‘Can you hear me?’ Mobile–radio interactions and governance in Africa

Researching mobile–radio interactions

The comparison is based on ethnography conducted in the radio stations and in the communities surrounding them. The fieldwork lasted for 2 months and took place between July and September 2011. In both sites, a combination of methods was employed to study interactions from the perspective of different actors. Participant observation in the radio stations’ studios and newsrooms revealed the practices used by journalists and presenters to collect and make use of information coming from people who are calling and texting. This also included people who are contacting journalists directly on their personal mobile phones and visiting the radio station to make sure their demands are given full attention. In the case of both radio stations, the focus progressively narrowed towards their breakfast shows, as they display the richest forms of interaction and regularly connect journalists, audiences and public authorities to debate issues of both local and national relevance.

In the case of both stations, I usually arrived in the studio in the early morning and sat with the presenters, taking notes, asking questions while songs were playing and observing the interactions with callers with the support of a research assistant. At the end of each show, I spent some time with the presenters, asking them to reflect on specific events, interactions they had with callers and public authorities. In some occurrences, I also observed journalists outside of the studio while they were collecting information and interacting with the residents of Kibera and Kisumu.

The observations collected in the radio stations were then compared with audiences’ perceptions of the exchanges they were taking part in and of their possible repercussions on governance processes. Particular attention was placed on the most active and engaged users, who were selected based on their frequent interactions with the station and then interviewed either in person or through the phone. To obtain more contextual information and develop a better sense of the significance of the radio stations beyond their active listenership, with the support of local facilitators we randomly selected individuals both in Kibera and in the outskirts of Kisumu, asking them some basic questions about their interactions with radio stations, their uses of mobile phones and their opinions about the ability of mobile–radio interactions to affect governance processes.

Finally, public authorities were contacted and interviewed in order to understand how they perceived the interactions and debates taking place on the radio. In both sites, individuals working at different levels of the local administration were interviewed (a DO, a chief and an assistant chief), as well as individuals working in formal and informal institutions involved in the provision of services identified as particularly relevant for the surveyed community. As a result, more than 50 people were interviewed. The interaction with different interviewees, however, varied significantly: from repeated interactions with radio presenters spanning across weeks, to sustained engagement with some regular callers and public authorities, who were interviewed in multiple occasions, to brief interviews (between 10 and 45 minutes) with members of the communities reached by the two radio stations.

Hybrid political orders and hybrid governance

Large-scale and comparative research programmes on governance in Africa (Booth, 2011) and on the impact of popular engagement on governance processes in developing countries (Gaventa and Barrett, 2010) have begun to indicate how participation may matter in ways that are significantly different from those experienced in (mostly idealized imaginings of) Western democracies. These differences depend on the types of actors and institutions participating in governance processes and the relationships between them, as well as on the kinds of strategies and opportunities meaningful to individuals seeking to influence decision making and access public goods and services.

In the case of Sub-Saharan Africa, the first notable difference with governance process characterizing Western democracies is found in the multiplicity of actors that often occupy the same socio-political spaces. As Boege et al. (2009) noted in their influential work on hybrid political orders, ‘the state is only one actor among others, and state order is only one of a number of orders claiming to provide security, frameworks for conflict regulation and social services’ (p. 6). While the state may claim authority over its territory, in vast parts of it only ‘outposts’ of the state may be found and ‘traditional societal structures (extended families, clans, tribes, religious brotherhoods, village communities) and traditional authorities (village elders, headmen, clan chiefs, healers, big men, religious leaders) determine the everyday social reality’ of the majority of the population, especially in rural areas (Boege et al., 2009: 7). In some cases, traditional bodies sit alongside non-state manifestations of modernity represented by national and international non-governmental organizations (NGOs), which derive their legitimacy from their ability to offer goods and services other institutions are unable to provide. Acknowledging how different actors coexist and possibly cooperate in the same socio-political space has led to approaches that do not simply emphasize what the African state is not, evaluating it through external standards, but approaches interested in understanding what in Africa works. Analysing governance and service delivery, Olivier de Sardan (2011), Blundo and Le Meur (2009) and Joshi and Moore (2004) have provided evidence of how the provision of a specific service, especially at the local level, is seldom the outcome of a single effective formal authority exercising its stated functions, but rather the result of co-production and co-delivery emerging from a confluence of organized groups of individuals, local nodes of the administration, NGOs, religious leaders and other entities (Olivier de Sardan, 2011). Hagmann and Hoehne (2009) have examined successful cases of peace and state-building which have emerged from below in the absence of a central monopoly of violence. In the break-away region of Somaliland, for example, in the absence of templates proposed from above, a hybrid form of statehood has emerged which ‘amalgamates customary, Islamic and statutory norms and practices’ (Hagmann in Boege et al., 2009: 13). In essence, hybrid political orders include a multiplicity of actors that coexist alongside the state, including networks of religious, cultural and family institutions (Boege et al., 2009). Hybrid governance refers to the resulting ways in which citizens experience hybrid political orders in the everyday in order to access the goods and services they need and value.

