Seeing Like Bureaucracies: Rearranging Knowledge and Ignorance in Somalia

Seeing Like Bureaucracies: Rearranging Knowledge and Ignorance in Somalia Abstract Development promotes bureaucratization, and bureaucracies are based on knowledge and produce knowledge. Failures of development are therefore regularly attributed to a lack of knowledge. The article argues that the quest for knowledge is embedded in the managerial rationality of interventions. This rationality also structures the developmental knowledge field and thereby generates ignorance. The example of a state-building program in Somalia is used to empirically explore how the generation, administration, and transfer of knowledge was intertwined with ignorance. It shows what knowledge missed, obfuscated, ignored, or even hid and how knowledge and ignorance were arranged in the daily state-building practice. This approach sheds light on relations and mechanism of power exerted in development and helps to explain its effects. In Somalia, omission, silence, secrecy, and strategic and bureaucratic ignorance enabled the program to delineate the interventionist terrain as technical and to depoliticize state-building. They also helped to expand liberal modalities of government to “remote” and “unruly” Somali villages. Development, as a way to conceptualize and theorize human progression, and as practice to support this progression, is under critique.1 A large body of literature attests to the failures of both. Initiated by the seminal assessment of Ferguson (1994), authors began to focus on the effects of development beyond its own objectives and have since emphasized its bureaucratizing and depoliticizing tendencies. International organizations, for example, have been analyzed as complex bureaucratic machineries (Barnett and Finnemore 1999, 2004), which roll out interventions using a managerial logic (Mosse 2004; Goetschel and Hagmann 2009; Winckler 2011). While they continuously expand their own mode of operation, developmental bureaucracies also render the reason for the intervention, its main objectives and its implementation practice, technical. In this way, interventions are removed from the political realm of contestation and decision making and presented as technical solutions to mainly technical problems (Ferguson 1994; Li 2011). The expansionist tendency of bureaucracies had already been recognized at the beginning of the last century by Max Weber. Weber (1980, 653–55) argued that modern bureaucracies are efficient forms of social organization because they operate formally and instrumentally, base decisions on knowledge rather than personal connections, and weigh means against ends instead of balancing favors. As development promotes bureaucratization, and bureaucracies are based on knowledge and generate knowledge, the failure of development is regularly attributed to a lack of knowledge. Ignorance or nonknowledge2 is usually identified in two dimensions: (local) people's lack of appropriate knowledge on how best to develop their societies, and international organizations’ lack of knowledge about the society in which they are intervening. The former has been targeted by a transfer of knowledge from the global North to the global South. Training, capacity building, and learning-by-doing exercises have evolved into a core activity of development (Eade 1997, 2–3, 9–10).3 The latter is tackled by a broad range of international organizations, which are producing knowledge on the global state of development and displaying it in global reports and indices. Beyond this highly aggregated form of knowledge, a lack of (micro)knowledge about the societies in which interventions are planned is still considered as among the main reasons for development failure (recently Autesserre 2014). The need to better contextualize interventions initiated the local turn in development in the 1990s. However, when development touched upon the local, it subjected it to its own bureaucratizing dynamic. While some authors continued their search for the “real,” untouched, and unaffected local (Richmond 2012),4 others criticized the (bureaucratic) practice of development interventions and identified the spatial and social distance between the intervener and the intervened upon as reason for the continued failure to gather adequate knowledge, and for development failure more broadly (Bayart 2000; Mosse 2004; Goetschel and Hagmann 2009; Autesserre 2014). This article contributes to these discussions from a “sociology of knowledge” viewpoint. It argues that the quest for better, more accurate and detailed knowledge is part of the managerial rationality of interventions. This rationality also structures the developmental knowledge field, and, while it delimits the type of knowledge that is generated, it also produces unknowns and ignorance. Put into practice, development then continues to generate and rearrange both knowledge and ignorance. The example of a state-building program is then used to empirically explore the generation, administration, and transfer of knowledge but puts its focus on ignorance and thus on what knowledge missed, obfuscated, ignored, or even hid and how knowledge and ignorance were intertwined in the daily state-building practice. This approach sheds light on the relations and mechanisms of power exerted in the way development lays claim to knowledge. At the same time, it avoids common categorizations of local versus external knowledge, the identification of the former as more accurate and the latter as simplistic, technocratic, or imposed, a claim that is usually followed by a call either for more and more accurate knowledge or for better suited (participatory) methodologies to capture the local. In spite of the focus on ignorance, the effects of development are, in this reading, not mainly attributed to a right or wrong reading of the local but analyzed through the ways in which knowledge and ignorance were arranged to make the local legible and accessible to state-building interventions. The argument is developed in three steps. Drawing selectively on the governmentality literature, I first outline how the neoliberal rearrangement, diffusion, and down-scaling of state regulatory competences emphasized distinction and revalued local and expanded community-based technologies. Executed as “government through community” (Rose 1996; Li 2007, 2011), neoliberal modes of government5 constituted communities as objects of knowledge and subjected them to interventions that activated self-regulatory techniques and practices. These interventions made the local accessible to a managerial rationality and expanded bureaucratic modes of thought and operation but delinked bureaucracies from the state, flattened bureaucratic hierarchies, and dispersed bureaucratic practices, a process I label, with Hibou (2015), as “neo-liberal bureaucratization.” This section also clarifies how bureaucracies lay claim to knowledge and discusses their (de)politicizing effects. The second section builds on insights from the recently revived field of “ignorance studies.”6 It draws theoretically on those aspects of the work of Foucault (1978), Mitchell (1991, 2002), and Taussig (1999) that help to build the argument that bureaucracies do not execute their power through knowledge but through rearranging knowledge and ignorance and through enacting the knowledge-ignorance assemblages in a way that subsumes the social to their own managerial rationality. The third section provides an in-depth empirical study of the way knowledge and ignorance were generated, arranged, and put into practice in a state-building project in Somalia. This section explores which unknowns were identified by the program while others were overseen or ignored (deliberately so or not), how the identified unknowns were made actionable for exploration, the technologies used to generate knowledge, and the ways in which new knowns were managed and initiated further actions (or not). It shows that while the intervention set out to generate knowledge on the local and to build on local knowledge to improve government in Somalia, it was also enabled by and generated ignorance, in particular omission, silence and secrecy, vagueness, and strategic and bureaucratic ignorance. These forms of ignorance were part of the interventionist design, enabling the program to delineate the interventionist terrain as technical and thus to expand managerial techniques, which depoliticized state-building. They also facilitated the “smooth” implementation of the program and helped to expand neoliberal modalities of government and with them a managerial rationality to the “remote” and “unruly” Somali districts and villages. Governing Through Communities: Bureaucratization beyond the State Max Weber's (1980, 126) famous finding that domination in everyday life is primarily executed as administration gained new evidence in the context of international development. More than two decades ago, James Ferguson (1994) concluded from his assessment of a development project in Lesotho that such projects have failed to bring about development but have significantly strengthened the state. They facilitated the expansion of a bureaucratic apparatus and thereby enabled the state to execute more “effective control and administrative supervision” (Ferguson 1994, 265) over people who were previously largely uncaptured. In the decades following Ferguson's assessment, international actors ceded to support the state. The neoliberal restructuring of state-society-market relations, which started in the 1970s, was promoted by a call toward less bureaucratic and more flexible and responsive structures of government. Communities were thereby constituted as mediators between the centrally and bureaucratically executed power of the state and the free exchange system of the market (Rose 1999 167–73), and community-based technologies promoted to replace the overexpanded state bureaucracy. However, “government through community” (Rose 1999; Li 2011)7 did not diminish bureaucracies but delinked them from the state and decentralized their mode of operation. Bureaucratic practices are not limited to the state.8 Instead, they make the social legible9 by generating a particular kind of “modern” or abstract knowledge, that is a knowledge that can be accumulated and instrumentally applied to intervene in, control, and improve nature and society. Bureaucracies enter into social processes and disaggregate the social into separate parts and functions, which they map, aggregate, and quantify. They then standardize the magnitude of data and transform it into abstract knowledge displayed in lists, tables, matrices, and diagrams (Spittler 1980). While emphasizing heterogeneity and distinction, government through community made the local accessible to bureaucratic knowledge and with it to standardization. Opinion polls, focus group discussions, roundtables, and participatory rural appraisals are examples of the increasing variety of participatory knowledge tools used to identify, map, measure, and categorize communities. Communities were thereby constituted as unitary social groups—women, minorities, youth, etc.—with common characteristics but also with unique concerns. They were thereby made thinkable “in a way that their ills appear susceptible to diagnosis, prescription and cure” (Rose and Miller 1992, 183) and thus to optimization through interventions and programming (Rose 1996, 189; Rose 1999 174; Cruikshank 1999, 78; Li 2007, 2011). While making the social legible, controllable, and governable, bureaucracies strengthen their own operational rationality and facilitate their continuous expansion (Spittler 1980; Barnett and Finnemore 1999, 707; Winckler 2011, 88–89). Bureaucracies are expansionist. They are also depoliticizing. Classical political sociologists, among them authors as diverse as Karl Marx, Max Weber, Pierre Bourdieu, and Michel Foucault, conceptualized the political as a realm of public dissent, contestation, hegemonic struggles, and conflicts (Schlichte 2012). If issues are made political or politicized, they are publicly discussed and challenged, sometimes even violently. Depoliticization, in contrast, reduces open contestation and conflict and conceals struggles. However, the practice of politics is, as Barry (2001, 205–9; 2002, 269–70) reminds us, not only about contestation and conflict but also about the containment of both. One way of containing conflicts is to shift the contested issues from the realm of public contestation to other societal domains, among them administration, and to hand them over to experts, who evoke their specialized knowledge, analyze, measure and evaluate, and come up with solutions10—a practice that was captured as rendering technical (Rose 1999, 175; Li 2011). Depoliticization is thus an important part of the repertoire of politics and, if we follow Ferguson (1994) and many others, also of the political repertoire of development. In order to empirically explore processes of politicization and depoliticization in contemporary politics, Lipschutz (2005) suggested differentiating between constitutional and distributional aspects of the political, the former concerned with the structural framework for political actions, the latter with procedural aspects and especially with the distribution of resources and benefits. With an increased density of political institutions, the externalization of state regulatory competences to nonstate actors, and the parallel expansion of a managerial culture, distributional and procedural aspects come to the fore of political contestation (Jaeger 2007). The quite technocratic focus of contemporary political debates (and the governance literature) on the most efficient arrangement of institutions and the most equitable management of resources can be interpreted as an effect of bureaucratization. They indicate the self-perpetuation of bureaucratic modes of domination and thus of modalities of rule that take the given order for granted while limiting the political to discussions about its most efficient execution (Schlichte 2012, 10). The Power of Bureaucracies: Rearranging Knowledge and Ignorance Development, even in its community-based form, is bureaucratic. Bureaucracies are set up as a particular way of knowing and ordering the social and thereby display expansionist and depoliticizing tendencies. Several authors have attested to both and have paid particular attention to bureaucratic knowledge practices. More recently, however, authors have also started to explore ignorance (McGoey 2007; Best 2012; Hull 2012). Ignorance is used as a summary term for everything that is not known, including false, imprecise, or incomplete knowledge. Instead of delineating a mere void, ignorance studies stress that knowledge and ignorance are interlinked, that ignorance comes in diverse forms and modalities and can be generated and used intentionally or unintentionally, and that it, too, is attached to power.11 Following Simmel, who identified the interplay between knowledge and nonknowledge, and the strategic use of both as crucial for social interactions,12 a number of authors have drawn attention to the intentional and strategic use of ignorance.13Trotha (1994), for example, examined how colonial officers and local intermediaries in Togo actively omitted or disguised information to enable the establishment of “rule from a distance” but also to strengthen their own position in the colonial administration (Trotha 1994, 379, 409-11). Mathews (2008), using the example of Mexican forest management, has shown how official state knowledge was negotiated between local stakeholders and state officials and how the latter discreetly ignored information that would have disturbed their work. Rappert (2012, 43) has additionally outlined how the UK government, during the 2003 invasion in Iraq, enacted ignorance to deflect political criticism. While ignorance was thus recognized as an important source of power, most authors focused on the intentional use of it. Aradau (2017, 330–31) additionally emphasized the generative effects of ignorance. Feminist studies have, for example, shown that knowledge engenders “epistemic injustice” and produces “epistemically disadvantaged subjects” while discrediting particular social groups as “not knowing” or as being “incredible knowers.” A similar observation was made by postcolonial studies. The historically specific experiences of Europe were set as the norm and mistaken for the universal, while non-European experiences were ignored, overseen, devalued, and discredited. A “geopolitics of knowledge” emerged, which continues to support interventionist approaches, as it provides the categories from which the rest of the world “can be described, classified, understood and ‘improved’” (Mignolo 2005, 36). Ignorance is thus enacted through, and functions in conjunction with, knowledge. Beyond its intentional use, ignorance can be explored as both a form and an effect of power. Taussig (1999) even characterized the tensions between concealment of knowledge and its revelation as the central feature of power. Accordingly, knowledge may be power, but the power of knowledge is only known through its disguise and the creation of public secrets, which then unfold their power through an ongoing “act of unmasking” (Taussig 1999, 105, 145). Taussig follows Foucault (1978) who, among others, used the example of sex to show that the power of secrets does not lie in the repression of truth but in their continuous revelation. Foucault discussed in his history of sexuality how the rising “will to knowledge” engendered a boom of discourses and truth claims that enabled the administration, management, and optimization of sex as part of a “political economy of population” (Foucault 1978, 12, 26). In a similar vein, Mitchell (2002) explored how new technologies and methods of knowledge, among them maps, cadaster, or statistics, contributed to the making of the (capitalist) economy in Egypt. These methods helped to rearrange the material, political, and social worlds according to the laws of the market. Both Foucault and Mitchell have pointed to the expansion of economic categories and knowledge practices into noneconomic domains. While these domains were now subjected to economic calculation and optimization, they were also made accessible to managerial techniques. Control is thereby not