From the paradigmatic to the practical battlefield : Southern development cooperation practices in a traditional aid hosting context

This article discusses the effects of the introduction of South-South Cooperation (SSC) practices in a traditional aid-hosting context. Using a Bourdieusian approach, it shows that SSC practices encountered traditional aid practices embodied in host actors’ engagement within SSC initiatives. Such encounters engendered a practical mismatch that affected SSC projects. The effects of the practical battle on project implementation underscores the determining role of host actors in shaping development cooperation practices. Finally, this article argues that changes in development cooperation practice, in contrast to changes in development cooperation paradigms, depend more on the hosts than on the donors and providers’ agenda.


Introduction
S outh-South Cooperation (SSC), although not a new practice, has gained momentum in the 2000s with an increasing presence in host countries through development projects and growing influence in traditional development cooperation arenas (Abdenur and Fonseca 2013;Amanor and Chichava 2016;Besharati and Esteves 2015;Eyben 2012;Eyben and Savage 2013;Mawdsley 2012;2019;Zoccal and Esteves 2018;Gray and Gills 2016). In this sense, Esteves and Assunção (2014), adopting a Bourdieusian approach, argue that the increased influence of the South in what was an acquired Northern territory has created a doxic, paradigmatic battle in the field of international development cooperation (IDC), as emerging countries challenged the structured, unbalanced power positions underpinning the traditional aid paradigm. However, while there is evidence of Southern providers transforming the traditional aid paradigms (Mawdsley 2018;Zoccal and Esteves 2018), it is not certain how SSC principles are translated into practice in host countries. The study of Brazilian and Chinese cooperation in Africa, for instance, demonstrates that Southern partners do bring new elements to the way programmes are negotiated and implemented, but equally so continuities from older experiences, reproducing unequal power relations within development encounters (Mawdsley 2012;2015;Scoones et al. 2016;Taela 2017;Shankland and Gonçalves 2016;Wolford and Nehring 2015). Referring to the Bourdieusian approach suggested by Esteves and Assunção (2014), in this article I argue that the introduction of SSC practices in traditional aid-hosting contexts engendered a practical battle explained by a mismatch between new practices and old expectations. By demonstrating that the mismatch of practices affected projects' processes and outcomes, this article also reveals the often underestimated role of host actors in shaping IDC practice.
To illustrate the challenges SSC practices encounter in traditional aid-hosting contexts, I focus on the history of IDC projects in the cotton sector of Benin. I draw on empirical data collected in Benin from October 2017 to August 2018 1 . The inventory of projects shows that SSC initiatives appear only from 2009 onwards, 45 years after the first cotton project was implemented. Before 2009, a tradition of hosting development projects was constituted, with shared understandings, norms, and values that were reproduced in subsequent encounters. Therefore, the introduction of SSC projects in such contexts represents an encounter with traditional practices. On one side, Southern providers introduce distinct managerial mechanisms and patterns of relationships, based on the SSC principles such as horizontality, mutual benefits, shared management and funding. On the other side, host actors' practices -embodied history as Bourdieu suggests -reproduce past experiences with traditional donors, marked by structured positions of donor and recipient, dominant and subordinated. As a result, a mismatch of practices arises and affects projects' processes and outcomes. Therefore, a deeper understanding of SSC practice requires a more thorough look at hosting contexts and their IDC history of roughly 60 years.
The remainder of this article is structured as follows. In the first section I present the conceptual framework based on previous research that demonstrate the growing influence of emerging powers in the IDC field. In the second section I present the consolidation of IDC practices in the cotton sector of Benin through successive cotton projects that presented similar characteristics throughout the decades. The third and fourth sections analyse the introduction of SSC practices in the hosting context and the outcomes of the encounter with traditional aidpractices. Finally, I discuss the possible outcomes of the practical battle and ways towards durable changes in the IDC field.