A second, related and crucial, element that differentiates political orders in parts of Africa is the nature of the networks connecting these actors. As Chabal and Daloz (1999) have claimed,

the state in Sub-Saharan Africa has not been institutionalized – in that it has not become structurally differentiated from society. […] The notion of civil society has been abused by those who would see in Africa social movements willing and able to challenge central power, [while] the substance of politics has to be found in the myriad networks which link the various levels of power, from top to bottom. (p. 24)

The links connecting different nodes of a network can variably indicate a relation between patron and clients, a shared belonging to a same ethnic group or clan, the connection between voters and their elected representative or, more often, a combination of these and other dimensions.

This emphasis on hybrid political orders and hybrid governance has represented a significant shift in development and African studies (while it has been less noticed in other disciplines) and has encouraged moving beyond mainstream conceptions of ‘fragile’ or ‘crisis’ states, suggesting how the fragility of one actor may be superseded by the strength of others or how a plurality of relations may be activated in the everyday that are not confined to the nexus between citizens and the state. The hybrid governance agenda, however, has also been subject to criticism. While it offers a counter-narrative to ‘pathological’ readings of Africa (Hagmann and Péclard, 2010), it has also inaugurated a tendency to over-celebrate hybridity. As Kate Meagher (2012) illustrated through the study of vigilante groups, hybridity can also lead to violent and destructive outcomes. While in some instances there may be a mutual tolerance of different orders, in others the coexistence of competing actors in the same space may lead to more or less violent conflicts. In others still, including the case of Kenya presented in this article, state actors may make use of extra-state violence in order to manipulate the political game to their own advantage (Mueller, 2008). Or, as Alex de Waal and Rachel Ibreck (2013) emphasized in their analysis of social movements in Africa, when client–patron relations prevail it is likely that

the patronage that sustains the ruler permeates civil society. While the state appears weak because of its limited bureaucratic capacity and its failure to exercise a monopoly over the means of violence, the ruler is in fact difficult to challenge because he (rarely, she) controls a host of informal networks that include clientage, ethnic loyalties and religious authority. (pp. 308–309)

Similar to the conflict among actors, there may be conflicts among different types of relationships within a network, where links do not simply add up but actually clash with one another.

This complexity has led to another criticism of hybridity, which has targeted the vagueness of the term – everything is a hybrid in one way or another – and its inability to serve as an analytical tool to distinguish among different configurations (Goodfellow and Lindemann, 2013). Hybridity can indeed be abused to give the impression of explaining something (sometimes adding a postmodern twist to an argument) without actually explaining much. The power of the discussed approach, however, is not that it offers a ready-made analytical toolkit. Rather, it creates a new space for empirical enquiry. It minimizes normativity and encourages the exploration of a plurality of dimensions within social life before deciding which ones are the most relevant. Hybridity can be both constructive and corrosive (Meagher, 2012), and it is through comparative empirical analysis that it is possible to progressively develop a body of knowledge that allows understanding under which conditions actors cooperate or clash and what the consequences are of activating certain types of relationships as compared to others.

