Skip to navigation – Site map

HomeIssues10Part 2 - Urban GovernanceInformal Governance: Comparative ...

Part 2 - Urban Governance

Informal Governance: Comparative Perspectives on Co-optation, Control and Camouflage in Rwanda, Tanzania and Uganda

Claudia Baez Camargo and Lucy Koechlin
p. 78-100
Translation(s):
Gouvernance informelle : Perspectives comparatives sur la cooptation, le contrôle et la dissimulation au Rwanda, en Tanzanie et en Ouganda. [fr]

Abstract

This article applies a novel conceptual framework to characterise and assess the repertoire of practices used by informal networks to redistribute power and access to resources. These distinct norms and practices are typologised as co-optation, control, and camouflage. Co-optation involves recruitment into the network by means of the reciprocal exchange of favours. Control is about ensuring discipline amongst network members by means of shaming and social isolation. Camouflage refers to the formal facades behind which informality hides and is about protecting and legitimising the network. All three are relevant to a more fine-grained understanding of corruption and its underpinnings. Findings from our comparative research in three East African countries (Rwanda, Tanzania and Uganda) suggest that these informal practices are relevant to an understanding of the choices and attitudes of providers and users of public services at the local level. Adopting this analytical lens helps to explain the limited impact of conventional anti-corruption prescriptions and provides a basis to develop alternative strategies that harness the potential of social network dynamics to promote positive anti-corruption outcomes.

Top of page

Index terms

Geographic keywords:

Africa Sub-Saharan, Tanzania, Rwanda, Uganda
Top of page

Full text

The research projects discussed in this paper are funded by UK aid from the UK government. However, the views expressed do not necessarily reflect the UK government’s official policies.

1. Introduction

1The literature dealing with corruption has typically based its analysis on a principal–agent model of decision-making, which assumes corruption happens as the result of flawed incentives on the part of the agent and weak monitoring capacity of the principal (Klitgaard, 1988; Rose-Ackerman, 1999). Policy solutions based on this conceptualisation prescribe measures such as a tougher criminalisation of corruption and stronger monitoring mechanisms in the public sector. The lack of success of conventional anti-corruption measures in many countries, including in sub-Saharan Africa, has subsequently led other scholars to conceptualise corruption as a collective action problem , which although providing a compelling framework to explain the unyielding persistence of high levels of corruption has nonetheless failed to provide alternative policy solutions. While the conceptual debate on this topic continues (Heywood, 2017; Mungiu-Pippidi, 2017), a growing consensus points to the need to collect empirical evidence in order to better understand the practices underpinning corruption, which evidently persist despite structural and institutional reform (Baez-Camargo and Passas, 2017).

2Regarding African countries, the lack of a context-sensitive approach to anti-corruption measures is not due to a lack of data. In fact, a vast body of literature on Africa has illustrated extensively particular features characterising African political regimes (Bayart, 1993; Bratton and Van de Walle, 1994; Chabal and Daloz, 1999) as well as particular social norms and practices among citizens that in one way or the other are associated with the prevalence of high levels of corruption (Hyden, 1980; Olivier de Sardan, 1999; Blundo and Olivier de Sardan, 2006). It would therefore be useful to work towards developing better systematic approaches linking the evidence on local drivers of corruption to the wider debate on anti-corruption. Such approaches should avoid a purely relativistic perspective while ensuring that generalizable dimensions are well specified and allow for contextual variation.

3Relevant questions to be asked include: How are the decisions and practices of key national and subnational actors shaped? What are the practices that individuals find useful in addressing their problems and needs? What are the networks linking up different actors, and how are they articulated? How do these networks and their practices underpin corruption?

  • 1 Empirical evidence presented in this paper stems from the findings from two research projects of th (...)

4In this chapter we present empirical evidence1 aimed at identifying those practices and norms that inform the agency of local actors, whether they be political and business elites, bureaucrats and providers of public services, or normal citizens, in three East African countries: Rwanda, Tanzania and Uganda.

  • 2 For a discussion of a non-normative, inductive approach to governance see Blundo and Le Meur (2008) (...)

5In the analysis of the empirical evidence from the case study countries we favour an approach to the characterisation of local governance processes that moves attention away from conventional categories and dichotomies (such as state vs. non-state actors, public vs. private, or sector-based assessments), shifting our analytical attention to the socially grounded practices that transcend the public sector. 2 That notwithstanding, we make use of the term ‘informal governance’ because, although it reproduces dichotomies that are empirically and conceptually debatable, it connects closely to the manner in which corruption is problematised among international development practitioners, who constitute a key audience for our research. In this sense, our use of ‘informal’ does not imply a fundamental divide between different spheres, but signifies the wide and contingent range of repertoires that actors have at their disposal, which transcend and straddle essentialist distinctions between ‘the public’ and ‘the private’.

  • 3 The projects used qualitative data collection methods with the support of a team of local researche (...)
  • 4 The concept of social networks is sociologically highly loaded—for a critique of social network ana (...)
  • 5 See Olivier de Sardan (2008), who coined the expression ‘real’ governance.

6In this chapter, we build on the evidence emerging from the ongoing research projects, discussing data collected from focus group discussions (FGDs) and interviews conducted across the three countries—namely, Tanzania, Uganda and Rwanda.3 The data strongly indicates the importance of social networks linking up actors that shape and are shaped by particular patterns of informal governance. Thus, social networks emerged inductively as heuristic entry points for a better understanding of prevalent practices and norms as well as lines of inclusion and exclusion of actors.4 Throughout the chapter we pay particular attention to the role of social networks as articulations through which social norms are enacted and reiterated. Through this analytical lens we seek to gain a better understanding of ‘real’ governance processes5 and the ways in which they underpin or reproduce corruption.

7The chapter is organised as follows: Section 2 provides general background information about the prevailing conditions across the three countries with regard to the fight against corruption. Section 3 outlines the conceptual approach used for uncovering locally determined drivers of corruption and the general traits characterising social networks in the three countries. Next, we describe how networks utilise the practices of co-optation (Section 4), control (Section 5) and camouflage (Section 6). Finally, in Section 7 we reflect on the implications of our findings.

2. Background Information on Corruption in Rwanda, Tanzania and Uganda

8East Africa presents excellent opportunities to explore what may be important factors impinging on the effectiveness of anti-corruption efforts. Regional trends point to sharp differences in performance that are reflected in the three cases included in our research.