From the paradigmatic to the practical battlefield: Southern development cooperation practices in a traditional aid hosting context Rev. Bras. Polít. Int.,63(2): e004, 2020 Moreira 3 The doxic battles in the international development cooperation field The consolidation of IDC as an international public policy in the aftermath of the Second World War meant the constitution of an autonomous social field in Bourdieusian terms. The social field consists, as Bourdieu suggests, of a social space structured along power relations, objects of struggle, and rules taken for granted (Pouliot and Mérand 2013). Throughout the second half of the twentieth century, agents consolidated their positions in the field as dominant or subordinated -or donor and recipient -according to the capital they endow and that are relevant in that specific field (Esteves and Assunção 2014;Zoccal 2018). As one of its main mechanisms, the field's doxa, or the prevalent meanings and perceptions that are taken for granted and which legitimate practices (Bourdieu 2005), was constituted simultaneously. Budget support, unilateral flows of knowledge, and asymmetrical power relations became taken for granted and unquestioned practices of IDC.
In that period of consolidation, Southern providers of IDC occupied a relevant but not transformative position in the field, being unable to challenge the prevailing doxa. Zoccal (2018) demonstrates that the 1978 United Nations Conference on Technical Co-operation among Developing Countries and the resulting Buenos Aires Plan of Action, represented an attempt to transform the IDC doxa, but was hampered by developing countries' lack of economic capital (which along with social, cultural and technical capitals determine one's position in the field) to challenge the structured and consolidated positions in the field (Zoccal 2018, 75). The emergence of Southern countries as economic powers in the 2000s transformed the doxic opposition into a doxic battle. This means, as suggested by Berling (2012), that strategic mobilizations within the field had the effect of changing the taken for granted, unquestioned, meanings and perceptions, or, in other words, the field's doxa. Evidences of the doxic battle could be found, first, in changes in the aid paradigm that acknowledged SSC as an alternative form of IDC during the 2008 OECD High-level Forum on Aid Effectiveness in Accra, Ghana, and in the following OECD's summit in 2011 in Busan, South Korea (Zoccal and Esteves 2018).
Changes in the field's positions and doxa implies changes in the IDC practice in hosting countries. As Bourdieu suggests, and Maton explains further, one's practice is the result of relations between one's habitus and one's current circumstances, determined by one's position in the field (Maton 2012). Habitus is 'embodied history', in which "early experiences have a particular weight, because the habitus tends to ensure its own consistency and its defence against change that might call it into question" (Bourdieu 1990, 60). However, this does not imply that our practices are immune to change, since the "habitus changes constantly in response to new experiences […] and dispositions [which constitute the habitus] are subject to a kind of permanent revision" (Bourdieu 2000, 161) -event though such changes are never radical. Yang (2014) proposes that change can be envisaged under Bourdieu's theory of practice when there is a mismatch between field and habitus. This happens when social agents present a 'deviant trajectory' from the collective trajectory of the social agent's original class. Thus, the repositioning of emerging countries in the From the paradigmatic to the practical battlefield: Southern development cooperation practices in a traditional aid hosting context Rev. Bras. Polít. Int., 63(2): e004, 2020 Moreira 4 IDC field constitutes one of these deviations that creates a gap between the field and the habitus through the introduction of divergent practices by new agents in the field. The increased influence of emerging powers on the IDC architecture is widely acknowledged. Yet, we know little about how such paradigmatic changes become lived experiences at the level of project implementation in host countries.
Early development cooperation experiences and the birth of a tradition: structuring positions in the development cooperation field Benin is a country of 11 million inhabitants neighbouring Nigeria, Niger, Burkina Faso and Togo, in the Gulf of Guinea, West Africa. Agriculture accounted for 23% of GDP in 2018, but employed more than 40% of the total employment. Cotton dominates this economic sector, being the main export good and cotton ginning the main industrial asset (Ministère d'Etat chargé du Plan et du Développement 2018). Cotton was planted in the region that corresponds to today's Benin since the eleventh century but attempts of intensifying production only took place during the French colonial period   (Levrat 2008). However, these first attempts failed in making cotton the main export good. In 1960, the Dahomey (today's Benin) became independent but maintained the same economic development plans drafted by the French and operationalised them through the signature of multiple cooperation agreements (Anignikin 2010;d'Almeida-Topor 2002;Sotindjo 2008).
