Regional organizations and democratic conditionality: Family resemblances and shaming

Framing the cases: Theory and context

This special issue provides a comparative analysis of the democratic conditionality practices towards their members of an impressive variety of major regional organisations (ROs) during an extended time period. During the first two decades after the Cold War the background conditions for such experiments were particularly favorable. This extensive and up-to-date assemblage of evidence invites broader reflection on a critical and under-theorized aspect of democratization as an international process, and on the normative as well as the realist dynamics involved.

Scholarly work on the international dimensions of democratization and democracy promotion has greatly flourished since 1989, both in comparative politics and in the parallel silo of international relations research. But the relevant activities of ROs have either been considered mainly on a case by case basis, or subsumed under much broader studies of international legal, normative, or geopolitical processes. The major exceptions are Cooper and Legler (2006), Davies (2018), Pevehouse (2005, 2016), and Poast and Urpelainen (2018).

ROs come in diverse varieties, and obtain whatever powers they have by delegation from member states. Even so they also function as privileged nodes and conduits for the processing and dissemination of information, norms, and practices diffused from extra-regional sources. Still, in comparison to their member states they are thin, or fragile entities, with a tendency to atrophy over time. They cannot be studied in isolation, but always need to be assessed in relation to their counterparts, their rivals, and as organizations lodged within a larger international ecosystem. ¹

Since no RO exists in isolation—the term denotes an entity that must belong within a larger context composed in part of other regions and their organizational forms—our topic is inherently comparative. But there has been little close comparative analysis of the divergences as well as the ‘family resemblances’ (see relevant following section) shaping the recent democratic conditionality and sanctioning activities (see RO disciplinary practices section) of the major ROs.

By contrast, this collection of studies focuses on the interplay between specifically regional organizations and their member states, paying careful attention to the informal as well as rules-based character of these disciplinary practices and interactions. Since member states undertake external policy commitments and incur financial costs when they set up ROs, they expect to gain offsetting benefits, which may include intermediation with dominant powers, reputational gains, enhanced collective voice, more secure borders, and perhaps other more specific objectives (European integration; Arabs against Israel; Africans against apartheid; and so on). But over time some of these initial spurs to membership lose traction, replaced by newer policy priorities (such as embracing liberal internationalism, even at some cost in terms of sovereignty). RO secretariats and leaders have an incentive to keep their organizations in existence when the demand for regional collective goods shifts. For example, as decolonization concerns faded and Cold War security imperatives relaxed, good governance and democratic norms offered new justifications for RO continuity. Indeed, across much of the Global South banding together to shelter a region from intrusive external impositions became an urgent matter of state security, as more insistent demands for liberalization and democratization became diffused. However, ROs that embraced these prescriptions, and that developed disciplinary processes to sanction member states that fell short on them, can run into mounting internal resistance (indeed there has always been a certain degree of ‘pushback’ from those targeted by sanctions). Beyond a certain level of friction, the cost of sanctions to an RO can prompt it to seek alternative justifications for its continued existence. This collection compares the sanctions policies in conditions of democratic crisis of these six RO operations (particularly when they were at their most ascendant), and by underscoring their ambiguous and informal features it helps explain their temporal and spatial variability.

The disciplinary procedures covered in this collection include such open sanctions as public condemnation and even suspension from the organization, but also the subtler behind the scenes practices of negotiations, monitoring, and peer reviewing. These articles show that such ‘soft’ power methods can carry significant weight by exerting a persuasive pressure that appeals to shared aspirations and identities, and so may alter collective expectations and understandings, with ‘realist’ effects. Such case studies highlight ‘shaming’ processes and stigmatization pressures, while paying due attention to the possible reactions of backlash, forum-shopping, and even retaliation (Soyaltin-Colella, this volume). ²

There is clearly a strong overlap between sanctioning and stigmatization, but the two are not identical. ³ For example (as noted in the Introduction to this issue, see Palestini and Hellquist, this volume), it is possible for an RO to issue a hurtful verbal condemnation without attaching any further consequences—what the editors term ‘deprivation of benefits’. They would classify that as a stigma, but not a sanction. At the other end of the spectrum, when severe punitive (even coercive) measures are imposed without any adequate justification, the target regime may be viewed as a victim rather than as a wrongdoer. Between these two extremes, both sanctions and stigmatization are terms that can encompass a multiplicity of disciplinary steps.

