Regional (dis)integration beyond governments: A comparison in social and civil society participation between Mercosur and SADC

Comparative regionalism, participatory regionalism and the role of non-state/civil society actors

Since the end of the Cold War, there has been a tendency towards increased civil society participation in international organizations and global governance. There is an evolving perception that international organizations should be open to non-state actors’ participation, considering trends of democratization and legitimacy building. In spite of this, one can verify that civil society participation in international organizations varies widely (Reinold, 2019). Furthermore, little is known about civil society interactions in the international organizations of the Global South (Fioramonti, 2014). If we consider that the density of a regional integration process depends on the level of accession of societies within the integration dynamics, it is crucial to better understand how civil society is engaging with regionalism and whether regionalist initiatives have fostered any increase in civil society participation. Some authors argue that civil society actors may stimulate new forms of regional governance, that is, participatory regional governance (Godsäter, 2013). In this sense, civil society organizations are, at least partly, involved in regional policy making and service provision. According to this view, civil society organizations can either act as problem solvers or critics. Additionally, in the face of the expansion of regional agendas to issues such as environment, human rights and migration, it is necessary to consider that social actors have the knowledge and capacity to voice societal demands, even in some cases contrary to governmental positions. Hence, regional organizations such as Mercosur and SADC may become arenas for political alliances and provide a new scope of action for civil society actors. Regional institutions can be susceptible to accusations of being undemocratic and illegitimate if they fail to open sufficient space for civil society involvement, which undermines their perception by both domestic and external actors (Vleuten and Hoffmann, 2010). Furthermore, the opening of institutional spaces for social participation can make regional integration less dependent or vulnerable to changes in political/partisan contexts. This makes regionalism less state-centric – something particularly relevant to regionalism in the Global South, characterized by low levels of economic interdependence (Ramanzini Junior and Luciano, 2020).

Acharya (2003: 382) has pointed out that participatory regionalism – that is, the involvement of civil society in regionalism – means not just ‘greater cooperation among the social movements leading to the emergence of a regional civil society, but also closer and positive interaction between the latter and the official regionalism of states’. We will demonstrate below that this is not always the case. It can be verified that participatory regionalism is able to generate new dynamics of conflict and opposition in relation to governments and policies. Nonetheless, when closer and cooperative interactions emerge, it is more likely that civil society inputs are incorporated into regionalism.

European theories of regional integration emphasize how contextual conditions affect civil society’s engagement in regionalism. According to Haas (1961: 149), ‘countries dominated by a non-pluralistic social structure are poor candidates for participation in the regional integration process’. According to this perspective, non-democratic states are not in favour of social participation in regional integration, and it is therefore unlikely to observe civil society engagement in a regional integration project partially or fully composed of non-democratic states. Neofunctionalist approaches emphasize that active social participation through elites and civil society is a relevant mechanism for the activation of spillover effects. Meanwhile, in liberal intergovernmentalism, as the bargaining process among nation states is shaped by national preference formation, domestic groups have a decisive impact on member states’ positions in regionalism (Moravcsik, 1998).

The way neofunctionalism and liberal intergovernmentalism delineate the role of civil society in regional integration is of little applicability when analysing processes that involve recently democratized and non-democratic states, or when civil society’s influence on decision-making processes is limited or non-existent. As stated by Söderbaum (2007: 319), ‘theories and conceptualizations of civil society rooted in the Western or European experience risk misunderstanding the logic of African civil society’. In this article we add that this risk also applies to South American civil societies, due to legacies of authoritarianism and dictatorship. Furthermore, in spite of specific differences, it can be considered that civil society in the Global South faces similar challenges in the construction of democratic institutions at the local, national and regional levels. In this sense, European integration theories have a limited applicability in understanding the dynamics of non-Western regions in terms of civil society’s engagement with regionalism.