Therefore, the Republic of Dahomey signed the first IDC cotton project in 1964 with France to multiply production by four. In fact, the Cotton Development project (1964)(1965)(1966)(1967)(1968)(1969) provided a legal framework to enable French companies to continue their enterprise in the former colony and to reduce France's dependence on US cotton (Levrat 2009). In this sense, hierarchical positions in the field were built, based on coloniser and colonised relations. The object of struggle was primarily the control over Dahomey's resources. For example, to achieve its objective the project restructured the whole agricultural sector as it eliminated previous arrangements organised around local cooperatives for accessing rural credits, to set-up a new system in which only cotton producers could have access to the input credit system (Mongbo and Dossou-Houessou 2000).
At the end of the first cotton project, cotton had become the main agricultural product. In 1972, the World Bank launched the Zou-Borgou Cotton Project (1972)(1973)(1974)(1975)(1976)(1977) aimed at continuing and scaling up the French first initiative (International Development Association 1972). The objective was to maintain the pace of intensification of cotton production by providing the resources (both human and financial) to enable the sector to operate. However, the coup d'état of Mathieu Kérékou in 1972 and the adoption of a Marxist-Leninist regime in 1974 would hamper the desired outcomes. Kérékou's government nationalised the cotton sector and prioritised food crops instead, leaving no other option to the Bank to suspend disbursements and terminate the project in 1977. As a consequence, the national production decreased and so did government 5 revenues. By the end of that decade, the government started negotiating a series of World Bank rural development projects (Zou, Borgou I, and Borgou II) which would inject the required resources to reorganise and further develop the cotton sector: building ginning plants, funding rural extension services, promoting the use of agrochemicals, and ensuring the functioning of the input credit system. At the same time, the Marxist-Leninist regime started to fade. Benin signed their first Structural Adjustment Plan in 1986 which anticipated the democratic renewal of 1990 and the liberalisation of the cotton sector from 1992.
In the 1990s, the cotton development cooperation was marked by the return of the French with a focus on cotton research and agricultural practices. The PARAB (1996)(1997)(1998)(1999)(2000) and its sequel, PARCOB (2001PARCOB ( -2006 introduced the processes for a cotton breeding programme within the Cotton Research Centre (CRA-CF). During the implementation phase, France placed CIRAD researchers in Benin to work daily with Beninese agronomists and also awarded scholarships for local researchers to pursue post-graduate degrees at French universities, therefore providing the resources for INRAB's human resources development. PADSE (1998PADSE ( -2005, while not working exclusively on cotton, left an important footprint because of the widespread use of staggered pest control in cotton farming systems. Although the French projects had a smaller scale than the World Bank projects, their support contributed to perpetuate the pattern of relationships that early projects introduced and thereby build up the doxa. Indeed, these early initiatives made taken for granted the idea of a project being an extra-ordinary financial resource that is locally managed and vital for the normal functioning of national public institutions. In the meantime, the World Bank focussed their support on institutional reforms towards the liberalisation of the economy. In the cotton sector, the liberalisation resulted in the multiplication of private actors engaged in ginning and provision of inputs. This led to the creation of the Association Interprofessionnelle du Cotton (AIC) in 1999 gathering ginners, input suppliers and farmers. In this context, a new World Bank project was signed to "provide the necessary support to private sector operators and their institutions to fill the void left by the privatisation of the SONAPRA's [then a public company] ginning activities" (World Bank 2001, 4). Alongside World Bank's project, France launched the Support to the Reform of the Cotton Sector (2004)(2005)(2006)(2007)(2008)(2009). The largest part of the project's budget was allocated to the realisation of technical studies on the functioning of the cotton sector, which were not realised after all. On the other hand, a smaller share enabled the AIC to set up essential services to run, such as a communication team. Although the World Bank and France used different funding mechanisms (credit for the former, grant for the latter), both projects consisted of direct financial support without which the AIC could not function. Indeed, these projects construed the idea of a project being the source of financial resources to be locally managed and which are vital for the functioning of institutions. By doing so, these projects reproduced the patterns of IDC practice that earlier projects set-up. This pattern has perdured with slight changes after the upsurge of projects post-2004.
