EU democracy projection in the Southern Mediterranean: A practice analysis

ABSTRACT

This special issue expands on the existing literature on the international dimension of democratization by focusing on democracy projection, defined as the projection of (democratic) norms in the every-day practice of interactions, beyond any donor-recipient relationship, between states and foreign civil society actors on issue areas where both have interests to defend. The SI examines the issue areas of trade, anti-corruption, applied research, gender and LGBTI, focusing on EU practices in its everyday dealings with civil society in the Southern Mediterranean. The authors conclude, based on comparative case studies relying on extensive interviews, direct observations and content analysis, that democracy projection varies according to four main factors: EU’s perceived interest, its ideational commitment to norms of dialogue and inclusion, the degree of institutional inertia and discourses/structures of meanings dominating in some policy areas which preclude EU engagement on substance.

KEYWORDS:

• Democracy projection
• democracy promotion
• practices
• European Union
• Southern Mediterranean

1. Introduction

In this special issue, we propose to expand on the existing literature on the international dimension of democratization, by focusing on what we call democracy projection. What we are concerned with is how powerful actors – and in this case the EU – engages with foreign – here Southern Mediterranean – civil society actors, including activist non-governmental organizations, religious groups, trade unions, researchers, and women. More specifically, we are interested in how they do so outside of the donor-recipient/project support relationship in which many of them are engaged. The starting point is a concern with whether the EU interacts with such actors, which are usually thought of as crucial for democratization to take root, and whether it does so in a way that is actually conducive to democratization.

This means that we focus on an aspect which is not currently given much attention in the literature on the international dimension of democratization. This literature, as will be further explained below, tends to focus either on active agency (democracy assistance, democratic conditionalities) or passive diffusion (through ‘contagion’, ‘emulation’, ‘policy transfer and learning’ or ‘demonstration’ effects). What we aim to capture with the concept of democracy projection lies somewhere in between: the search light is on EU practices in its everyday dealings with politically relevant groups such as activist NGOs, religious groups, researchers, women and so on, and whether such exchanges are marked by democratic norms. Thus, the interactions under examination are not those where the EU is pursuing specifically designed policies of democracy promotion per se, as is the case with its vast democracy assistance arsenal or its (considerably more sparingly applied) democratic conditionalities. Instead, we are looking at day-to-day interactions and practices in areas where the EU is pursuing a series of other goals, such as strengthening applied research across the Mediterranean, negotiating trade agreements and anti-corruption measures, working with/on women as a category important for the social stability and modernization of Southern Mediterranean countries etc. As has been noted elsewhere, states (and in this case the EU as an association of states) are at the present day dealing more extensively and directly with non-state actors active inside other states than was the case in the past (Grimm, 2019; Korosteleva, 2016). We are examining if, in those contexts, EU interaction with its interlocutors is based on democratic norms or not – i.e. whether and to what extent it is projecting democratic norms and practices.

The focus on direct interactions beyond the donor – recipient relationship also brings with it a concern with the ‘recipient’ side: democracy is in its very essence reciprocal, and an understanding of the extent to which EU relations with typical recipients of democracy assistance also outside the strict donor – recipient relationship has democratic characteristics will necessarily imply examining the agency of the ‘recipient’/target side of democracy assistance and democracy promotion efforts writ large. Thus, instead of studying solely the EU and EU member state policies, agencies, and programmes, the focus is also on recipient institutions and organizations and how they interact – in various ways, with various aims and effects – with EU policy making, both in terms of the democratization agenda and beyond. Similarly, some of the contributions in this special issue pay attention to how non state actors perceive the relations with EU institutions as they develop in practice.

It should emerge from the above that we are honing in on practices of interaction. By analysing EU practices of interaction with Southern Mediterranean non-state actors, we can learn about the norms and values (democratic or otherwise) underpinning them. So here, we are interested in understanding if (and if so how) democracy is projected through the performance of everyday practices which emerge out of specific contexts (Christensen, 2017; Korosteleva, 2016). In other words, we are asking: How and in what manner is the EU engaging with these Southern Mediterranean non-state actors? Are they included in consultations on EU – third country policy agendas in a wide variety of policy areas? If so, which actors are included/excluded and under what conditions? To what extent do the Southern Mediterranean non-state actors feel (dis-)empowered in and through their interactions with the EU? What kind of a say do they have in policy formulation in practice? Is the EU, through its practices, instrumentalizing these actors for its own purposes? These are the questions that guide this special issue at the empirical level.

