conflict at hand to 'tensions between Poland and one commissioner ' -Frans Timmermans (in Karnitschnig, 2017).
Next, to these principal sources, the analysis also makes use of individual public appearances by other key policymakers from European institutions and member states. Media reports appear as complementary sources. Unless otherwise specified, the findings represent recurrent patterns of reasoning. All sources quoted in the text are listed in the reference list (Council of the European Union n = 25; European Council n = 6; European Commission n = 12; European Parliament n = 5; Mogherini n = 14; Timmermans n = 11; Reding n = 4; Juncker n = 4; Macron n = 2; Merkel n = 2; Tusk n = 2; Barroso n = 2; Van Rompuy n = 1). 5 Drawing on the earlier discussion of ostracism, the following investigates whether and how three significant traits of ostracism emerge in justifications of (non)-use of sanctions. First, ostracism is indicated by symbolic and material distance-taking from the target's wrongdoing. Second, the concrete expression of this distance-takingsocial exclusion, predominantly rupture with normal terms of conductis taken to signify ostracism.
Third, in a framework of ostracism, the impact of sanctions in terms of inflicting harm is prioritised over their effectiveness in terms of changing the target. E contrario, justifications of non-use of sanctions would highlight these aspects as negative factors.

EU foreign policy sanctions
In its communications on decisions, the Council justifies its sanctions as necessary, inevitable reactions to intolerable wrongdoings (e.g. Council of the European Union, 2002European Council, 2000;2010, p. 14;Merkel, 2014b). It is in view of the severity of the norm violation that sanctions are necessary: the EU 'take[s] the necessary measures' (Council of the European Union, 2004), 'the necessary action' (European Council, 2004), 'necessary decisions' (European Council, 2004 'as a matter of urgency' (Council of the European Union, 2014b), 'as needed' (Council of the European Union, 2015b, p. 242). The policy is tied to the norm violations themselves (ex ante), rather than to expected policy results (ex post).
Hence, the norm violationsfrom nuclear proliferation to misappropriation of state funds, from human rights crimes to fraudulent elections, from (financing of) terrorism activities to breaches of international law (Council of the European Union, 2011a, 2011b, 2012b, 2014d)are at the core of the EU's justifications. The Council routinely starts with condemning, often in emotional terms (it is 'appalled and deeply dismayed', 2014b, 'disappointed', 'extremely worried' 2011d, or 'alarmed' 2012a, 2014e), the acts committed by the target (see Council of the European Union, 2011b, 2014cEuropean Council, 1989). The decision to impose (or modify) sanctions is presented as the logical continuation of the wrongdoing: to sanction is what the EU does under such circumstances. This is a 'causal argumentation' (Garssen, 2001, p. 92) where the link between the causethe norm violation, and the effectthe sanction, is provided by a causal conjunction (e.g. 'therefore', 'because of', 'in view of', 'given') or a causative verb (e.g. 'led to', 'prompted', 'respond to').
In a universe of almost endless norm violations, the inclusion of some violations under sanctions is justified with them being particularly 'serious' (European Commission, 2011, p. 16). Commission President Juncker highlights that Russia has surpassed the EU's limits of toleration: '[i]t is not acceptable for President Putin to redraw borders in Europe. That is why we imposed sanctions, which are not the smartest form of political reaction' (Juncker, 2015). By using sanctions, the EU shows that it 'stands ready to support Ukraine in more or less all the matters' (Juncker, 2016). Despite conflicting interests, 'noone here, I deeply believe, will ever accept that a state invades and occupies the territory of another state or kills its citizens' (Tusk, 2015). Mogherini (2015c) likewise links the 'impossibility to divide us' on Russia to 'a principle that we cannot accept the violation of international rules': 'the basic core principle of the European Union'. The norm violation obliges the EU's reaction: 'You do not change borders by force: not in this continent, not in this century, not in this millennium. You do not do it, and we will never accept anything like that' (Mogherini, 2015a). Sanctions remain in place since 'we cannot pretend that there was nothing wrong going on in 2014' [the annexation of Crimea] (Mogherini, 2017a).