Certainly, stigmatization is a very broad label. Which are the relevant target audiences, and how do they change over time and between ROs? On the temporal dimension, for example, the focus could be on immediate impact, but over time traces of the initial stigma will tend to linger, with cumulative effects, and these may affect the timing of eventual sanctions relief. Likewise, on targeting, is stigmatization about appealing to domestic audiences in the non-compliant state? If so, is it the appeal to mass or elite opinion that is most important (or to business)? Or is it to an international audience? Again if so, is that to ‘democratic opinion’ at large, or to particular key audiences (sponsoring great powers, international non-governmental organizations, diasporas)? All these effects are linked to some extent, as information flows between them. But each audience is also open to counter-arguments—nationalism, in the case of domestic mass opinion; loss of market privileges for business; strategic ties for sponsoring governments, and so forth. So—as constructivists would predict—stigma produces debate, controversy, counter-claims, allegations of selectivity and hypocrisy. For example, it may not be straightforward to determine whether (or how) sanctions work. Success could mean satisfying the expectations of the sanctioners (though these may not be clear and stable). But when sanctions involve public shaming, in addition to lingering unintended effects (e.g. polarizing opinion in the target country), there is a distinction to be made between a reluctant change of conduct and an authentic change of heart (Hafner-Burton, 2008). Finally, regional sanctions are not just a transaction between a sender and a target. They may also be directed to bystanders (keeping other members in line, or potentially inducing other members to rally to the side of the target). All this makes causal links between stigma and change of stance hard to trace and even harder to prove, either in a single case, or in larger clusters. This is even more the case when enforcement routines are largely presentational rather than results-dependent.

This concluding overview highlights the relevance of such metaphors as ecosystem, family resemblance, and peer review; directs attention to the temporal and spatial scope conditions of the cases considered; and to the bargaining involved in these specifically regional sanctioning practices. Overall, when taken together, these heterogeneous, context-bound, and internally negotiable features of RO democratic conditionality raise big questions over future prospects. There can be no assurance of further RO normative convergence—either on sanctions activations, or even around the underlying theme of democratic conditionality. Under currently unfavorable international conditions such norms are particularly vulnerable to distortion and backlash—although so long as an RO remains in place and does not repeal its formal commitments, they stay available for eventual reactivation.

‘Family resemblances’

With these framing considerations in mind, we can now turn to the specifics of the six major RO case studies presented in this special issue. They display a considerable range of common features, and yet they can also be differentiated on a series of important dimensions. There is an analogy here with the similarities and individual differences that enable us to generalize about a given family, while also recognizing the individual attributes of its members. For example, these six major ROs encompass diverse regions, some composed solely or mainly of democracies, but also some where hybrid and even non-democratic regimes prevail, leaving little scope for democratic conditionality proper. Even so, these organizations share some major ‘family resemblances’. ⁴

This section firstly sketches each of the six cases separately, identifying some of their major particularities (differences within this family of major ROs) and then secondly, explores the extent of such configurative resemblances between cases, under the following six rubrics—like-mindedness; the ‘add on’ of democratic conditionality; variable conceptions of democracy; overlapping boundaries with other ROs; normative divergences; and procedural differences.

The first of our six ROs—the EU—has always been composed entirely of democracies (notwithstanding some current backsliding), and has the strongest supranational institutions and disciplinary powers (as Closa, this volume, demonstrates in his careful analysis of article 7 on the Treaty of the European Union) and most elaborate specification of democratic conditionality—notably spelled out in the 1993 Copenhagen Criteria that set the political standards for accession to the Union. ⁵ But in practice, inter-governmentalism and informal intra-institutional bargaining often trump supra-nationalism. As Closa explains, the EU’s complex and divided internal procedures limit its capacity to discipline member states. It actually relies more on its powers of persuasion than on its means of sanction.