Our focus on practices is obviously – if quite loosely – inspired by the practice turn in IR in recent years. The authors of this special issue examine practices – such as the selection of interlocutors by European External Action Service personnel – which are, on a deeper level linked to habits as a central part of the theorization of practice (Cornut, 2015; Hopf, 2010). Likewise, they de facto look at communities of practice (e.g., the EU civil service and EU diplomacy (Bicchi, 2011, Lequesne, 2017, pp. 14–15)). However, the goal of the authors of this special issue has not necessarily been to explore theoretically such habits or communities, and they have not been bound by any particular practice approach or framework. What they all have in common, instead, is going beyond studying formal institutions and elite discourses, instead focusing on ‘everyday agents’ and paying attention to routines and ways of doing that are quotidian. It is, in other words, a pragmatic focus on practices, defined simply in terms of ‘standard ways of doing’. Defining practices is the subject of longstanding debates (Schatzki, Adler and Pouliot, Adler-Nissen). Similarly to Bicchi and Bremberg (2016) we adopt a rather broad understanding of practices, which enables us to simply place the emphasis on practices ‘as the place to study human activity’.

This focus on practice obviously brings with it a strong empirical commitment. The articles in this special issue have all adopted the case study approach (although in various guises such as comparative case studies and detailed single case studies). Methodologically, there is a wide diversity of approaches used to evaluate and analyse practices: “practices are ‘seen’ (e.g., ethnography, participant observation), ‘talked about’ (interviews) or ‘read’ (textual/discourse analysis)’ (Adler‐Nissen, 2016). It can be about interaction or performance of practices. This diversity is reflected in this special issue, where the articles by Anna Khakee and Ragnar Weilandt and by Esra LaGro and Hakan Cavlak rely to a large extent on interviews, whereas, in addition to interviews, Ragnar Weilandt uses direct observation of consultation meetings in Tunisia in his article. Assem Dandashly relies on a combination of interviews and observations of day-to-day interactions between the EU and LGBTI groups and, finally, the article by Sarah Wolff relies on content analysis. So, as just stated, our approach is strongly anchored in empirical analysis.

However, in addition to the empirical commitment, we also aim to add to the conceptualization of the international dimension of democratization, even if in a modest way, given that we are here restricting our analysis to the EU and practices in the Mediterranean region. Thus, we argue for the usefulness of adding the concept of democracy projection to the theoretical toolbox for the analysis of the external dimension of democratization. As will be further discussed below, we define democracy projection as the projection of (democratic) norms in the every-day practice of interactions, beyond any donor-recipient relationship, between states and foreign civil society actors on issue areas where both have interests to defend.

In policy terms, an approach focusing on relations between powerful state actors and civil society¹ is particularly important at the current juncture, globally and in the southern Mediterranean in particular. We have witnessed several waves of citizens’ mobilization seeking more active participation in democratic processes. Anti-austerity and anti-authoritarian movements and uprisings in the streets of Athens, Algiers, Beirut, Cairo, London and Minsk have been found to reject representative democracy in favour of a more participative understanding of what democracy means (Ishkanian & Glasius, 2017, p. 1006). Common across the ‘square movements’ in Europe and the Middle East and North Africa is the call for ‘active citizenship’ (ibid, 1007). These movements are contesting the traditional engagement of would-be democracy promoters with states institutions and long-established civil society organizations (CSOs). They push for a broader understanding of what citizens’ participation in a democracy means. This should lead to a reconceptualization of democratization and, importantly for this special issue, of relations between the EU and politically relevant actors in the southern Neighbourhood states.

This introductory chapter is of the Special Issue divided into five subsections. After having set the scene in the introduction, section two defines the concept of democracy projection in more detail and sets out the argument for its usefulness. Section three aims to situate the notion in the literature on the external dimension of democratization. These two sections are followed by an overview of the special issue, covering both the depth and breadth of the research undertaken. The concluding section outlines the contributions of this special issue to scholarship, stressing that democracy projection varies according to four main factors: EU’s perceived interest to reinforce democracy agency, EU’s ideational commitment to the norms of dialogue, inclusion and mutuality, the degree of institutional inertia that can enhance or inhibit democracy projection and finally the meanings dominating some policy areas such as trade, which may also be a brake to real EU engagement on substance.