With unacceptability as premise and unity as method, sanctions cut the target's access to some aspect of friendly, cooperative relations with the EU. Norm-breaking behaviour disqualifies the culpable from the EU's preferred approach of 'win-win'-solutions, partnership, and dialogue (e.g. Mogherini, 2015f, 2016aMogherini, 2015f, , 2016bMogherini, 2015f, , 2016c. To have a strategic partnership with Russia would be 'quite surreal' as long as there are sanctions: '[t]his is not the state of play that partners normally have' (Mogherini, 2017a). The severity of the norm violation precludes 'business as usual' (Council of the European Union, 2015a, p. 242;Mogherini, 2015b;Merkel, 2014a). Similarly, Mogherini (2015d) speaks of how cooperation with Belarus in the Eastern Partnership framework was 'put on the back burner' following sanctions against post-electoral violence in 2010.
When the EU imposes sanctions, it defines with one voice what is unacceptable, and thereby establishes its normative difference to the target. The EU's foundational identity as a 'community of values' (Art. 2) is the declared basis of its foreign policy (Art. 21): 'The Union's action on the international scene shall be guided by the principles which have inspired its own creation, development and enlargement.' The EU 'act[s] abroad because our interests at homesecurity, prosperity, and the values on which we have built the European Uniondemand it' (Council of the European Union, 2011c, p. 10). At the European Parliament, MEP Kamiński (2006) supports sanctions since 'as a community of values […] the Union must set an unequivocal example by condemning all infringements of human rights'. MEP Korhola (2008) sees sanctions as 'a moral message from the European community of values' and MEP Van den Bos (2002) as one cure against 'powerlessness' for 'Europe, the mother of moral values'. Commissioner Kovàcs (2006) declares that as a 'community of shared values', the EU 'has to make it explicitly clear that we do not accept the repressive regime of President Lukashenko', and therefore 'introduce and maintain sanctions'. Similarly, according to MEP Brok (2000), sanctions are a way for the EU to 'prove itself to be a community of values'. MEP Kelam (2011) thinks that the EU should be ready to 'cause pain' through 'real sanctions', since 'it is our joint responsibility to make authoritarian rulers believe that we take our basic values seriously'.
EU justifications demonstrate that ostracism operates in a dialectic between the exclusionary measure and 'the normal'. The horizon of engagement, of readmittance to normal relations and escape from the stigma of the target, is the necessary contrast to the relational sanction. This corresponds to what Mogherini calls the EU's 'integrated approach' (e.g. 2016d), its 'unique' (e.g. 2016e) mixing of tools of hard and soft power. Indeed, the lifting of sanctions reopens political, economic, and physical access to Europe. The nuclear deal with Iran has 'a positive impact on trade and economic relations with Iran including benefits for the Iranian people. It strengthens cooperation and allows for continuous dialogue with Iran' (Council of the European Union, 2017b).
Despite repeated assurances from Mogherini (e.g. 2017b and 2018) and other key actors (e.g. Merkel, 2014b;Van Rompuy, 2014) that 'sanctions are a tool and not a goal in themselves', justifications pertaining to expectations of how the target will react are strikingly absent. When sanctions are found to not be 'working', the standard conclusion is that more sanctions are needed: 'the argument of the "ineffectiveness" of sanctions cannot be used in favour of lifting them' (European Parliament, 2008, §11, p. 10, 19). As a dominant pattern, the Council declares its incessant readiness to strengthen, toughen, reinforce, or tighten (to give just a few examples of expressions used) measures. With the ineffectiveness of the medicine justificatory ground for increasing the dosage, the treatment is often protracted. As of 2018-11-01, CFSP sanctions cases in force have on average been in place for 12 years and 10 months. 6 This is where the story of the ostracism of CFPS sanctions comes full circle. EU sanctions often have a far-reaching harmful impact on targeted individuals, sectors, and societies (see Dreyer et al., 2015;Giumelli & Ivan, 2013). Arguments for sanctions are oriented towards realising this harm to the fullest by means of cutting access to normal interactions with the EU. Ostracism does not anticipate the target's reaction, it imposes a consequence connected to the committed crime.