9At one extreme, Uganda represents a modal pattern in the region, involving endemic corruption in spite of extensive reforms and anti-corruption interventions implemented by the government of President Yoweri Museveni. Instances of bribery, favouritism, absenteeism and gift-giving are widespread, with significant negative impacts on the provision of basic public services. The situation is confirmed by research findings in which Ugandan participants corroborate the perception that corruption is rampant and has become a normalised feature of daily life.

10In Rwanda, the government led by President Paul Kagame is considered to have achieved significant success in controlling corruption by means of radical public sector reform, enforcement of traditional accountability mechanisms, and the introduction of locally designed governance tools (Golooba-Mutebi and Booth, 2013; Bozzini, 2014). Strong political commitment is operationalised in a ‘zero-tolerance’ approach to corruption based on harsh punishments and their strict enforcement, coupled with mass campaigns sensitising the public about the value of anti-corruption measures and integrity, which have resulted in Rwanda being categorised among the least corrupt countries in Africa.

  • 6 A citizens’ assessment and expectations survey published in September 2016 reported that ‘96 per ce (...)

11In the case of Tanzania, high levels of corruption have pervaded all aspects of public administration for decades, becoming a constant feature in the experience of citizens in their interactions with the state, especially when it comes to the delivery of public services. However, since the 2015 election, the government led by President John Magufuli has vowed to renew anti-corruption efforts, and has taken significantly bold actions to punish and prevent illicit behaviours. Although some of the measures undertaken by the current administration have generated national and international criticism on the grounds of being perceived as detrimental to essential democratic rights and civil liberties, it is nonetheless remarkable that President Magufuli’s promise to fight corruption has received overwhelming support from Tanzanian citizens.6

12These three distinct anti-corruption outcomes are consistent with the recent East African Bribery Index (EABI), which found that 81 per cent of respondents in Uganda describe the level of corruption in their country as high, whereas in Tanzania 44 per cent of respondents describe levels of corruption in their country as medium. A majority of Rwandan respondents—61 per cent—describe corruption in their country as low. Furthermore, the EABI also indicates that Ugandans share a grim assessment of past and future performance, as 59 per cent of respondents believe that corruption has increased in comparison to the previous year and will continue to increase during the following 12 months. In contrast, a positive outlook was shared among a majority of Tanzanian and Rwandan respondents (70 per cent and 61 per cent, respectively), who believe that corruption has decreased in comparison to the previous year and will continue to decrease in the subsequent 12 months (70 per cent of respondents in each country) (Transparency International, 2017).

3. Conceptual Approach to Social Networks: Solidarity, Lock-in and Protection

13Research findings from the three countries confirm a vast body of literature regarding the role of social networks in East Africa and beyond: strong personal connections constitute an effective mechanism with which to pool scarce resources and provide access to goods, services, resources and even career opportunities in contexts of resource scarcity and weak state performance (Jauregui, 2014; Ruud, 2000; Chang et al., 2001; Grodeland et al., 1998; Ledeneva, 1997; Sneath, 2002). The findings concerning social networks deserve to be further emphasised in the context of the present chapter because they provide valuable insights about local social practices that are linked to the prevalence of corruption but are nevertheless not normally taken into consideration in conventional anti-corruption programming.

14Whereas it is clear that individuals belong to multiple, overlapping networks anywhere, the type of networks that are most relevant for this research are those to which the individual feels bound by obligations that are enforced and sanctioned by stringent social controls. Such networks are the social equivalent of what political science refers to as the leader’s ‘winning coalition’ (Mesquita et al., 2005), meaning those constituencies whose support is most essential for maintaining support and legitimacy in the case of the political leader, and status and social recognition within the community in the case of the individual. They are the core networks, or reference groups (Koni-Hoffmann and Navanit-Patel, 2017), in which strict social norms are enforced and the individuals feel obliged to satisfy the expectations of the group.

15Indeed, our data shows how informal social networks generate strong moral imperatives, including a sense of obligation to provide mutual assistance and reciprocity for favours given, which go above and beyond any respect for the formal legal framework. Individuals are thus expected to share with their next of kin, who in turn feel strongly entitled to a share of the wealth. Obligations vis-à-vis networks outside the family are based on strong norms of reciprocity. Thus, social networks rely on a transactional logic characterised by the exchange of a gift and then a counter-gift, sequentially. These imperatives translate, in practice, in that an individual who has a position in the public sector has the duty to utilise that position for the benefit of his or her network and that the economy of favours penetrates the public arena at the point of service provision in the form of practices of bribing, gift-giving and favouritism.

16Significantly, the research suggests that inclusion criteria for entry into the core networks may be fluid and instrumentally constructed. This means that networks are sometimes expanded in a deliberate manner to link up individuals increasingly far removed from each other on the basis of pragmatic considerations. As one Tanzanian research participant remarked, ‘when you have a problem and you do not have people to help you with it, then you go ahead and build the network to the person who can help you’. Thus, we find that the size of the core networks varies considerably across the three cases and this variation appears to be related to state effectiveness in terms of accessibility to essential public services and enforcement of the rule of law.

17Thus, the research findings suggest that extensive social networks are more prevalent in Uganda and Tanzania, where formal rights and entitlements are seen as having little bearing on the actual provision of essential services such as healthcare and where there is a perception that law enforcement is weak. Contrarily, core networks appear to be much smaller in Rwanda, where the perception is that services are provided according to the formal legal stipulations and that detection of corrupt actions is effective and followed by sanction enforcement.

18In the subsequent sections of this chapter we zoom into the practices of the social networks in Rwanda, Tanzania and Uganda through the analytical lens of three distinct patterns of governance practices, which we have previously identified (Baez-Camargo and Ledeneva, 2017) and by means of which political, business and social networks pursue their goals, whether these be to stay in power, maximise financial gains, access resources and services or simply ‘get things done’. In this chapter we discuss these informal practices—which we have termed co-optation, control and camouflage—and argue that they are associated with high levels of corruption because they ultimately enact an informal redistribution of power and resources.

19- Co-optation is associated with recruitment or strategic appointments into public office of allies and potential adversaries, who are granted impunity in exploiting the power and resources associated with public office in exchange for mobilising support and maintaining loyalty to the regime.