executed centrally but through the dispersion of norms and rules, measurements and calculations, performance assessments, indicators, and audits, in short through bureaucratic practices (Hibou 2015, 22). Bureaucracies thus unfold as a nearly ideal-typical form of disciplinary power. They disaggregate societal elements and processes, transform them into intelligible domains in need of improvement, and increase efficiency and precision while reassembling the parts “into more productive and powerful combinations” (Mitchell 1991, 93). Foucault and Mitchell also stressed that the rise of scientific methods, detailed analyses, and continuous investigation neither necessarily enhanced the truth about the studied phenomena nor made knowledge more accurate. Instead, they constituted objects of knowledge (sex, the economy, the community) by rearranging the relationship between the known and the unknown, the said and the unsaid, and thus through a web of discourses and practices in which forms of knowledge and ignorance were redistributed and the normal and the abnormal rearranged (Mitchell 2002, 91; Foucault 1978, 24–26, 56–57). The next section explores the unknowns and how knowledge and ignorance were arranged and set into practice in a community-based state-building intervention in Somalia. This approach sheds light on the social relations and flows of power that are ingrained in the production of knowledge, in the case of this article, knowledge on government in Somalia. Building States from the Bottom-Up: Experiments in Somalia The Somali state apparatus collapsed in 1991 during the course of an ongoing civil war that started in 1988. Violence has since been concentrated in the southern and central parts of the country. The clan-based militias that fought against the military dictatorship (1988–91) and then against themselves (1991–92) fragmented in the course of a failed UN intervention (1992–95). Afterward, the country was divided into a patchwork of localized forms of rule, but violence decreased (Menkhaus 1997; Bakonyi 2013). The political dynamics changed again after a Transitional Federal Government (TFG) was established in 2005. The rise of an Islamist movement, US counter-terrorism operations, and the military involvement of neighboring and international troops in Somalia initiated new rounds of severe violence. An Islamist umbrella organization, the Islamic Court Union (ICU), successfully mobilized resistance against the TFG but was, at the turn of 2006/07, ousted by the US-backed military intervention of Ethiopia. The TFG established itself in Mogadishu but was challenged by an Islamist insurgency, spearheaded by Harakat al-Shabaab al-Mujahideen (al-Shabaab).14 By 2010, al-Shabaab controlled most of south-central Somalia, while the TFG clung to a few districts in Mogadishu. International donors eventually stepped up their financial and military aid, and the African Union Mission to Somalia (AMISOM)15 intensified the war against al-Shabaab. AMISOM, together with the partially reconstructed Somali National Army and allied clan militias, pushed al-Shabaab back to the rural areas. From 2011 onward, the TFG, later transformed into the Federal Government of Somalia, or one of its allied militias, took control of most urban centers in south-central Somalia. Al-Shabaab, however, continues with guerrilla-style warfare and regular suicide attacks against the Somali government and its allies. Much less affected by violence were the northern regions of the country. The northwestern Republic of Somaliland declared independence in 1991, but is not internationally recognized, although it managed to re-establish a central state and embarked on democratization. Its eastern neighbor, Puntland, declared its semi-autonomous status in 1998 but became later a member-state in the Federal Republic of Somalia. International attempts toward peace- and state-building initially focused on the war-torn southern regions but soon had to acknowledge that power is dispersed in Somalia and that the reconstructed state institutions did not reach far beyond Mogadishu. Since 2008, UN organizations have additionally supported state-building from the “bottom-up” and implemented a “Joint Program on Local Government and Decentralized Service Delivery” (JPLG) in Somaliland and Puntland. JPLG supports the establishment of district and regional offices and builds the capacity of its staff. Between 2012 and 2015, JPLG was complemented by a program that aimed at expanding state structures from the regional and district level to remoter rural areas. This program was funded by DFID and implemented by a consortium of three international organizations with longstanding experience in community-based development. The consortium adapted its approaches to the reconstruction of village-level governance and implemented its program in sixty villages (some of which were actually towns) in three districts in the east of Somaliland and the southwest of Puntland. Funded by DFID, and carried out from October 2014 to September 2015, an evaluation of the program used a mixed-method approach to assess if it had reached its main objectives. It conducted seventy-nine semistructured interviews and forty-seven focus group discussions and administered 1,604 household surveys in fifty-seven of sixty program villages and in thirteen control villages. The evaluation team also reviewed a large number of program documents, among them the proposal, the original and revised theory of change (ToC) and logical framework (logframe), the monitoring and evaluation framework, training and learning handbooks, community action planning documentation (village reports), training manuals, and internal and external baseline studies. At the beginning and end of the evaluation, workshops with program staff were organized in Hargeysa (Somaliland) and Nairobi (Kenya). Communication with staff from the offices in New York, London, Nairobi, Hargeysa, and Garowe was ongoing throughout the evaluation. This article will draw more on the documents and communications than on the findings of the evaluation,16 which will mainly be used to illustrate the argument. State-Building as Management: Bureaucratic Knowledge Practices Development provides an example of how management has set itself up as a way of knowing and learning about societies. The need to continuously collect information is built into the different sequences of the program management cycle, a tool commonly used in development programming. The cycle encourages a trial and error approach and emphasizes knowledge gathering as a means to evidence assumptions, monitor progress, avert undesired effects, and mitigate against risks. It suggests the implementation of programs in an incremental helix of sequences ranging from problem identification → program design → program implementation → monitoring (preliminary) results → and eventually to evaluating results and longer-term impacts. Learning is an integral part of the cycle, and monitoring and reflection loops cross-cut through the sequences. They aim to safeguard the newly gathered information, especially knowledge on how the program impacts on local social dynamics, and utilize it to ensure that the program progresses in the right direction. Program knowledge is collated in reports, among them feasibility and baseline studies, monitoring reports, and mid-term and end-line evaluations. Program experiences, as well as monitoring and evaluation results, are further aggregated to “lessons learnt,” “best practices,” or “guiding principles,” which are collectively endorsed by a broad range of development actors (Suhrke 2007, 1292). The Somali program followed the sequences of the management cycle, embraced its trial and error logic, and utilized a variety of participatory tools to ensure that the planning, design, and implementation of the program were based on locally grounded knowledge and allowed for (joint) learning. In an initial proposal, donors and implementing organizations agreed on the main representation of the local and established a framework for their interaction with it. The structure of the program proposal was quite conventional and included a contextual analysis, a (preliminary) ToC, a logframe, and a work plan. The contextual analysis mapped out the interventionist terrain by identifying particular challenges and problems, in the Somali case, of local governance. This was followed by the ToC, a planning and monitoring tool used to develop strategies on how identified problems can best be addressed. It consists of a set of hypotheses (if-then statements) on how change is likely to unfold if certain inputs are provided and if certain conditions are met (main assumptions). A ToC underscores the experimental character of interventions and uses prediction (hypothesis), manipulation (program inputs), testing (implementation), and control (monitoring and evaluation) to explore whether the deficient present can be steered into a better future. The ToC was further translated into a logframe, another management tool that breaks down the change hypothesis into smaller tasks and sequences and aligns them to a specific time frame. During the program implementation, the logframe is further dissected into work plans in which the day-to-day activities of program staff are specified. The proposal and program implementation thus provided a nearly ideal-typical example of the above-described way in which bureaucracies enter into, disaggregate, and reassemble the social, aiming at optimization. This process is based on and generates knowledge. It is also, as the following will show, based on and generates ignorance, including vagueness, omission, strategic ignorance, active secrecy, and bureaucratic ignorance. The Interventionist Terrain: Omission, Silence, and Strategic Ignorance In its context analysis, the consortium examined local governance in Somalia and described the village as “the heart of the Somali community” (Consortium 2012, 23). Key changes at village-level institutions were considered likely to contribute to stabilization, conflict resolution, and civil and economic development (Consortium 2012, 23). The proposal described how villages are governed by different institutions, among them the Village Council (VC), sectoral subcommittees, elders, and religious authorities. It particularly emphasized the importance of elders in conflict-resolution and the management of everyday village affairs (Consortium 2012, 8, 10, 11). On the downside, however, it was noted that “traditional structures” tend to marginalize particular social groups, among them “minority clans, women and youth” (Consortium 2012, 22; 34). The proposal additionally identified a lack of governance efficiency due to overlapping roles and the duplication of responsibilities, as well as a general lack of capacity. Accordingly, local authorities lack the capacity to plan village development and to administer villagers’ needs, while citizens lack the capacity to articulate demands and to participate in decision making (Consortium 2012, 5, 12, 22). Similar to Green's (2000) observations on participatory approaches in Tanzania, the Somali program simultaneously asserted and denied local practices and expertise. It emphasized the success and importance of hybrid institutional arrangements but only to diagnose their inadequacy, inefficiency, and need for improvement. This diagnosis opened up the space for interventionist action and attested to the superiority of interventionist practices and knowledge, which can help to improve local governance. Additionally, however, the proposal reduced the role of government to the (bureaucratically organized) management of community needs and the delivery of services. Questions such as “who rules?” or, alternatively, “who should rule?”; “on which basis?”; “about what or whom?”; and “how so?”—questions which invite an exploration of the constitutional aspects of politics—were absent in the proposal. These omissions precluded discussions on alternative forms of authority and promoted a view of government as the management of resources. The program discussed state-building with villagers but channeled the debate to the distributional aspects of rule, an approach that was matched by the conceptualization of citizens as beneficiaries and thus consumers of government services (see also next section). The transformation of state officials into managers and citizens into consumers is, however, a crucial part of the depoliticized and “bureaucratized vision of the political” that neoliberalism engenders (Hibou 2015, 52). The avoidance of constitutional debates and thus of substantive aspects of politics became most pronounced in the ignorance of clan relations. The Somali society is structured by a segmentary lineage system that subdivides ethnic Somalis in patrilinear descent groups and structures them in a pyramidal order in clan families, clans, subclans, and the smallest segment, the household. Similar to ethnicity in other postcolonial contexts, clan affiliations provide a channel through which wealth, power, and status are expressed and through which many aspects of social, political, and economic life are structured. The ongoing formation of central, regional, and local governments is, for example, primarily negotiated as clan-based representation. When asked why clan relations were not taken into consideration while designing and (re-)building government at the village level, program staff referred to the difficulty and sensitivity of obtaining clan-related information. Although the evaluation team did not face these difficulties, an in-depth assessment of existing representational mechanisms and an open discussion on their legitimacy may well have amplified political tensions. After all, the formation of governments, and thus the decision of who rules and how, is a conflictive political process, and one that is often negotiated violently in Somalia. The program decided to strategically ignore the clan-based and politically contested character of both the village community and the village government. The omission of clan-related information and the waiving of information related to existing mechanisms and practices of representation enabled the interventionists to approach local governance as technical terrain and thus as suitable for an intervention. It also enabled the international state-builders to present themselves as politically neutral. After all, international organizations claim legitimacy for their actions by presenting themselves as “not exercising power but instead serving others” (Barnett and Finnemore 1999, 21). The depoliticization of constitutional aspects of government in Somalia did not play out smoothly. (Strategically) unrecognized or at least unacknowledged by the international state-builders, clan affiliation affected nearly every aspect of the program, be it participation in community meetings, attendance at training, the organization of community contributions, or the distribution of program resources (Bakonyi, Cohen, and Bédard 2015). While clan relations structure socioeconomic and political life in Somali villages, a silent consensus emerged between villagers and program staff to not touch upon the political character of governance. We know from Foucault (1978, 27) that the production of silences does not limit what is said and known but functions “alongside the things said, with them and in relation to them within over-all strategies.” Silences are part of negotiated discretions between speakers who determine (consciously or not) “which type of discourse is authorized, or which form of discretion is required” (Foucault 1978, 27). In Somalia, and in order to act upon the local in a way that impedes political struggle and conflict, “uncomfortable knowledge” (Rayner 2012) was eliminated and silences were produced and reproduced in the communication between state-building actors. The proposal also emphasized the need to improve the modalities of government, in particular to optimize the division of labor in order to increase efficiency. The improvement strategy, however, focused mainly on the VC. Elders, religious authorities, even the sectoral subcommittees were downgraded to supporting actors, and the village government discursively centralized around the institution that (at least ideally) complied most with the interveners’ views on how governments should be organized: in a formal-bureaucratic way, through proper offices, and with clearly demarcated tasks and responsibilities. The omission of elders and religious authorities became a striking feature of the program. Shortly after the program's inception, the strength, weakness, opportunities, and threats (SWOT) of village-level governance were analyzed in each of the sixty villages. However, the SWOT analysis focused exclusively on the VC, while elders and religious authorities were not even considered as part of village governments. The Government Improvement Plans that were designed on the basis of the analysis, therefore, dealt only with this institution. Somehow, on the way from the context analysis to the strategy development and state-building practice, the VC was conceptualized as the center of village government, despite being characterized as particularly weak and largely dependent on elders. Omission, silences, and strategic ignorance enabled the program to promote a bureaucratic way of thinking about, talking about, and practicing government as formally regulated, practiced by a dominant institution, and exercised in a managerial manner. Transforming (Some) People into Active Citizens: Simplifications and Secrecy The Somali state-building program problematized particular aspects of local government, in particular the lack of active involvement of villagers in decision making. Women, youth, and minority clans were in this context identified as particularly vulnerable, without voice, unheard, and passive. This diagnosis was to be corrected by “empowering” villagers “to take on a larger role in claiming their rights and services and engaging in decision-making and planning processes” (Consortium 2012, 22). Villagers were conceptualized as political citizens, active agents who contribute to the execution of government. However, similar to the above-outlined omissions of constitutive aspects of government, the modalities of citizens’ participation in decision making