Sanctions against members
Two intertwined notions are central for understanding the peculiar status of the Article 7 in EU affairs: the 'Copenhagen dilemma' and the 'nuclear option'. When a chain of rule of law crises implode around 2011 (Hungary 2011cont., Romania 2012, Poland 2015cont.) key policymakers are frustrated. They start to speak of a gap between the high standards of the Copenhagen criteria -'the essential conditions all candidate countries must satisfy to become a member state' (European Commission, 2016d)and how members behave after accession. The Copenhagen dilemma is born. The expression pinpoints the 'fact' that the EU is 'very strict with regard to fundamental rights standards of candidate countries but lack tools once countries have joined' (European Parliament, 2012Parliament, , 2013Reding, 2014b).
Yet, the Article 7the legal basis tailored at systematic violations of fundamental article 2 values at the domestic level, including the rule of law (Council Legal Service, 2014, § 17)existed since the Amsterdam Treaty (European Union, 1997/1999). 7 First created in preparation of Eastern enlargement (Sadurski, 2009), a risk-provision (7.1) had been added in Nice (European Union, 2001/2003, following the bilateral sanctions experience against Austria, to allow for early action against not yet materialised breaches of values. Despite the continuous existence of normative problems within the Union, the envisaged role of the provision was as a formal safeguard against a distant worst-case-scenario. The young democracies that joined in the big bang enlargement were thought to be internalised into the existing community of values (created by the fulfillment of the Copenhagen criteria), where there is no need for sanctions. The reintegration of Europe, 'healing history's deep scars' (Barroso, 2013), was told as a success story of EU 'transformative power' through enlargement (see Grabbe, 2014;cf. Kochenov, 2008). In a communication from 2003, the Commission is confident that 'in this Union of values it will not be necessary to apply penalties pursuant to Article 7', since values 'are the hard core of the Union's identity and enable every citizen to identify with it' (European Commission, 2003, p. 12).
With the rule of law crises and the ensuing formulation of the Copenhagen dilemma, it becomes difficult to dismiss the Article 7 as unnecessary. Instead, the provision receives its current framing as a 'nuclear option' (Barroso, 2012(Barroso, , 2013, which 'one should think twice or better even three times before resorting to' (Reding, 2014a, p. 2). Other descriptions in a similar vein are 'die große Keule' [the big cudgel] (Westerwelle, in Müller, 2013, p. 17), and 'ultima ratio remedy' (Council of the European Union, 2014f).
The 'nuclear option' presents sanctions as the opposite of the 'soft power' of political persuasion (see Barroso, 2012). Just as nuclear weapons are supposed to not be used but serve a security function by deterrence (Brodie, 1959), the magnitude and exceptionality of sanctions should deter members from misbehaving. By assigning the Article 7 a place outside of what is politically possible under normal circumstancesit is 'very heavy to handle', 'in practice almost impossible to use' (Reding, 2013b) the Copenhagen dilemma is confirmed. Consequently, innovations defined as different from sanctions have flourished: the Commission's 'Rule of Law Framework' (2014b), 'Justice Scoreboard' (2013), and anticipated budget-conditionality (2018), the Council's (2014a) 'rule of law dialogue' 8 , and the Parliament's (2016) proposed 'EU mechanism on democracy, the rule of law and fundamental rights' as well as 'European Values Instrument' (European Parliament, 2018). Apart from the EP's proposed mechanism (European Parliament, 2016, §1, 3-4; see the Commission's lukewarm reaction, European Commission, 2017b), all innovations circumvent the designated sanctions provision in the Treaty, the Article 7. As noted by Kochenov andPech (2016, p. 1071), 'every time the EU institutions have been confronted with a seriously "backsliding state" they have decided to add yet another pre-stage before actual sanctions may be considered let alone adopted'.
The conventional tool of normative discipline within EU lawinfringementhas been employed in reaction to specific problems related to the systemic transformation in Poland and Hungary. Infringement involves some naming and shaming through 'incriminating press releases' (Tallberg, 2002, p. 617). Yet, the ostracising potential is moderated by the fact that all members are regularly taken to Court and almost always lose (Börzel, Hofmann, & Panke, 2012, p. 456). In contrast to political sanctions, infringement is a legal process which concludes the matter by offering a clear road to reparation: the verdict itself (which may be an economic penalty).