Second, the Council of Europe (CoE) is a larger and looser RO with an overlapping but more heterogeneous membership, and a more focused political purpose. It was created in 1949 by ten fully democratic west European states, and it was given a mandate and structure strongly geared toward the promotion of democracy—but even in this case the next two members (Greece and Turkey) signaled a more mixed agenda. Its current membership embraces all the 27 European Union (EU) states plus 20 others (ranging from Albania and Azerbaijan to Russia, Serbia, Turkey, and Ukraine). The eastward expansion took place after 1990, with only Belarus as an exception. ⁶ The Council’s core priority has long been the application of the 1950 European Convention for the Protection of Human Rights and Fundamental Freedoms, which by the late 1980s had come to include abolition of the death penalty and the outlawing of torture. The associated European Court of Human Rights applies this Convention and has compulsory jurisdiction and the right of individual petition. When Yeltsin’s Russia agreed to recognize these provisions as part of Moscow’s application for membership, a major legal precedent was established. Still, the Council’s Parliamentary Assembly (PACE) is more broadly responsible for upholding human rights, democracy and the rule of law within its region. Russia was stripped of its voting rights in PACE following Putin’s 2014 incursion into eastern Ukraine and re-annexation of the Crimea. But in mid 2019 Moscow was readmitted to the Assembly, over-riding objections from Kiev. This is only the most recent bargaining episode to reveal how, in practice, the Council understands its mandate. Soyaltin-Colella (this volume) reconstructs a parallel history concerning the imposing, and then lifting, and then re-imposing, of membership suspension in the case of Turkey. (Greece was also suspended from 1967 to 1974). Solyaltin-Colella (this volume) describes its expert ‘monitoring’ processes and its record of ‘naming and shaming’ democratic underperformers, but also shows (contrasting the Turkey experiences of 1996 and 2017) that results depend upon political conditions in the receiving state as much as on the intentions of the RO.

Third, Palestini (this volume) deals with another major and longstanding RO with a distinctive and chequered career, the Organisation of American States (OAS). Here it is not always the regional hegemon that decides on RO sanctions, even though the USA provides most of its budget and houses the Organization in Washington. Both the White House, and the US Senate have sometimes exercised crucial influence over key controversies, notably in the case of Honduras and Venezuela. Changing US administrations impact both on OAS priorities and on its room for maneuver. Palestini’s article uses the qualitative comparative analysis (QCA) method to search out causal factors accounting for recent OAS sanction decisions (a method designed to cope with few cases and multiple causes), and complements this with process tracing in two contrasting cases—Haiti in 2014 and Venezuela in 2017 (see also Ribeiro-Hoffmann, 2019).

The fourth case study concerns the most active practitioner of regional sanctions, the African Union (AU). Hellquist (this volume) shows that the main focus of AU democracy promotion concerns the prevention and reversal of outright military takeovers. She stresses the like-mindedness of African states and their acceptance of regional ‘peer review’ in contrast to what is perceived as external intervention when sanctions are imposed from outside the region (notably by the West against Zimbabwe). In addition to displaying zero tolerance for coups (even including the ousting of President Morsi by the Egyptian military in 2013), the AU also deploys sanctions against mercenary power grabs; rebel uprisings; incumbent rejection of adverse election outcomes; and forced mandate extensions. It might extend this rejection of unconstitutional outcomes to include civil war, though as yet no decision has been taken. This is an impressive list of bad governance practices subject to RO activism, but it also indicates the acute political instabilities characteristic of this major region, far more severe than any other. So the pertinent conception of ‘democracy’ has to be tailored to this especially adverse context. Collective regional acceptance of such AU activities reflects an awareness that without the shelter of this RO Africa could be exposed to much more intrusive and durable forms of external intervention. As Hellquist notes, although there were 18 sanction exercises between 2003 and 2018 their average duration was only 18 months, and they were also in other respects quite ‘soft’. The AU has undertaken a disproportionate share of all the RO sanctioning actions recorded anywhere.

The fifth case is the League of Arab States (LAS), composed almost entirely of autocracies, at least until the Arab Spring, when it did temporarily respond to the region-wide outburst of bottom-up citizen demands for democracy. But, as Debre shows (this volume) this proved a re-positioning in order to protect as much as possible of the status quo, not a shift towards popular participation. Given the marginal status of democratic polities within the region (only Tunisia and perhaps Lebanon can be placed there), other considerations were bound to shape the sanctioning policies of this RO. Where democratization would empower Shia populations (as it had in Iraq and Lebanon) this was seen as threatening not only by incumbent autocrats but also by a wider stratum of majority Sunni opinion. If it meant electoral success for Muslim Brotherhood influenced parties, that too would provoke an anti-democratic backlash. On the other hand, if Arab Spring idealism could be turned against such troublesome republican autocrats as Assad and Gaddafi then those in control of the League would be likely to seize a tactical advantage. Nothing could highlight this contrast in prevailing RO orientations towards democracy better than the gulf separating the LAS from the AU over the 2013 coup in Egypt. If democracy spreads in the Arab world it will surely be despite the LAS, not with its backing.