20- Control mechanisms are instrumental to the management of clashes of hidden interests, to ensuring elite cohesion and to enforcing discipline among allies, which is why the unwritten rules of co-optation depend on the existence of such informal control practices. Common examples involve the selective enforcement of anti-corruption laws against opponents or renegades, underscoring the validity of the saying attributed to former Brazilian dictator Getulio Vargas: ‘For my friends everything, and for my enemies the law’.

21- Camouflage refers to the manner in which the realities of political co-optation and control are hidden underneath institutional facades and policies consistent with a commitment to good governance and democratic accountability. Thus, punishment of a detractor may be presented as evidence of the leaders’ commitment to anti-corruption.

4. Co-optation

22The first of the informal practices we have identified is co-optation, which is the means by which recruitment into the network takes place. A key feature of co-optation is that it is transactional; it involves establishing a relationship that is based on strong bonds of trust, reciprocity and loyalty, which enable and sustain a system of mutually beneficial exchanges. Thus, a gift, an appointment, a bribe or a favour are used to establish a relationship, which generates indebtedness on the part of the accepting party and a gift economy arises. The balance of indebtedness never gets settled; otherwise there would be no relationship. Practices of co-optation happen in a top-down, horizontal, or bottom-up fashion.

23Co-optation can occur in a top-down fashion, initiated by higher-level officials (e.g. at the Ministry or District Council level) or mid-level service managers (for example, hospital accountants) and targeting lower-level staff and service providers (for example, nurses and teachers). Top-down, actors engage in co-optation in order to promote and consolidate their own positions. They do so by rewarding supporters and by neutralising opposition through a discretionary allocation of opportunities and resources, thus nurturing and expanding the networks that ensure support. In turn, the beneficiaries can typically extract rents in return for their unconditional support. Top-down co-optation can extend to the grassroots level incorporating important social groups into the ruling coalition, and is observed in the prevalence of patronage and clientelistic networks.

24Practices of top-down co-optation in the public sector result in situations where service providers are confronted with the demands of hierarchical networks in the workplace in addition to their formal work responsibilities and the demands and expectations of family, friends and acquaintances. According to our research findings, vertical networks can be found in the Tanzanian and Ugandan health sectors as the result of corruption in the recruiting process. Research participants from both countries indicated that employment is often obtained through network connections where a relative, friend or other social acquaintance of the appointing official recommends a particular individual for a position. Under those circumstances, it is understood that the job giver is granting a favour to the recipient, who thereafter enters the workplace with an unwritten, but well understood, indebtedness vis-à-vis the sponsor. Corruption in recruitment also happens when substantial sums are paid to secure a position. In such cases the debt is both a moral one (for being considered for the position, for which the bribe is just the precondition) as well as a real one (as invariably one will have to raise money with family and friends to generate enough funds to pay for such a position).

25In such contexts, where employment is obtained not by virtue of skills and professional qualifications, but rather as a personalised transaction cemented by social connections and/or money, loyalty tends to be much more valued than on the job performance when it comes to being eligible for promotion or other benefits. The research data further suggests that vertical networks are conducive to collusive behaviours in which corrupt service providers enjoy, if not the full support and complicity, at least passive concealment on the part of their colleagues.

26Co-optation can also be exercised horizontally when two or more networks engage in practices that serve the mutual benefit of all those involved. Thus, political elites frequently not only obtain the support of key non-state actors but also become recruited into exercising public authority to inordinately favour particular interests. For example, powerful business interests may be co-opted into supporting the interests of the political elites (e.g. financing electoral campaigns), receiving in exchange substantial benefits (e.g. large government contracts). Horizontal co-optation also happens when public officials use their decision-making authority over the provision of services, access to opportunities or distribution of resources instrumentally to grow their own networks. In Uganda, head teachers or principals at prestigious schools have been known to informally earmark some slots in the enrolment process in order to please higher ranking public figures or wealthy families whose children would not otherwise be accepted on the basis of their grades alone. Other accounts from Tanzania indicate that health staff may choose to provide preferential treatment strategically to particular patients, say a teacher, with a view to receiving a favour in return through help with a child’s enrolment in the future.

27Bottom-up co-optation is associated with informal networks at the community level, where membership of particular social groups binds individuals through ties of reciprocity and moral obligation into supporting each other and pooling resources. Co-optation practices at this level are associated with corruption because when a member gets a position in public office then the network expects this person to extract rents for the benefit of the group. In Uganda and Tanzania, like in many other African countries, a fair amount of social pressure is exerted on individuals who are formally employed or somehow perceived to be relatively affluent, pushing them to acquire resources for the benefit of their group—thereby legitimising certain corrupt acts and practices for the sake of the community, family and kin.

28Social pressures may also be associated with more extortive modalities of bribing and even embezzlement. This can happen when the opportunities for material redistribution that a provider’s salary allows do not adequately meet the expectations of the networks. This was vividly narrated in an interview with a medical doctor in Tanzania, who explained that the prevalent socially constructed preconception is that holding a medical degree is associated with a high salary, which is drastically contradicted by the reality of prevailing wages in the public health sector.

29The research data from Tanzania and Uganda shows that social networks tend to be extensive and fluid. In this regard, exchanging favours is a key practice conducive to expanding social connections, which works to both initiate and corroborate co-optation into the network. ‘There is a reciprocal relationship, meaning today you have helped me, tomorrow I must find a way to return the favour’, as one Tanzanian Focus Group Discussion participant described it. Thus, building a network involves evolving relationships on the basis of a reciprocal exchange of favours, goods and resources. These hinge on the mutually reinforcing conditions of indebtedness on the part of the recipient of the favour and entitlement on the part of the giving party. These practices are observed in the case of the delivery of public health services in that patients actively seek to co-opt health workers into their networks based on the belief that otherwise it will not be possible to obtain the medical treatment required. This is one reason why patients often bring unsolicited gifts of different kinds to their appointments at the health facility with the expectation that by doing so they can create a privileged relationship with the provider.

30In Uganda, service providers estimated that in about half of the interactions concerned users offer unsolicited ‘gifts’ to providers. Tanzanian service providers mentioned that they are used to receiving gifts from people in the form of material goods, such as maize, rice, goats, beans or through a credit transfer via mobile phone. Due to the ambivalent nature of the gift, one service provider stated that one has to try to figure out what the gift means in the current context and if it would create an environment in which one is suddenly indebted—when the motive of the gift was a form of co-optation—which raises the question, as one participant asked: ‘How do you deal with such matters, when you have already eaten the fish he bought?’ Further examples of the different types of co-optation are described below (Table 4.1).