Hybridity and ICTs: making the connection

Acknowledging hybrid political orders can lead to a deeper understanding – rather than to an idealized imagining – of the roles individuals actually play, and the tactics they develop, within these complex and chaotic networks. It has also important repercussions on how ICTs are conceptualized and ICT-based projects are envisioned. New ICTs such as mobile phones, which are increasingly pervasive and able to integrate with other locally relevant means of communication – from poetry (Lamoureaux, 2010; Stremlau, 2012) to songs (Nyairo and Ogude, 2005) and word of mouth (Obadare, 2006) – do not need a Weberian state to affect governance.

As much as ICTs can alter relationships between individuals and public authorities in spaces characterized by high state legitimacy and by the ability of formal institutions to perform their stated functions, ICTs can have transformative effects even when these conditions are lacking. The problem is that most research on ICTs and governance in Africa has focused on the former rather than the latter, or, more critically, when it has focused on the latter – on spaces where formal and informal coexist – it has analysed them as if they were the former – privileging the citizen–state nexus and bureaucratic rationality (e.g. emphasizing accountability and transparency).

As indicated earlier, it is not that ICT for development scholarship (less often practice) has lacked self-criticism or failed to seek more locally grounded perspectives. Numerous authors have proposed ways to build on local knowledge, and allow users to progressively shape and master innovation, rather than be forced into new programmes from above (Grieve, 2004; James, 2005; Madon, 2002; Miller and Slater, 2000; Walsham, 2001). These critiques echoed the arguments previously put forward by the advocates of technological blending (Bhalla and James, 1988) and appropriate technologies (Stewart, 1987), insisting that new artefacts should embrace, rather than displace, older ones, making use of traditional knowledge and habits and be appropriated as part of a more harmonious path towards progress. They also resonate with the appeals in media studies to resist the fascination for the ‘new’ in new media, and rather than considering the latest communication technologies as incommensurably different from what has come before, understand them as part of broader media ecologies (Postman, 2000; Scolari, 2012) or study how they function in hybrid media systems where the new and the old are intertwined (Chadwick, 2013).

The main weakness of most of these positions, however, is that while they have sought to overcome techno-determinism, they have remained techno-centric, failing to extend their critical perspective beyond ICTs. If similarities do exist between literatures on hybrid governance and hybrid forms of communication, they have not yet been fully exploited.

In the next sections, by exploring and comparing how radio journalists, audiences and public authorities interacted in two different settings in Kenya, I empirically illustrate how the insights and critiques deriving from the literature on hybrid political orders can inform the study of how ICTs are affecting governance in Africa. I focus first on the language and competence required for callers to go and remain on air, then on the actors that are called upon by audiences and finally on the networks that are activated to connect audiences and public authorities. The findings lead to elaborating an empirically informed framework that emphasizes understanding, rather than assuming, which type of governance matters in a specific locality and how ICTs contribute both to its reproduction and reconfiguration.

Communicating in Kibera and Kisumu

Pamoja FM and Nam Lolwe are both local radio stations (Fardon and Furniss, 2000), catering to a limited audience and geographic area. Pamoja FM was launched in 2007 as a community radio station for the people living in Kibera, the largest slum in Nairobi. ¹ In Kibera, almost all Kenyan ethnic groups are represented, but they occupy different spaces. While Luo are the majority, most landlords are Nubian and Kikuyu, creating problematic fault lines between economic survival and ethnic divisions. Pamoja means ‘together’ in Kiswahili and, according to its funders, was established with the goal of promoting unity in this complex landscape. Nam Lolwe, which means Lake Victoria in Luo, was started as a political radio by Jakoyo Midiwo, a member of parliament (MP) for the Orange Democratic Movement (ODM) and caters to the Nyanza province, where Luo represent the dominant ethnic group. It initially served as a platform for the party to support its political campaign in an area recognized as an ODM stronghold, but after the 2007 elections it gained greater independence, adopting a fresher approach and trying to take a more active stance towards local political actors.