Among all instruments, the article 7 is uniquely associated with exclusion, shaming and inappropriate emotionality. Banishing a member from community practices is understood as a 'far reaching sanction mechanism' (Timmermans, 2015c), known in the literature as a 'material sanction' (Sedelmeier, 2017). Whereas a real nuclear bomb exercises ostracism in its most brutal form: shutting the enemy out by means of extinction, the nuclear option metaphor in EU affairs reflects an understanding of sanctions as 'a kind of normative quarantine' (Müller, 2016, p. 212, original emphasis), where the target would be turned into a 'pariah state' (Scheppele, 2016a, p. 106). Asked about a prospective use of sanctions, Juncker (in Macdonald, 2016) answered: 'Let's not overdramatize … We have to have friendly and good relations with Poland so our approach is very constructive. We are not bashing Poland.' Around the same time, Council President Tusk remarked that the Polish situation would only make it onto the agenda of the European Council through an activation of the Article 7, which he did 'not believe is a good idea'. Instead, he sought to 'calm down the hotheads', 'tone down the emotional debate', and questioned 'this persistent emphasis on bad narrative'. The one who 'uses too-strong words or takes actions that do not entirely match the Western European community of values' 'has to be ready for a strong reaction', but 'this reaction cannot be hysterical' (Tusk, 2016, translator's version from Polish). In March 2018, the three Baltic countries declared (in Bodalska, 2018) that they would not support sanctions against Poland if the Article 7 procedure were to continue: 'We would be against any punishments imposed on Poland', Latvian Prime Minister Maris Kuczinskis stated. The Estonian Prime Minister Jurim Ratas elaborated: 'Any problems related to voting and taking away the right to vote -I do not think that it should happen at all, it would be a step too far'.
Notably, the Article 7 requires the Council to 'take into account the possible consequences of such a suspension on the rights and obligations of natural and legal persons' (TEU 7:3). Any EU action in the sensitive domain of rule of law needs to have an identified 'added value', including in terms of 'a real positive impact on the lives of ordinary persons' (Council of the European Union, 2013). In March 2017, Timmermans (2017a) tells the European Parliament that he is sometimes tempted 'to be the biggest hero of the day, and just invoke Article 7'. Yet, he concludes that '[e]voking article 7 now would be self-defeating and would not help us in the wider context of what is going on'.
Although the Council legal service (2014) questioned the legal basis for introducing a new procedure circumventing the article 7, the Commission's Rule of Law initiative was launched in March 2014 and first tried out since January 2016 against Poland (European Commission, 2016a. The framework explicitly aims to avoid the use of sanctions (European Commission, 2014b, p. 7;Reding, 2014b). As put by Timmermans (2017b): 'We have always said that our stated objective is not the imposition of sanctions or resorting to Article 7.' That sanctions are outside of the scope of this 'pre-Article 7 procedure' (Reding, 2013a;see Kochenov & Pech, 2016) is visualised in the Commission's graphical illustration of the framework (in Appendix 1; European Commission, 2014a).
A counterimage to the nuclear option of sanctions, the rule of law framework is consistently justified with reference to properties that stand in contrast to ostracism. The 'aim is not to accuse, to go into a polemic' (Timmermans, 2016a), but to engage in a 'constructive and facts-based dialogue' 'in a spirit of cooperation not of confrontation' (Timmermans, 2016b;see Reding, 2013a). As an 'independent and objective referee' (Barroso, 2013), an 'objective arbiter that is never blinded by political colours' (Reding, 2014a), the Commission's action is to be 'non-ideological', 'fair (Timmermans, 2015a). This corresponds to the Council's insistence that any action in the field of rule of law must be consensus-oriented and follow a 'transparent' and 'objective' evidence-based methodology (Council of the European Union, 2013). The issued recommendations should 'help the Polish authorities to find a solution', 'offer some advice', 'encourage progress', 'solve the issues in a rational way based on our legal obligations' (Timmermans, 2016a(Timmermans, , 2016b(Timmermans, , 2016c, and 'remedy the current situation' (European Commission, 2017d). As noted by Pech (2016), 'the whole point of the Commission's Rule of Law Framework is to promote the resolution of relevant problems via a discursive approach'.