Table 4.1 Types of co-optation

Table 4.1 Types of co-optation

Source: Compiled by Claudia Baez-Camargo.

31The case of Rwanda provides both contrast to and confirmation of the practices described above. According to evidence from our research, in Rwanda favouritism is the modality of petty corruption that is still most prevalent in the health sector, where preferential (faster) treatment is given to family and friends of service providers. However, the instrumental use of gift-giving or bribing to build useful networks appears to be less utilised in Rwanda. Whereas this certainly has to do with the potential risk of detection and punishment that accompany such activities (discussed in the following section), an important element to underscore is that the Rwandan regime, by significantly improving the quality and accessibility of public services, has made the functionality of network building for the most part redundant, in the sense that no informal strategy such as gift-giving or appealing to social connections is needed in order to receive adequate treatment.

  • 7 These rules emanate from Ubudehe, a home-grown solution and poverty-reduction initiative and one of (...)

32The generalised experiences among Rwandan research participants indicate that for the most part service provision is delivered as stipulated by formal rules and entitlements7. In other words, one of the drivers for the growth and and nourishment of social networks identified in Tanzania and Uganda—that is, overcoming the challenges associated with the poor quality and accessibility of essential services—is mitigated in Rwanda with improved public services.

5. Control

33The second pattern of informality is control, which relates to how informal networks enforce discipline among their members. This can happen by means of the selective enforcement of formal rules (e.g. anti-corruption laws) or informally (e.g. by means of veiled threats or exclusion from the network). Interestingly, informal control also occurs bottom-up, whereby powerful actors are locked into the system and expected to tend to the demands and expectations of their networks, otherwise risking negative consequences ranging from social shaming to even violence.

34There are several kinds of sanctions that can be expected if one does not adhere to the norms of the networks. Practices of control among hierarchical networks in the workplace were illustrated by two Ugandan rural health workers, according to whom people not only become corrupt to get better positions, but also because those not participating in the prescribed corruption schemes or those who denounce incidences of corruption may face serious repercussions. As one health worker experienced: ‘If you refuse, you miss the job [opportunity] or your name will not appear on the payroll.’ Therefore, fear over losing one’s job or being bullied can translate into a ‘don’t ask, don’t tell’ policy, as explained by a nurse who had witnessed incidences of corruption involving fellow health workers: ‘Although you know what is happening, you stay quiet because it’s beyond your understanding.’

35As previously mentioned, the effectiveness of social networks is related to the manner in which they embody deeply ingrained values of reciprocity, solidarity and moral obligation towards one’s group. Based on these values, certain behaviours are expected. Concretely, when it comes to the family, the entitlement to partake of the gains of one’s next of kin is an essential, unquestionable premise of social life in the three countries. Consequently, one of the worst transgressions occurs when an individual is seen to turn his or her back on a family member. This can result in social isolation and shaming, although in some contexts the consequences might be more extreme as explained by a service user in Kampala: ‘If my brother earns 15 million and I ask him to help me pay twenty thousand shillings for my son to register for exams and he tells me about his cow, which is sick, and that he needs 100,000 for it to get treated yet he can’t help me with only twenty thousand shillings, then what do I need from him apart from beating him to death?’

36Outside the family, social networks rely on an expectation of reciprocity in the ongoing exchange of favours and gifts. Service providers recounted the difficulties they face rejecting unsolicited gifts from users; because the act of presenting and receiving a gift is socioculturally anchored, rejecting a gift can be interpreted as disrespectful.

37The significance of the control that can be exercised by social networks is underscored by the notable extent to which fulfilling one’s duties and obligations vis-à-vis the network is closely linked to social status. In the FGDs participants were asked about the actions that gain individuals respect and good social standing in their communities versus those that cause a person shame and to fall into disgrace. Some of the FGD responses to this question are summarised below (Table 4.2).

Table 4.2 Actions conducive to gaining and losing status

Table 4.2 Actions conducive to gaining and losing status

Source: Compiled by Claudia Baez-Camargo.

38In Tanzania and Uganda, the expectation of help and a contribution to the network’s welfare described above influences the behaviours of service providers profoundly. For example, if a service provider does not favour a relative in service provision or rejects a small bribe or a gift, the perception among research participants was that this person’s adherence to formal rules destroys his or her trust-based relationships, and such a person might be ultimately abandoned by the community. The infringement of socially embedded norms is therefore seen as affecting the moral economy of the community, because it threatens the socially constituted values and practices of economic redistribution that are critical to meeting essential needs.

39The responses from Rwanda are particularly interesting as they suggest the tensions generated by a context in which there are strict constraints in place inhibiting service providers from engaging in those practices that are nonetheless still regarded as the gateway to social status and respect. Rwandan service providers can therefore be characterised as being subject to bottom-up control in a similar manner as their Tanzanian and Ugandan counterparts, as attested by the following statement from a Rwandan health worker: ‘Actually, when you refuse to render the service, people you are associated with will say that “having him is like not having someone”. The effect of this is that the worker is put aside and is looked upon as a bad person, even losing friends or the appreciation of their extended family.’

40However, social pressures exist in significant tension with formal control mechanisms in Rwanda, which because of its enforcement of a zero-tolerance approach to corruption has distinguished itself significantly from Tanzania (at least until recently) and Uganda. In Rwanda, instances of petty corruption involving bribing or favouritism are harshly sanctioned, including by means of a policy of naming and shaming whereby individuals guilty of corruption are publicly identified in published lists that not only bear their names but also those of their parents and community. Our research indeed suggests that, because the sanctions against cases of corruption are widely perceived to be extremely harsh and effectively enforced, compliance is commonly associated with fear with regard to the possible reporting of negative performance or to legal repercussions.

41Our research also shed light on creative solutions by means of which health workers in Rwanda negotiate the tensions that exist between the formal and the informal norms they must abide by. In particular, health workers gave account of receiving family members for treatment outside the opening hours of the health facility to avoid openly disrupting the first come, first served rules of service provision. Rwandan FGD participants also expressed that bribery still can occur but requires intricate arrangements to avoid disclosure and may only be arranged with people who are well known and trusted by the service provider for fear of being denounced to the law-enforcement authorities.

42For those reasons, Rwandan health providers concluded that corruption has increasingly become a private affair, meaning that it is restricted to the innermost circle of the informal network, due to the harsh costs associated with detection and the risk of being shamed extending to the whole family. Consequently, the strong enforcement of anti-corruption laws in Rwanda appears to be resulting in a situation where core networks are shrinking.