Sixth, and last, comes Association of Southeast Asian Nations (ASEAN)—the organization with the least explicit commitment to political conditionalities. In contrast to the other regions just considered, in this case there is an explicit alternative theoretical argument (‘Asian values’ as superior to ‘western’ democracy). Moreover, as Schembera (this volume) underlines, the foundational norm of the Association has always been non-interference in internal affairs, and even though there has been some shift towards the global democracy script since then, that remains a weak and not fully institutionalized tendency.

In synthesis, then, even though these six major ROs share an important range of common characteristics, and even fit together as the set of institutions available for diffusing democratic norms (potentially backed by sanctions against delinquent members) in their regions when prevailing international orthodoxy requires it, nevertheless there are also major divergences. Like a heterogeneous set of family members they are each highly individual. Their shared conditions and differing responses can be traced under six broad configurative headings, as follows:

All our six major ROs, were created before the fall of the Berlin Wall, with common goals shaped by the Cold War. Even the CoE (which originated in 1949 as a democracy-monitoring organization) was significantly repurposed after 1990, when its membership nearly doubled, generating the need to accommodate a range of member states with relatively illiberal political proclivities. At the end of the Cold War minority rights were added to the Council’s agenda, and became a prominent feature of its activities in the Balkans. The OAS originated at the same time, and also claimed to support democracy (in its region), but it took the eclipse of the Soviet bloc to uprate that objective and enshrine it in a charter with substantive content. The EU developed incrementally from an economic to a more explicitly political RO, only fully spelling out its democratic commitments after the reunification of Germany (Whitehead, 1993). In 2001 the AU actually changed its name from Organization of African Unity to signal its new democratic priorities. Likewise, the LAS and ASEAN were created by states with few democratic credentials and for other purposes (opposition to Israel, and co-operation for development, respectively); they moved only hesitantly towards democratic conditionality after about 2000.

In summary, then, the major ROs of interest here have extensive and mixed pre-histories. They were created with purposes other than support of democracy in mind, and they continue to operate under mandates and expectations that extend beyond that goal alone. In consequence they are liable to face trade-offs between democracy supportive policies and, say, regional security, or economic integration, or comparable competing regional objectives.

Even taking at face value their current pro-democracy commitments, they have varied pasts to live down. They have inherited alternative purposes and practices that still color expectations of their underlying reaction functions as the international climate changes. This makes it misleading to model their democracy promotion policies in strict principal-agent or democratic treaty enforcement terms.

Each of these ROs claims to be politically representative of the states in its region, but all have to contend with rival claimants. For example, the EU overlaps with the CoE as regards democratic conditionality, and indeed both the Organization for Security and Co-operation in Europe (OSCE) ⁷ and North Atlantic Treaty Organization also figure in this regard. Likewise the OAS has to contend with various competitor ROs—Mercosur (Southern Common Market) and Caricom (Caribbean Community) both have democracy promoting track records, and more recently the Union of South American Nations and the Community of Latin American and Caribbean States also posed a challenge (see Palestini, this volume). The overlap (and divergence) between the LAS and the AU over such key tests as the coup in Egypt is well described in those two case studies. Moreover, the AU also coexists with Economic Community of West African States, Southern African Development Community, and other African Regional Economic Communities (Hellquist, this volume), as does the LAS with the Cooperation Council for the Arab States of the Gulf, (Debre, this volume). Only ASEAN has its region more or less to itself, and even there the alternative of Asia-Pacific Economic Cooperation could eventually pose an issue. In addition to these regional competitors there are also extra-regional potential rivals (the Commonwealth, the global democracy sanctions promoted by the EU, the US, and ultimately even the UN). Overall, then, these ROs must jostle for space and adjust their ambitions according to the contextual conditions they encounter from time to time and across varied geographies.

Our six regions vary in composition from the EU (27 democratic regimes) to the LAS (dominated by absolute monarchies, and containing no more than two weakly democratic regimes—Tunisia and maybe Lebanon). In between there are quite heterogeneous regions (in the western hemisphere and Africa), and then ASEAN (which is basically hybrid). The CoE has expanded from an all-democratic to a much more divided grouping. Democratic conditionality is bound to vary considerably across such diverse entities.