As expressed by Barroso (2013), how to discipline normative backsliding among members 'is a debate that is key to our idea of Europe'. The weight attached to suspension of membership rights follows from deep-rooted idea of the European project as one which is not made to exclude but to include (Timmermans, 2015c). 9 The insistence on a 'permanent dialogue which must be based on mutual understanding and respect' is contrasted with the pre-integration 'confrontational mode based on false ideological differences' (Timmermans, 2015c(Timmermans, , 2015d. 10 As put by Juncker (2017): 'Europe has been divided: she must never again be redivided'. 11 Article 7 sanctions, defined through their exclusionary hence ostracisingproperties, do not fit well with this core understanding of European cooperation.
However, with European integration in an extraordinary state of (poly)crisis, and with the Polish authorities not always interested in dialogue under the premises offered by the Commission (Timmermans, 2017c), the debate has opened new frontiers. On 20 December 2017, 'after trying all humanly possible' (Timmermans, 2017d), the Commission activated Article 7.1 for the first time. Although a historical step, admittedly taken with 'a heavy heart', this is not a resort to the nuclear optionaka sanctionsbut 'again an attempt to start a dialogue to resolve the situation' (Timmermans, 2017c). Without member state consensus that a breach of Article 2 values has occurred (Article 7.2), sanctions under 7.3 cannot materialise.
Nonetheless, the activation of Article 7.1, together with increasing calls for actual sanctions, illustrate a widening cleavage in the Union. In line with the argument of this article, this cleavage concerns whether Poland has disqualified itself from the normal approach of solidary, rational cooperation. In November 2017, following the Polish vote on the reform of judiciary, German then minister of Justice (from spring 2018 Foreign Minister) Heiko Maas declared that 'a state that shows such scant respect for the rule of law must be prepared to accept political isolation' (Bundesregierung, 2017). The conservative EPP-group at the European Parliament called for 'measures' against the Polish governments, since '[a] red line was crossed', '[w]ith this vote, the PiS is putting an end to the rule of law and democracy in Poland and leaving the European community of values' (Weber & González Pons, 2017). Months earlier, Emmanuel Macron had promised a tougher line on Poland if elected President: 'I want sanctions to be imposed against the lack of respect of the rights and values of the European Union. This is provided for in the Treaties' (Ouest France, 2017). 12 Much of the engagement with Poland has highlighted the country's central belonging in Europe. If Poland no longer qualifies as an insider, if it has 'decided to isolate itself' (Macron in Tsolova & Sobczak, 2017) at its own will and through its own sovereign actions, this invites a redrawing of the EU's limits of toleration. Commenting on the Commission's activation of article 7.1, Council President Tusk declared that Poland 'has turned its back to the whole EU and, more broadly, to the West' (in Rettman, 2017). As an 'outsider' within the Union, the premise of its protection from sanctions under the normal way of European cooperation would be shaken. At the same time, the concept of suspension only works if there is an integral community to return to after penance. In times where the EU's legitimacy as a normative authority is particularly frail, and with its borders changing through Brexit, using sanctions at home would be a high gamble with unpredictable outcome.

Conclusion
The EU's dual-face when it comes to sanctions is counter-intuitive. At home, there is a voluntary contractual bond between member states and the EU, and a mutual interest that normative standards are respected across borders for the daily micro-practices of European integration to function. Abroad, the EU opts for sanctions without any contractual obligation and in situations that often lack obvious relevance for the member states.