Speaking the language

The breakfast programmes of Radio Nam Lolwe and Pamoja FM follow a similar structure. They ‘wake up’ their communities at 5 a.m. and 6 a.m., respectively, offer them a space to exchange greetings and discuss the issues of the day. There is a balance between local, national and sometimes international news, mostly picked up from national newspapers and the Internet – and occasionally the opportunity to talk with leaders. The presenters act as brokers among various constituencies in their communities. In line with the rhetoric highlighting how the convergence between radio and other ICTs is supporting ‘participatory development processes’ (Gilberds and Myers, 2012), presenters often remark how they are offering a platform for people to raise their voice. As one of the hosts of Pamoja’s morning programme, Tuamke Pamoja, explained,

At the beginning, when we started our morning show, we could go on for three hours without receiving any call. And when people were calling they were afraid, they did not feel comfortable talking on the radio. Now they are much more confident, they know what they want and how to say it. Before people were talking about just a few issues, but now they can talk about politics, about corruption, they can accuse and say what they do not like in the community. ²

At first this seems to reinforce the image of the radio empowering ordinary citizens, as advanced by donor agencies and NGOs concerned about transparency and accountability. However, as it became increasingly clear during fieldwork at Pamoja first and later at Nam Lolwe, while these interactive spaces are open to all listeners with access to a phone, they are in practice inhabited by a much smaller cohort of recurrent characters. Presenters recognize their voices and some callers are addressed by their nicknames: ‘the bull fighter’, ‘senior citizen’, ‘the governor’. One of the hosts of Ye Lolwe, Nam Lolwe’s show, acknowledged that on a normal day two-thirds are recurrent callers. Outside of the radio, the prevalence of some voices over others has led people to think calls are arranged in advance (as one listener in Kibera alleged, ‘the presenters know before who is calling, because they recognize the number’), but the view from the studio illustrates this is not the case, simply because the technology used to manage calls is too rudimentary to show the caller’s number. Rather, in resonance with other research on radio talk shows (Chignell, 2011; Dori-Hacohen, 2012; Willems, 2013; Ytreberg, 2004), this phenomenon is rooted in the constructed and managed nature of audience participation.

Over time, the radio stations have created new public figures, people who know what language to use on air, who respect the format and are able to adopt a ‘participant role’ (Ytreberg, 2004). Their emergence challenges the simplistic rhetoric of democratic engagement, at various levels. In Ghana, some recurrent callers have been labelled ‘serial callers’ (Bob-Milliar, 2012), and they are activated and paid by political parties to call radio stations and advance a particular view. Other callers may simply see participation as an opportunity to build their reputational network. As one of Nam Lolwe’s recurrent characters noted, ‘I feel that my role in my community has increased. People around can tell me that they heard me on the radio and I said something that was important, that they agreed with’. ³ Another caller made a similar point, drawing on a more vivid imagery ‘Now I can be heard. Before people did not know me. But now my presence is felt in the community. If I go to a funeral now my presence is acknowledged’. ⁴ In rural Kenya, funerals are not only moments of mourning, but they are also a political space. For the funeral of a respected member of the community, an MP is likely to come back to his constituency, allowing people to reach him and express their claims. They can also become spaces where other individuals test their political weight and where the popularity they acquired through the media can be assessed among peers.

Across different cases, relationships between recurrent callers and radio stations are often mutually beneficial, as they help sustain a format that is both engaging and that supports the idea of radio as a forum for public participation. Callers are given, and seize, the opportunity to challenge public authorities, but, as explained below, they have to follow certain norms, and inappropriate and irrelevant calls are screened out. While the radio–mobile convergence has been touted as particularly powerful in Africa because of low literacy levels and widespread access to mobile phones (Gilberds and Myers, 2012), the reality of who calls and formulates a claim raises the issue of the competence that is required to first get on, and then to stay on, the air.

Talking about the state

Scholarship on hybrid governance stresses how multiple actors may operate on the same terrain, cooperating or competing for the provision of specific public goods and services (Olivier de Sardan, 2011; Rocha Menocal and Sharma, 2008). This potentially has important consequences for how people may use ICTs to relate to these actors. A first repercussion relates to accountability, as in some cases it can be more meaningful to use the media to address, and possibly challenge, NGOs or traditional authorities, rather than formal state institutions. The second consequence relates to coordination, as there may be instances when it is more effective to self-organize and bypass formal institutions in order to obtain a specific outcome. Both are now addressed in turn.