Even so, each of these six ROs displays some unifying historical, linguistic and geographically interactive characteristics that set them apart from other regions (Ahram et al., 2018). They are composed of states that are fated by geography to continue operating alongside each other, whether or not their domestic political arrangements prove convergent and mutually supportive. States that voluntarily join their local RO, and that in consequence promise to conform to the rules of the club, retain their national autonomy. In principle such commonalities can be expected to generate a certain sense of common identity and ‘like-mindedness’ and this is frequently invoked by RO leaders. However, our case studies indicate that, as with divided families, propinquity and shared reference points can also produce their own intense sources of disagreement. As a contemporary illustration, every secretary-general of the OAS since its foundation has been unanimously elected for a second term—until 2020 (when deep differences over Venezuela produced a 23-10 split). More generally, big internal disagreements over democratic conditionality are also evident within both the EU and the CoE, and seem latent elsewhere.

As the editors argue in their Introduction (see Hellquist and Palestini, this volume) the six major ROs in this special issue display some important differences in both the conceptualization and the practical application of their ideas concerning democracy across the cases as well as over time. Perhaps at the most abstract level they can all be said to pay lip service to the very general (but actually quite demanding) democratic principles expressed in the UN Charter, and further elaborated in the ensuing Universal Declaration of Human Rights and its subsequent elaborations. ⁸ The key headings—human rights/rule of law/peaceful dispute resolution/national sovereignty/ democratic elections—are all present, although as we have seen they appear in different combinations and with variable emphasis across these large regions and over time. But on closer inspection, especially when the concern is with practical applications of these principles, these RO comparisons indicate that the appearance of discursive homogeneity is deceptive. Rather than revealing a monolithic and unitary concept of democracy they display branching, overlapping, and even contrasting normative commitments.

Africa provides a good example here. Among the OAU’s foremost political concerns was the termination of white minority rule in Southern Africa, and the achievement of universal suffrage there. Democracy, understood in these terms, was achieved in Namibia and then South Africa in the early 1990s. The AU was founded as a successor in 2002, with a much stronger and sharper focus on democracy (notably stressing civilian rule and competitive elections overseen by non-partisan institutions). The new AU also promotes good governance, human rights (including gender equality and the rights of the child), and the rule of law (via an African Court on Human and Peoples’ Rights). These are highly aspirational objectives given the regional context. According to Ani (2018), ‘the AU has helped to reduce coups, but steers clear of other tactics used to subvert democracy’. The AU’s electoral assessment and monitoring procedures look promising, but are rarely operative.

Even the CoE, with its specialized democratic mandate, has in practice developed a flexible and negotiable conception of the democracy it claims to promote. Beyond realism and democratic purism what has emerged here is the importance of upholding a strong and enforceable suite of individual human rights, as underlined by the central role assigned to the European Court of Human Rights (ECHR). The inter-governmental as opposed to supra-national basis of its activities requires considerable deference to the national sovereignty strand in democratic thinking. This may have diluted its attachment to a strictly ‘liberal democratic’ agenda, given the need to accommodate a range of member states with relatively illiberal proclivities, including Russia and Turkey. The most critical component of its agenda has been defense of rule of law principles. Here it can deploy not just the ECHR but its respected team of legal experts under the Venice Commission (as described by Soyaltin-Colella, this volume). Over time and space these multiple elements of a democracy support conception have shifted and adapted.

The OAS has also changed its focus over time, (although like the CoE some core stability is provided by the Inter American Court of Human Rights). The Canada-led creation of a Unit for Democracy Promotion in 1992 (Poast and Urpelainen, 2018: 178) represented a significant shift in outlook as compared to this RO’s first four decades of non-performance on that front. An amendment to article 9 of the charter introduced the threat of suspension where a member government has been overthrown by force, although articles 19 and 20 (which debar external intervention in member’s sovereignty, and 20 even adds that ‘no state may use or encourage use of coercive measures of an economic or political character to force the sovereign will of another state’) remained in force. Even so the traditional Latin American notion of sovereignty was altered so that only freely elected governments were to benefit from these protections. As Cooper and Legler observed even at the highpoint of this new fashion: ‘In practical terms, a great deal of unevenness can be found in the extension and application of the democratic solidarity doctrine’ (Cooper and Legler, 2006: 29).

As for the LAS, Debre (this volume) identifies domestic regime preservation (especially the maintenance of the Gulf monarchy autocracies) rather than democracy promotion, or concern about human rights under the UN’s Responsibility to Protect rubric, as the League’s main priority. This is something of a contrast with the AU, which also includes many non-democratic member regimes, but which can react strongly—at least in the case of coups—although it too has a strong pro-incumbent bias and does not act against stable autocratic regimes.