2. Conceptualizing democracy projection

This SI examines how state actors that purportedly aim to defend and advance democracy abroad interact in practice with non-state actors in countries targeted by such democracy promotion efforts, outside the direct democracy promotion relationship. Unlike much of the existing literature on this topic, this issue hones in not on democracy assistance or democratic conditionalities (together usually referred to as democracy promotion – see further details below) per se, but on everyday interactions beyond direct democracy promotion between representatives of such state actors, on the one hand, and, on the other, faith-based groups, human rights NGOs, labour unions, local community groups, think tanks/academic institutions, women’s and youth associations, etc. in traditional ‘target’ countries. Thus, analytically, our aim is to add to the conceptual toolbox of the academic literature on the external dimension of democratization. We have labelled this new notion ‘democracy projection’, as we wish to stress that it involves practices where states – often by habit or convention rather than by a conscious choice – project certain (democratic or non-democratic) norms in their day-to-day interactions with foreign non-state actors. Thus, in short, the democracy projection aspect of the external dimension of democratization can be defined as the projection of (democratic) norms in the every-day practice of interactions, beyond any donor-recipient relationship, between states and foreign civil society actors on issue areas where both have interests to defend.

We believe in the usefulness of the concept of democracy projection in understanding how and why democracy may – or may not – travel across borders: If the relations that the EU, in our case, entertains on a continuous basis with civil society actors in third countries are marked by norms of consultation, exchange, respect for difference and inclusivity this will, we presume, be beneficial for the democractic agency of such actors. If, on the other hand, interactions are based on top-down, exclusionary, or ‘token’ practices, the outcome will be that civil society actors will not be thus strengthened: it may even (though this would need further empirical investigation) lead to their weakening vis-a-vis national governments, as the latter may emulate their foreign counterparts in disregarding civil society actors.

It is important to stress that we are not arguing that democracy is always, or indeed as a rule, thus projected in day-to-day interactions between actors from states purportedly aiming to advance democracy abroad and non-state actors in ‘target’ countries. Rather, the goal of the various contributions to this Special Issue has been to examine this empirically, and our conclusions are, as will become clear in subsequent sections of this introduction, mixed. This, it should be noted, makes it similar to other concepts in the democratization toolbox. For instance, the extent of democratic diffusion/learning is uneven at best and depends on a variety of factors (for a review, see Åberg & Denk, 2020). Thus, in our view, the fact that democracy projection is sometimes present, and in other cases absent, adds to its analytical value, and the contributions to this SI are devoted to better understanding why and when this occurs.

3. Situating democracy projection

Conceptually, the notion of democracy projection is, as already noted, situated somewhere in between the ‘number of factors of the international context “without agency” that could positively influence democratization, i.e. all forms of imitation, contagion, [and] learning’, on the one hand, and ‘all overt and voluntary activities adopted, supported, and (directly or indirectly) implemented by (public or private) foreign actors explicitly designed to contribute to the political liberalization of autocratic regimes, democratization of autocratic regimes, or consolidation of democracy in specific recipient countries’, on the other (Schmitter and Brouwer 1999, 12).

The factors ‘without agency’ are widely discussed in the literature on the international dimension of democratization. They include diffusion, demonstration effects, contagion, emulation, and learning. This brief literature review cannot hope to do them justice in any way, not least since there is considerable disagreement on their definition: for instance, a recent article identified at least four different understandings of diffusion (Åberg & Denk, 2020). The idea behind demonstration/emulation effects is, very crudely put, that the successful democratic transitions in neighbouring states or in ‘self-identified peers’ and the transmission of information influence elites in autocratic states. Learning, in turn, can be achieved through transnational networks of various types (for an overview, see Levitsky & Way, 2010, pp. 38–9; Diamond, 2008).