With regard to accountability, evidence emerging from Pamoja FM and Nam Lolwe appears to counter the initial hypothesis. Even when the state is perceived as corrupt and fractured, it continues to represent the simulacrum, the imagined figure to which listeners address most of their demands. In the slum of Kibera, the state has either left voids that have been occupied by a variety of actors belonging to both the civil and uncivil societies (Fatton, 1995) or has established ties with some of these actors, as indicated in the case of the water cartels illustrated below. In the case of Kisumu, the frustration of the repeated attempts to seize the state’s presidency by representatives of the Luo ethnic group that constitute the majority in the city and surrounding province has created a sense of distance and acrimony towards the central state (Mueller, 2008). And yet, through interactions in radio talk shows, the state is enacted as if it could be able to eventually provide the services it is currently unable to offer. Some calls and texts may simply be an opportunity for people to vent their frustrations, but in other instances audiences address their MPs or DOs directly, reminding them of their promises and of their roles. State authorities may even accept the challenge and respond on air, offering not only their own version of a particular case but also long lists of their achievements, which may connect more or less directly to the issue at stake.

It is not that other figures are absent from the debate on air. As a texter contributing to a discussion on Nam Lolwe regarding a proposal to raise the educational requirements for MPs wrote, ‘All leaders should be educated, including pastors’, bringing religious authorities into a discussion that had initially addressed only elected representatives. ⁵ Or in Pamoja, there are frequent references to water cartels that control the illegal provision of water in Kibera. These cartels, connecting officials of the Nairobi Water Authority, landlords, religious leaders and gangs (Birongo and Le, 2005), profoundly affect the life of the resident of Kibera, as water is mostly sourced at illegal kiosks for a price that is much higher than what is paid in richer neighbourhoods. ⁶ Some of the faces of these water cartels are well known, but they are never addressed directly, even when the main theme of a radio programme is water provision in the slum. As it emerged from interviews with journalists and audiences, one of the reasons for this omission is rooted on the violent nature of some of these cartels and the fear of retaliation. But even NGOs seem to receive a similar treatment. NGOs are a crucial presence in Kibera and often step in for the state in essential fields such as health, sanitation and education and yet they occupy a very little space in the debate on air. NGOs are present in the radio as they buy airtime to promote specific agendas, but they are not usually addressed by citizens in non-sponsored programmes as governance actors. One resident of Kibera responded during a focus group when asked why he did not use the radio to challenge the arbitrary nature in which an NGO had selected the beneficiaries of a project: ‘The NGO does not have to be here. They come freely and I cannot criticize if I am not part of a project. I don’t know what they are doing now, so I cannot denounce [them] on the radio’. ⁷

The literature on hybridity challenges normativity and offers a more comprehensive picture of how different networks of individuals and institutions do work to provide a specific service or good. When the ways in which people who are part of these networks publicly articulate their views and claims through calling and texting local radio stations, however, it appears that normativity, and not hybridity, is on their mind (and on their tongues). The state is not seen for what it is, but for what it should be. Other institutions are rarely called upon, despite their possible role in providing specific services and goods. This finding leads to considerations that are beyond the scope of this article and could be further developed elsewhere, but it does offer a concrete example of how research on ICTs and governance in the context of hybridity can enrich each other, presenting and testing new hypotheses that are relevant for both fields.

When it comes to coordination, however, which the literature on hybrid governance discussed earlier emphasizes as an essential resource to access goods and services (Booth, 2011; Olivier de Sardan, 2011), and that ICTs have the potential to dramatically improve, assumptions seem to be confirmed. This is particularly true in the case of emergencies. When asked when the combination of mobile phones and radio is the most effective, different groups, from regular callers to public authorities, referred to emergencies. In Kibera, these are epitomized by fire outbreaks and children getting lost in the slum. As one listener put it, ‘It is much better to go to a place and help when you hear from a radio there is a fire because if you wait for the police, they will never come’. ⁸ This extends to security threats, both in Kibera and in Kisumu. In some occasions, alerting the community about an ongoing event by calling the radio can be a particularly effective way to eventually get public authorities involved, even if, as explained below, there may be significant differences in the modes in which authorities respond.