were never discussed. In its implementation practice, the program instead promoted a politically impoverished concept of citizenship, portraying villagers mainly as customers and consumers of government services and assigning them with tasks such as communicating their needs, overseeing the distribution of services, and ensuring that they receive their “fair” share of resources and benefits. Political enunciation was to happen in a formalized institutional setup and to be reduced to a discussion of distributional aspects of the political. The deviation from substantive, and the focus on distributive, aspects of citizenship built on the representation of village governance as nonparticipatory and noninclusive. It is worth noting that this diagnosis was not shared by many villagers, who emphasized the participatory and inclusive nature of village-level decision making and described diverse participatory practices ranging from representation through elders to consultation and joint decision making, the latter particularly in situations of crisis (Bakonyi et al. 2015, 27–31). Female and younger respondents’ views did not differ significantly from those of older men and, with the exception of people from minority clans, the simplified representation of marginalization in the proposal was not reflected in the self-perception of people from these groups (Bakonyi et al. 2015, 31–40).17 Generational and gender relations are more complex than the proposal suggested (see El-Bushra and Gardner 2016). The binary juxtaposition of those with power and the powerless (and the subsequent attempt to “empower” those without power) is itself mediated by power. It places expertise, knowledge, and agency in the hands of experts who are able to know who is powerless and marginalized, a knowledge that allows them to speak for, act upon, and act in the interest of the powerless (Cruikshank 1993, 38-40). To be politically empowered was equated with being able to participate in the VC, and the identified lack of participation was then rationalized as a problem of capacity. Capacity transfer from those who know (international and local experts and trainers) to those who do not know (villagers, village authorities) became the obvious solution. Training in “civil education” or “dialogue processes” for villagers more generally was accompanied by training in leadership skills and advocacy for the marginalized groups. The latter were additionally supported to practice their newly acquired skills in public meetings, while villagers were encouraged “to hold Village Councils to account” (Consortium 2012, 34). The way in which the problem of government was codified as a problem of capacity is not merely a failure to accommodate complexities or caused by a lack of knowledge about local practices. Problems and features of government, including marginalization and exclusion, were individualized as lack of skills, leadership capacities, or the ability to speak publicly. Responsibility for their inabilities, unequal status, and lack of power was thus assigned to the villagers themselves—of course with the prospect that the deficiencies were to be corrected by skills development. The simplified representation of villagers as passive, and of some groups as marginalized and vulnerable, in need of empowerment, was followed by the propagation of a mode of government that is executed through the active participation and choice of (empowered) community members. The program promoted what Cruikshank (1993, 1999) labeled “technology of citizenship,” that is the rearticulation of relations between government and citizens, which make a (neo)liberal form of government operable in Somali villages. It set out to support villagers to play a more active role in village governance, to participate in meetings and decision-making, and to transform marginalized people into “new actors” (Consortium 2012, 22), engaged and vocal enough “to ensure their issues and concerns are addressed” (Consortium 2012, 22). The reconstruction of local level government thus involved the strengthening of the “self-management aspirations” (Rose 1996, 41, 61) of villagers and their transformation from powerless into “active citizens,” capable of promoting themselves and taking their “rightful place as self-actualizing and demanding subjects” (Rose 1996, 60; also Dean 2010, 82-86). The program thus aimed at reconfiguring the subjects of governments and propagated the same kind of “active citizenship” that has been accompanying the “re-invention of government” in advanced liberal democracies (Cruikshank 1993; Rose 1996). This required and built upon simplified versions of gender and generational relations, a depoliticized rationality of rule as management of (human) resources, and of citizenship as resource consumption. Interestingly, however, minority clans,18 the only group that clearly described themselves as marginalized and discriminated against, were not targeted by the program. The strategic ignorance of clan affiliation and the silences that emerged around the clan were matched by secrecy from the villagers when touching upon minority clans. While the Somali program originally aimed at empowering and enhancing the status of minorities, it failed to even identify members of this group. Village authorities often denied the existence of minorities in their villages, claiming that they live mainly in cities. While both the international and Somali members of the consultancy team were able to identify minorities in many villages, the program staff took the local authorities’ denial at face value. The reason for this became clearer when two Somali enumerators in the consultancy team expressed their discomfort with actively seeking out minorities for interviews. They argued that it would be “shameful” for minorities to be identified as such and insisted that enumerators should not openly address them. This type of active disguise was embedded in culturally anchored sentiments and taboos. The secrecy of village authorities reinforced these sentiments but also provided an escape route for Somali staff and released them from engaging in a culturally uncomfortable situation. Internationals, however, did not investigate closely either. As long as a program moves on, employees often decide to discreetly ignore issues that could interfere with the implementation. Again, ignorance became a resource of action, displayed here as secrecy and as a decision not to look too closely to avoid disturbing the everyday practices of the program. This example also accentuates how both knowledge and ignorance are coproduced through negotiations in which power plays an important role. Constituting Communities through Knowledge Tools: Bureaucratic Ignorance Community-driven programs have developed a broad range of knowledge and planning tools to contextualize programming, among them, in Somalia, the mapping of village institutions (VENN diagram), seasonal calendars, socioeconomic analysis, and livelihood analysis. The participatory approach is aimed at generating local knowledge, ensuring that the “community [is] in the driver's seat in the process of designing and implementing their own development plans,” and enabling “the community to understand and overcome their problems in order to rebuild their lives” (Village Report). Villagers were encouraged to discuss and prioritize needs and to develop adequate community action plans (CAPs). The findings of the participatory exercises were compiled in village reports, which included information about history, institutional setup, and the challenges facing villages. The reports were complemented by a number of further documents, among them minutes of meetings, lists of participants, signed CAPs, or training manuals. These documents were centrally stored, but the information they contained was not analyzed, systematized, standardized, or in any other way imbued with meaning. Beyond the CAPs, which identified surprisingly uniform projects (Bakonyi at el. 2015, 79), village-related information did not influence the program's practice. The participatory tools generated information that was, however, rendered inactive and which contributed to what Mathews (2008) has labeled “bureaucratic ignorance.” Indeed, it seems that many reports were never read, as several contained easily detectable copy-and-paste errors.19 The production of information which is then rendered inactive seems futile. The knowledge generating exercises, however, had further functions: they generated legitimacy and were used to demonstrate the program's anchorage in the local. Employees often referred to the files and reports to demonstrate their attentiveness to matters in even the remotest village, to show that local specifics were taken into account, and to evidence the partnership between the program and the local population. The exercises also contributed to the constitution of communities as object and subject of government. Communities are made governable when they are made legible, “objectified by positive knowledge” and subjected “to truth claims by expertise” (Rose 1999, 189; also Cruikshank 1999, 40). The surveys, lists, maps, reports, and diagrams that were produced from the participatory exercises were not simple tools to stimulate discussion, generate knowledge, or “empower” people. They are better understood as technologies through which the village is established as a community, transformed into a governable object, and subjected to government in the form of community self-management. The newly constituted and empowered community is then supposed to act in a way that optimizes government by making it “more effective” (Consortium 2012, 12, 17, 26). With respect to the activation of self-management aspirations, the program was quite successful. It channeled villagers’ expectations and shifted their perceptions of government responsibilities from the regional and district government to the VC and thus to the village and “community” level. The trust and reliance of villagers in the VC increased significantly, although villagers continued to consider the governance capacities of the VC as weak (Bakonyi et al. 2015, 74–75). Translating Concepts into Practice: Vagueness and Tricks Although often criticized as among the reasons for development failure, the Somali program built its strategy on fairly vague concepts and failed to provide conceptual specificity or to adapt concepts to local conditions.20 The term community was, for example, used to characterize the implementation method (community driven), the main beneficiaries (the community), participants in meetings (community meeting), and as a substitute for the village in which the program was implemented. In a similar vein, a large number of program activities were labeled as participation, be it meetings between program staff and local authorities, meetings among villagers or groups of villagers, or the contribution of money and labor, even if people were hired to do the job. It is usually assumed that vagueness and the inability to adapt concepts is driven by a lack of contextual knowledge. The Somali experience, however, suggests the opposite, that vagueness enabled the translation of concepts into different contexts. The concept of community as it has emerged in development theory and practice since the 1980s seems to resist precision and specification. Instead, its attractiveness lies in its ability to integrate ambiguous, even contradictory, meanings. While community assumes the existence of a culturally specific microcosmos, it delineates a moral rather than a social field, structured by emotional, affect-laden, and durable relations that are built on shared norms and values (Rose 1999, 172). It is this morality that provides participatory approaches with the necessary allure of optimism and purpose (Cornwall and Brock 2005, 1045) needed to activate social relations among people willing to cooperate and manage themselves as communities, through which government can then be executed. There was no need to explore how villagers understand community or to adapt the concept to the local—taking clan-relations or practices of representation into account, for example. Instead, the program invoked moral notions of community and participation and facilitated participatory exercises in order to establish a set of relations between those villagers able and willing to become the village community, that is, to become a “group of people who could be acted upon” and who are themselves “acting together to form ‘communities’” (Cruikshank 1999, 78). The “community” thus embodies a set of relations between people willing (and allowed to, excluding, in the Somali case, minorities) to take over the “driver's seat” and administer themselves in the name of community. This also explains why the program worked mainly with people who were already quite active in the villages, usually people who were organized in community-based organizations and had previously cooperated with international organizations (Bakonyi et al. 2015, 76). The program's attempt to establish communities as self-regulatory bodies—independent of the social reality of already existing communities—became obvious when the program turned its focus to the three district capitals and thus larger cities. Instead of adapting the program practice to the context of the city, the cities were adapted to the rationality of the program: they were first divided into three or four “villages,” the people living within these newly drawn boundaries were then labeled as communities and treated in the same way as the “communities” in the villages. Rather than being caused by a lack of (local) knowledge, conceptual vagueness provided the necessary flexibility to put the program's approaches into practice. Vagueness also allows a broad interpretation of program effects. The sketchy understanding of participation, for example, enabled the program to observe an increase in participation, to use this observation as an indicator of “empowerment” and thus to demonstrate the program's positive impact. Villagers often confirmed this impact. Asked about the effects of the program, an increase in participation and an awareness of the need for consultation were regularly mentioned. Here, for example, is the answer of one elder in a focus group discussion, when asked about impacts of the program: “As one of the benefits . . . [we learnt that] the man who consults can do a lot for the community.” When further prompted by the rather surprised interviewer, the elder admitted that consultations were already commonly practiced: Interviewer: “Let me ask you as Somalis, you knew this before? You used to consult with each other?” Respondent: “Yes.” Other villagers explained that they have learned the importance of joint contributions, albeit that such contributions are a constitutive part of the economic and moral fabric of Somali society and widely practiced (Bakonyi 2009). The program may have channeled collective resources in a specific direction, but it neither introduced contributions nor created awareness of their benefits. Conceptual vagueness enabled negotiations among various stakeholders and the adaptation of meaning to different contexts without having to engage in lengthy discussions. However, it also invited bricolage and tricks. The intervenor's tricks to demonstrate success were mirrored in the tendency of villagers to privilege the program's narrative over the reality of local practices and to fake amazement or applaud the program for reaching its objectives and promoting practices that were already quite common.21 Vagueness allows concepts to travel between different locations, opens up space for negotiations, and enables a variety of stakeholders to translate these concepts to the terms of their main audiences without having to engage seriously with different motivations or worldviews.22 The staff could record an increase in participation and report that the community is indeed in the driving seat of state-building. Local authorities, prominent among them elders and religious authorities, could continue to operate the VC and distribute the benefits of the program on the basis of clan affiliation, and already established women's groups could receive a share of these benefits. None of the involved actors had to endanger the continuation of the program by revealing information or meanings that might be “uncomfortable.” In short, conceptual vagueness allowed actors at all levels in the development chain to “muddle through” the everyday program requirements. Instead of holding vague concepts responsible for development failure, vagueness may well be interpreted as a prerequisite for interventions. It makes concepts flexible enough to travel and to be translated and retranslated to the variety of audiences in the development chain, without having to engage with potentially disruptive or conflictive interpretations. Vagueness is, however, inherently risky, as the continuous translation and retranslation can overburden newly established relations and networks, and uncomfortable information can rise above the surface, rupture institutional arrangements (Rayner 2012, 112–13), or uncouple arenas that were set up during the intervention and linked through community exercises. Conclusion Development aims to support people to improve their lives. In order to do so, development lays claim to knowledge of people and the societies it intervenes in, and it does so, as the above sections outlined, in a managerial rationality that promotes bureaucratization. Above all, development identifies deficits, gaps, and problems—in the Somali case, in the ways villages were governed. Problems are then framed in ways that make societies amenable to improvement through interventions. The failure of development to fulfil its objectives, that is to really build states, empower the weak and marginalized, reduce poverty, or bring about peace, is linked to a lack of, inadequate, or even false knowledge, a diagnosis that is usually followed by a request to generate more or better knowledge and by a search for methodologies that are better suited to do so. However, this article has demonstrated that the line between knowledge and ignorance is a thin one. While management has set itself up as a way of knowing and learning about the local, it has also been enabled by, generated, and enacted ignorance. The article