In the 1990s, the so-called new regionalism approach (NRA) contributed to the enlargement of the research agenda on regionalism. It represented a change in traditional explanations derived from both neofunctionalism and liberal intergovernmentalism. Among the main focuses of this research agenda was the study of the relationship between regionalism and globalization, attention for the multidimensionality of the regionalist phenomenon and the analysis of different cases of regionalism in the world, besides Europe (Fawcett and Hurrell, 1995). In this perspective, regionalism came to be understood as a complex process, interconnected with global trends towards the shifting of power to the transnational level as a response to the end of the Cold War, which has led to the creation of a post-Westphalian regionalized world order (Hettne, 2005). In this respect, Söderbaum (2015: 54) highlighted that ‘the NRA critically assesses state–society complexes in the formation of regions, and opens up a broader understanding of how regions are formed and what characterizes regionalism and regionalization in various parts of the world’. In sum, new regionalism studies focus on regionalization dynamics such as transboundary and transnational regional cooperation. Thus, a wider perspective on the constituent actors of regionalism and the emphasis on formal and informal networks of social interaction allow for the opening of a discussion about civil society’s involvement in regionalism.

However, even though it has amplified the research agenda and scope of regionalism studies, new regionalism’s assumption that the decreasing centrality of states is the main driving force behind regional processes may be misleading. Notwithstanding the limits of new regionalism for understanding regionalism in the Global South, the notion of regionness is an important insight that helps to better understand the contribution of civil society to regionalism. According to Hettne and Söderbaum (2000: 461), regionness is the process whereby a geographical area ‘is transformed from a passive object into an active subject capable of articulating the transnational interests of the emerging region. Regionness thus implies that a region can be a region “more or less”. The level of regionness can both increase and decrease.’ Therefore, it can be argued that civil society involvement in regionalism is an important path towards increasing regionness.

In the 2000s, the questioning of the paradigmatic status of European integration theories increased. Acharya (2012) has pointed to the Euro/EU-centrism of regional integration theories and studies. In this sense, the agenda of comparative regionalism should not rely on either neofunctionalism or intergovernmentalism as universal approaches. Therefore, contrary to NRA, comparative regionalism has not considered regionalism as dependent on the pooling of sovereignty, and has privileged the development of more contextual and systematic parameters of comparison among regional cases (Fioramonti and Mattheis, 2016; Ramanzini Junior and Luciano, 2018).

The notion of civil society has been used in various contexts in IR and political sciences. There is no consensus on the meaning of the concept, and the increasing role and diverse universe of non-governmental organizations turns the conceptualization even more complex. In this article, following Edwards (2012: 1–2) and Reinold (2019), we depart from Walzer’s (1998) definition of civil society as a ‘sphere of uncoerced human association between the individual and the state, in which people undertake collective action for normative and substantive purposes, relatively independent of governments and the market’.

An additional conceptual clarification relevant for this article is the difference between participation and influence. Even though our analysis focuses on the participation of civil society in Global South regionalism, we do not equate participation with influence and changes in regionalism and foreign policy. Participation entails a communicative process where several actors interact. Lasker and Guidry (2008: 201) pointed out that players ‘have a voice when the opportunity enables them to express their ideas to others. They have an influential voice when their ideas are used to make something happen (for example, when actions are taken to address issues they raise or when their ideas shape actions that are taken).’

Considering previous discussions on the role of civil society actors in regionalism, our analytical framework will focus on examining the role of three variables in the participation of social actors in the cases of Mercosur and SADC: institutional design, member states’ support of social participation, and civil society resources. We aim to assess which of these factors are the most relevant to explain the degrees of social participation in civil society in the two case studies, by highlighting key similarities and differences between the cases. Moreover, we expect that the trends identified in the two cases may offer relevant generalizations to better understand the involvement of civil society in Global South regionalism.

Grounded on this framework, this article will empirically demonstrate that in the case of Mercosur, the ideology of national governments was the key explaining factor of both the historical increase and contemporary decrease of social participation in this bloc. Meanwhile, in the SADC, the main drive behind participatory regionalism has been the resources of specific civil society organizations and networks, which succeed in dialoguing with SADC executives in certain policy fields. Nonetheless, the intergovernmental institutional design of both blocs – an aspect also widely observed in other regional organizations in the Global South – has contributed to the marginalization of civil society actors in regional decision making. Therefore, we aim to stress that although civil society actors have been given more opportunities to express their concerns in Global South regionalism, their influence on decision-making processes remains limited.