43Interestingly, the research findings from Tanzania suggest how the incentives of service providers are being reshaped with increased enforcement of sanctions against corrupt behaviours under the current regime. As in Rwanda, this exacerbates the tensions between the demands of the formal and informal normative frameworks to which the service providers are expected to respond. One participant in the service providers’ FGD explained that they ‘only employ you as an individual and not as a family. However, they [the family] do not understand such a thing […] they do not know the regulations the employed have to follow’.

44Thus, the perception that anti-corruption enforcement is tightening means that the opportunity space to engage in practices of petty corruption is shrinking. Interestingly, service providers are now developing a new language to convey the situation. According to anecdotal evidence, public officials who cannot grant a favour or provide the amounts expected by their networks resort to simply stating ‘Magufuli’ to communicate the idea that under the current president things have changed.

6. Camouflage

45The third informal pattern is camouflage, which refers to how informal practices of co-optation and control take place beneath a facade of commitment to formal procedures, such as the rule of law, anti-corruption measures or democratic accountability. Camouflage also takes place under the guise of respecting ‘traditional values’. Practices of camouflage protect the network from external scrutiny by external actors and play on the coexistence of multiple normative frameworks (Olivier de Sardan, 1999; Blundo et al., 2006).

46Thus, informality becomes difficult to grasp to an outsider precisely because it makes use of the formal order (whether based on legal or traditional norms) in order to pursue hidden agendas. For example, camouflage is routinely used by corrupt political regimes to create a smooth facade that legitimates them not only vis-à-vis donors and the international community but also with domestic audiences. Seemingly upholding certain pledges with regard, for example, to elections or legal reform, de facto the reforms are undertaken to facilitate relations with external actors, often succeeding in hiding ‘real’ processes, practices and power from closer scrutiny.

47Whereas camouflage is normally seen to be a practice exercised from the top down, evidence suggests that it might also be practiced from the bottom up. For instance, lowly public officials may use ‘correct’ (i.e. formalised) procedures to hide malfeasance and extraction from the law enforcement authorities or citizens may convey bribing as an expression of the gift-giving customs cherished by society.

48In our research data one of the most conspicuous forms of camouflage relates to the ambivalent nature of gift-giving. FGD participants elaborated on how although a gift can be offered simply as a gift, the expectation of a counter-gift—in the form of a service or a favour—is inherent, understood by all parties involved although it may never be explicitly articulated. Thus the act of gift-giving is intrinsically ambivalent and a bribe can always be disguised as a gift. Rwandan FGD participants explained that a gift is perceived as part of the social routine, an inherent social practice, and as such is described as a public rather than a private transaction. A bribe, in contrast, is an essentially private transaction. Therefore, it comes as no surprise that if the bribe becomes exposed (taken out from the private into the public realm), it should immediately morph into a gift.

49Kampala health facility staff agreed that there is nothing wrong with giving preferential treatment to family members—‘Your relative must be given priority; if there is not an emergency, my mother, father, son or daughter come first’—because this is the way things should proceed ‘according to culture’. In this context the same health staff gave other examples of practices of camouflage that make use of social norms to cover up illicit activities. Some research participants recounted how support staff at the health facility often ask them to treat their family members, when in reality they are soliciting bribes from health facility users in order to pose as a relative and thus get them expedited treatment. There were stories told of support staff that sometimes requested speedy treatment for up to five sick relatives in one day. This practice is referred to as ‘clan soccer’, and according to the FGD participants it takes advantage of the fluid manner in which the terms ‘brother’ and ‘sister’ are used in Uganda.

50Camouflage is also proactively used in other ways in order to get access to a network. Participants of FGDs in Tanzania and Uganda described how service providers often assess the situation to determine whether it is worth co-opting the service seeker into their network. Thus, service seekers recount how they prepare before going to the facility; strategies include dressing up in order to give the impression that they have resources that will make them attractive, or—conversely—dressing down and bringing small children along to encourage feelings of compassion on the part of the provider. On their part, FGD participants from Rwanda acknowledged that in order to get away with bribing one needs to carefully plan in order to mask the bribe as a gift, or make sure the exchange never happens directly but rather through a mutual and trusted acquaintance, because one cannot even try to engage in any remotely suspicious activity with strangers out of fear of being denounced.

51An interesting example of camouflage among public officials in Rwanda occurs when they resort to a practice known as ‘techniquer’. In Rwanda, to ‘techniquer’ a report (‘gutekinika Raporo’) or statistics mean to falsify or present false reports or figures to one’s superiors (Beaz Camargo and Gatwa, 2018). In a first instance, resorting to ‘techniquer’ may be used to hide corruption favouring individuals or percolating up (or down) the network. However, that may not be the only use of this practice as Beaz Camargo and Gatwa (2018) found that public officials may also use such practices pre-emptively when they have not been successful in meeting their professional performance targets and fear tough sanctions from above. Therefore, what is particularly interesting in the case of Rwanda is that these practices are not condoned by or led from the top, as is the case in the other East African countries. Rather, camouflage is used to hide such practices from the top leadership.

7. Concluding Remarks

52In this chapter, we have sought to outline the patterns shaping social networks in three East African countries. Evidence from two related research projects shows that governance processes are informed by the norms and practices through which social networks are configured. The findings confirm that there are certain patterns that can be identified, each with a distinct set of practices and norms. Co-optation, camouflage, and control are all practised both to secure the moral economy from which the network draws its legitimacy and to enable the extraction of concrete material gains for individual members. These practices are ‘star-shaped’, so to speak, in that they radiate in all directions: bottom-up, horizontally, as well as top-down. The distinction between ‘formal’ and ‘informal’ governance becomes obsolete, as these practices draw on a diverse and contingent range of norms. Whereas certain practices may be more iterative and habitual, others emerge from calculated judgements of particular situations and contexts.

53Some of the most interesting findings are emerging from ‘liminal’ contexts, where established patterns of locking into and reproducing networks are breaking down and new ones remain in the making. Such open-ended transformations are likely present in all countries, even though in some contexts they may be happening within more circumscribed spaces than in others. Our findings suggest that these transformations are the result of changes of political regimes—particularly in the case of Rwanda under Paul Kagame, where social networks have had to reconfigure and react to top-down sanctions against corruption in the public sector. Whereas in Tanzania and Uganda the networks are more fluid and are built instrumentally, in Rwanda there seems to be a trend whereby people are trimming rather than expanding their networks as formal rules and procedures are being vigorously enforced by the regime. Thus, while health workers continue to be subject to their networks’ expectations of preferential treatment, these are increasingly restricted to encompass only family and close friends.