has argued that an exploration of ignorance, the way ignorance is intertwined with knowledge, and how both are put into action can open up productive ways to explore the daily practice of development and help shed light on some of its effects. The article detailed the modes of ignorance that accompanied the delineation of a discursive terrain in which local governance in Somalia was made intelligible and actionable for state-building interventions that could present themselves as politically neutral. It additionally outlined the forms of ignorance that enabled the depoliticization of government by channeling attention from questions of legitimacy and representation to distributional and managerial aspects of rule. Omission allowed the program to discuss problems of government as questions of management. And while citizenship and the right of citizens to be involved in decision making was emphasized, the program simplified power relations, approached power differences as problems of capacity, and thereby individualized responsibilities and assigned them to those “without” power. The moralizing simplifications, the individualization of responsibilities and the emphasis on empowerment are, however, characteristic features of neoliberal bureaucratization. While conceptual vagueness is often blamed for development failure, the article showed how vagueness facilitated interventions. It helped to put potentially contested concepts into practice but concealed conflicts and hegemonic struggles. The forms and modalities of ignorance were negotiated through a series of contingent social practices. The empirical findings, however, demonstrate that both knowledge and ignorance are constitutive parts of the developmental ordering of the social. This brings us back to Max Weber's claims that everyday domination is mainly executed as administration and that bureaucracies dominate through knowledge. While these claims have not lost validity, they can be nuanced and expanded by exploring the ways in which bureaucracies rely on and generate ignorance and through the ways they enact and intertwine knowledge and ignorance while expanding their own operational rationality and outreach. Bureaucracies are generating an enormous amount of information on some aspects of the social, while other aspects are overseen, ignored, hidden, and forgotten. In their daily practice, and here I come back to Foucault and Mitchell, they unfold their expansionist, depoliticizing, and disciplinary power by enacting and rearranging knowledge and ignorance in ways that make the social accessible to an economic and managerial rationality. The specific forms of knowledge and ignorance, and the ways in which they are assembled in interventions, remain an empirical question. If nothing else, the focus on ignorance prevents yet another call toward more, better, or more accurate knowledge, the usual response to development failure. Footnotes 1 Development is here used to characterize interventions aimed at the betterment of a society, irrespective of the intervention supporting economic growth, poverty reduction, state-building, or peacebuilding. While objectives vary, international interventions operate in the same structural field, and operational options, constraints, and knowledge practices are therefore similar. 2 Daase and Kessler (2007, 412) suggest using the term nonknowledge to avoid the derogatory implications of ignorance and to limit ignorance to the strategic use of nonknowledge. Nonknowledge is translated from the German term Nichtwissen, but is neither commonly used in English nor is it possible to clearly differentiate nonknowledge from strategic ignorance (Taussig 1999, 118–26). I therefore continue to use ignorance but emphasize the relational character of knowledge and ignorance. 3 For a critique of knowledge-transfer, see Narayanaswamy (2013). 4 For a profound critique see Sabaratnam (2013). 5 Foucault (1979, 132, 133) labels these modes of government as “intervening liberalism” or neoliberalism, while Rose (1999, 139–42) prefers “advanced liberal” to stress the complexification of liberalism and not the succession from it. I stick to neoliberal, mainly because I discuss, with Hibou, (2015) “neoliberal bureaucratization.” 6 See, in particular, the special issue of Economy and Society 41, no. 1 (McGoey 2012) and the Routledge Handbook of Ignorance Studies (McGoey and Gross 2015). 7 Development has used community-based methods since the 1970s, but they were significantly expanded after the World Bank decided to promote them in the 1990s (Mansuri and Rao 2013, IX). 8 Max Weber instead emphasized the deep intertwinement of bureaucracies with capitalism. 9 Legibility is a central concept in James Scott's (1998) analysis of the state. This article argues that the bureaucratic ordering described by Scott is not limited to the state. 10 Hoag (2011, 83–84) points to the shared ethos and affinity of bureaucracy and science. 11 A detailed overview is provided by McGoey and Gross (2015), a critical summary by Aradau (2017). 12 Simmel's contribution toward a sociology of ignorance is outlined by Gross (2012). 13 A review of strategic ignorance is provided by McGoey (2012). 14 The sad absurdity of some interventionist actions and categories became blatantly obvious in Somalia. Many of the ICU Islamists that were ousted by Ethiopia were, during subsequent peace negotiations, integrated into the TFG, classified as moderate, and distinguished from the radicals, for example al-Shabaab. 15 AMISOM grew from sixteen hundred to twenty-two thousand troops between 2007 and 2016. Details are provided by Williams and Hashi (2016). 16 Evaluations are an important part of the managerial rationality of development and likely contribute to further types of ignorance. The evaluation in Somalia was accompanied by so many constraints that it requires a separate assessment. 17 The lack of self-perception as marginalized does not mean that people are formally equal; they clearly are not. 18 Minority clans are, due to their occupation as blacksmiths, barbers, etc., assigned an inferior status in the clan system, lack political representation, and often live in dire deprivation and poverty. 19 This could be interpreted as bad practice relating to a particular program. However, similar effects were found in other contexts (see Heimer 2012; Hull 2012). 20 Vagueness refers to the use of the same terms in different contexts and with different meanings. The failure to properly contextualize interventions or to clarify core concepts, among them participation, community, ownership, or empowerment, were regularly considered as reasons for development failure (Cornwall and Brock 2005; Mansuri and Rao 2013, 288-90). 21 While such tricks could be interpreted as a strategic response of subaltern actors, they are not restricted to the subaltern, but are a rather common feature of development (see Schlichte and Veit 2007). 22 Strunz (2012) argued that conceptual vagueness can enhance creativity, facilitate inter- and trans-disciplinary communication, and promote problem-solving. A similar argument was made by Best (2012) with respect to ambiguity. Acknowledgements I would like to thank my colleagues Gidon Cohen, Markus Hoehne, Raphaela Kormoll, Klaus Schlichte, Leaonard Seabrooke, Kirsti Stuvoy, and the two anonymous reviewers for their constructive critique of earlier versions of the text. References Aradau Claudia . 2017 . “ Assembling (Non)Knowledge: Security, Law, and Surveillance in a Digital World .” International Political Sociology 11 ( 4 ): 327 – 42 . Autesserre Séverine . 2014 . Peaceland. Conflict Resolution and the Everyday Politics of International Intervention . Cambridge : Cambridge University Press . Bakonyi Jutta . 2009 . “ Moral Economies of Mass Violence: Somalia 1988–1991 .” Civil Wars 11 ( 4 ): 434 – 54 . Google Scholar Crossref Search ADS Bakonyi Jutta . 2013 . “ Authority and Administration beyond the State: Local Governance in Southern Somalia 1995–2006 .” Journal of Eastern African Studies 7 ( 2 ): 272 – 90 . Google Scholar Crossref Search ADS Bakonyi Jutta , Cohen Gidon , Bédard Pierre-Olivier . 2015 . “ An Evaluation of Hogaan Iyo Nabad: A Community Driven Governance Programme in Somalia/Somaliland .” Accessed March 13, 2018, https://www.dur.ac.uk/dgsi/cdgsomalia/ . Barnett Michael , Finnemore Martha . 1999 . “ The Politics, Power, and Pathologies of International Organizations .” International Organization 53 ( 4 ): 699 – 732 . Google Scholar Crossref Search ADS Barnett Michael , Finnemore Martha . 2004 . Rules for the World. International Organizations in Global Politics . Ithaca, NY : Cornell University Press . Barry Andrew . 2001 . Political Machines: Governing a Technological Society . London : Athlone Press . Barry Andrew . 2002 . “ The Anti-political Economy .” Economy and Society 31 ( 2 ): 268 – 84 . Google Scholar Crossref Search ADS Bayart Jean-Francois . 2000 . “ Africa in the World: A History of Extraversion .” African Affairs 99 ( 395 ): 217 – 67 . Google Scholar Crossref Search ADS Best Jacqueline . 2012 . “ Bureaucratic Ambiguity .” Economy & Society 41 ( 1 ): 84 – 106 . Google Scholar Crossref Search ADS Consortium . 2012 . Proposal to UK Department for International Development (Unpublished) . Cornwall Andrea , Brock Karen . 2005 . “ What Do Buzzwords Do for Development Policy? A Critical Look At ‘Participation,’ ‘Empowerment’ and ‘Poverty Reduction.’ ” Third World Quarterly 26 ( 7 ): 1043 – 60 . Google Scholar Crossref Search ADS Cruikshank Barbara . 1993 . “ The Will to Empower: Technologies of Citizenship and the War on Poverty .” Socialist Review 23 ( 4 ): 29 – 55 . Cruikshank Barbara . 1999 . The Will to Empower. Democratic Citizens and Other Subjects . Ithaca, NY : Cornell University Press . Daase Christopher , Kessler Oliver . 2007 . “ Knowns and Unknowns in the ‘War on Terror’: Uncertainty and the Political Construction of Danger .” Security Dialogue 38 ( 4 ): 411 – 34 . Google Scholar Crossref Search ADS Dean Mitchell . 2010 . Governmentality: Power and Rule in Modern Society . London : Sage . Eade Deborah . 1997 . “ Capacity-Building: An Approach to People-Centred Development .” UK Oxfam. Accessed September, 10, 2016, http://policy-practice.oxfam.org.uk/publications/capacity-building-an-approach-to-people-centred-development-122906 . El-Bushra Judy , Gardner Judith . 2016 . “ The Impact of War on Somali Men: Feminist Analysis of Masculinities and Gender Relations in a Fragile Context .” Gender & Development 24 ( 3 ): 443 – 58 . Google Scholar Crossref Search ADS Ferguson James . 1994 . The Anti-politics Machine. Development, Depoliticization, and Bureaucratic Power in Lesotho . Minneapolis, MN : University of Minnesota Press . Foucault Michel . 1978 . The History of Sexuality Volume I: An Introduction . New York : Pantheon Books . Foucault Michel . 1979 . “ German Neoliberalism III .” In The Birth of Biopolitics. Lectures at the College De France , edited by Senellart Michael , Ewald Francois , Fontana Allesandro , 129 – 57 . New York : Palgrave Macmillian . Goetschel Laurent , Hagmann Tobias . 2009 . “ Civilian Peacebuilding: Peace by Bureaucratic Means? ” Conflict, Security & Development 9 ( 1 ): 55 – 73 . Google Scholar Crossref Search ADS Green Maia . 2000 . “ Participatory Development and the Appropriation of Agency in Southern Tanzania .” Critique of Anthropology 20 ( 1 ): 67 – 89 . Google Scholar Crossref Search ADS Gross Matthias . 2012 . “ ‘Objective Culture’ and the Development of Nonknowledge: Georg Simmel and the Reverse Side of Knowing .” Cultural Sociology 6 ( 4 ): 422 – 37 . Google Scholar Crossref Search ADS Heimer Carol A. 2012 . “ Inert Facts and the Illusion of Knowledge: Strategic Uses of Ignorance in HIV Clinics .” Economy & Society 41 ( 1 ): 17 – 41 . Google Scholar Crossref Search ADS Hibou Beatrice . 2015 . The Bureaucratization of the World in the Neoliberal Era. An International and Comparative Perspective . New York : Palgrave Macmillan . Hoag Colin . 2011 . “ Assembling Partial Perspectives: Thoughts on the Anthropology of Bureaucracy .” PoLAR: Political and Legal Anthropology Review 34 ( 1 ): 81 – 94 . Google Scholar Crossref Search ADS Hull Matthew S. 2012 . “ Documents and Bureaucracy .” Annual Review of Anthropology 41 ( 1 ): 251 – 67 . Google Scholar Crossref Search ADS Jaeger Hans-Martin . 2007 . “ ‘Global Civil Society’ and the Political Depoliticization of Global Governance .” International Political Sociology 1 ( 3 ): 257 – 77 . Google Scholar Crossref Search ADS Li Tania . 2007 . The Will to Improve: Governmentality, Development, and the Practice of Politics . Durham, NC : Duke University Press . Li Tania . 2011 . “ Rendering Society Technical: Government through Community and the Ethnographic Turn at the World Bank in Indonesia .” In Adventures in Aidland: The Anthropology of Professionals in International Development , edited by Mosse D. , 57 – 80 . Oxford : Berghahn . Lipschutz Ronnie D. 2005 . “ Power, Politics and Global Civil Society .” Millennium—Journal of International Studies 33 ( 3 ): 747 – 69 . Google Scholar Crossref Search ADS Mansuri Ghazala , Rao Vijayendra . 2013 . Localizing Development: Does Participation Work? Washington, DC : World Bank . Accessed March 13, 2018, https://openknowledge.worldbank.org/handle/10986/11859 . Mathews Andrew S. 2008 . “ State Making, Knowledge, and Ignorance: Translation and Concealment in Mexican Forestry Institutions .” American Anthropologist 110 ( 4 ): 484 – 94 . Google Scholar Crossref Search ADS McGoey Linsey . 2007 . “ On the Will to Ignorance in Bureaucracy .” Economy & Society 36 ( 2 ): 212 – 35 . Google Scholar Crossref Search ADS McGoey Linsey . 2012 . “ The Logic of Strategic Ignorance .” The British Journal of Sociology 63 ( 3 ): 533 – 76 . Google Scholar Crossref Search ADS PubMed McGoey Linsey , Gross Matthias . 2015 . “ Introduction .” Routledge International Handbook of Ignorance Studies . New York : Routledge . Menkhaus K. L. 1997 . “ Somalia: Political Order in a Stateless Society .” Current History 97 ( 6 ): 220 – 4 . Mignolo Walter D. 2005 . The Idea of Latin America . Malden : MA: Blackwell Publishing . Mitchell Timothy . 1991 . “ The Limits of the State: Beyond Statist Approaches and Their Critics .” American Political Science Review 85 ( 1 ): 77 – 96 . Mitchell Timothy. 2002 . Rule of Experts. Egypt, Techno-Politics, Modernity . Berkley : University of California Press . Mosse David . 2004 . Cultivating Development: An Ethnography of Aid Policy and Practice . London : Pluto Press . Narayanaswamy Lata . 2013 . “ Problematizing ‘Knowledge-for-Development.’ ” Development and Change 44 ( 5 ): 1065 – 86 . Google Scholar Crossref Search ADS Rappert Brian . 2012 . “ States of Ignorance: The Unmaking and Remaking of Death Tolls .” Economy & Society 41 ( 1 ): 42 – 63 . Google Scholar Crossref Search ADS Rayner Steve . 2012 . “ Uncomfortable Knowledge: The Social Construction of Ignorance in Science and Environmental Policy Discourses .” Economy and Society 41 ( 1 ): 107 – 25 . Google Scholar Crossref Search ADS Richmond Oliver P. 2012 . “ A Pedagogy of Peacebuilding: Infrapolitics, Resistance, and Liberation .” International Political Sociology 6 ( 2 ): 115 – 31 . Google Scholar Crossref Search ADS Rose Nikolas . 1996 . “ Governing ‘Advanced’ Liberal Democracies .” In Foucault and Political Reason. Liberalism, Neo-Liberalism, and Rationalities of Government , edited by Barry Andrew , Osborne Thomas , Rose Nikolas , 37 – 64 . Chicago : University of Chicago Press . Rose Nikolas . 1999 . Powers of Freedom. Reframing Political Thought . Cambridge : Cambridge University Press . Rose Nikolas , Miller Peter . 1992 . “ Political Power Beyond the State: Problematics of Government .” The British Journal of Sociology 43 ( 2 ): 173 – 205 . Google Scholar Crossref Search ADS Sabaratnam Meera . 2013 . “ Avatars of Eurocentrism in the Critique of the Liberal Peace .” Security Dialogue 44 ( 3 ): 259 – 78 . Google Scholar Crossref Search ADS Schlichte Klaus . 2012 . “ Der Streit Der Legitimitäten. Der Konflikt Als Grund Einer Historischen Soziologie Des Politischen .” ZeFKo Zeitschrift für Friedens- und Konfliktforschung 1 ( 1 ): 9 – 43 . Google Scholar Crossref Search ADS Schlichte Klaus , Veit Alex . 2007 . “ Coupled Arenas: Why State-Building is So Difficult .” In Working Papers Micropolitics of Armed Groups No. 3 . Berlin : Humbold University . Scott James C. 1998 . Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed . New Haven, CT : Yale University Press . Spittler Gerd . 1980 . “ Abstraktes Wissen Als Herrschaftsbasis: Zur Entstehungsgeschichte Bürokratischer Herrschaft Im Bauernstaat Preußen .” Kölner Zeitschrift für Soziologie und Sozialpsychologie 32 ( 3 ): 574 – 604 . Strunz Sebastian . 2012 . “ Is Conceptual Vagueness an Asset? Arguments from Philosophy of Science Applied to the Concept of Resilience .” Ecological Economics 76 : 112 – 8 . Google Scholar Crossref Search ADS Suhrke Astri . 2007 . “ Reconstruction as Modernisation: The ‘Post-Conflict’ Project in Afghanistan .” Third World Quarterly 28 ( 7 ): 1291 – 308 . Google Scholar Crossref Search ADS Taussig Michael T. 1999 . Defacement: Public Secrecy and the Labor of the Negative . Stanford, CA : Stanford University Press . Trotha Trutz von . 1994 . Koloniale Herrschaft: Zur Soziologischen Theorie der Staatsentstehung am Beispiel des “Schutzgebietes Togo.” Tuebingen, Germany : Mohr Siebeck . Weber Max . 1980 . Wirtschaft Und Gesellschaft. Grundriss der Verstehenden Soziologie . Hrsg. Von Johannes Winckelmann . Tübingen, Germany : J.C.B. Mohr (5. rev. Aufl.) . Williams Paul , Hashi Abdirashid . 2016 . Exit Strategy Challenges for the AU Mission in Somalia . Mogadishu : The Heritage Institute for Policy Studies . Accessed March 13, 2018, http://www.heritageinstitute.org/wp-content/uploads/2016/02/Exit-Strategy-Challenges-for-the-AU-Mission-in-Somalia.pdf . Winckler Joel . 2011 . “ Managing the Complexities of Intervention: United Nations Peace Operations as Organisational Action .” Journal of Peace, Conflict & Development 18 : 83 – 103 . © The Author(s) (2018). Published by Oxford University Press on behalf of the International Studies Association. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Political Sociology Oxford University Press

Seeing Like Bureaucracies: Rearranging Knowledge and Ignorance in Somalia

Loading next page...