Social actors in Mercosur: The rise and fall of Mercosur Social Summits

Social participation was not a constitutive principle of Mercosur during its formation. The Treaty of Asunción that set up Mercosur in 1991 was essentially a trade agreement linked to economic liberalization of the founding member states (Argentina, Brazil, Paraguay and Uruguay ¹ ). The bloc is driven by three main intergovernmental bodies: the Common Market Council (CMC), the Common Market Group (GMC) and the Mercosur Trade Commission, all of which are composed of representatives from national governments. Their decisions are all made through intergovernmental and consensus-based mechanisms. In none of the three bodies is there a formalized presence of civil society representatives. The Advisory Economic and Social Forum of Mercosur, created by the Ouro Preto Protocol of 1994, was a result of economic actors’ influence, such as businesses, and to a lesser extent trade unions. In fact, it was only in 1995, with the creation of the Labour, Employment and Social Security Subgroup (SGT-10), that civil society issues were incorporated into Mercosur’s policies. What is more, during the 1990s, social issues discussed in Mercosur organs were mainly related to labour standards (SGT-10) and health issues (SGT-11, created in 1996).

In spite of this, social actors, particularly trade unions, have been mobilizing and engaging with the bloc. In some cases, the articulation of civil society organizations at the regional level precedes the creation of institutional spaces in Mercosur. For instance, the Southern Cone Trade Unions (CCSCS) was created in 1986, prior to the establishment of SGT-10. Jelin (2001: 94) points out that although civil society actors are for the most part absent in Mercosur’s institutional negotiations, ‘the field of collective actors and movements is one in which the impact and consequences of decisions made at the formal level of negotiations on regional integration is clearly felt’.

In the first decade of the 2000s the social dimension of Mercosur broadened. Issues of human rights, migration, family agriculture, the fight against hunger, culture and solidarity economy were inserted in regionalism. This turn towards social issues in Mercosur occurred in tandem with the increasing participation of social actors in domestic politics related to these topics in the member states, within the political context of centre-left and left-wing governments in the region in the first decade of the 2000s. These themes are associated with the notion of participatory regionalism, since they refer to issues in which there is a greater propensity for social participation. A corollary of this dynamic was the creation of agreements, rules and regulations that favour social cooperation. This dynamic has increased with the establishment of the Social Summits in 2006, and subsequently via the implementation of the Social Institute in 2007, the Strategic Social Action Plan (PEAS), the Unit for Social Participation (UPS) and the Action Plan for the Creation of a Citizenship Statute in Mercosur, all created in 2010.

In order to understand the changes observed in Mercosur in the 2000s, some authors have referred to the notion of ‘post-hegemonic regionalism’ (Riggirozzi and Tussie, 2012). According to this concept, regional initiatives began to put more emphasis on aspects of development, social participation and national and regional autonomy. Governments’ support for civil society involvement is related to this new perception of regionalism, which focuses not exclusively on trade, but also on a more social dimension. For instance, these renewed ambitions are identified in the positions of Argentine and Brazilian governments, which defended that regionalism – in addition to an economic dimension – should also assume social and regional governance facets (Briceño-Ruiz, 2012; Briceño-Ruiz et al., 2017). In fact, there was a common recognition by member states that regional integration and cooperation need more society involvement, which led to the expansion of opportunities for civil society participation in the structures of the bloc (Hoffmann, 2015). According to Briceño-Ruiz (2012: 185), the construction of the social dimension of Mercosur ‘has been a process in which state actors and civil society have worked together’. By opening channels of social participation within the framework of the bloc’s structures, this would make the process less identified with the preferences of political and economic elites (Martins, 2014) – something particularly relevant for the left-wing governments.

The first Mercosur Social Summit was held in Brasília in 2006, ² gathering five hundred participants from civil society (Silva and Martins, 2018). The Tucuman meeting, in 2008, had 940 participants and the Isla del Cerrito Chaco, in 2010, had the largest number of participants (1,500). Since 2006, 18 Social Summits have taken place every semester, simultaneously with Mercosur’s CMC Summits. The final declaration of every Social Summit was submitted to the chiefs of state in the context of the CMC meetings.

It is interesting to note that most of the demands presented in the Final Declaration of the First Mercosur Social Summit, such as that the Social Summits should be supported by the pro tempore presidencies and that mechanisms should be incorporated to incorporate social actors as observers in the GMC and CMC, were successfully addressed through the following years, until 2015. The last Social Summit in 2015 had as one of its results the development of the Mercosur Socio-Labour Declaration, subsequently subscribed by the presidents during the 48th Mercosur Summit of Heads of States.