Likewise, activities and measures ‘explicitly designed to contribute to … democratization’ – that is, democracy promotion – have garnered substantial attention over recent years. In line with commonly accepted definitions, democracy promotion is a usually operationalized to include both democracy assistance (i.e. foreign aid funding specifically aimed at building and strengthening those institutions and groups which are considered key for democracy to emerge/consolidate) and positive/negative conditionalities (foreign policy sticks such as ‘essential elements clauses’² and carrots such as increased aid and closer economic ties, aimed at nudging states in a democratic direction) as well as public pronouncements in support of democratic actors and aims in third countries (Khakee, forthcoming). Such explicit democracy promotion policies have been a central aspect of European external relations since the fall of the Berlin Wall. Whether resisted, criticized, hailed as a mark of progress, or actively championed, politicians and scholars would agree on its significance in shaping – directly or indirectly, advertently or inadvertently – developments in Europe, its neighbourhood and beyond. Not surprisingly then, the democracy promotion agenda has received considerably scholarly attention. However, theorizing and conceptualization has, surprisingly, been rather limited (Smith, 2000). The focus of much of the academic literature on democracy promotion generally has been on policy effectiveness or what Milja Kurki has called ‘the “problem-solving” mainstream of democracy support analysis’ (Kurki, 2013, p. 215). Thus, scholarly inquiry has largely focused on the questions: ‘Does democracy assistance work?’; Do democratic conditionalities work (K. E. Smith, 2015; Velluti, 2016)? ‘Why (not)?’ and ‘Can we make it work better?’ (see e.g., Schimmelfennig & Scholtz, 2008; Youngs, 2009).³

This ‘problem-solving’ focus of the EU democracy promotion literature is evident in the Mediterranean context as well. Academic writing has effectively shadowed the evolution of EU policy making: every new policy initiative from the Barcelona Process in the mid-1990s, via the Advanced Status of the 2000s to the post-Arab Uprising ‘more for more’ principle has led to its crop of new policy papers, journal articles and books (see e.g., Del Sarto & Schumacher, 2005; Teti, 2012). In other words, academic output – and, it must be underlined, writings by several of us authors of this special issue are no exceptions in this regard – has tended to closely follow the output of EU institutions, thus reacting to new EU policy documents and initiatives, without necessarily critically analysing them from a broader, more conceptual perspective. Here too, the ‘problem-solving’ character manifests itself in the conundrums tackled: how to better deal with Islamist political parties and civil society organizations (Pace & Wolff, 2017)? How can democracy promotion tackle crackdowns on civil society (Youngs & Echagüe, 2017)? Can democracy support be made more receptive to the needs and preferences on the recipient side? These are problems and hurdles faced by the democracy promoting practitioners in their work, and, again, exemplify the ‘problem-solving’ trend in academic texts.⁴ We argue in this special issue that it is important to take an analytical step back when it comes to understanding conceptually European democracy promotion. We do so by dissociating ourselves from the conceptual grids and policy classifications created by the European institutions and instead proposing the conceptualization of democracy projection which is anchored in practice.

So most of the literature has focused on better understanding the effectiveness of traditional EU policies and instruments of democracy promotion such as for instance, the European Instrument for Democracy and Human Rights (EIDHR), political conditionality within the European Neighbourhood Policy (ENP), ENP sub-committees, ENP benchmarking procedures, and the like. In other words, democracy promotion has mostly been looked at from the perspective of EU policy. The approach remains Eurocentric and centred on policy, analysing mostly how the EU organizes the governance of democracy promotion through instruments, agreements, action plans, progress reports etc. Only to a much lesser extent does it examine how the policy is implemented in practice, including at the level of the delegations and how it meshes – or not – with interactions between EU representatives and groups and individuals in recipient states in other issue areas.

It must be pointed out that, in the field of democracy promotion, some institutional practices have been analysed, in particular when it comes to the granting and disbursement of aid (Carapico, 2014). Academics have come to the conclusion that in spite of an increase in the number of EU actors involved in EU foreign policy towards Arab countries after the Uprisings and the reform and boosting of instruments such as the European Neighbourhood Policy Instrument (ENPI), practices have changed to a lesser extent. Democracy assistance is still disbursed predominantly to state institutions, even though NGOs receive a larger slice of the pie. There is an effort to be nimbler, with new institutions such as the European Endowment of Democracy granting aid using a ‘lighter’ set of practices and procedures (Leininger & Richter, 2014). Thus, this means that, while funding practices are analysed, there is little understanding of practices beyond the recipient-donor relationship and what it entails in different policy areas, beyond pure democracy assistance programmes.