 
/lp/ou_press/seeing-like-bureaucracies-rearranging-knowledge-and-ignorance-in-oCIL99zXj0
Publisher
Oxford University Press
Copyright
© The Author(s) (2018). Published by Oxford University Press on behalf of the International Studies Association.
ISSN
1749-5679
eISSN
1749-5687
D.O.I.
10.1093/ips/oly010
Publisher site
See Article on Publisher Site

Abstract

Abstract Development promotes bureaucratization, and bureaucracies are based on knowledge and produce knowledge. Failures of development are therefore regularly attributed to a lack of knowledge. The article argues that the quest for knowledge is embedded in the managerial rationality of interventions. This rationality also structures the developmental knowledge field and thereby generates ignorance. The example of a state-building program in Somalia is used to empirically explore how the generation, administration, and transfer of knowledge was intertwined with ignorance. It shows what knowledge missed, obfuscated, ignored, or even hid and how knowledge and ignorance were arranged in the daily state-building practice. This approach sheds light on relations and mechanism of power exerted in development and helps to explain its effects. In Somalia, omission, silence, secrecy, and strategic and bureaucratic ignorance enabled the program to delineate the interventionist terrain as technical and to depoliticize state-building. They also helped to expand liberal modalities of government to “remote” and “unruly” Somali villages. Development, as a way to conceptualize and theorize human progression, and as practice to support this progression, is under critique.1 A large body of literature attests to the failures of both. Initiated by the seminal assessment of Ferguson (1994), authors began to focus on the effects of development beyond its own objectives and have since emphasized its bureaucratizing and depoliticizing tendencies. International organizations, for example, have been analyzed as complex bureaucratic machineries (Barnett and Finnemore 1999, 2004), which roll out interventions using a managerial logic (Mosse 2004; Goetschel and Hagmann 2009; Winckler 2011). While they continuously expand their own mode of operation, developmental bureaucracies also render the reason for the intervention, its main objectives and its implementation practice, technical. In this way, interventions are removed from the political realm of contestation and decision making and presented as technical solutions to mainly technical problems (Ferguson 1994; Li 2011). The expansionist tendency of bureaucracies had already been recognized at the beginning of the last century by Max Weber. Weber (1980, 653–55) argued that modern bureaucracies are efficient forms of social organization because they operate formally and instrumentally, base decisions on knowledge rather than personal connections, and weigh means against ends instead of balancing favors. As development promotes bureaucratization, and bureaucracies are based on knowledge and generate knowledge, the failure of development is regularly attributed to a lack of knowledge. Ignorance or nonknowledge2 is usually identified in two dimensions: (local) people's lack of appropriate knowledge on how best to develop their societies, and international organizations’ lack of knowledge about the society in which they are intervening. The former has been targeted by a transfer of knowledge from the global North to the global South. Training, capacity building, and learning-by-doing exercises have evolved into a core activity of development (Eade 1997, 2–3, 9–10).3 The latter is tackled by a broad range of international organizations, which are producing knowledge on the global state of development and displaying it in global reports and indices. Beyond this highly aggregated form of knowledge, a lack of (micro)knowledge about the societies in which interventions are planned is still considered as among the main reasons for development failure (recently Autesserre 2014). The need to better contextualize interventions initiated the local turn in development in the 1990s. However, when development touched upon the local, it subjected it to its own bureaucratizing dynamic. While some authors continued their search for the “real,” untouched, and unaffected local (Richmond 2012),4 others criticized the (bureaucratic) practice of development interventions and identified the spatial and social distance between the intervener and the intervened upon as reason for the continued failure to gather adequate knowledge, and for development failure more broadly (Bayart 2000; Mosse 2004; Goetschel and Hagmann 2009; Autesserre 2014). This article contributes to these discussions from a “sociology of knowledge” viewpoint. It argues that the quest for better, more accurate and detailed knowledge is part of the managerial rationality of interventions. This rationality also structures the developmental knowledge field, and, while it delimits the type of knowledge that is generated, it also produces unknowns and ignorance. Put into practice, development then continues to generate and rearrange both knowledge and ignorance. The example of a state-building program is then used to empirically explore the generation, administration, and transfer of knowledge but puts its focus on ignorance and thus on what knowledge missed, obfuscated, ignored, or even hid and how knowledge and ignorance were intertwined in the daily state-building practice. This approach sheds light on the relations and mechanisms of power exerted in the way development lays claim to knowledge. At the same time, it avoids common categorizations of local versus external knowledge, the identification of the former as more accurate and the latter as simplistic, technocratic, or imposed, a claim that is usually followed by a call either for more and more accurate knowledge or for better suited (participatory) methodologies to capture the local. In spite of the focus on ignorance, the effects of development are, in this reading, not mainly attributed to a right or wrong reading of the local but analyzed through the ways in which knowledge and ignorance were arranged to make the local legible and accessible to state-building interventions. The argument is developed in three steps. Drawing selectively on the governmentality literature, I first outline how the neoliberal rearrangement, diffusion, and down-scaling of state regulatory competences emphasized distinction and revalued local and expanded community-based technologies. Executed as “government through community” (Rose 1996; Li 2007, 2011), neoliberal modes of government5 constituted communities as objects of knowledge and subjected them to interventions that activated self-regulatory techniques and practices. These interventions made the local accessible to a managerial rationality and expanded bureaucratic modes of thought and operation but delinked bureaucracies from the state, flattened bureaucratic hierarchies, and dispersed bureaucratic practices, a process I label, with Hibou (2015), as “neo-liberal bureaucratization.” This section also clarifies how bureaucracies lay claim to knowledge and discusses their (de)politicizing effects. The second section builds on insights from the recently revived field of “ignorance studies.”6 It draws theoretically on those aspects of the work of Foucault (1978), Mitchell (1991, 2002), and Taussig (1999) that help to build the argument that bureaucracies do not execute their power through knowledge but through rearranging knowledge and ignorance and through enacting the knowledge-ignorance assemblages in a way that subsumes the social to their own managerial rationality. The third section provides an in-depth empirical study of the way knowledge and ignorance were generated, arranged, and put into practice in a state-building project in Somalia. This section explores which unknowns were identified by the program while others were overseen or ignored (deliberately so or not), how the identified unknowns were made actionable for exploration, the technologies used to generate knowledge, and the ways in which new knowns were managed and initiated further actions (or not). It shows that while the intervention set out to generate knowledge on the local and to build on local knowledge to improve government in Somalia, it was also enabled by and generated ignorance, in particular omission, silence and secrecy, vagueness, and strategic and bureaucratic ignorance. These forms of ignorance were part of the interventionist design, enabling the program to delineate the interventionist terrain as technical and thus to expand managerial techniques, which depoliticized state-building. They also facilitated the “smooth” implementation of the program and helped to expand neoliberal modalities of government and with them a managerial rationality to the “remote” and “unruly” Somali districts and villages. Governing Through Communities: Bureaucratization beyond the State Max Weber's (1980, 126) famous finding that domination in everyday life is primarily executed as administration gained new evidence in the context of international development. More than two decades ago, James Ferguson (1994) concluded from his assessment of a development project in Lesotho that such projects have failed to bring about development but have significantly strengthened the state. They facilitated the expansion of a bureaucratic apparatus and thereby enabled the state to execute more “effective control and administrative supervision” (Ferguson 1994, 265) over people who were previously largely uncaptured. In the decades following Ferguson's assessment, international actors ceded to support the state. The neoliberal restructuring of state-society-market relations, which started in the 1970s, was promoted by a call toward less bureaucratic and more flexible and responsive structures of government. Communities were thereby constituted as mediators between the centrally and bureaucratically executed power of the state and the free exchange system of the market (Rose 1999 167–73), and community-based technologies promoted to replace the overexpanded state bureaucracy. However, “government through community” (Rose 1999; Li 2011)7 did not diminish bureaucracies but delinked them from the state and decentralized their mode of operation. Bureaucratic practices are not limited to the state.8 Instead, they make the social legible9 by generating a particular kind of “modern” or abstract knowledge, that is a knowledge that can be accumulated and instrumentally applied to intervene in, control, and improve nature and society. Bureaucracies enter into social processes and disaggregate the social into separate parts and functions, which they map, aggregate, and quantify. They then standardize the magnitude of data and transform it into abstract knowledge displayed in lists, tables, matrices, and diagrams (Spittler 1980). While emphasizing heterogeneity and distinction, government through community made the local accessible to bureaucratic knowledge and with it to standardization. Opinion polls, focus group discussions, roundtables, and participatory rural appraisals are examples of the increasing variety of participatory knowledge tools used to identify, map, measure, and categorize communities. Communities were thereby constituted as unitary social groups—women, minorities, youth, etc.—with common characteristics but also with unique concerns. They were thereby made thinkable “in a way that their ills appear susceptible to diagnosis, prescription and cure” (Rose and Miller 1992, 183) and thus to optimization through interventions and programming (Rose 1996, 189; Rose 1999 174; Cruikshank 1999, 78; Li 2007, 2011). While making the social legible, controllable, and governable, bureaucracies strengthen their own operational rationality and facilitate their continuous expansion (Spittler 1980; Barnett and Finnemore 1999, 707; Winckler 2011, 88–89). Bureaucracies are expansionist. They are also depoliticizing. Classical political sociologists, among them authors as diverse as Karl Marx, Max Weber, Pierre Bourdieu, and Michel Foucault, conceptualized the political as a realm of public dissent, contestation, hegemonic struggles, and conflicts (Schlichte 2012). If issues are made political or politicized, they are publicly discussed and challenged, sometimes even violently. Depoliticization, in contrast, reduces open contestation and conflict and conceals struggles. However, the practice of politics is, as Barry (2001, 205–9; 2002, 269–70) reminds us, not only about contestation and conflict but also about the containment of both. One way of containing conflicts is to shift the contested issues from the realm of public contestation to other societal domains, among them administration, and to hand them over to experts, who evoke their specialized knowledge, analyze, measure and evaluate, and come up with solutions10—a practice that was captured as rendering technical (Rose 1999, 175; Li 2011). Depoliticization is thus an important part of the repertoire of politics and, if we follow Ferguson (1994) and many others, also of the political repertoire of development. In order to empirically explore processes of politicization and depoliticization in contemporary politics, Lipschutz (2005) suggested differentiating between constitutional and distributional aspects of the political, the former concerned with the structural framework for political actions, the latter with procedural aspects and especially with the distribution of resources and benefits. With an increased density of political institutions, the externalization of state regulatory competences to nonstate actors, and the parallel expansion of a managerial culture, distributional and procedural aspects come to the fore of political contestation (Jaeger 2007). The quite technocratic focus of contemporary political debates (and the governance literature) on the most efficient arrangement of institutions and the most equitable management of resources can be interpreted as an effect of bureaucratization. They indicate the self-perpetuation of bureaucratic modes of domination and thus of modalities of rule that take the given order for granted while limiting the political to discussions about its most efficient execution (Schlichte 2012, 10). The Power of Bureaucracies: Rearranging Knowledge and Ignorance Development, even in its community-based form, is bureaucratic. Bureaucracies are set up as a particular way of knowing and ordering the social and thereby display expansionist and depoliticizing tendencies. Several authors have attested to both and have paid particular attention to bureaucratic knowledge practices. More recently, however, authors have also started to explore ignorance (McGoey 2007; Best 2012; Hull 2012). Ignorance is used as a summary term for everything that is not known, including false, imprecise, or incomplete knowledge. Instead of delineating a mere void, ignorance studies stress that knowledge and ignorance are interlinked, that ignorance comes in diverse forms and modalities and can be generated and used intentionally or unintentionally, and that it, too, is attached to power.11 Following Simmel, who identified the interplay between knowledge and nonknowledge, and the strategic use of both as crucial for social interactions,12 a number of authors have drawn attention to the intentional and strategic use of ignorance.13Trotha (1994), for example, examined how colonial officers and local intermediaries in Togo actively omitted or disguised information to enable the establishment of “rule from a distance” but also to strengthen their own position in the colonial administration (Trotha 1994, 379, 409-11). Mathews (2008), using the example of Mexican forest management, has shown how official state knowledge was negotiated between local stakeholders and state officials and how the latter discreetly ignored information that would have disturbed their work. Rappert (2012, 43) has additionally outlined how the UK government, during the 2003 invasion in Iraq, enacted ignorance to deflect political criticism. While ignorance was thus recognized as an important source of power, most authors focused on the intentional use of it. Aradau (2017, 330–31) additionally emphasized the generative effects of ignorance. Feminist studies have, for example, shown that knowledge engenders “epistemic injustice” and produces “epistemically disadvantaged subjects” while discrediting particular social groups as “not knowing” or as being “incredible knowers.” A similar observation was made by postcolonial studies. The historically specific experiences of Europe were set as the norm and mistaken for the universal, while non-European experiences were ignored, overseen, devalued, and discredited. A “geopolitics of knowledge” emerged, which continues to support interventionist approaches, as it provides the categories from which the rest of the world “can be described, classified, understood and ‘improved’” (Mignolo 2005, 36). Ignorance is thus enacted through, and functions in conjunction with, knowledge. Beyond its intentional use, ignorance can be explored as both a form and an effect of power. Taussig (1999) even characterized the tensions between concealment of knowledge and its revelation as the central feature of power. Accordingly, knowledge may be power, but the power of knowledge is only known through its disguise and the creation of public secrets, which then unfold their power through an ongoing “act of unmasking” (Taussig 1999, 105, 145). Taussig follows Foucault (1978) who, among others, used the example of sex to show that the power of secrets does not lie in the