This article has argued that an understanding of sanctions as ostracism helps explain the duality in the EU's approach to sanctions. Cutting norm-violating third countries off normal relations furthers the EU's self-identification as an actor that cares for the preservation of international norms. Sanctions are justified focusing on the severity of the crime, leaving the target's expected reaction unspecified. In terms of norm promotion, it is a cul-de-sac, which feeds into the EU's self-perception as a well versed 'force for good ' (e.g. Manners, 2002), but is disconnected from reasoned expectations of sanctions' ability to foster normative change on the ground. At home, already the initial post-world War strategy to integrate economically to prevent future conflicts encapsulated a preference for cooperation over isolation, solidarity over judgmentalism. The big-bang enlargement of 2004 coupled the assumption that membership would lock-in democratic institutions with emotive arguments about belonging in Europe. As the 'nuclear option', Article 7 sanctions would only be used in extreme situations when this belonging is threatened. Even if the rule of law crises could be seen as extreme enough to warrant sanctions in principle, the reactions of potential targets predict that sanctions would in practice not help fix the problem. Hungarian Fidesz, and Polish Law and Justice have a trump card on hand which the Commission as well as most targets of foreign policy sanctions lack: they are democratically elected. As long as these governments enjoy broad popular support, a turn to sanctions risks being not only ineffective, but counterproductive.
In conclusion, the duality in the EU's use of sanctions constitutes a genuine dilemma. On the one hand, the EU has deliberately built its external sanctions policy around the premise that the Union is a value community. It is from this internal identity that the EU derives a moral mandate to police international norm violations through sanctions. In addition, coherence between internal and external action is an expressed priority of EU institutions and individual leadersnot the least of H.R. Mogherini. In view of its claim to be a holistic normative actor, the flagrant discrepancy in the EU's view on sanctions at home and abroad invites accusations of hypocrisy and double standards from targets of sanctions as well as critical bystanders. Whether merited or not, such accusations leave a non-trivial mark on the EU's reputation as an international actor.
On the other hand, it could be argued that politics areby their very naturea repeated story of different shades of double standards, of which only some are destructive. As much as the policy-and scholarly community has been calling for firmer action against value-breaking member states, lessons from foreign policy suggest that sanctions advocates at home should be cautious. Change does not come about solely through the infliction of harm but requires that the target is involved. However, there is one decisive difference between foreign policy sanctions and measures used against members. If foreign policy sanctions can be in place for years and yearsmanifestly at least partly ineffective, the life cycle of sanctions against members would need to be much shorter. The intensity of cooperation and density of legal commitments make any protracted isolation of a member unpractical and politically hard to defend. If the EU pulls the trigger for its 'nuclear option' it has to be prepared to face the day after. If sanctions do not deliver, the only remaining option is expulsion: an outlook so drastic that it is not even provided for by the Treaties.  Reding, 2013b). 8. The three rounds of 'rule of law dialogue' held by the Council so far have not dealt directly with the crises at hand but with sharing national 'best practices' and discussing digitalization (Council of the European Union, 2015c), migration (Council of the European Union, 2016a) and media pluralism 'in the digital age' (Council of the European Union, 2017c). 9. Author's translation. French original: 'n'est pas faite pour exclure mais pour inclure'. 10. Author's translation. French original: 'un dialogue permanent qui doit être fondé sur la compréhension et le respect mutuels' 11. Author's translation. French original: 'L'Europe a été divisée; elle ne doit plus jamais être redivisée'. 12. Author's translation. French original: 'Je veux que sur le non-respect des droits et des valeurs de l'Union européenne, des sanctions soient prises. Les traités le prévoient' at the annual workshop on research on European politics between Stockholm University and Södertörn University (November 2016). I would like to thank all colleagues taking part in these events, in particular, Nicholas Aylott, Inken von Borzyskowski, Tanja Börzel, Tyra Hertz, and Jonas Tallberg, for their thought-provoking feedback and constructive criticism. Special thanks also go to Josef Hien, who has been an excellent sparring partner throughout the process of writing this article. Finally, I am grateful to the two anonymous reviewers of Contemporary Politics for their most helpful and relevant comments on the manuscript.

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
This research was funded by the Swedish Research Council and generated in the framework of DFGsupported Kolleg-Forschergruppe 'The Transformative Power of Europe' at Freie Universität Berlin.

Notes on contributor
Elin Hellquist is postdoctoral researcher at the Department of Political Science, Stockholm University. This article was written during her time as an associated fellow at the research college 'The Transformative Power of Europe', Freie Universität Berlin. She is interested in the evolving role of sanctions and other institutionalised forms of punishment in international relations. Her research, funded by the Swedish Research Council, seeks to explain why regional organizations have developed diverging approaches to sanctions.