54Tanzania is also an important case to follow up on given that similar top-down approaches to anti-corruption are being favoured by the government of John Magufuli, the president—for example—using personal discretion and presidential authority to short-circuit ‘official’ procedures, albeit to sanction corrupt senior officials. In these cases, although at first glance it may seem as if the regime is imposing ‘official’ norms, the picture on the ground is somewhat messier. Although our findings may not be sufficient to make a strong argument, it seems as if new forms of camouflage are emerging that are not top-down (as in the case of Uganda, for instance), but bottom-up. This could also be observed in Rwanda, where camouflage practices are being used to hide both inefficiencies and failures to achieve the centrally determined performance targets from the feared eye of the government. This raises important questions as to shifts in informal patterns: whether a unilinear move from greater trust in informality networks to formal rules and institutions is really underway in Rwanda, or whether the transformation process is more complicated, with new and different forms of informal practices emerging—one being new practices of camouflage. Our findings also highlight some significant side effects of enforcing anti-corruption measures in a strictly top-down manner, whereby social networks based on values of solidarity and reciprocity break down, the wider consequences of which in the long run are not yet clear.

55From a political economy perspective, it is conventionally argued that the eradication of corruption hinges not merely on institutional reform, but on the political will to change norms and practices. However, our findings suggest a far more contextualised and complicated relationship. Firstly, leadership indeed matters. Ugandans refer to the scandalous behaviour of political leaders to justify their own acts of corruption. In Tanzania, on the other hand, Magufuli sacked his close associates, even some who were known to be close friends of his. Shared images and ideas about social roles—including what constitutes being a ‘good’ politician, service provider and public official—influence how people come to expect themselves and others to behave in different situations (Kotzian, 2011). However, the evidence from our research projects suggests very strongly that leaders, even those at the very top, are not as autonomous and powerful as they are sometimes portrayed in the literature. Our focus on social networks carves out the many ways in which they are locked into configurations from which they cannot easily or readily break away. All members of a network, including powerful actors, are tied into relations of reciprocal exchange through co-optation and control. These practices are all the stronger as they are frequently underpinned by social norms and values that are used to legitimise extraction and redistribution (Olivier de Sardan, 1999).

56Furthermore, as indicated above, even in the cases where to all intents and purposes the strong political will of a president exists to enforce official norms and procedures, the effects on the actual patterns of governance are empirical questions. Whether and where new forms of ‘gaming the system’ emerge, or indeed a Weberian public ethos is nurtured, is subject to specific contexts and networks. These effects are highly contingent upon the structuring of the society in question, the patterns of the networks and the practices of governance (Koechlin 2013). These need to be understood in more detail if we are to shed light on the transformations effected by political will. In what way urbanisation and the transformation of urban spaces affect such processes is a further pertinent topic of future research.

57What do our findings imply for anti-corruption policies? We join the large choir of scholars and practitioners who call for more nuanced and evidence-based approaches to the anti-corruption field. Although our findings are still emerging, there are some pointers that can be drawn from the two research projects.

58Social norms and values matter—and they can and do change. The patterning of networks is not just imposed by structures or by calculated self-interest; our findings show how actors orient themselves and their practices explicitly (i.e. not merely habitually) along particular norms and values, and seek to create moral coherence by drawing on specific norms and values when legitimising or underpinning their practices. ‘Being part of a family’, ‘being one of us’ or ‘looking after each other’ are frequent expressions that are employed by actors when they explain why and how they interact within social networks. In this sense, they are evidently also used to legitimise ‘corrupt’ practices, by drawing on alternative normative repertoires that emphasise reciprocity or social obligations.

59However, it is absolutely central to underline that these norms and values themselves are contingent upon context. Of course all actors draw on multiple repertoires that speak to different normative frameworks—we speak and act differently depending on the role we are playing and the situation we are in. But more pertinently for anti-corruption policies, the key point emerging from our research is that societal norms and values can and do change in interaction with changes in the political and economic environment.

60Our research findings underline that these changes need to be understood far better than has been the case to date. In Rwanda, for instance, such change is being effected by the regime, and enforced rigorously throughout the state and society. The effects may not be as coherent and consistent as some authors portray them—that is to say, as marching towards a rule-and-procedure-based state. But there is no doubt that actors are evaluating their practices in the light of these changes and that certain norms and values may indeed be changing fundamentally, the ultimate outcome of which remains to be seen. Also, the jury is still out with regard to the current situation in Tanzania; the effects of Magufuli’s very top-down and personal anti-corruption battle are yet to be fully ascertained. One interesting insight is that, much like in Rwanda, new practices of camouflage seem to be emerging bottom-up.

61To conclude, the three governance patterns identified in this chapter—namely, control, co-optation, and camouflage, all play with the ambivalence of ‘formal’ or ‘official’ norms, on the one hand, and ‘informal’ or ‘traditional’ norms, on the other, depending on the context and the most promising strategy for the intended purpose. In practice they cannot be separated, but are part and parcel of a range of repertoires at the disposal of the actors to address specific problems and pursue specific interests. At the same time, our findings underline that ingrained expectations and social sanctions can lead to a situation in which actors are ‘locked’ into particular patterns. However, governance patterns are changing all the time, sometimes as a result of changing political regimes. The many ambivalences characterising these patterns and their dynamics may represent potential entry points to promoting positive behavioural changes and advancing anti-corruption agendas if understood in-depth and by working closely with the social actors concerned.

Top of page

References

Baez Camargo, C. and T. Gatwa (2018) Rwanda Country Report, Research report on behalf of research project ‘Informal Governance and Corruption: Transcending the Principal-Agent and Collective Action Paradigms’, (Basel: Basel Institute on Governance), http://informalgovernance.baselgovernance.org/sites/default/files/2018-07/Rwanda.InformalGovernance.Country%20Report.pdf (accessed on 30 August 2018).