repression of truth but in their continuous revelation. Foucault discussed in his history of sexuality how the rising “will to knowledge” engendered a boom of discourses and truth claims that enabled the administration, management, and optimization of sex as part of a “political economy of population” (Foucault 1978, 12, 26). In a similar vein, Mitchell (2002) explored how new technologies and methods of knowledge, among them maps, cadaster, or statistics, contributed to the making of the (capitalist) economy in Egypt. These methods helped to rearrange the material, political, and social worlds according to the laws of the market. Both Foucault and Mitchell have pointed to the expansion of economic categories and knowledge practices into noneconomic domains. While these domains were now subjected to economic calculation and optimization, they were also made accessible to managerial techniques. Control is thereby not executed centrally but through the dispersion of norms and rules, measurements and calculations, performance assessments, indicators, and audits, in short through bureaucratic practices (Hibou 2015, 22). Bureaucracies thus unfold as a nearly ideal-typical form of disciplinary power. They disaggregate societal elements and processes, transform them into intelligible domains in need of improvement, and increase efficiency and precision while reassembling the parts “into more productive and powerful combinations” (Mitchell 1991, 93). Foucault and Mitchell also stressed that the rise of scientific methods, detailed analyses, and continuous investigation neither necessarily enhanced the truth about the studied phenomena nor made knowledge more accurate. Instead, they constituted objects of knowledge (sex, the economy, the community) by rearranging the relationship between the known and the unknown, the said and the unsaid, and thus through a web of discourses and practices in which forms of knowledge and ignorance were redistributed and the normal and the abnormal rearranged (Mitchell 2002, 91; Foucault 1978, 24–26, 56–57). The next section explores the unknowns and how knowledge and ignorance were arranged and set into practice in a community-based state-building intervention in Somalia. This approach sheds light on the social relations and flows of power that are ingrained in the production of knowledge, in the case of this article, knowledge on government in Somalia. Building States from the Bottom-Up: Experiments in Somalia The Somali state apparatus collapsed in 1991 during the course of an ongoing civil war that started in 1988. Violence has since been concentrated in the southern and central parts of the country. The clan-based militias that fought against the military dictatorship (1988–91) and then against themselves (1991–92) fragmented in the course of a failed UN intervention (1992–95). Afterward, the country was divided into a patchwork of localized forms of rule, but violence decreased (Menkhaus 1997; Bakonyi 2013). The political dynamics changed again after a Transitional Federal Government (TFG) was established in 2005. The rise of an Islamist movement, US counter-terrorism operations, and the military involvement of neighboring and international troops in Somalia initiated new rounds of severe violence. An Islamist umbrella organization, the Islamic Court Union (ICU), successfully mobilized resistance against the TFG but was, at the turn of 2006/07, ousted by the US-backed military intervention of Ethiopia. The TFG established itself in Mogadishu but was challenged by an Islamist insurgency, spearheaded by Harakat al-Shabaab al-Mujahideen (al-Shabaab).14 By 2010, al-Shabaab controlled most of south-central Somalia, while the TFG clung to a few districts in Mogadishu. International donors eventually stepped up their financial and military aid, and the African Union Mission to Somalia (AMISOM)15 intensified the war against al-Shabaab. AMISOM, together with the partially reconstructed Somali National Army and allied clan militias, pushed al-Shabaab back to the rural areas. From 2011 onward, the TFG, later transformed into the Federal Government of Somalia, or one of its allied militias, took control of most urban centers in south-central Somalia. Al-Shabaab, however, continues with guerrilla-style warfare and regular suicide attacks against the Somali government and its allies. Much less affected by violence were the northern regions of the country. The northwestern Republic of Somaliland declared independence in 1991, but is not internationally recognized, although it managed to re-establish a central state and embarked on democratization. Its eastern neighbor, Puntland, declared its semi-autonomous status in 1998 but became later a member-state in the Federal Republic of Somalia. International attempts toward peace- and state-building initially focused on the war-torn southern regions but soon had to acknowledge that power is dispersed in Somalia and that the reconstructed state institutions did not reach far beyond Mogadishu. Since 2008, UN organizations have additionally supported state-building from the “bottom-up” and implemented a “Joint Program on Local Government and Decentralized Service Delivery” (JPLG) in Somaliland and Puntland. JPLG supports the establishment of district and regional offices and builds the capacity of its staff. Between 2012 and 2015, JPLG was complemented by a program that aimed at expanding state structures from the regional and district level to remoter rural areas. This program was funded by DFID and implemented by a consortium of three international organizations with longstanding experience in community-based development. The consortium adapted its approaches to the reconstruction of village-level governance and implemented its program in sixty villages (some of which were actually towns) in three districts in the east of Somaliland and the southwest of Puntland. Funded by DFID, and carried out from October 2014 to September 2015, an evaluation of the program used a mixed-method approach to assess if it had reached its main objectives. It conducted seventy-nine semistructured interviews and forty-seven focus group discussions and administered 1,604 household surveys in fifty-seven of sixty program villages and in thirteen control villages. The evaluation team also reviewed a large number of program documents, among them the proposal, the original and revised theory of change (ToC) and logical framework (logframe), the monitoring and evaluation framework, training and learning handbooks, community action planning documentation (village reports), training manuals, and internal and external baseline studies. At the beginning and end of the evaluation, workshops with program staff were organized in Hargeysa (Somaliland) and Nairobi (Kenya). Communication with staff from the offices in New York, London, Nairobi, Hargeysa, and Garowe was ongoing throughout the evaluation. This article will draw more on the documents and communications than on the findings of the evaluation,16 which will mainly be used to illustrate the argument. State-Building as Management: Bureaucratic Knowledge Practices Development provides an example of how management has set itself up as a way of knowing and learning about societies. The need to continuously collect information is built into the different sequences of the program management cycle, a tool commonly used in development programming. The cycle encourages a trial and error approach and emphasizes knowledge gathering as a means to evidence assumptions, monitor progress, avert undesired effects, and mitigate against risks. It suggests the implementation of programs in an incremental helix of sequences ranging from problem identification → program design → program implementation → monitoring (preliminary) results → and eventually to evaluating results and longer-term impacts. Learning is an integral part of the cycle, and monitoring and reflection loops cross-cut through the sequences. They aim to safeguard the newly gathered information, especially knowledge on how the program impacts on local social dynamics, and utilize it to ensure that the program progresses in the right direction. Program knowledge is collated in reports, among them feasibility and baseline studies, monitoring reports, and mid-term and end-line evaluations. Program experiences, as well as monitoring and evaluation results, are further aggregated to “lessons learnt,” “best practices,” or “guiding principles,” which are collectively endorsed by a broad range of development actors (Suhrke 2007, 1292). The Somali program followed the sequences of the management cycle, embraced its trial and error logic, and utilized a variety of participatory tools to ensure that the planning, design, and implementation of the program were based on locally grounded knowledge and allowed for (joint) learning. In an initial proposal, donors and implementing organizations agreed on the main representation of the local and established a framework for their interaction with it. The structure of the program proposal was quite conventional and included a contextual analysis, a (preliminary) ToC, a logframe, and a work plan. The contextual analysis mapped out the interventionist terrain by identifying particular challenges and problems, in the Somali case, of local governance. This was followed by the ToC, a planning and monitoring tool used to develop strategies on how identified problems can best be addressed. It consists of a set of hypotheses (if-then statements) on how change is likely to unfold if certain inputs are provided and if certain conditions are met (main assumptions). A ToC underscores the experimental character of interventions and uses prediction (hypothesis), manipulation (program inputs), testing (implementation), and control (monitoring and evaluation) to explore whether the deficient present can be steered into a better future. The ToC was further translated into a logframe, another management tool that breaks down the change hypothesis into smaller tasks and sequences and aligns them to a specific time frame. During the program implementation, the logframe is further dissected into work plans in which the day-to-day activities of program staff are specified. The proposal and program implementation thus provided a nearly ideal-typical example of the above-described way in which bureaucracies enter into, disaggregate, and reassemble the social, aiming at optimization. This process is based on and generates knowledge. It is also, as the following will show, based on and generates ignorance, including vagueness, omission, strategic ignorance, active secrecy, and bureaucratic ignorance. The Interventionist Terrain: Omission, Silence, and Strategic Ignorance In its context analysis, the consortium examined local governance in Somalia and described the village as “the heart of the Somali community” (Consortium 2012, 23). Key changes at village-level institutions were considered likely to contribute to stabilization, conflict resolution, and civil and economic development (Consortium 2012, 23). The proposal described how villages are governed by different institutions, among them the Village Council (VC), sectoral subcommittees, elders, and religious authorities. It particularly emphasized the importance of elders in conflict-resolution and the management of everyday village affairs (Consortium 2012, 8, 10, 11). On the downside, however, it was noted that “traditional structures” tend to marginalize particular social groups, among them “minority clans, women and youth” (Consortium 2012, 22; 34). The proposal additionally identified a lack of governance efficiency due to overlapping roles and the duplication of responsibilities, as well as a general lack of capacity. Accordingly, local authorities lack the capacity to plan village development and to administer villagers’ needs, while citizens lack the capacity to articulate demands and to participate in decision making (Consortium 2012, 5, 12, 22). Similar to Green's (2000) observations on participatory approaches in Tanzania, the Somali program simultaneously asserted and denied local practices and expertise. It emphasized the success and importance of hybrid institutional arrangements but only to diagnose their inadequacy, inefficiency, and need for improvement. This diagnosis opened up the space for interventionist action and attested to the superiority of interventionist practices and knowledge, which can help to improve local governance. Additionally, however, the proposal reduced the role of government to the (bureaucratically organized) management of community needs and the delivery of services. Questions such as “who rules?” or, alternatively, “who should rule?”; “on which basis?”; “about what or whom?”; and “how so?”—questions which invite an exploration of the constitutional aspects of politics—were absent in the proposal. These omissions precluded discussions on alternative forms of authority and promoted a view of government as the management of resources. The program discussed state-building with villagers but channeled the debate to the distributional aspects of rule, an approach that was matched by the conceptualization of citizens as beneficiaries and thus consumers of government services (see also next section). The transformation of state officials into managers and citizens into consumers is, however, a crucial part of the depoliticized and “bureaucratized vision of the political” that neoliberalism engenders (Hibou 2015, 52). The avoidance of constitutional debates and thus of substantive aspects of politics became most pronounced in the ignorance of clan relations. The Somali society is structured by a segmentary lineage system that subdivides ethnic Somalis in patrilinear descent groups and structures them in a pyramidal order in clan families, clans, subclans, and the smallest segment, the household. Similar to ethnicity in other postcolonial contexts, clan affiliations provide a channel through which wealth, power, and status are expressed and through which many aspects of social, political, and economic life are structured. The ongoing formation of central, regional, and local governments is, for example, primarily negotiated as clan-based representation. When asked why clan relations were not taken into consideration while designing and (re-)building government at the village level, program staff referred to the difficulty and sensitivity of obtaining clan-related information. Although the evaluation team did not face these difficulties, an in-depth assessment of existing representational mechanisms and an open discussion on their legitimacy may well have amplified political tensions. After all, the formation of governments, and thus the decision of who rules and how, is a conflictive political process, and one that is often negotiated violently in Somalia. The program decided to strategically ignore the clan-based and politically contested character of both the village community and the village government. The omission of clan-related information and the waiving of information related to existing mechanisms and practices of representation enabled the interventionists to approach local governance as technical terrain and thus as suitable for an intervention. It also enabled the international state-builders to present themselves as politically neutral. After all, international organizations claim legitimacy for their actions by presenting themselves as “not exercising power but instead serving others” (Barnett and Finnemore 1999, 21). The depoliticization of constitutional aspects of government in Somalia did not play out smoothly. (Strategically) unrecognized or at least unacknowledged by the international state-builders, clan affiliation affected nearly every aspect of the program, be it participation in community meetings, attendance at training, the organization of community contributions, or the distribution of program resources (Bakonyi, Cohen, and Bédard 2015). While clan relations structure socioeconomic and political life in Somali villages, a silent consensus emerged between villagers and program staff to not touch upon the political character of governance. We know from Foucault (1978, 27) that the production of silences does not limit what is said and known but functions “alongside the things said, with them and in relation to them within over-all strategies.” Silences are part of negotiated discretions between speakers who determine (consciously or not) “which type of discourse is authorized, or which form of discretion is required” (Foucault 1978, 27). In Somalia, and in order to act upon the local in a way that impedes political struggle and conflict, “uncomfortable knowledge” (Rayner 2012) was eliminated and silences were produced and reproduced in the communication between state-building actors. The proposal also emphasized the need to improve the modalities of government, in particular to optimize the division of labor in order to increase efficiency. The improvement strategy, however, focused mainly on the VC. Elders, religious authorities, even the sectoral subcommittees were downgraded to supporting actors, and the village government discursively centralized around the institution that (at least ideally) complied most with the interveners’ views on how governments should be organized: in a formal-bureaucratic way, through proper offices, and with clearly demarcated tasks and responsibilities. The omission of elders and religious authorities became a striking feature of the program. Shortly after the program's inception, the strength, weakness, opportunities, and threats (SWOT) of village-level governance were analyzed in each of