In addition, according to Martins (2014: 113), further proof of some successful involvement of civil society in Mercosur is that ‘it was at the Social Summit in Córdoba, in 2006, that the idea of   developing a Social Action Plan for Mercosur was discussed for the first time. Submitted to the presidents, the proposal was approved and officially incorporated into the official agenda, giving rise to the PEAS.’ Among the PEAS objectives are: eradicating hunger, poverty and fighting social inequalities; guaranteeing human rights, humanitarian assistance and racial, ethnic and gender equality; and establishing mechanisms for regional cooperation to implement and finance social policies. PEAS and the Mercosur Citizenship Statute of 2010 could be considered as the two most relevant plans related to civil society at the Mercosur level. Both were conceived in a context of cooperation between civil society and governments. The role of experts as agenda setters in specific policy fields was reduced as a driving force behind social participation in Mercosur. Most of the dynamics that involved specialized social technologies were induced by the states or by processes initiated at the domestic level with state support. Contrary to the dynamic prevalent at SADC, experts have exerted limited influence in guiding social participation in Mercosur. The processes evidenced by the Social Summits indicate that the evolution towards a more social Mercosur originates from governments’ recognition of the challenges faced by the bloc and the intent to construct a more positive agenda through the involvement of domestic actors relatively more favourable to integration (Ramanzini Junior and Farias, 2014).

It is possible to note a constant support for Mercosur and its deepening by the actors participating in Social Meetings. In most cases, social actors have supported the strengthening of Mercosur and the convergence of sectoral policies among member states in a context where a sceptical vision of the bloc’s centrality to state foreign and economic policies arose (Martins, 2014). One can also verify that the Social Summit was an arena for critical positions on the trade agenda of the bloc, in particular with regard to the negotiations of a free trade agreement with the EU. Another topic observed in most Social Summits’ final declarations was regarding the Structural Convergence Fund of Mercosur (FOCEM), as social actors demanded governments to increase contributions to this fund, aiming to reduce economic asymmetries among the member states.

Since 2015, with the election of Mauricio Macri in Argentina and the impeachment of Brazilian President Dilma Rousseff in 2016, the building of a social Mercosur has lost momentum. One can verify a disintegration of civil society channels in Mercosur, evidenced by the fact that the last Social Summit was in 2015 and the 2017 Social Summit to be organized in Argentina was cancelled by President Macri (La Diaria, 2017). In this new context, the autonomy of the Social Participation Unit was reduced, becoming more subordinated to Mercosur’s Secretariat. One common feature of the right-wing political forces that emerged in Argentina and Brazil was a criticism of the post-hegemonic model adopted in the first decade of the 2000s, considered a left-wing ideological project, incompatible with the strengthening of Mercosur’s trade agenda. In recent years, national governments have deactivated spaces for civil society in regionalism, which demonstrates that the development of a social dimension in Mercosur is strongly dependent on the willingness of member states.

Civil society actors in Southern Africa and SADC Civil Society Forums

The Southern African Development Community (SADC), founded in 1992 and composed of 16 member states, ³ is the major sub-regional organization in Southern Africa. Driven by an intergovernmental decision-making system, its mandate has traditionally focused on both socio-economic integration – with prominence for the Southern African Customs Union (SACU), formed by five SADC countries (Botswana, Lesotho, Namibia, South Africa and Eswatini) – as well as political and security cooperation. Even though SADC is a state-driven organization comprised of a number of authoritarian states, ⁴ over time it has supported in key documents the strengthening of democratic governance and the participation of citizens when it comes to regional integration (Louw-Vaudran, 2019: 10). As stated in article 23 of the SADC foundation treaty, ‘SADC shall seek to involve fully the people of the region and key stakeholders in the process of regional integration’, indicative of the intention to include civil society sections in regional decision making.

Nonetheless, the involvement of civil society in Southern African regionalism predates SADC’s foundation. In the early 1980s – still in the context of its predecessor organization SADCC – an NGO Liaison Desk was established, set up as an initial channel of communication with civil society sectors regarding regional policies and development projects (Moyo et al., 2007). However, it was only in the late 1990s/2000s that stronger civil society channels were established at SADC level. Two main channels have been created to guarantee civil society engagement in Southern Africa regionalism: the Southern Africa Development Community Council of Non-Governmental Organizations (SADC-CNGO), formed in 1998 as a coordinator for NGOs at the regional level, and the SADC Civil Society Forums, which began to be organized in 2007 parallel to SADC high-level summits. In addition, at the national level, SADC national committees (SNC) have also been focal points for civil society involvement (Mukumba and Musiwa, 2016).