Lastly, any attempt to pin down the conceptual basis for ASEAN’s democracy supporting activities must proceed inductively, with an emphasis on implicit rather than explicit factors. It is only when extreme cases arise, and precipitate open conflict between competing norms, that such implicit regional assumptions are tested and displayed. On this basis the critical cases of Cambodia and Myanmar deserve close attention, and can in part be understood in terms of a clash between the norms of non-intervention and good governance (Schembera in this volume). But the outcomes were somewhat indeterminate, and the details of the cases seem at least as important as the underlying norms. On this basis it is doubtful whether any specific ASEAN concept of democratic conditionality can be discerned.

Rules also matter. ASEAN is more flexible than the OAS, or the AU, and the EU is particularly precise (and legalistic), while the CoE has developed elaborate consultative procedures and draws on technical expertise. Over time, as new challenges arise, all the relevant ROs may shift ground at least a little in response to both the overall international context and to specific immediate regional pressures. Given the scope for resistance and retaliation, ROs interested in self-preservation have reason to prefer informal compromise, postponement of deadlines, and the flexible interpretation of democratic norms. For this reason, as the contributors to this issue show, the informal and negotiated features of RO sanctioning practices require as much attention as the formal rules.

The ROs just surveyed take their decisions not simply on the basis of their stated purposes and normative commitments, but in accordance with highly specific procedural rules. As Closa’s (this volume) account of the complex tripartite rule-governed procedures of the EU makes clear, outcomes can in no way be simply read off from stated intentions. The implementation of most EU decisions is many steps removed from the principles that Brussels necessarily invokes. When it comes to the CoE, a very different set of decision rules produces a highly contrasting pattern of negotiations (e.g. even when suspended, Russia was paying about a tenth of the Council’s budget—if it withdrew, its citizens would cease to have individual redress via the European Court). While it is true that the unanimity rule confers veto power on any member who wishes to slow or redirect the main thrust of CoE policy, in the case of the LAS that rule was simply disregarded when Libya and Syria were sanctioned. As regards the OAS, there is a huge imbalance between the size and weight (including budgetary role) of its largest contributor and the demography and influence of its many very small members. Yet crucial decisions are taken by a majority vote of the membership, making the sensitivities of small Caribbean islands more consequential than is often realized. In the 54 member AU, it turns out that the 15 member Peace and Security Council exercises crucial initiative powers (independent from the Department of Political Affairs). In the LAS the formal rule is that all key decisions must be unanimous, and no decision is binding on a dissenting state. Finally, ASEAN relies on consensus for its formal procedures, but in any case, its key decisions on democracy-related issues are taken informally, rather than through explicit rules. Given this wide array of procedural practices, it would be most surprising if these six major ROs produced any uniform pattern of outcomes, even if they had been more normatively convergent than we have seen to be the case.

Overall then, our six ROs are a heterogeneous, permeable, somewhat fluid, and contextually bounded spread of institutions. They range from structures with well-developed institutional processes for collective decision-making to others where one or two dominant members act with almost unrestrained discretionality (e.g. the LAS). Not all members carry equal weight, even when all are formally democratic—e.g. the OAS might sanction Honduras for democratic backsliding, but the same procedure could hardly be applied to Brazil—and still less to the USA, no matter what electoral abuses might arise there (Closa and Palestini, 2018).

Sanctions may feature prominently in their operational toolkit, but even when this is so, the decision to apply (or suspend) them will be discretionary and context dependent. These normative commitments and enforcement practices may branch out from a common origin, and can project structurally similar profiles, but—following Wittgenstein’s metaphor of ‘family resemblances’— they also present physiognomically contrasting faces to the world. Over time these appearances evolve, potentially converging, although it is equally possible for them to diverge. RO conduct and the reception of these normatively informed behavioral patterns can vary widely depending on how these initiatives are designed, intended, received, and understood.