the sixty villages. However, the SWOT analysis focused exclusively on the VC, while elders and religious authorities were not even considered as part of village governments. The Government Improvement Plans that were designed on the basis of the analysis, therefore, dealt only with this institution. Somehow, on the way from the context analysis to the strategy development and state-building practice, the VC was conceptualized as the center of village government, despite being characterized as particularly weak and largely dependent on elders. Omission, silences, and strategic ignorance enabled the program to promote a bureaucratic way of thinking about, talking about, and practicing government as formally regulated, practiced by a dominant institution, and exercised in a managerial manner. Transforming (Some) People into Active Citizens: Simplifications and Secrecy The Somali state-building program problematized particular aspects of local government, in particular the lack of active involvement of villagers in decision making. Women, youth, and minority clans were in this context identified as particularly vulnerable, without voice, unheard, and passive. This diagnosis was to be corrected by “empowering” villagers “to take on a larger role in claiming their rights and services and engaging in decision-making and planning processes” (Consortium 2012, 22). Villagers were conceptualized as political citizens, active agents who contribute to the execution of government. However, similar to the above-outlined omissions of constitutive aspects of government, the modalities of citizens’ participation in decision making were never discussed. In its implementation practice, the program instead promoted a politically impoverished concept of citizenship, portraying villagers mainly as customers and consumers of government services and assigning them with tasks such as communicating their needs, overseeing the distribution of services, and ensuring that they receive their “fair” share of resources and benefits. Political enunciation was to happen in a formalized institutional setup and to be reduced to a discussion of distributional aspects of the political. The deviation from substantive, and the focus on distributive, aspects of citizenship built on the representation of village governance as nonparticipatory and noninclusive. It is worth noting that this diagnosis was not shared by many villagers, who emphasized the participatory and inclusive nature of village-level decision making and described diverse participatory practices ranging from representation through elders to consultation and joint decision making, the latter particularly in situations of crisis (Bakonyi et al. 2015, 27–31). Female and younger respondents’ views did not differ significantly from those of older men and, with the exception of people from minority clans, the simplified representation of marginalization in the proposal was not reflected in the self-perception of people from these groups (Bakonyi et al. 2015, 31–40).17 Generational and gender relations are more complex than the proposal suggested (see El-Bushra and Gardner 2016). The binary juxtaposition of those with power and the powerless (and the subsequent attempt to “empower” those without power) is itself mediated by power. It places expertise, knowledge, and agency in the hands of experts who are able to know who is powerless and marginalized, a knowledge that allows them to speak for, act upon, and act in the interest of the powerless (Cruikshank 1993, 38-40). To be politically empowered was equated with being able to participate in the VC, and the identified lack of participation was then rationalized as a problem of capacity. Capacity transfer from those who know (international and local experts and trainers) to those who do not know (villagers, village authorities) became the obvious solution. Training in “civil education” or “dialogue processes” for villagers more generally was accompanied by training in leadership skills and advocacy for the marginalized groups. The latter were additionally supported to practice their newly acquired skills in public meetings, while villagers were encouraged “to hold Village Councils to account” (Consortium 2012, 34). The way in which the problem of government was codified as a problem of capacity is not merely a failure to accommodate complexities or caused by a lack of knowledge about local practices. Problems and features of government, including marginalization and exclusion, were individualized as lack of skills, leadership capacities, or the ability to speak publicly. Responsibility for their inabilities, unequal status, and lack of power was thus assigned to the villagers themselves—of course with the prospect that the deficiencies were to be corrected by skills development. The simplified representation of villagers as passive, and of some groups as marginalized and vulnerable, in need of empowerment, was followed by the propagation of a mode of government that is executed through the active participation and choice of (empowered) community members. The program promoted what Cruikshank (1993, 1999) labeled “technology of citizenship,” that is the rearticulation of relations between government and citizens, which make a (neo)liberal form of government operable in Somali villages. It set out to support villagers to play a more active role in village governance, to participate in meetings and decision-making, and to transform marginalized people into “new actors” (Consortium 2012, 22), engaged and vocal enough “to ensure their issues and concerns are addressed” (Consortium 2012, 22). The reconstruction of local level government thus involved the strengthening of the “self-management aspirations” (Rose 1996, 41, 61) of villagers and their transformation from powerless into “active citizens,” capable of promoting themselves and taking their “rightful place as self-actualizing and demanding subjects” (Rose 1996, 60; also Dean 2010, 82-86). The program thus aimed at reconfiguring the subjects of governments and propagated the same kind of “active citizenship” that has been accompanying the “re-invention of government” in advanced liberal democracies (Cruikshank 1993; Rose 1996). This required and built upon simplified versions of gender and generational relations, a depoliticized rationality of rule as management of (human) resources, and of citizenship as resource consumption. Interestingly, however, minority clans,18 the only group that clearly described themselves as marginalized and discriminated against, were not targeted by the program. The strategic ignorance of clan affiliation and the silences that emerged around the clan were matched by secrecy from the villagers when touching upon minority clans. While the Somali program originally aimed at empowering and enhancing the status of minorities, it failed to even identify members of this group. Village authorities often denied the existence of minorities in their villages, claiming that they live mainly in cities. While both the international and Somali members of the consultancy team were able to identify minorities in many villages, the program staff took the local authorities’ denial at face value. The reason for this became clearer when two Somali enumerators in the consultancy team expressed their discomfort with actively seeking out minorities for interviews. They argued that it would be “shameful” for minorities to be identified as such and insisted that enumerators should not openly address them. This type of active disguise was embedded in culturally anchored sentiments and taboos. The secrecy of village authorities reinforced these sentiments but also provided an escape route for Somali staff and released them from engaging in a culturally uncomfortable situation. Internationals, however, did not investigate closely either. As long as a program moves on, employees often decide to discreetly ignore issues that could interfere with the implementation. Again, ignorance became a resource of action, displayed here as secrecy and as a decision not to look too closely to avoid disturbing the everyday practices of the program. This example also accentuates how both knowledge and ignorance are coproduced through negotiations in which power plays an important role. Constituting Communities through Knowledge Tools: Bureaucratic Ignorance Community-driven programs have developed a broad range of knowledge and planning tools to contextualize programming, among them, in Somalia, the mapping of village institutions (VENN diagram), seasonal calendars, socioeconomic analysis, and livelihood analysis. The participatory approach is aimed at generating local knowledge, ensuring that the “community [is] in the driver's seat in the process of designing and implementing their own development plans,” and enabling “the community to understand and overcome their problems in order to rebuild their lives” (Village Report). Villagers were encouraged to discuss and prioritize needs and to develop adequate community action plans (CAPs). The findings of the participatory exercises were compiled in village reports, which included information about history, institutional setup, and the challenges facing villages. The reports were complemented by a number of further documents, among them minutes of meetings, lists of participants, signed CAPs, or training manuals. These documents were centrally stored, but the information they contained was not analyzed, systematized, standardized, or in any other way imbued with meaning. Beyond the CAPs, which identified surprisingly uniform projects (Bakonyi at el. 2015, 79), village-related information did not influence the program's practice. The participatory tools generated information that was, however, rendered inactive and which contributed to what Mathews (2008) has labeled “bureaucratic ignorance.” Indeed, it seems that many reports were never read, as several contained easily detectable copy-and-paste errors.19 The production of information which is then rendered inactive seems futile. The knowledge generating exercises, however, had further functions: they generated legitimacy and were used to demonstrate the program's anchorage in the local. Employees often referred to the files and reports to demonstrate their attentiveness to matters in even the remotest village, to show that local specifics were taken into account, and to evidence the partnership between the program and the local population. The exercises also contributed to the constitution of communities as object and subject of government. Communities are made governable when they are made legible, “objectified by positive knowledge” and subjected “to truth claims by expertise” (Rose 1999, 189; also Cruikshank 1999, 40). The surveys, lists, maps, reports, and diagrams that were produced from the participatory exercises were not simple tools to stimulate discussion, generate knowledge, or “empower” people. They are better understood as technologies through which the village is established as a community, transformed into a governable object, and subjected to government in the form of community self-management. The newly constituted and empowered community is then supposed to act in a way that optimizes government by making it “more effective” (Consortium 2012, 12, 17, 26). With respect to the activation of self-management aspirations, the program was quite successful. It channeled villagers’ expectations and shifted their perceptions of government responsibilities from the regional and district government to the VC and thus to the village and “community” level. The trust and reliance of villagers in the VC increased significantly, although villagers continued to consider the governance capacities of the VC as weak (Bakonyi et al. 2015, 74–75). Translating Concepts into Practice: Vagueness and Tricks Although often criticized as among the reasons for development failure, the Somali program built its strategy on fairly vague concepts and failed to provide conceptual specificity or to adapt concepts to local conditions.20 The term community was, for example, used to characterize the implementation method (community driven), the main beneficiaries (the community), participants in meetings (community meeting), and as a substitute for the village in which the program was implemented. In a similar vein, a large number of program activities were labeled as participation, be it meetings between program staff and local authorities, meetings among villagers or groups of villagers, or the contribution of money and labor, even if people were hired to do the job. It is usually assumed that vagueness and the inability to adapt concepts is driven by a lack of contextual knowledge. The Somali experience, however, suggests the opposite, that vagueness enabled the translation of concepts into different contexts. The concept of community as it has emerged in development theory and practice since the 1980s seems to resist precision and specification. Instead, its attractiveness lies in its ability to integrate ambiguous, even contradictory, meanings. While community assumes the existence of a culturally specific microcosmos, it delineates a moral rather than a social field, structured by emotional, affect-laden, and durable relations that are built on shared norms and values (Rose 1999, 172). It is this morality that provides participatory approaches with the necessary allure of optimism and purpose (Cornwall and Brock 2005, 1045) needed to activate social relations among people willing to cooperate and manage themselves as communities, through which government can then be executed. There was no need to explore how villagers understand community or to adapt the concept to the local—taking clan-relations or practices of representation into account, for example. Instead, the program invoked moral notions of community and participation and facilitated participatory exercises in order to establish a set of relations between those villagers able and willing to become the village community, that is, to become a “group of people who could be acted upon” and who are themselves “acting together to form ‘communities’” (Cruikshank 1999, 78). The “community” thus embodies a set of relations between people willing (and allowed to, excluding, in the Somali case, minorities) to take over the “driver's seat” and administer themselves in the name of community. This also explains why the program worked mainly with people who were already quite active in the villages, usually people who were organized in community-based organizations and had previously cooperated with international organizations (Bakonyi et al. 2015, 76). The program's attempt to establish communities as self-regulatory bodies—independent of the social reality of already existing communities—became obvious when the program turned its focus to the three district capitals and thus larger cities. Instead of adapting the program practice to the context of the city, the cities were adapted to the rationality of the program: they were first divided into three or four “villages,” the people living within these newly drawn boundaries were then labeled as communities and treated in the same way as the “communities” in the villages. Rather than being caused by a lack of (local) knowledge, conceptual vagueness provided the necessary flexibility to put the program's approaches into practice. Vagueness also allows a broad interpretation of program effects. The sketchy understanding of participation, for example, enabled the program to observe an increase in participation, to use this observation as an indicator of “empowerment” and thus to demonstrate the program's positive impact. Villagers often confirmed this impact. Asked about the effects of the program, an increase in participation and an awareness of the need for consultation were regularly mentioned. Here, for example, is the answer of one elder in a focus group discussion, when asked about impacts of the program: “As one of the benefits . . . [we learnt that] the man who consults can do a lot for the community.” When further prompted by the rather surprised interviewer, the elder admitted that consultations were already commonly practiced: Interviewer: “Let me ask you as Somalis, you knew this before? You used to consult with each other?” Respondent: “Yes.” Other villagers explained that they have learned the importance of joint contributions, albeit that such contributions are a constitutive part of the economic and moral fabric of Somali society and widely practiced (Bakonyi 2009). The program may have channeled collective resources in a specific direction, but it neither introduced contributions nor created awareness of their benefits. Conceptual vagueness enabled negotiations among various stakeholders and the adaptation of meaning to different contexts without having to engage in lengthy discussions. However, it also invited bricolage and tricks. The intervenor's tricks to demonstrate success were mirrored in the tendency of villagers to privilege the program's narrative over the reality of local practices and to fake amazement or applaud the program for reaching its objectives and promoting practices that were already quite common.21 Vagueness allows concepts to travel between different locations, opens up space for negotiations, and enables a variety of stakeholders to translate these concepts to the terms of their main audiences without having to engage seriously with different motivations or worldviews.22 The staff could record an increase in participation and report that the community is indeed in the driving seat of state-building. Local authorities, prominent among them elders