As noted, our goal is to bring more conceptual clarity to that area in between all overt and voluntary activities adopted, supported, and implemented by foreign actors explicitly designed to contribute to democratization, on the one hand, and the factors of the international context ‘without agency’’, on the other. As mentioned, we argue that it is important to examine the extent to which basic democratic norms infuse modes of EU direct interaction with politically relevant actors in the southern Neighbourhood in policy areas outside democracy promotion such as, in this special issue, anti-corruption, trade, social science research, gender, and LGBTI rights. Lavenex and Schimmelfennig (2011) touch upon this intermediate level with their theorizing around ‘functional cooperation’. They argue that EU democratic principles are ‘embedded in the governance of individual policy fields’ such as the environment, trade, migration, security cooperation, fisheries, safety standards and competition policy (Lavenex & Schimmelfennig, 2011, p. 887). These democratic principles then unfold ‘through the deepening of transgovernmental, horizontal ties between the EU and third countries’ public administrations’ (ibid). In other words, the EU has built democratic governance (and in particular transparency, accountability and participation) into its regulatory framework and procedures. As and when these are exported as part of the acquis communautaire, democratic governance principles are included in the package. While this theory presumes inter alia that the acquis is democratic in character, the concept of democracy projection leaves this question to be settled by empirical investigation – as noted in the previous section, we do not postulate that democracy projection is always, or even mostly, present in interactions between the EU and non-state interlocutors in third countries. In other words, we shed the assumption that the EU is necessarily a ‘force for good’ (Poli, 2016). Also, this theory to a certain extent conceives of the EU as a passive and so to speak en passant promoter of democratic governance, while what we focus on is the character and values embedded in ‘normal’, everyday, and routinized interactions between the EU and extra-European actors in policy areas other than democracy/governance – including those discussed by Lavenex and Schimmelfennig. In line with thinking around practices, such interactions are based on habits, but while they offer continuity, they do not preclude conscious agency.

Elena A. Korosteleva’s research comes perhaps the closest to our own. Although she in part focuses on direct foreign aid, she explores how in principle technocratic instruments act ‘as a non-linear process of social empowerment which works with local issues on an individual level’ (Korosteleva, 2016, p. 680). Investigating the case of Belarus, and through an analysis of ‘citizens’ daily narratives’, she highlights how, through daily practices, individuals may “develop new knowledge of ‘good governance’ or simply share ‘good practices’ to become better able to solve and less tolerant of the existing inadequacies in their daily lives, including mismanagement, inequality, corruption, or abuse’ (Korosteleva, 2016, p. 688). Taking such budding research further, this special issue argues that democracy can be projected in interaction with populations in third countries – civil society, NGOs, political parties, women, LGBTI activists, etc. – and not necessarily only through the provision of democracy assistance. This relational element allows us to explore further how democracy may, or may not, travel across borders.

4. Overview of the special issue

In this special issue, we propose to take a different view from the classical democracy promotion literature, focusing on how and to what extent the EU projects democracy across issue areas in the Southern Mediterranean. Khakee and Weilandt’s (2021) contribution centres on interactions between the EU/European states and Moroccan/Tunisian non-state actors in two particular issue areas: trade negotiations and international anti-corruption policies. Both have been considered of crucial importance politically and economically by the EU and Tunisia/Morocco alike since the Arab Uprisings and the importance of implicating civil society in these areas has also been highlighted across the Mediterranean. The contribution thus examines, through a structured comparison, to what extent Moroccan and Tunisian non-state actors are included and can influence policy making processes in these two policy domains. Khakee and Weilandt find that the EU is more likely to listen to civil society where the latter’s arguments are in line with EU preferences, and that there is divergence across policy areas with more inclusivity on anti-corruption than on trade, and more timidity in the Moroccan case than in the Tunisian.

Moving to the area of gender equality, Wolff (2021) studies the EU’s processes and interaction with (in) direct recipients (i.e. women, civil society, state). Investigating the case of the Tunisian democratic transition, Wolff shows that the practice of EU interaction with civil society organizations (CSOs) in the field of gender equality has been mostly driven by the priority of trust-building amongst the different partners including the EU. Like Khakee and Weilandt in their joint article and Weilandt in his singe-authored piece, Wolff also comes to the conclusion that in spite of the innovation of the practice of tripartite dialogue, as a new venue to perform EU narratives on gender equality, this innovation has not led to a major change in the EU’s narrative and practice on gender equality which still lacks the intersectionality dimension. Yet one of the major contributions of the tripartite dialogue on gender has been to build trust amongst the different stakeholders (newer/older women organizations, the state and the EU).