As already noted above (like international ecosystem), the ‘family resemblance’ metaphor deserves the same consideration as other analogies that have already entered this literature—‘target audience’, ‘sanctions as signals’, ‘norms cascades’, and of course ‘shaming’. Indeed, even very standard terms such as ‘actor’, ‘performance’, ‘credibility’, and ‘dialogue’ also carry anthropomorphic overtones that sharpen their imagery without disqualifying them from social scientific applications. One merit of the ‘family resemblance’ approach to ROs is as a corrective to any impression of them as a homogeneous set of entities, and as a reminder that they may be viewed quite differently from contrasting perspectives, and in successive contexts. At the same time, they do have subtle shared features (‘resemblances’), hence the need for a term combining parallelism with heterogeneity. And crucially, like the other anthropomorphic analogies listed above, they are agentic—i.e. these ROs can assess how they are being received and adjust their profiles to improve their presentations. ⁹

Moreover, the ‘family resemblance’ perspective could also have a second application here—within ROs as well as between them. The constitutive member states of an RO can also be analyzed as a heterogeneous interactive community with some shared identity, together with divergences and rivalries that can even lead to ostracism and sanction (to curb the reputational damage to the whole family occasioned by a miscreant member, or ‘black sheep’).

RO disciplinary practices: Sanctions, stigmatization, ‘shaming’, peer review, and negotiation

Every organization requires some structure, and an at least ostensible purpose. That requires the establishment of some at least minimal practices for disciplining unacceptable conduct by members. In the case of our six major ROs the core sanction may be suspension from membership (Von Borzykowski and Vabulas, 2019), but since the member state will remain within the region even when ostracized, lesser disciplinary practices may better serve regional purposes. Where some degree of regional identity with shared interests and norms are involved, an RO can resort to ‘shaming’ miscreant members. Analytically, however, ‘shaming’—a frequent feature of family relationships—is inherently subjective and hard to pin down. The more objective term ‘stigmatization’ is both more exact and also more exacting. Regardless of the sentiment of the outcast, a stigma is an empirically observable discredit directed by a community toward a relevant target audience, intended to impose a cost on the miscreant, and perhaps to induce a rectification. But what then, in our cases, is the relevant audience, (or audiences)? Do they change over time and between ROs? Debre (this volume) explains sanctions against Syria and Libya as responses to the mass of regional opinion on display during the Arab Spring, and yet in response to Bahrain, an opposite course was chosen, catering for some significantly different Arab audiences. Admittedly, all such audiences are linked to some extent, as information and debate flows between them through the current international ecosystem. Also, as already noted in the ‘family resemblances’ section, each audience is open to counter-arguments so that, as interpretivists would expect, stigma produces debate, controversy, counter-claims, allegations of selectivity and hypocrisy etc. Soyaltin-Colella (this volume) records an apparently effective example of ‘naming and shaming’ in Turkey in 1996, but then relates a diametrically opposite outcome when the same RO applied the same disciplines to the same regime two decades later. More generally, the causal links between stigma and change of stance are often hard to prove, either in a single case, or in clusters. So even the relatively objective practice of stigmatization is interactive, dialogic, and judgmental, like ‘shaming’ (Adler-Nissen, 2014). Both necessarily involve cycles of negotiation.

Poppe et al. (2019) have recently highlighted the role of ‘negotiations’ in democracy promotion, going on to sketch what they term a ‘causal model’:

By focusing our attention on the communicative interaction between external democracy promoters and local ‘recipients’. . . the study of negotiation processes allows us to analyze when and how democracy (promotion) is contested by local actors, how democracy promoters respond to such challenges, and whether and how the ensuing controversy is resolved (766–767).

In particular, they seek to uncover the often overlooked agency of those they call ‘recipients’ (‘targets’ in the terminology of this special issue). A significant empirical finding of their project is that

the supposedly normatively weaker party—the recipient or target of democracy promotion—not seldom is quite strong in normative terms, even sometimes has the edge, whereas the supposedly normatively stronger party. . . often avoids engaging in normative debate, is relatively silent, or even on the defensive. p767

They suggest this defensiveness may be partly because not only resisters but also promoters may have concealed agendas ‘[i]n that structural sense and despite the promoters holding the normative edge, all parties in democracy promotion negotiation often act in a similar way’ (767). In their outline of a causal theory they consider not just power asymmetries and normative settings, they also (under ‘cultural traits’) recognize that the negotiating parties may be working with somewhat competing conceptions of democracy: ‘With increasing normative divergence, the likelihood that negotiations will touch upon fundamental (normative) issues will increase’ (786). Neither purely normative, nor solely realpolitik interpretations can suffice: the dynamics of each negotiation shapes both its output and the broader outcome.