In 2004, the main African cotton producing countries created the Cotton 4 group to express their concerns at the World Trade Organisation (WTO) regarding market distortions created From the paradigmatic to the practical battlefield: Southern development cooperation practices in a traditional aid hosting context Rev. Bras. Polít. Int., 63(2): e004, 2020 Moreira 6 by developed countries' subsidies. Benin, Burkina Faso, Chad and Mali came together for the first time at the WTO Cancún Ministerial Conference in 2003, leading to the creation of the Cotton Sub-Committee in 2004. This has put cotton at the core of aid-for-trade topics and, as a consequence, led to an unprecedented increase in the number of projects and partners in host countries from 2005 onwards. In this context, GMZ-COMPACI (2006, SNV-PROCOTON (2006, and USAID-WACIP (2006-2013) became the main cotton initiatives in Benin. The three projects aimed at developing the cotton sector either by creating parallel chains of production or by enabling the existing organisations to perform their tasks through financial and technical support. WACIP, and PROCOTON, especially, aimed at supporting -but in fact, enablingprocesses that already existed or were supposed to exist in order for the sector to function: WACIP disseminated conservation agriculture practices to farmers, thereby supporting the national rural extension services (United States Agency for International Development/West Africa 2010); and PROCOTON supported the village cooperatives by providing material resources and organising trainings on the effective management of organisations (Wennink et al. 2013).
These projects left a mark in the local context. For example, host actors linked USAID-WACIP to the dissemination of integrated soil management techniques -even though these techniques were known by agronomists and some farmers, the project enabled its widespread. Cooperatives' representatives also remembered the 'good times' of SNV-PROCOTON, when the headquarters were equipped, and the staff had opportunities to be trained. Although still inscribed within the traditional aid paradigm, these projects were different from earlier initiatives. On the one hand, the cotton sector of Benin was in a different stage of development that required a different kind of support, with more focus on capacity-building than on direct budget support. On the other hand, the aid architecture had also evolved, moving the focus of the initiatives in the rural sector from state-led initiatives in the 1970s, to other themes such as poverty alleviation, sustainable livelihoods, gender equity, good governance, from the 1990s (Ellis and Biggs 2001). This implied that project support would go directly to the private sector, farmers' organisations, and ginning operators, instead of public entities.
Yet, the Northern projects post-2004 still meant the provision of necessary resources that lacked in the hosting context; but the receiver end changed. This means that the change in project approaches did not affect the agents' positions in the field. While there has been a change in the recipient side, the power relations between dominant and subordinated continued. As Sneyd suggests, the upsurge of projects following the C-4 initiative at the WTO was a way to compensate for the continuation of agricultural subsidies in the North, a claim that developing countries have been expressing at least since the 1960s (see Sneyd 2011). In this sense, IDC practice at the local level continued to be the reverberation of economic struggles at the global level. As such, the reconfigurations in the international arena were also reflected in hosting contexts, as the cotton sector of Benin began to host projects brought by providers from the Global South. South-South Cooperation project processes, which claim to be different, were then inserted in a context that already had a tradition of development projects and associated expectations. In the next section The SSC arrangements introduced by Southern providers were different from what Beninese host actors were acquainted to. A common characteristic of the Southern cotton projects was a focus on capacity-building and knowledge sharing, while financial support remained exclusive to North-South projects. This distinction brought forward new practices. For example, study tours in the provider country emerged as a novelty, since this was not an activity ever carried out by traditional donors. Interestingly, the United States being a major producer of cotton could have conducted study-tours in their cotton belts as part of USAID-WACIP project. However, study tours under USAID projects took place in Benin's neighbouring countries. This could be understood as a search for similar contexts; nevertheless, it remains different from the SSC context-similarity claim, which is tied to the principle of horizontality in the provider-host relationship. The WACIP project did not aim to share or exchange the US experience. Instead, the project kept dissociating donor and hosts, maintaining a cleavage between the hierarchical positions in the field. The Southern projects, by contrast, created new provider-host relationships that differed from the traditional donor-host interactions in the field. Even though these relations were not necessarily horizontal, such practice represented a rupture with traditional ones.