Drawing upon the study of the practice turn in international relations, Weilandt (2021) examines the EU’s interactions with Tunisian civil society as a case study and the practice of regular consultations with Tunisian activists. It provides a textured mapping and a critical interpretation of the processes, mechanisms and patterns that constitute this practice. In spite of the EU efforts to listen to civil society during the democratic transition, Weilandt shows how this practice may lead to exclusionary practices by selecting certain partners over others. For instance, as in the case of women’s organizations for Wolff, the EU has some trouble moving beyond its comfort zone and tends to listen more to CSOs based in Tunis and part of the set of pre-2011 activists. Investigating the case of the LGBTI in Lebanon, Dandashly (2021) shows that EU engagement beyond (modest) funding and low-key interaction is not always wanted or beneficial. Building on semi-structured interviews and engaging with secondary literature, Dandashly seeks to answer the questions: to what extent does the EU engage with local actors to improve minority rights? And how do the various EU actors engage with LGBTI rights’ CSOs in Lebanon in particular? His main findings show that the EU is ambiguous when it comes to LGBTI rights in the MENA and tend to treat the matter within broader a human rights context. While most initiatives come from the local CSO activists themselves, the EU delegation and EU member states are following the lead of such local actors, which, given the sensitivity of the issue, is appropriate.

LaGro and Cavlak (2021) look at the role expert communities play in democracy projection. Following a body of 205 interviews with experts from networks such as Euromesco or FEMISE, they demonstrate that mutual transfer of knowledge and norms such as pluralism and participation to policy-making, inclusiveness of civil society is enhanced through the cooperation amongst Euro-Mediterranean expert communities. Expert networks provide socialization venues across a multitude of policy areas and could become new mediums to amplify democratic projection if new modalities would be put in place.

5. Contribution of this special issue to the scholarship

One of the main findings of this special issue is variance. Democracy projection is uneven across policy areas and countries. Covering a wide range of issue areas such as gender, LGBTI, anti-corruption, trade, research, and expert communities, our contributions all point to differences in the extent of projection of democratic norms by the EU. This varies across a number of factors.

One factor at play is clearly perceived interest. For instance, as discussed by Khakee and Weilandt, the EU is much more reticent to effectively include civil society in a dialogue reinforcing their democratic agency where it could disrupt the swift conclusion of negotiations on the Deep and Comprehensive Free Trade Agreement in Tunisia or jeopardize the extension of the agricultural and fisheries agreements to Western Sahara. In contrast, they are quicker to do so in the case of anti-corruption initiatives in Tunisia, where EU and civil society interests converge.

But interest does not explain everything. Another factor is ideational: Pioneering Tunisia has thus definitely seen the most innovative set of consultative practices which in some areas have helped to create a constructive environment of trust-building between the EU and different Tunisian stakeholders during the delicate time of the democratic transition (Weilandt; Wolff). Thus, the EU is trying harder in a case such as the Tunisian to ensure that its engagement with civil society is in accordance with democratic norms of dialogue, inclusion, and mutuality, even if the results, as Weilandt shows, are sometimes mixed. Arguably, the same is true in the case of LGBTI rights in Lebanon. Through dialogue with civil society, the EU has concluded that it is best for it to stay out of the internal discussion on LGBTI rights, lest it jeopardizes its struggle. Instead, EU engagement has had an impact in helping CSOs to speak with one voice, as explained by Dandashly in his contribution.

A third factor may be habit and institutional inertia. LaGro and Cavlak show that in relation to expert communities such as EUROMESCO, FEMISE and EMNES, the EU is quite receptive to dialogue and inclusion, with a mutual transfer of knowledge and norms. In contrast, the inability of EU institutions to ‘translate’ NGO concerns and suggestions (which tend to be less technical and not in conformity with applicable terminology and procedures) could be put in the column of institutional inertia (ibid).