This insight seems applicable to the sanctions and ‘shaming’ experiences analyzed in our special issue as well. In particular, the ‘negotiation’ approach directs attention to whether sanctions not simply induce compliance but also change the normative outlook of the targeted regime. Our case material indicates that the normative shift can actually prove greater at the transmitting than at the receiving end of the negotiations (e.g. see Closa, this volume, between Brussels and Budapest). Extending the ‘family resemblance’ metaphor one might say that the core relationship involves a two-way tug as the RO aims to ‘shame’ miscreants into conforming more closely to the group’s reputational standards, and the ‘black sheep’ of the family attempts to redefine those standards. Thus, for example, the African Union paper records that when Egypt was suspended for the 2013 coup its ambassador to the AU expressed his country’s wish ‘to continue to take part in this family’, and once coup leader Sisi had staged his legitimizing election, Cairo ‘celebrated Egypt’s return to its African fold . . . our African brothers are more understanding now’. (So much so that within six years Sisi was chairing the AU mission to Khartoum following the coup in Sudan). Under either rubric the potential exists for the identified miscreant not only to remain out of line, but also to ‘push back’, seeking to win over other group members to its standpoint, or even to recast group standards in accordance with its dissident stance. Whereas an enforcement model presumes one way causality, the negotiation approach must allow for two way patterns of adjustment.

Hellquist (this volume) also proposes a related metaphor—‘peer review’—as in academic publishing. This too is a disciplinary practice, but without the stigma of a sanction. In principle both parties are collaborating in a common endeavor—to produce the best quality outcome for both individual and collective reputational benefit. To the extent that practice matches the ideal no ‘shame’ is involved, so this would be a very collegial form of democratic conditionality. This model is appealing, but it only applies if all participants are equally liable to equitable peer review. In the more typical case, when some leading members of the RO can pass judgment on peripheral neighbors without fear of reciprocal evaluation, the metaphors of ‘family resemblance’ and ‘black sheep’ are more pertinent.

Prospects for RO democracy promotion

Twenty years ago, with liberal internationalism in the ascendant, it may have seemed as if it were only a matter of time before all ROs became effective promoters of democratic standards in their respective regions. Subsequent experience of repeated setbacks in diverse contexts has highlighted the over-confidence of this initial outlook. The factors obstructing enforcement and encouraging resistance and pushback have proved more widespread and durable than anticipated. Adler-Nissen (2014) recently outlined three variants of ‘stigma management’ that situate such pushbacks within international relations theory. At the practical level many ROs face the dilemma that if their political commitments are taken too literally their other functions may be put at risk. On the other hand, they will not wish to lose face by explicitly discontinuing their democratic conditionality provisions. So they are tempted to persist with a largely fictional (or at least an unrealizable) commitment that risks undermining their credibility when they fall short.

The present international panorama alters such calculations. Starting in 2001 slow multilateral charter-based collective actions peaked and then lost traction. Informal ‘coalitions of the willing’ put together by leading western democracies took the lead with coercive sanctions, overshadowing softer RO approaches. They were seen as effective, fast-acting, and not inhibited by too many legalistic scruples (e.g. about torture). In due course the ‘war on terror’ and the ‘war on drugs’ were supplemented by renewed frictions between the great powers, and then by an increasing flow of refugees trying to escape the resulting mayhem. None of this fits the previous script of ever more authoritative RO convergence around shared international democracy norms and disciplines. Then, in 2016 President Trump entered the White House committed to riding roughshod over inconvenient international obstacles, and prioritizing transactional deals with (usually undemocratic) foreign counterparts. Brexit had a similar impact in Europe. Consequently, at least in the short run, international reliance on RO ‘shaming’ of miscreant members in their own regions has largely evaporated. In this context the notion of democratic conditionality has become unstable. What standards are to be employed? Who is to judge or monitor compliance? Perhaps the most to be expected for now of an RO conditionality clause is that it might serve to shelter some member states from harsher and more arbitrary and uncontrolled pressures from without.

At best, as we have seen, most RO disciplinary structures are quite weak, and key actors are often well-placed to manipulate or to sidestep them. The norms cascade can flow in both directions. Linkage and leverage presumes a core international commitment to global democracy that is no longer visible. But this is not to say that all RO democratic conditionality has become necessarily inconsequential. There can still be strength in weak ties. Charter commitments may be temporarily shelved rather than permanently repealed. The underlying prospect for such democratic conditionality remains open.