SADC Civil Society Forums, which have convened annually since 2005 and organized their 15th edition in August 2019, are the most recent avenue constituted by civil society actors to engage with regional decisions taken by SADC heads of state in the context of SADC high-level summits. Organized parallel to and independently from intergovernmental meetings, the SADC Civil Society Forum ‘provides space for NGOs, donors, churches, trade unions, social movements and other civil societies to act collaboratively on a wide range of development issues affecting the region’ (Mukumba and Musiwa, 2016: 1). By issuing resolutions to be submitted to SADC member states and the Secretariat, civil society actors aimed to also engage with the deliberations conducted at SADC level, providing their own input and perceptions about critical issues under discussion by governmental actors. The latest resolutions have covered a wide range of topics, including democracy and elections, decent working conditions, agriculture and nutrition, land reform, climate change, disaster response, health, youth employment, education and migration. Moreover, since 2012, the forum has promoted a campaign (‘The SADC we want’) to support a more people-centred approach to regional integration. Among its key pillars are the strengthening of key SADC institutions the SADC Secretariat, the national committees, and the establishment of a SADC Tribunal and Parliament (SADC Civil Society Forum, 2019).

Not only have SADC civil society actors been involved in SADC internal policies, they have also shown interest in SADC’s relations with third countries/blocs, exemplified by the Economic Partnership Agreement (EPA) negotiated with the European Union. For instance, an SADC-EU Civil Society Forum was organized in South Africa in October 2017, as a way of communicating with the civil society in both regions (SADC, 2017). Even though initiatives of this nature are relevant to familiarizing civil society actors with important aspects of trade negotiations, these actors have never been seriously consulted by SADC executives in the negotiation process. In fact, some civil society sectors in Africa have been very critical of trade liberalization with the EU (Godsäter, 2013). Furthermore, likewise across the negotiation stages, SADC civil society actors have remained detached from the EPA’s implementation process, initiated in February 2018. ‘While the EU has a mechanism for government–civil society engagement in the agreement, SADC countries do not, nor have they demonstrated a willingness to set one up’ (Mawowa, 2018: 1). Although concerns from civil society are seen in the case of trade negotiations with the EU, no significant complaints from civil society were identified when it comes to previous trade agreements, such as the one signed between SACU and Mercosur. This is for the most part due to its more technocratic nature and low range of products (Nkomo and Olmos, 2012).

The above example illustrates some of the challenges faced by civil society actors in the SADC, the main one being the fact that civil society demands that materialized in the resolutions issued during SADC Civil Society Forums are in fact marginalized from the SADC decision-making system. This is due to ‘the deeply rooted statism in Southern Africa that tends to not place a high level of importance on non-state actor engagement’ (Mukumba and Musiwa, 2016: 1). Given that the SADC was initially designed as a state-driven initiative, there is very little room for non-state actors to participate in the decisions taken by national governments’ representatives, even when the topic is particularly relevant for these societal agents. As pointed out by Khan (2006: 4), ‘most decisions are made in the Summit – of heads of state and government – without any information concerning the discussions being provided to any other structures in SADC’.

Differences between civil society and government agendas have also overlooked the output of the SADC Civil Society Forums, which has tended to focus on ‘soft issues’ – more relevant and closer to civil society sectors – instead of on the same issues discussed by SADC executives (Moyo et al., 2007). Furthermore, the fact that some civil society groups have adopted a confrontational approach regarding national governments – especially concerning human rights conditions and electoral processes – have made SADC executives suspicious of further engagement with civil society actors (Moyo et al., 2007). Finally, apart from the Civil Society Forums, civil society actors may also engage with SADC authorities via their national committees. However, not only are societal agents largely unaware of the existence of these national representations, these bodies often lack the capabilities to properly meet civil society demands (Mukumba and Musiwa, 2016).