Despite their primary focus on knowledge sharing, Southern providers also introduced material resources, thereby reproducing traditional aid practices -although at a smaller scale. In Benin, this was mainly the case for the Chinese, Turkish, and Indian projects. The Indian C-TAP project built a ginning factory for the exclusive use of Cotton Research. Brazil's C4+Togo project also planned to provide equipment to enable mobile rural extension units. Nevertheless, this common trait with Northern providers disappears at the level of the availability and management of financial resources, as a project coordinator explained: [SSC] is not the same approach. Often, they come with their money and they spend their money. The Chinese come with their money, they spend it. The Turkish, they come with their money, they spend it. The Indian, they come with their money, they spend it... the Brazilians, they are even more serious, because they do not even come with money [laughs] it's from [Brazil] that they say what to do with the money. It's more serious. 2

9
The differing availability of resources and the extent to which Benin had to financially contribute to the project activities, drew the attention of host actors, who, based on past experiences, took for granted that a project entails the provision of all necessary resources. As such, a participant enjoyed the training he attended in India, but regretted the long travels and the lack of resources of the project: "not even the money of the Brazilians […] it is not a lot." 3 Similarly, a project coordinator had to deal with the dissatisfaction of participants who criticised the small size of the vehicle for a monitoring mission in the bush as part of the C4+Togo project. According to a project expert who navigated across multiple paradigms of aid, the Beninese counterpart expected the project to provide the vehicle, even though they were the ones who had to, as stated in the project document. He explained that the resistance to financially contribute to projects usually makes donors concede to give the funds. This strategy did not work with Brazilians though, to the surprise of host actors.
In addition to having comparably fewer resources than traditional donors, the principles of management also distinguished SSC practices. Because of past experiences, host actors were used to having a budget to manage for conducting project activities in the field that gave them more room for manoeuvre. The fact that none of the Southern partners transferred a budget to the host institution represented a rupture to what had been done previously. Similarly, Southern projects did not set up project teams in Benin or hired local staff to the same extent as other projects did. China's STCP project team was small and composed of Chinese expatriates that interacted seldomly with the Beninese counterparts and Brazil's C4+Togo, hired a financial assistant in 2018, one year before the end of the implementation period. The managerial approach went against their 'tradition', as a project participant argued: From a logistical point of view, [SSC] style of management does not consider our traditions here [...] I take a simple example, in a training, there are trainers whom you invited, there are also producers whom you invited. Now, in our traditions here, the trainer will be paid more than the producer. Already as a basic principle. But in their project, we found that it's the opposite. Producers are paid more than trainers. That 's unacceptable. 4 Interestingly, when the interviewee above mentions a disrespect of traditions, he did not refer to Beninese culture. The tradition mentioned here is the tradition of development projects. In such, it provides a testimony of the traditional IDC doxa and corresponding practices, nurtured in Benin since the 1960s. According to the project managers, farmers were paid more because they were coming from their villages to the city. By contrast, the participation of Beninese agronomists as trainers was included in the budget as a national contribution in the form of working hours. Yet, the trainers expected to be paid separately, according to the arrangements in other projects: "the other projects are not like that. The payment to the trainers is put aside in the budget. The trainer needs to be paid more than the trainee. It is like that." 5 With regards to the diversity of SSC projects, host actors do not view the Southern providers as a homogenous bloc. In part, this stems from their pre-conceived perception on country partners. While Brazilians and Indians were praised by their expertise and methodologies, the Chinese "worked too much in isolation" 6 and their experts lacked legitimacy from the perspective of host actors because of the misunderstandings regarding scientific methods applied 7 . In a specific activity, host actors complained that the Chinese experts were not open to their suggestions nor willing to consider the local knowledge of Beninese researchers 8 . Similarly, the Turkish experts were criticized for bringing prêt-à-porter packages disregarding the intricacies of the local context. Such practices contradicted the SSC principle of horizontality, meaning that Southern providers also reproduce traditional aid patterns of asymmetrical relationships.