‘We mostly belong all to one tribe’

Textbook approaches to citizen action tend to focus on critical moments such as demonstrations and revolutions to frame ICTs as tools that can be appropriated by citizens to mobilize against an unresponsive central authority, but in the everyday it is more likely that changes occur through the institutions holding formal and informal authority, rather than through their removal or substantial reconfiguration (Booth, 2011; Kelsall, 2013). Innovations in ICTs may increase the ability of citizens to participate in public debates and take an active role in the provision of public services and goods, but whether this increased participation has the power to eventually affect decision making and processes on a large scale depends on how other actors interpret it and incorporate it into their agendas. In the two communities of Kibera and Kisumu, interesting differences emerged in the ways in which public authorities respond to citizens’ claims, some of which can be traced back to the nature of the relationship connecting different nodes in a network. As indicated earlier, it may not simply be a demand of greater transparency or accountability to encourage authorities to act, but other types of links may be relevant (and needed) in order to trigger a response.

While both in Kibera and in Kisumu most of the actors involved in governance processes recognized the new spaces created by mobile–radio interactions as significant and useful for collecting information important for their jobs, substantial differences emerged in the ways in which actors in the two sites indicated they were making use of this information. In the case of Kibera, public authorities such as DOs, chiefs and assistant chiefs tended to show greater resistance to following up on claims made by citizens calling radio stations. They reclaimed the role of the state as the primary agent towards which citizens should address their demands and present their complaints. As one of them explained,

If someone calls the radio and denounces someone else for having done this and that, it is not an official communication. People now tend to forget this. They think that they can just call the radio, denounce something and that is it. But in many cases it is not enough, you need to come to an office of the administration and denounce what has happened, provide evidence, so that we can follow up. The residents may think that the media have the power to act, but they do not. ⁹

On the contrary, public authorities in Kisumu demonstrated a greater willingness to make use of the information collected through the radio and to integrate it in their daily activities. They saw the relationship between ICTs and the state as less adversarial. As a local chief described,

We can act through the radio. When I hear something on the radio I do act. I may decide to send my people there. […] I can learn for example that an area in Kisumu was raided by people who are known in the community. So why would I wait before acting on that? Even if what I hear in the radio is just an allegation, I have to act on what I hear. Then I can decide to have a baraza, so that people can talk to us. A baraza can allow us to get the facts from the ground. ¹⁰

Differences in the approaches adopted by public authorities in Kibera and in Kisumu were corroborated by the accounts collected from other actors in the two locations, including journalists and audiences, who often pointed to ethnicity as an explanation for the different behaviours registered among public authorities. The complex ethnic composition of Kibera makes it harder for chiefs and their assistants to react to some of the claims collected by radio stations for the fear that more deep-rooted grievances are behind these cases. In Kisumu, the more homogeneous nature of the networks linking citizens to public authorities increased the likelihood that the latter would act. While in one case ethnicity was an obstacle preventing certain claims from being picked up by public authorities and producing concrete outcomes, in the other, ethnicity increased the ability of certain types of information to produce concrete results. As a journalist in Kisumu summarized,

Here in Kisumu we mostly belong all to one tribe, so when something happens the authorities feel that they have to follow up, they are expected to, they may know who is involved, and that there is less risk that this event is simply made up. But if you are in Kibera where there are a lot of different tribes, you don’t really know if what is said may be the result of an ethnic tension, a tension between different groups, so you have to be more careful. ¹¹

Conclusion

In spaces where the idealized civic conception of the relation between citizens and the state that undergirds most ‘good governance’ programmes is only one of the many modes in which individuals interface with power, understanding how ICTs can influence governance should begin through exploring, rather than assuming, which actors, networks and languages actually matter. Comparing two interactive spaces emerging at the intersection between mobile phones and radio – the two most widespread communication technologies in Africa – and using the concept of hybridity to bring two disciplines in dialogue, this article has illustrated how such mapping and analysis can work in practice.