Despite all the structural and political challenges for civil society’s involvement in regionalism, some civil society actors have succeeded in contributing to certain SADC policy agenda points. This is mainly due to two aspects: the level of expertise in the policy field, and the degree of confrontation towards SADC member states. On the one hand, civil society actors with strong knowledge in certain fields have been more successful at influencing SADC policies over time. As SADC governments sometimes lack evidence-based knowledge on some policies, they are more prone to communicate with non-state actors willing to contribute to these areas. For instance, SADC health and gender policies have been strongly influenced by civil society actors. In health governance, SADC executives often need assistance from non-state actors regarding HIV/AIDS-related issues. Therefore, civil society agents experienced in health issues can be quite influential in the formulation of SADC health policies (Reinold, 2019). Likewise, regarding gender policies, the SADC Secretariat has acknowledged the expertise of certain civil society organizations in this area, noting ‘that gender-based NGOs tend to bring great contributions and have a well-developed capacity for evidence-based research’ (Moyo et al., 2007: 54).

Although the SADC-CNGO and SADC Civil Society Forums have been singled out as SADC’s focal point for civil society participation, in practice, social organizations with real expertise on specific issues have exercised considerably more influence on SADC policy formulation (Godsäter, 2015). SADC-CNGO is not the only actor coordinating civil society participation at SADC level. Other civil society networks, such as the Southern African Treatment Access Movement and the Southern African Poverty Network, have also forwarded input from social actors to SADC authorities (Mukumba and Musiwa, 2016).

Acting in parallel to SADC-CNGO, networks involved in issues such as poverty elimination, gender and HIV/AIDS policies have pushed for debates and the exchange of ideas among key stakeholders, in order to disseminate knowledge and good practices that aim to provide relevant policy inputs to SADC authorities (Hulse et al., 2018). For instance, one of the initiatives co-organized by the Southern Regional African Poverty Network (SARPN), ‘Look, Listen & Learn’, has put strong emphasis on ‘promoting the use of Civil Society Organizations’ (CSOs) evidence in policies for food security’ (SARPN, 2005: 1), in particular policies adopted in SADC countries. This is grounded in the organization’s perception that ‘One of the important roles of CSOs is to hold international bodies accountable. CSOs need to ensure that they apply well-informed pressure to ensure that government commitments are acted upon and not approached at a leisurely pace, or even abandoned’ (SARPN, 2005: 2).

In this sense, SADC governments are more willing to dialogue with those civil society groups that have adopted a less confrontational approach. As stated by Nzewi and Zakwe (2009: 43), ‘there is the general view that CSOs that government interacts with are either those that are viewed by government to be less adversarial, or organizations that have more resources and capacity to engage with government’. Moreover, in some cases civil society leaders are themselves members of the region’s ruling elites, also lowering the chances of a confrontational stance against national governments. ‘(M)any Southern African NGOs are staffed by relatives or close associates of the ruling political elites. This phenomenon is particularly evident in Botswana, Swaziland and Lesotho, but occurs across the region’ (Söderbaum, 2007: 330).

Finally, it is also important to highlight that SADC member states vary in terms of their openness to engage with civil society, which is a reflex of the coexistence of democratic and autocratic regimes within the SADC. Therefore, while countries such as South Africa, Botswana, Namibia and Zambia have established a strong relationship with civil society actors at the national level, other SADC countries, such as Angola and Zimbabwe, are more reticent to open communication channels with social actors, given the strong level of internal contestation of their political regimes (Reinold, 2019). In fact, in some countries civil society has not only been marginalized, but harassed by national governments, undermining their overall involvement in regional policies (Söderbaum, 2007).

Civil society and regionalism in the Global South: A comparative assessment

After examining the development and challenges for civil society actors in Mercosur and the SADC, this section provides a comparative assessment of the two cases by offering an adjustable comparative framework to understand the particular dynamics of regionalism in the Global South (Ramanzini Junior and Luciano, 2020). Inspired by previous comparative analysis of civil society in African regionalism (Reinold, 2019), our comparison will draw on the following elements: regional institutional design, member states’ support for societal participation, and civil society resources. Key differences and similarities in civil society participation in regional integration will be pointed out for both cases. Lastly, we pay particular attention to the current dynamics in both regions, highlighting any evidence of regional (dis)integration that may impact the future of civil society participation in these regional organizations.