These perspectives of SSC practice emerging from host countries mean that the volume of resources and the ways of managing projects introduced by Southern providers did not meet the expectations of host actors. Particularly because their expectations reflected past experiences in similar settings. From the perception of Beninese actors, a project represents the provision of resources. This was also referred to as 'no project, no job' -an expression that appeared during the first projects between 1960 and the 1980s (Mongbo and Dossou-Houessou 2000). The introduction of resources was what Beninese practitioners expected before engaging in an IDC project, but they did not find it in Southern projects.

When Southern practices meet the prevailing doxa
The introduction of IDC practices by Southern providers distinct from usual practices in the field had an impact on projects' processes and outcomes. The mismatch between SSC practices and the prevailing doxa led to delays in project implementation, misunderstandings about arrangements and even resistance to participate in projects' activities. Paradoxically, this mismatch between hosts and providers' practices also created opportunities for course correction and eventually, to fitting implementation through unexpected ways.
Southern projects expected a greater financial contribution from Benin; a feature that goes against past experiences and the expectations of host actors. For instance, as part of the construction of a ginning plant under India's C-TAP Project, the government of Benin was responsible for the provision of electricity. However, the host's request for funds encountered difficulties to be approved by the government and the operationalisation of the research ginnery had been delayed. From the 5 Focus group with host actors on 13/11/2017 6 Interview with a host actor on 02/08/2018 7 Focus group with host actors on 13/11/2017 viewpoint of project management, the problem occurred because the Beninese contribution was not written down in the project document. In order to enable the Beninese agronomists to start using the ginnery right the way, the C-TAP Project bought a generator to be used as a temporary solution. Both parties decided to extend project's implementation period to give time to the Beninese counterpart to finish equipping the ginnery.
For the same reasons, the construction of China's STCP Project training centre was delayed. In the end, the Beninese counterpart provided the land and the connection to water and electric suppliers, which costed approximately 197 million FCFA (roughly USD 335,000). However, the delays created an opportunity for parties to reassess the overall objective of the project, as a project stakeholder explained: [The Chinese] quickly realised that the transfer of technology is not easy, that we, we have our technology and they, they have their technology and that it is not necessarily a problem of technology, it is much more about other problems that would be good to solve. After the analyses, on their side as well as on our side, it was about how to improve the technologies that exist here, trying to see what their shortcomings are, the strengths, and weaknesses. Then comes the focus on the mechanisation of agriculture. How are we going to mechanize agriculture so as to preserve soils and achieve the expected results in a sustainable way? 9 In the second phase of the project, the Beninese counterpart succeeded in shaping the project objectives according to their own interests. They would still use the facilities of the training centre, but the project leadership has shifted, since China and Benin envisioned the mechanisation differently: We are going to recruit a mechanic and a specialist in mechanisation to work with the [Chinese] to inculcate our new vision of mechanisation. It is no longer the systematic ploughing; it is the minimum ploughing. Our mechanic will work with them to still get something out of this experience. 10 The implementation of Brazil's C4+Togo Project also went through unplanned ways due to a mismatch of expectations. The project planned to build five demonstration fields across cotton producing areas to function as vitrines for the farmers. This was to assess the benefits of the adoption of direct seeding, cover plants, and crop rotation in cotton farming systems. It represented a big step from the first phase of the project. In addition, this made the C4+Togo become the most important SSC initiative in the cotton sector of Benin, with a large presence in the field. The demonstration fields needed to be secured with fences -which are unusual in Beninese rural landscape -because the cattle of pastoralists particularly appreciate the cover plants. To ensure that the animals would not compromise the experiments, the fence was a necessary condition From the paradigmatic to the practical battlefield: Southern development cooperation practices in a traditional aid hosting context Rev. Bras. Polít. Int.,63(2): e004, 2020 Moreira 12 for the successful implementation of the project. However, the mechanisms put in place for the execution of activities were inadequate. The funds for the construction of the fences should have come