The exploration of the actors that populate different localities and the ways in which they are addressed by audiences calling and texting are where the mutual contribution of scholarship on ICTs and hybrid governance emerges more evidently. The literature on hybrid governance allows developing a deeper understanding of how various actors involved in governance processes operate every day, beyond dramatic junctures like elections or popular uprisings. And the analysis of how audiences, journalists and public authorities communicate in new interactive spaces is offering new perspectives to distinguish between how hybridity is experienced and how it is represented. As the comparison between Pamoja FM and Radio Nam Lolwe has indicated, even when the state is fractured or absent, and other actors step in its place, it continues to be called upon by callers advancing their demands for better services and greater responsiveness, while other actors remain in the background. This does not mean, however, that requests always appeal to bureaucratic types of relations, and often networks of ethnicity and patronage linking individuals and public authorities do make a difference in triggering responses.

The focus on exploring the networks that link audiences, journalists and public authorities also helps reconsidering the emphasis placed on ‘voice’ as the most important factor affected by the increasing availability of ICTs. As Tacchi (2012) pointed out, building on the work of Bickford (1996) and Couldry (2010), the interest in the ‘politics of speaking’ has prevailed on the necessity to understand how and by whom voice is actually listened to and valued. As she pledges,

the notion of voice as valued rather than simply facilitated […] focuses us on issues of attention and response, not simply about access and participation, but also the quality of relationships between speakers and listeners mediated by technologies and institutions. (Tacchi, 2012: 655)

As the analysis of the responses of public authorities in Kibera and Kisumu indicates, however, research that is open to exploring how networks function in specific localities should be ready to accept that the mechanisms leading public authorities to listen may lie beyond the realm of what is civic. Within the cases analysed in this article, ethnicity, for example, has emerged as one of these mechanisms. A lot has been written on how identities that are – potentially – in competition with purely political affiliations, including ethnicity, have been used as instruments for mobilization in Africa (Ajulu, 2002; Berman, 1998). Much less attention has been given to how ethnicity, but the same argument can also be extended to other types of relations (e.g. patron–client), may be used as an instrument for claim making. When these dimensions have been taken into consideration in research looking at ICTs, it has often been to highlight their disruptive potential, indicating, for example, how vernacular radio stations have incited ethnic hatred, or how the fear that one ethnic group could prevail over another has acted as a powerful catalyser in turning rumours – often spread by SMS – into violence (Abdi and Deane, 2008; Kellow and Steeves, 1998; Somerville, 2011). The normativeness of the ICT and good governance agenda has prevented the proliferation of studies exploring whether these dimensions actually play a role in creating pressure from below, eliciting responses from public authorities. The fact that claims resting on these types of identities are considered suboptimal when compared to more civic forms of ‘raising voice’ (mostly because they exploit in-group cohesion while they are likely to enhance out-group tensions) should not prevent studies from understanding how they actually work.

Finally, and also in partial contradiction with the rhetoric framing popular ICTs such as radio and mobile phones as necessarily democratizing, the cases presented in this study underline how a specific competence may be needed to first get on, and then to stay on, the air. Interactive spaces are open to all listeners with access to a phone, but they are in practice inhabited by those who know how to ‘speak the language’. This language does not have clear rules and contours but emerges from negotiation between journalists and audiences, both interested in sustaining a format that is engaging and that supports the idea of radio as a forum for public participation. Some people are even able to turn this competence into a regular occupation, becoming public figures to which other members of the community can refer to (sometimes for a price) if they want something they care about aired – and possibly listened to – on the radio. These practices can be seen as challenging the ideal of democratic participation. And in some cases they are, as a ‘serial caller’ may be paid to skew the debate in a particular direction while maintaining the appearance of engaging a genuine dialogue. However, as Brisset-Foucault (in press) points out in the case of serial callers in Uganda, many of them are emerging as new intermediaries carrying with them ambitions to educate themselves and others through the radio and whose presence and role in mediating voice demand greater attention.

These findings should encourage adopting a different outlook when exploring whether and how ICTs are affecting governance, challenging idealized narratives that have so far dominated both the theory and practice of ICT for development, but without losing the ability of understanding the transformative potential of new communication technologies from the ground up.