First, concerning the institutional design, the decision-making system of both regional organizations – which is in fact an overall trend in Global South regionalism – is marked by the dominance of intergovernmental and even interpresidential characters (Fioramonti and Mattheis, 2016; Malamud, 2005). ‘Mercosur represents a typical state-led regional integration process. Domestic drivers of integration are politically and geographically limited by the excessive emphasis on the role of the executive branches of member states. They remain the most important – and often the sole – actors defining the path of integration, via presidential diplomacy’ (Caichiolo, 2017: 133). Likewise, SADC regionalism was also conceived as a state-led project, in which national governments never expected to share competences with non-governmental actors (Aeby, 2019; Louw-Vaudran, 2019). In fact, the SADC’s decision-making system has remained in the hands of the Summit of the Heads of State and the Council of Ministers. Due to this key feature in both regions, non-state actors were in principle excluded from regional policy formulation.

Moreover, the context in which both organizations were established in the first place also stresses the limited channels for social participation. Both regional organizations were constituted in the context of open regionalism, which has essentially prioritized economic/trade integration. The only actors able to influence regional integration apart from national governments were economic actors, such as businesses confederations, mostly in consultative arenas such as Mercosur’s Economic and Social Consultative Forum. Nonetheless, in the 2000s a new regionalist wave (post-hegemonic regionalism) brought a window of opportunity for the inclusion of new dimensions in regional integration initiatives. Thus, non-commercial topics, including a social agenda, started to be discussed at the regional level. At the same time, member states acknowledged that civil society participation was a way to reduce the so-called ‘democratic deficit’ of regional organizations (Briceño-Ruiz, 2012; Riggirozzi and Tussie, 2012).

In this sense, in both cases a trend of establishing limited civil society channels in the 1990s and developing new and broader social participation spaces in the 2000s can be observed. In the latter decade, Mercosur supported the development of several instruments for social participation, such as the Social Summits and the Social Institute. Besides, Mercosur’s Secretariat received stronger support and relative autonomy to communicate with civil society actors, and the creation of the Social Participation Unit under the umbrella of the High Representative-General of Mercosur was a relevant instrument in contributing to the organization of the Social Summits until 2017. Meanwhile, although the SADC has been more timid in this respect, civil society initiatives were also established in the 2000s, such as the SADC-CNGO and the Civil Society Forums, even though these are not formally part of the SADC institutional framework. Nonetheless, in both cases it could be verified that the increase of rhetorical objectives related to civil society participation was not accompanied by the building of strong institutional channels to achieve such objectives. This means that executive actors have set high political ambitions to enable civil society engagement, without establishing the necessary instruments and regional institutions to achieve the objective of developing participatory regionalism.

Secondly, an additional element for comparison concerns member state support for civil society involvement in regionalism, which is associated with the level of democracy in member states, as well as the openness of executive actors to civil society participation. Tallberg et al. (2013) stated that the level of domestic democracy constitutes an important explanatory variable for understanding variations in the receptiveness of international organizations to civil society. Additionally, Acharya (2003: 388) has pointed out that ‘the emergence and role of regional institutions are often the product of domestic political institutions and structures’. In this sense, the transformations seen in both Mercosur and the SADC with regard to civil society participation derive from changes in domestic politics of member states, particularly the recent democratization process in Southern African and South American countries.

However, important variations can be seen when it comes to democratization and engagement with civil society in the two regions. While all current Mercosur members are considered either full or flawed democracies, only five out of 16 SADC countries are considered democratic nations (The Economist, 2019). The lack of stronger democratic institutions at the national level in most member states has made the SADC less open to the establishment of strong communication channels with civil society actors. This explains the timid channels established in the 2000s, especially when contrasted with Mercosur’s social participation institutions. Although most SADC member states have seen some benefits in building civil society channels in terms of their internal and external legitimacy, some countries – especially the more authoritarian ones – do not tolerate confrontational positions. For instance, in the SADC, some governments ‘have an innate distrust of civil society and often undermine its ability to play a significant role in democracy and development. Therefore, a number of SADC member states do not welcome collaboration with civil society organizations’ (Reinold, 2019: 70).

Thirdly, another element that has impacted the involvement of civil society in regional politics in Mercosur and the SADC is the level of influence and resources of civil society organizations in both regions. This dimension contains a strong difference between the two cases. Whereas in Mercosur political ideology has played a key role in enabling civil society participation, civil society involvement in SADC regional policies is primarily driven by knowledge-based expertise of NGOs in certain policy fields.