from Brasilia, following complicated administrative procedures involving UNDP due to a lack of specific legislation in Brazil to allow transfer of funds via IDC projects (Ferreira and Moreira 2018). The Beninese counterpart had to proceed with the construction of the fences and present the invoices to the Brazilian Cooperation Agency. But the non-conformity of the paperwork sent to Brasilia has created an imbroglio that compromised the realisation of activities and undermined the relationships between partners. In August 2018, the reimbursement of the services was still outstanding. The Beninese counterpart acknowledged the importance of the techniques that the project was supposed to share. Nevertheless, the project was in a dead end. The solution found was to incorporate the costs in the internal budget, despite resistance from host actors to do so. I asked my administrative advisor to allow us to add project activities as internal activities. Everyone recognized that the [project activity] is good. If the provider does not want to fund because of these administrative problems, we forget that and we take it nonetheless [...] anyways, the activities will bear the logo of the C4 + Togo, but in reality, there is no such thing. So, we need to find a way to mobilize funds to finance ourselves and let the project do whatever they want [...] but we must not remain dependent on the project […] We are going to do it anyways. We have the means to do it, so we'd better do it. 11 Through unexpected ways, host actors incorporated project activities, signalising ownership of the project's main product. In addition, Benin intended to have 20 instead of five demonstration plots in the country sharing Brazilian techniques, thereby scaling up project outreach. The combination of relevant techniques and favourable political context made it possible. In June 2018, the Government of Benin had published a decree that increased the funds for agricultural research, giving more room of manoeuvre for researchers within the cotton sector and autonomy from international partners. The C4+Togo Project continued its own path towards completion. In the WTO's 31 st Round of Consultations on Cotton Development assistance, the Beninese delegation praised Brazilian support and recognised the contribution of Brazil to the development of the cotton sector. With regards to the C4+Togo Project, the coordinator expected to finish the project by the end of 2019 and turn the page with a new bilateral project setting a focus on lint quality standards, thereby continuing to write the history of IDC projects in the cotton sector of Benin.

The practical battle and durable changes in development cooperation practice
To summarise, the economic development of emerging economies enabled them to change their positions in the IDC field. They shifted from positions of recipients to providers and From the paradigmatic to the practical battlefield: Southern development cooperation practices in a traditional aid hosting context Rev. Bras. Polít. Int.,63(2): e004, 2020 Moreira 13 altered the field's structures. At the local level, this meant the arrival of new agents in the field, bringing novel practices that diverged from hosting actors' expectations shaped by the field's doxa. The appearance of new practices in the hosting context generated a mismatch between agents' habitus and the field. This situation is one of the conditions under which Yang (2014) envisages change within a Bourdieusian approach. This mismatch engendered a 'practical battle' between traditional development practices reproduced within SSC initiatives by hosting actors, and actions inscribed under SSC principles, promoted by Southern providers. The problems that Southern providers encountered during project implementation, discussed above, are effects of the practical battle and could be observed in every SSC project in the cotton sector of Benin.
Whether the practical battle will lead to durable changes in development practice remains an unsolved question, especially under a Bourdieusian approach. First, as suggested by Yang, the agents whose habitus mismatches with the field's structures tend to re-align their practices to what is already taken for granted. This means that Southern providers tend to reproduce traditional aid practices at the local level, as it was observed in Benin and in many other contexts (Shankland and Gonçalves 2016;Scoones et al. 2016;Taela 2017;Wolford and Nehring 2015). Secondly, the historical analysis of cotton projects demonstrates that host actors have more power in shaping projects' processes and outcomes than the donor, provider, or the nature or design of a project alone. Hence, durable changes in practices cannot happen without the participation of hosting countries. Yet, with the emergence of Southern powers, the restructuring of positions in the field has had little effect on the position of countries who did not make the move from recipient to provider but remained exclusively hosts of IDC projects. Certainly, their relationships with other agents within the field changed, but they remain dependent on external resources and occupying subordinated positions at the global level.