Baez-Camargo, C. and A. Ledeneva (2017) ‘Where Does Informality Stop and Corruption Begin? Informal Governance and the Public/Private Crossover in Mexico, Russia and Tanzania’, The Slavonic and East European Review, 95(1), pp. 49–75, DOI: 10.5699/slaveasteurorev2.95.1.0049

Baez-Camargo, C. and N. Passas (2017) ‘Hidden Agendas, Social Norms and Why We Need to Re-Think Anti-Corruption’ (OECD/ 2017 OECD Global Anti-Corruption & Integrity Forum), https://www.oecd.org/cleangovbiz/2017-integrity-forum-research-posters.htm (accessed on 17 October, 2017).

Bayart, J. F. (1993) The State in Africa: The Politics of the Belly (New York: Longman).

Bevir, M. (2013) A Theory of Governance (Berkeley:University of California Press).

Blundo, G. and P.-Y. Le Meur (eds) (2008) The Governance of Daily Life in Africa: Ethnographic Explorations of Public and Collective Services (Leiden and Boston: Brill Academic Publishers).

Blundo, G. and J.-P. Olivier de Sardan (2006) Everyday Corruption and the State: Citizens and Public Officials in Africa (London and New York: Zed Books).

Blundo, G., J.-P. Olivier de Sardan, N. B. Arifari and M. T. Alou (2006) Everyday Corruption and the State: Citizens and Public Officials in Africa (Cape Town: Zed Books).

Bozzini, A. (2014) ‘Fighting Corruption While Undermining Accountability: The Case of Rwanda’ Working Paper (Hamburg: German Institute of Global and Area Studies), http://anticorrp.eu/wp-content/uploads/2014/03/Rwanda-Background-Report_final.pdf (accessed on 24 September, 2017).

Bratton, M. and N. Van de Walle (1994) ‘Neopatrimonial Regimes and Political Transitions in Africa’, World Politics, 46(4), pp. 453–489, DOI: 10.2307/2950715.

Chabal, P. and J.-P. Daloz (1999) Africa Works: Disorder as Political Instrument (Indiana: Indiana University Press).

Chang, C. S., N. J. Chang and B. T. Freese (2001) ‘Offering Gifts or Offering Bribes? Code of Ethics in South Korea’, Journal of Third World Studies, 18(1), pp. 125–139, https://trove.nla.gov.au/version/72287458 (accessed on 2 May 2018).

Emirbayer, M. and J. Goodwin (1994) ‘Network Analysis, Culture, and the Problem of Agency’, American Journal of Sociology, 99(6), pp. 1411–1454, DOI: 10.1086/230450

Golooba-Mutebi, F. and D. Booth (2013) Bilateral Cooperation and Local Power Dynamics: The Case of Rwanda, ODI Research Reports and Studies, (London: Overseas Development Institute), https://www.odi.org/publications/7888-bilateral-cooperation-and-local-power-dynamics-case-rwanda (accessed on 2 May 2018).

Grodeland, A. B., T. Y. Koshechkina and W. L. Miller (1998) ‘‘Foolish to Give and Yet More Foolish Not to Take’. In-Depth Interviews with Post-Communist Citizens on Their Everyday Use of Bribes and Contacts’, Europe-Asia Studies, 50(4), pp. 651–677, http://www.jstor.org/stable/153781

Heywood, P. M. (2017) ‘Rethinking Corruption: Hocus-Pocus, Locus and Focus’, The Slavonic and East European Review, 95(1), pp. 21–48, DOI: 10.5699/slaveasteurorev2.95.1.0021

Hyden, G. (1980) Beyond Ujamaa in Tanzania: Underdevelopment and an Uncaptured Peasantry (Berkeley and Los Angeles: University of California Press).

Jauregui, B. (2014) ‘Provisional Agency in India: Jugaad and Legitimation of Corruption’, American Ethnologist, 41(1), pp. 76–91, DOI: 10.1111/amet.12061.

Klitgaard, R. (1988) Controlling Corruption (Berkeley, Los Angeles and London: University of California Press).

Koechlin, L. (2015) ‘The Conceptual Polysemy of Governance’, in L. Koechlin and Förster, T. (eds) The Politics of Governance: Actors and Articulations in Africa and Beyond (New York: Routledge).

Koechlin, L. (2013) Corruption as an Empty Signifier: Politics and Political Order in Africa (Leiden: Brill).

Koni-Hoffmann, L. and R. Navanit-Patel (2017) Collective Action on Corruption in Nigeria: A Social Norms Approach to Connecting Society and Institutions (London: Chatham House - The Royal Institute of International Affairs), https://www.chathamhouse.org//node/29640 (accessed on 7 August 2017).

Kotzian, P. (2011) ‘Cosi Fan Tutte: Information, Beliefs, and Compliance with Norms’, Zeitschrift Für Soziologie, 40(4), pp. 158–173. DOI: 10.2139/ssrn.1497922.

Latour, B. (2005) Reassembling the Social: An Introduction to Actor-Network-Theory (Oxford: Oxford University Press).

Ledeneva, A. V. (1997) ‘Practices of Exchange and Networking in Russia’, Soziale Welt, 48(2), pp. 151–170.

Marquette, H. and C. Peiffer (2015) Corruption and Collective Action, Research Paper 32, Developmental Leadership Program and U4 Anti-Corruption Resource Centre, (Birmingham: University of Birmingham). http://publications.dlprog.org/CorruptionandCollectiveAction.pdf (accessed on 5 January 2017).

Mesquita, B. B. de, A. Smith, R. M. Siverson and J. D. Morrow (2005) The Logic of Political Survival (Cambridge and London: MIT Press).

Mungiu-Pippidi, A. (2017) Corruption as Social Order, World Development Report Background Paper (Washington D.C.: World Bank), https://openknowledge.worldbank.org/handle/10986/27046 (accessed on 16 January 2018).

Nesoba, R. (2015) ‘Tanzania’s John Magufuli in Profile’, BBC News, 24 November, http://www.bbc.com/news/world-africa-34670983 (accessed on 30 October 2017).

New African Magazine (2017) ‘John Magufuli: Tanzania’s Rising Star’, 8 June, http://newafricanmagazine.com/john-magufuli-tanzanias-rising-star/ (accessed on 30 October 2017).

Olivier de Sardan, J.-P. (2008) Researching the Practical Norms of Real Governance in Africa, APPP Discussion Paper No. 5 (London: Department for International Development), https://www.gov.uk/dfid-research-outputs/appp-discussion-paper-no-5-researching-the-practical-norms-of-real-governance-in-africa (accessed on 30 October 2017).