In the case of Mercosur, one can observe a strong and historical connection of left-wing parties with social actors at the domestic and regional levels. Given this prior ideological affinity, the ascension of left-wing governments in the 2000s in South America, denominated by some as the pink tide (Murray et al., 2009), and the reformulation of Mercosur institutions in the same period led to the establishment of stronger formal channels of civil society participation in the bloc. However, the more recent rise of right-wing leaders in South America, particularly in Argentina and Brazil, has undermined civil society actors’ engagement in the bloc, given that the new leaders consider them as ‘natural allies’ of left-wing political parties. Since Argentine President Mauricio Macri cancelled the Mercosur Social Summit in 2017, no similar events were organized by the bloc, which highlights how civil society participation was interrupted at the Mercosur level. On the other hand, in the SADC it is not ideology but expertise that has played a key role in civil society involvement in SADC policies. While the SADC-CNGO and SADC Civil Society Forums have been ignored by SADC executive bodies, civil society actors who possess significant expertise in relevant policy areas have been able to influence SADC decision making in their respective fields of specialization (Godsäter, 2015).

In fact, both cases demonstrated how participatory regionalism is affected by rhetorical contestation and disputes concerning the priorities of regional initiatives. As highlighted in the conceptual section, participatory regionalism is not synonymous with the convergence of visions among civil society and state actors. Notwithstanding the fact that when levels of contestation from civil society regarding national governments are higher – which has been the case in Mercosur since the emergence of right-wing presidents, and of some civil society actors in the SADC (mainly those more vocal on human rights and the electoral situation) – national executives either erode civil society channels or simply ignore the demands of social actors. Contrarily, when contestation levels are lower and either ideological affinity or knowledge-based expertise are identified – such as the cases of Mercosur under left-wing leadership, and the SADC in its relations with NGOs in certain policy fields (health, gender) – civil society actors are more able to influence regional policies and participate to some extent in the formulation of decisions.

Lastly, it is important to point out that regional organizations are not static bodies, since current developments may change the course of civil society channels at the regional level. In this regard, the latest developments in Mercosur have shown a process of disintegration in terms of civil society participation. The cancellation of Social Summits and the downgrading of the Social Participation Unit have reduced the formal channels of civil society participation in the bloc. If we understand that regional integration projects are inserted in a spectrum between ‘full-scale integration vs full-scale disintegration’ (Börzel, 2018: 478) and are essentially conceived by institution-building (Wunderlich, 2012), it can be argued that Mercosur has gone through a disintegration process when it comes to participatory regionalism. Meanwhile, the SADC – even though not fully composed of democratic member states – remains at an inertial pace in this respect, as the SADC-CNGO is still active and Civil Society Forums continue to be regularly organized, despite their low level of influence on regional decision making. This challenges previous scholarly assumptions that democracy at the domestic level is crucial for the involvement of civil society at the regional/international scale.

Our analysis has indicated that although other non-state actors are increasingly relevant, national governments remain the central actors in the Global South. While examining the dynamics involving the participation of civil society in regionalism, we have observed that social actors only engage successfully with regional organizations when there is a clear support from national governments. In this sense, the interaction between regional institutions and civil society does not disempower member states, as neofunctionalists would expect. At the same time, the position of member states regarding social participation in regionalism is not purely the outcome of domestic bargaining grounded in national preferences, as liberal intergovernmentalism would predict, but of specific characteristics that involve social participation, which have even made authoritarian countries agree with some levels of social participation in regionalism. In other words, social participation in regionalism in the Global South is associated with the goal of strengthening the capacity of member states and national governments, either by assisting in the provision of regional public goods, as seen in the case of the SADC, or via allowing the involvement of like-minded social movements, as is the case of Mercosur.

This discussion aims to contribute to the debates about the construction and consequences of the so-called regionalism from below. Our research corroborates Fioramonti’s (2014) argument that civil society can perform three roles not necessarily mutually exclusive in regional governance: legitimization, manipulation and contestation. At the same time, the observation was made that we cannot overestimate the impact of civil society as promoters of changes in regional policies that member states would never initiate. Actually, by performing two different case studies, our research has demonstrated that the chances of civil society actors to bypass national governments and exploit existing institutional channels in order to promote social changes in regionalism are quite limited. In this sense, one cannot expect a weakening of national governments vis-à-vis non-state actors within regional organizations in the near future. Social participation in Global South regionalism has in fact been both preserved and conditioned by the sovereignty of member states. In that respect, our empirical findings diverge from the second part of Acharya’s definition of participatory regionalism (2003), given that a decrease in the centrality of sovereign member states has not yet been observed in Global South regionalism.