Going further, the defining influence of host agents on projects' processes indicates that the IDC field may be structured differently at the global and local level. As we move the analysis from high-level meetings to project implementation, the weight of economic capital in structuring the field and practices seems to dwindle. As research on the doxic battle suggests, the increase of economic capital of emerging economies allowed them to redraw the field's structures (Esteves and Assunção 2014;Zoccal and Esteves 2018;Zoccal 2018). However, more locally contextualised forms of capital, such as social capital, appear to be more important at the local level to ensure both project implementation and impact beyond project timeframe. For instance, because host actors are part of more durable networks of relationships in the local context than donors and providers, they have more influence in shaping project processes and outcomes. This underscores the often-underestimated agency of host actors in project implementation. In addition, this perspective argues in favour of a greater integration of host countries in IDC arenas at the global level to make both paradigmatic and practical changes become durable.

Conclusion
In this article I discuss possible outcomes regarding the implementation of SSC projects in traditional aid-hosting contexts. Although not new in the IDC field, SSC gained importance in the 2000s, reaching more partners and covering more topics. The expansion of SSC enabled a spread of an alternative set of IDC practices in traditional aid-hosting countries that had only known the traditional aid paradigm hitherto. Such was the case of the cotton projects in Benin, which hosted the first Southern cotton project 45 years and 18 projects after the first initiatives took place in the cotton sector. The making of cotton as an export crop in Benin was a colonial project that international donors continued beyond independence by means of development aid. The first initiatives Benin hosted were thus marked by colonial relationships. The international development field was thereby structured with the dominant and subordinated positions of donor and recipient, whereby such power asymmetries resonated in projects' processes and practices. In the cotton sector of Benin, this is illustrated by a pattern of dependence on external funding and the motto 'no project, no job' that emerged among the institutions of the cotton sector from the 1970s. As suggested by Bourdieu's theory of practice, early experiences are of particular importance in shaping subsequent actions. Benin continued to host North-South projects that incorporated the transformations in the aid paradigm but did not completely modified the structured positions settled in its origins. Hence, in spite of changes in the aid doxa at the international level, these structured positions and unbalanced relationships remained unchallenged in the field until the emergence of the Global South as an economic global power and the widespread of SSC practice in traditional aid-hosting contexts from the 2000s.
This article demonstrates the difference pace and configurations between the doxic battle at the global level and the practical battle, at the level of project implementation in traditional aid-hosting contexts. Herein, the introduction of different IDC norms and practices by Southern providers implied the encounter and conflict with practices that had become self-evident, taken for granted, and thereby formed traditions in the field. In the cotton sector of Benin, various projects implemented by Brazil, China, India, and Turkey brought new types of activities with distinct institutional arrangements. One main rupture with traditional donors concerns the financial management of projects. While North-South projects have traditionally provided the necessary resources to enable the normal functioning of the cotton sector, SSC projects created structures with different aims and mechanisms, which required a further engagement, including a financial one, of the host country. As a result, the transposition of SSC principles from high-level meetings to everyday project management practices was affected by aid-hosting traditions. These traditions were externalised by host actors when engaging in SSC projects. Doing so led to misunderstandings and sometimes resistance to participate in projects' activities, making projects' outcomes even more unpredictable than in traditional aid practice. On the other hand, the introduction of new processes in the development practice landscape in Benin also enabled windows of opportunities for course corrections and indicates a certain flexibility and adaptiveness of SSC projects.
Finally, the outcomes of the encounter of SSC with traditional hosting practices highlights the determining role of host actors in shaping projects' processes and impact. More precisely, it suggests that the dominant-subordinated positions may invert as we go from the level of international meetings to the scale of project implementation. This means that changes in IDC practice, in contrast to changes in IDC paradigms, depend more on the hosts than on the donors and providers' agenda. Consequently, from the hosts' perspectives, North-South and South-South projects, although distinct, remain seen as opportunities to access necessary resources that are deployed according to enduring objectives whose timeframes transcend projects' periods of implementation. In this sense the widespread of SSC actions in traditional aid-hosting contexts is welcome because it supplements the repertoire of dispositions from which hosts actors draw to respond to their everyday challenges.