Olivier de Sardan, J. P. (1999) ‘A Moral Economy of Corruption in Africa?’, The Journal of Modern African Studies, 37(01), pp. 25–52.

Persson, A., B. Rothstein and J. Teorell (2013) ‘Why Anticorruption Reforms Fail—Systemic Corruption as a Collective Action Problem’, Governance, 26(3), pp. 449–471, DOI: 10.1111/j.1468-0491.2012.01604.x

Rose-Ackerman, S. (1999) Corruption and Government (New York and Cambridge: Cambridge University Press).

Ruud, A. E. (2000) ‘Corruption as Everyday Practice. The Public—Private Divide in Local Indian Society’, Forum for Development Studies, 27(2), pp. 271–294, DOI: 10.1080/08039410.2000.9666142

Sneath, D. (2002) ‘Reciprocity and Notions of Corruption in Contemporary Mongolia’, Mongolian Studies, 25, pp. 85–99.

Transparency International (2017) The East African Bribery Index 2017 (Nairobi: Transparency International), https://actionfortransparency.org/view/accountability/the-east-african-bribery-index-2017/ (accessed on 19 September 2017).

Twaweza East Africa (2016) The People’s President? Citizens’ Assessment and Expectations of the Fifth Phase Government, Sauti Za Wananchi Brief No. 35 (Dar es Salaam: Twaweza East Africa), http://twaweza.org/uploads/files/PeoplesPresident-EN-FINAL-A4.pdf (accessed on 9 February 2017).

Top of page

Notes

1 Empirical evidence presented in this paper stems from the findings from two research projects of the Basel Institute’s Public Governance Division. The first project, entitled ‘Informal Governance and Corruption - Transcending the Principal-Agent and Collective Action Paradigms’, is funded by the joint UK Department for International Development (DFID) and British Academy Anti-Corruption Evidence (ACE) programme. For more information about the ACE programme and the project please visit http://www.britac.ac.uk/anti-corruption and http://www.britac.ac.uk/node/4660. The second project is entitled ‘Corruption, Social Norms and Behaviours in East Africa’ and is funded by the East Africa Research Fund (EARF) of DFID. For more information about EARF and the project please visit http://www.earesearchfund.org/ and http://www.earesearchfund.org/research-corruption-social-norms-and-behaviors-east-africa (accessed on 2 May 2018)

2 For a discussion of a non-normative, inductive approach to governance see Blundo and Le Meur (2008), Bevir (2013) or Koechlin (2015).

3 The projects used qualitative data collection methods with the support of a team of local researchers. The data presented in this paper was collected during research activities conducted between January 2016 and August 2017, activities that involved focus group discussions (FGDs) at two sites (one urban and one rural area) in each country. In the FGDs we used the health sector as an indicative entry point to study the interface where corruption risks arise. The participant profiles of the focus groups were a) community members, b) providers of health services, and c) managers of health facilities. A total of 18 FGDs were conducted. The local researchers are Prof. Tharcisse Gatwa and Mr Abel Dufitumukiza of the Protestant University of Rwanda, Dr Richard Sambaiga and Mr Egidius Kamanyi of the University of Dar es Salaam, Prof. Paul Bukuluki of Makerere University and Mr Robert Lugolobi, independent anti-corruption consultant in Uganda.

4 The concept of social networks is sociologically highly loaded—for a critique of social network analysis see (Emirbayer and Goodwin, 1994); for the seminal contribution to the actor–network approach see (Latour, 2005). Here, the significance of social networks as a frame of reference emerged inductively from the data, and still needs to be theorised adequately, although our interpretations are closer to the constructivist approach of actor–network theory.

5 See Olivier de Sardan (2008), who coined the expression ‘real’ governance.

6 A citizens’ assessment and expectations survey published in September 2016 reported that ‘96 per cent of Tanzanians approve or strongly approve of President Magufuli’s performance since taking office’ (Twaweza East Africa, 2016). Numerous local and international media have also reported on the popularity of President Magufuli. See for example the BBC’s profile of the president (Nesoba, 2015) and the New African’s analysis of ‘John Magufuli: Tanzania’s rising star’ (New African Magazine, 2017).

7 These rules emanate from Ubudehe, a home-grown solution and poverty-reduction initiative and one of the country’s core development programmes. Ubudehe involves a system of categorisation of all households according to income level, which forms the basis on which entitlements are determined and subsidies allocated.

Top of page

List of illustrations

Title Table 4.1 Types of co-optation
Credits Source: Compiled by Claudia Baez-Camargo.
URL http://journals.openedition.org/poldev/docannexe/image/2646/img-1.png
File image/png, 88k
Title Table 4.2 Actions conducive to gaining and losing status
Credits Source: Compiled by Claudia Baez-Camargo.
URL http://journals.openedition.org/poldev/docannexe/image/2646/img-2.png
File image/png, 114k
Top of page

Cite this article

Bibliographical reference

Claudia Baez Camargo and Lucy Koechlin, Informal Governance: Comparative Perspectives on Co-optation, Control and Camouflage in Rwanda, Tanzania and UgandaInternational Development Policy | Revue internationale de politique de développement, 10 | 2018, 78-100.

Electronic reference

Claudia Baez Camargo and Lucy Koechlin, Informal Governance: Comparative Perspectives on Co-optation, Control and Camouflage in Rwanda, Tanzania and UgandaInternational Development Policy | Revue internationale de politique de développement [Online], 10 | 2018, Online since 01 October 2018, connection on 16 April 2024. URL: http://journals.openedition.org/poldev/2646; DOI: https://doi.org/10.4000/poldev.2646

Top of page

About the authors

Claudia Baez Camargo

Claudia Baez Camargo holds a PhD in Political Science from the University of Notre Dame, USA and a graduate degree in Economics from the University of Cambridge, England. She is Head of Governance Research at the Basel Institute on Governance/University of Basel, where she is responsible for the development, oversight and management of the Institute’s research activities as well as delivering technical assistance projects aimed at preventing corruption in the public sector.

Lucy Koechlin

Lucy Koechlin is Senior Lecturer at the Institute of Social Anthropology, University of Basel. Her research and teaching focuses on political transformations in Africa, with a particular focus on urban politics, democratic spaces and governance. As a consultant, Dr Koechlin has been involved in a wide range of cross-sectoral mandates on anti-corruption, accountability and governance.

Top of page

Copyright

CC-BY-NC-4.0

The text only may be used under licence CC BY-NC 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search