A fourth factor is social construction (and as such cannot be dissociated from interests to the extent that these are socially constructed). If freer trade is constructed as intrinsically good, then engaging with actors on such topics becomes superfluous, and even potentially counterproductive or outright harmful. This can explain why the European Commission has been so reluctant to engage on substance in the area of trade (see Khakee and Weilandt): it would go against the episteme of freer trade as ultimately a good thing (Hannah 2016).

Interestingly, most of the contributions point to the problematic issue of the meaning given to norms and power asymmetries as structurally constraining progress in the field of democracy projection. In various policy areas such as gender or trade, the authors raise the issue of how democratic norms are understood by the EU. In the area of gender policies, for instance, the author stresses that the EU has at times an instrumental understanding of engagement with women, regarding them mostly as economic resources in a neoliberal environment. Another issue is the ‘one-size fits all’ approach that prevails in many of the EU policy documents and which tend to regard for instance women as a single entity sharing the same problems, thus leading to an undifferentiated engagement. In the case of gender in Tunisia for instance, there is only a too fragile recognition of the intersectionality issues.

Thus, this special issue points to a varied picture: the EU has sometimes engaged in a dialogue with politically important actors beyond states in the Southern neighbourhood, and sometimes chosen not to, for a variety of reasons. Sometimes, such as in the area of LGBTI rights, it has acted sensitively, while in others, such as e.g., in the area of trade in Western Sahara, it has not. It is our contention that this messy empirical reality is likely to evolve over time – while retaining its messy elements – with the EU engaging more democratically with some actors in the coming years, while remaining top-down, aloof and rhetoric in other areas. What we have not so far touched upon is how far democratically infused interactions (increasing process legitimacy) has led to increased output legitimacy. Clearly, more research is needed.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by Malta Enterprise; International Institute of Social Studies, The Hague; Queen Mary University of London.

Notes

1. We are employing the same definition of civil society as elsewhere in this special issue, namely the space between the state, market, and family “where voluntary associations deliberately seek to shape the rules [in terms of specific policies, more general norms, and deeper social structures] that govern one or the other aspect of social life. Operationally, it excludes organizations (parties) seeking public office or commercial profit. In practice, it includes a wide range of registered and non-registered organizations of different political/societal persuasions and goals, e.g., environmental movements, ethnic/regional lobbies, faith-based groups of various types, human rights NGOs, labour unions, local community groups, philanthropic foundations, professional bodies, think tanks/academic institutions, women’s and youth associations, etc (Scholte, 2002, p. 283).

2. Since the mid-1990s, the EU has included the respect for democracy, human rights and the rule of law as ‘essential elements’ in bilateral trade agreements, including the Cotonou Partnership Agreement and its predecessor, the Lomé IV bis Convention, which both cover EU cooperation with African, Pacific and Caribbean mostly former colonies, the Association Agreements with Eastern and Southern neighbours from Ukraine to Morocco, as well as the cooperation agreement with Mercosur signed in 1995.

3. The conclusion has often been that democracy support in the European neighbourhood has not worked, as it has often gone hand in hand with support for authoritarian but stable regimes which serve Western strategic, economic, and political interests (Börzel, 2015; Brownlee, 2012; Schlumberger, 2006). Post-Arab uprisings, relatively little has changed, although there have been some lessons learned, including increased pragmatism, flexibility and reactivity as well as increased engagement with CSOs (Youngs 2014). Strategically, democracy assistance has been admonished for missing out on the agents of democratic change, including the central role played by social networks and citizens (and not necessarily CSOs) (Khondker, 2011) as well as the roots of the uprisings: socio-economic inequalities, the gap between centres and peripheries, and the lack of prospects for youth (Kamel & Huber, 2015). In general, policy effectiveness is seen as limited at best.

4. There are of course exceptions to this relative lack of theorization. Notably, there is an ongoing debate around the types of democracy promoted externally and what this says about the nature of democracy promotion and its role in world ordering (Kurki, 2013; Wetzel & Orbie, 2011). Moreover, the democracy promotion literature has largely followed the long-standing general theoretical debate on how to understand EU foreign policy generally towards its neighbourhood and beyond, pitting proponents of a ‘normative power Europe’ against more realist/rational approaches (see e.g., Pace, 2009). There are also a number of post-colonial and neo-marxist analyses (Guilhot, 2005; Hanieh, 2006).