and religious authorities, could continue to operate the VC and distribute the benefits of the program on the basis of clan affiliation, and already established women's groups could receive a share of these benefits. None of the involved actors had to endanger the continuation of the program by revealing information or meanings that might be “uncomfortable.” In short, conceptual vagueness allowed actors at all levels in the development chain to “muddle through” the everyday program requirements. Instead of holding vague concepts responsible for development failure, vagueness may well be interpreted as a prerequisite for interventions. It makes concepts flexible enough to travel and to be translated and retranslated to the variety of audiences in the development chain, without having to engage with potentially disruptive or conflictive interpretations. Vagueness is, however, inherently risky, as the continuous translation and retranslation can overburden newly established relations and networks, and uncomfortable information can rise above the surface, rupture institutional arrangements (Rayner 2012, 112–13), or uncouple arenas that were set up during the intervention and linked through community exercises. Conclusion Development aims to support people to improve their lives. In order to do so, development lays claim to knowledge of people and the societies it intervenes in, and it does so, as the above sections outlined, in a managerial rationality that promotes bureaucratization. Above all, development identifies deficits, gaps, and problems—in the Somali case, in the ways villages were governed. Problems are then framed in ways that make societies amenable to improvement through interventions. The failure of development to fulfil its objectives, that is to really build states, empower the weak and marginalized, reduce poverty, or bring about peace, is linked to a lack of, inadequate, or even false knowledge, a diagnosis that is usually followed by a request to generate more or better knowledge and by a search for methodologies that are better suited to do so. However, this article has demonstrated that the line between knowledge and ignorance is a thin one. While management has set itself up as a way of knowing and learning about the local, it has also been enabled by, generated, and enacted ignorance. The article has argued that an exploration of ignorance, the way ignorance is intertwined with knowledge, and how both are put into action can open up productive ways to explore the daily practice of development and help shed light on some of its effects. The article detailed the modes of ignorance that accompanied the delineation of a discursive terrain in which local governance in Somalia was made intelligible and actionable for state-building interventions that could present themselves as politically neutral. It additionally outlined the forms of ignorance that enabled the depoliticization of government by channeling attention from questions of legitimacy and representation to distributional and managerial aspects of rule. Omission allowed the program to discuss problems of government as questions of management. And while citizenship and the right of citizens to be involved in decision making was emphasized, the program simplified power relations, approached power differences as problems of capacity, and thereby individualized responsibilities and assigned them to those “without” power. The moralizing simplifications, the individualization of responsibilities and the emphasis on empowerment are, however, characteristic features of neoliberal bureaucratization. While conceptual vagueness is often blamed for development failure, the article showed how vagueness facilitated interventions. It helped to put potentially contested concepts into practice but concealed conflicts and hegemonic struggles. The forms and modalities of ignorance were negotiated through a series of contingent social practices. The empirical findings, however, demonstrate that both knowledge and ignorance are constitutive parts of the developmental ordering of the social. This brings us back to Max Weber's claims that everyday domination is mainly executed as administration and that bureaucracies dominate through knowledge. While these claims have not lost validity, they can be nuanced and expanded by exploring the ways in which bureaucracies rely on and generate ignorance and through the ways they enact and intertwine knowledge and ignorance while expanding their own operational rationality and outreach. Bureaucracies are generating an enormous amount of information on some aspects of the social, while other aspects are overseen, ignored, hidden, and forgotten. In their daily practice, and here I come back to Foucault and Mitchell, they unfold their expansionist, depoliticizing, and disciplinary power by enacting and rearranging knowledge and ignorance in ways that make the social accessible to an economic and managerial rationality. The specific forms of knowledge and ignorance, and the ways in which they are assembled in interventions, remain an empirical question. If nothing else, the focus on ignorance prevents yet another call toward more, better, or more accurate knowledge, the usual response to development failure. Footnotes 1 Development is here used to characterize interventions aimed at the betterment of a society, irrespective of the intervention supporting economic growth, poverty reduction, state-building, or peacebuilding. While objectives vary, international interventions operate in the same structural field, and operational options, constraints, and knowledge practices are therefore similar. 2 Daase and Kessler (2007, 412) suggest using the term nonknowledge to avoid the derogatory implications of ignorance and to limit ignorance to the strategic use of nonknowledge. Nonknowledge is translated from the German term Nichtwissen, but is neither commonly used in English nor is it possible to clearly differentiate nonknowledge from strategic ignorance (Taussig 1999, 118–26). I therefore continue to use ignorance but emphasize the relational character of knowledge and ignorance. 3 For a critique of knowledge-transfer, see Narayanaswamy (2013). 4 For a profound critique see Sabaratnam (2013). 5 Foucault (1979, 132, 133) labels these modes of government as “intervening liberalism” or neoliberalism, while Rose (1999, 139–42) prefers “advanced liberal” to stress the complexification of liberalism and not the succession from it. I stick to neoliberal, mainly because I discuss, with Hibou, (2015) “neoliberal bureaucratization.” 6 See, in particular, the special issue of Economy and Society 41, no. 1 (McGoey 2012) and the Routledge Handbook of Ignorance Studies (McGoey and Gross 2015). 7 Development has used community-based methods since the 1970s, but they were significantly expanded after the World Bank decided to promote them in the 1990s (Mansuri and Rao 2013, IX). 8 Max Weber instead emphasized the deep intertwinement of bureaucracies with capitalism. 9 Legibility is a central concept in James Scott's (1998) analysis of the state. This article argues that the bureaucratic ordering described by Scott is not limited to the state. 10 Hoag (2011, 83–84) points to the shared ethos and affinity of bureaucracy and science. 11 A detailed overview is provided by McGoey and Gross (2015), a critical summary by Aradau (2017). 12 Simmel's contribution toward a sociology of ignorance is outlined by Gross (2012). 13 A review of strategic ignorance is provided by McGoey (2012). 14 The sad absurdity of some interventionist actions and categories became blatantly obvious in Somalia. Many of the ICU Islamists that were ousted by Ethiopia were, during subsequent peace negotiations, integrated into the TFG, classified as moderate, and distinguished from the radicals, for example al-Shabaab. 15 AMISOM grew from sixteen hundred to twenty-two thousand troops between 2007 and 2016. Details are provided by Williams and Hashi (2016). 16 Evaluations are an important part of the managerial rationality of development and likely contribute to further types of ignorance. The evaluation in Somalia was accompanied by so many constraints that it requires a separate assessment. 17 The lack of self-perception as marginalized does not mean that people are formally equal; they clearly are not. 18 Minority clans are, due to their occupation as blacksmiths, barbers, etc., assigned an inferior status in the clan system, lack political representation, and often live in dire deprivation and poverty. 19 This could be interpreted as bad practice relating to a particular program. However, similar effects were found in other contexts (see Heimer 2012; Hull 2012). 20 Vagueness refers to the use of the same terms in different contexts and with different meanings. The failure to properly contextualize interventions or to clarify core concepts, among them participation, community, ownership, or empowerment, were regularly considered as reasons for development failure (Cornwall and Brock 2005; Mansuri and Rao 2013, 288-90). 21 While such tricks could be interpreted as a strategic response of subaltern actors, they are not restricted to the subaltern, but are a rather common feature of development (see Schlichte and Veit 2007). 22 Strunz (2012) argued that conceptual vagueness can enhance creativity, facilitate inter- and trans-disciplinary communication, and promote problem-solving. A similar argument was made by Best (2012) with respect to ambiguity. Acknowledgements I would like to thank my colleagues Gidon Cohen, Markus Hoehne, Raphaela Kormoll, Klaus Schlichte, Leaonard Seabrooke, Kirsti Stuvoy, and the two anonymous reviewers for their constructive critique of earlier versions of the text. References Aradau Claudia . 2017 . “ Assembling (Non)Knowledge: Security, Law, and Surveillance in a Digital World .” International Political Sociology 11 ( 4 ): 327 – 42 . Autesserre Séverine . 2014 . Peaceland. Conflict Resolution and the Everyday Politics of International Intervention . Cambridge : Cambridge University Press . Bakonyi Jutta . 2009 . “ Moral Economies of Mass Violence: Somalia 1988–1991 .” Civil Wars 11 ( 4 ): 434 – 54 . Google Scholar Crossref Search ADS Bakonyi Jutta . 2013 . “ Authority and Administration beyond the State: Local Governance in Southern Somalia 1995–2006 .” Journal of Eastern African Studies 7 ( 2 ): 272 – 90 . Google Scholar Crossref Search ADS Bakonyi Jutta , Cohen Gidon , Bédard Pierre-Olivier . 2015 . “ An Evaluation of Hogaan Iyo Nabad: A Community Driven Governance Programme in Somalia/Somaliland .” Accessed March 13, 2018, https://www.dur.ac.uk/dgsi/cdgsomalia/ . Barnett Michael , Finnemore Martha . 1999 . “ The Politics, Power, and Pathologies of International Organizations .” International Organization 53 ( 4 ): 699 – 732 . Google Scholar Crossref Search ADS Barnett Michael , Finnemore Martha . 2004 . Rules for the World. International Organizations in Global Politics . Ithaca, NY : Cornell University Press . Barry Andrew . 2001 . Political Machines: Governing a Technological Society . London : Athlone Press . Barry Andrew . 2002 . “ The Anti-political Economy .” Economy and Society 31 ( 2 ): 268 – 84 . Google Scholar Crossref Search ADS Bayart Jean-Francois . 2000 . “ Africa in the World: A History of Extraversion .” African Affairs 99 ( 395 ): 217 – 67 . Google Scholar Crossref Search ADS Best Jacqueline . 2012 . “ Bureaucratic Ambiguity .” Economy & Society 41 ( 1 ): 84 – 106 . Google Scholar Crossref Search ADS Consortium . 2012 . Proposal to UK Department for International Development (Unpublished) . Cornwall Andrea , Brock Karen . 2005 . “ What Do Buzzwords Do for Development Policy? A Critical Look At ‘Participation,’ ‘Empowerment’ and ‘Poverty Reduction.’ ” Third World Quarterly 26 ( 7 ): 1043 – 60 . Google Scholar Crossref Search ADS Cruikshank Barbara . 1993 . “ The Will to Empower: Technologies of Citizenship and the War on Poverty .” Socialist Review 23 ( 4 ): 29 – 55 . Cruikshank Barbara . 1999 . The Will to Empower. Democratic Citizens and Other Subjects . Ithaca, NY : Cornell University Press . Daase Christopher , Kessler Oliver . 2007 . “ Knowns and Unknowns in the ‘War on Terror’: Uncertainty and the Political Construction of Danger .” Security Dialogue 38 ( 4 ): 411 – 34 . Google Scholar Crossref Search ADS Dean Mitchell . 2010 . Governmentality: Power and Rule in Modern Society . London : Sage . Eade Deborah . 1997 . “ Capacity-Building: An Approach to People-Centred Development .” UK Oxfam. Accessed September, 10, 2016, http://policy-practice.oxfam.org.uk/publications/capacity-building-an-approach-to-people-centred-development-122906 . El-Bushra Judy , Gardner Judith . 2016 . “ The Impact of War on Somali Men: Feminist Analysis of Masculinities and Gender Relations in a Fragile Context .” Gender & Development 24 ( 3 ): 443 – 58 . Google Scholar Crossref Search ADS Ferguson James . 1994 . The Anti-politics Machine. Development, Depoliticization, and Bureaucratic Power in Lesotho . Minneapolis, MN : University of Minnesota Press . Foucault Michel . 1978 . The History of Sexuality Volume I: An Introduction . New York : Pantheon Books . Foucault Michel . 1979 . “ German Neoliberalism III .” In The Birth of Biopolitics. Lectures at the College De France , edited by Senellart Michael , Ewald Francois , Fontana Allesandro , 129 – 57 . New York : Palgrave Macmillian . Goetschel Laurent , Hagmann Tobias . 2009 . “ Civilian Peacebuilding: Peace by Bureaucratic Means? ” Conflict, Security & Development 9 ( 1 ): 55 – 73 . Google Scholar Crossref Search ADS Green Maia . 2000 . “ Participatory Development and the Appropriation of Agency in Southern Tanzania .” Critique of Anthropology 20 ( 1 ): 67 – 89 . Google Scholar Crossref Search ADS Gross Matthias . 2012 . “ ‘Objective Culture’ and the Development of Nonknowledge: Georg Simmel and the Reverse Side of Knowing .” Cultural Sociology 6 ( 4 ): 422 – 37 . Google Scholar Crossref Search ADS Heimer Carol A. 2012 . “ Inert Facts and the Illusion of Knowledge: Strategic Uses of Ignorance in HIV Clinics .” Economy & Society 41 ( 1 ): 17 – 41 . Google Scholar Crossref Search ADS Hibou Beatrice . 2015 . The Bureaucratization of the World in the Neoliberal Era. An International and Comparative Perspective . New York : Palgrave Macmillan . Hoag Colin . 2011 . “ Assembling Partial Perspectives: Thoughts on the Anthropology of Bureaucracy .” PoLAR: Political and Legal Anthropology Review 34 ( 1 ): 81 – 94 . Google Scholar Crossref Search ADS Hull Matthew S. 2012 . “ Documents and Bureaucracy .” Annual Review of Anthropology 41 ( 1 ): 251 – 67 . Google Scholar Crossref Search ADS Jaeger Hans-Martin . 2007 . “ ‘Global Civil Society’ and the Political Depoliticization of Global Governance .” International Political Sociology 1 ( 3 ): 257 – 77 . Google Scholar Crossref Search ADS Li Tania . 2007 . The Will to Improve: Governmentality, Development, and the Practice of Politics . Durham, NC : Duke University Press . Li Tania . 2011 . “ Rendering Society Technical: Government through Community and the Ethnographic Turn at the World Bank in Indonesia .” In Adventures in Aidland: The Anthropology of Professionals in International Development , edited by Mosse D. , 57 – 80 . Oxford : Berghahn . Lipschutz Ronnie D. 2005 . “ Power, Politics and Global Civil Society .” Millennium—Journal of International Studies 33 ( 3 ): 747 – 69 . Google Scholar Crossref Search ADS Mansuri Ghazala , Rao Vijayendra . 2013 . Localizing Development: Does Participation Work? Washington, DC : World Bank . Accessed March 13, 2018, https://openknowledge.worldbank.org/handle/10986/11859 . Mathews Andrew S. 2008 . “ State Making, Knowledge, and Ignorance: Translation and Concealment in Mexican Forestry Institutions .” American Anthropologist 110 ( 4 ): 484 – 94 . Google Scholar Crossref Search ADS McGoey Linsey . 2007 . “ On the Will to Ignorance in Bureaucracy .” Economy & Society 36 ( 2 ): 212 – 35 . Google Scholar Crossref Search ADS McGoey Linsey . 2012 . “ The Logic of Strategic Ignorance .” The British Journal of Sociology 63 ( 3 ): 533 – 76 . Google Scholar Crossref Search ADS PubMed McGoey Linsey , Gross Matthias . 2015 . “ Introduction .” Routledge International Handbook of Ignorance Studies . New York : Routledge . Menkhaus K. L. 1997 . “ Somalia: Political Order in a Stateless Society .” Current History 97 ( 6 ): 220 – 4 . Mignolo Walter D. 2005 . The Idea of Latin America . Malden : MA: Blackwell Publishing . Mitchell Timothy . 1991 . “ The Limits of the State: Beyond Statist Approaches and Their Critics .” American Political Science Review 85 ( 1 ): 77 – 96 . Mitchell Timothy. 2002 . Rule of Experts. Egypt, Techno-Politics, Modernity . Berkley : University of California Press . Mosse David . 2004 . Cultivating Development: An Ethnography of Aid Policy and Practice . London : Pluto Press . Narayanaswamy Lata . 2013 . “ Problematizing ‘Knowledge-for-Development.’ ” Development and Change 44 ( 5 ): 1065 – 86 . Google Scholar Crossref Search ADS Rappert Brian . 2012 . “ States of Ignorance: The Unmaking and Remaking of Death Tolls .” Economy & Society 41 ( 1 ): 42 – 63 . Google Scholar Crossref Search ADS Rayner Steve . 2012 . “ Uncomfortable Knowledge: The Social Construction of Ignorance in Science and Environmental Policy Discourses .” Economy and Society 41 ( 1 ): 107 – 25 . Google Scholar Crossref Search ADS Richmond Oliver P. 2012 . “ A Pedagogy of Peacebuilding: Infrapolitics, Resistance, and Liberation .” International Political Sociology 6 ( 2 ): 115 – 31 . Google Scholar Crossref Search ADS Rose Nikolas . 1996 . “ Governing ‘Advanced’ Liberal Democracies .” In Foucault and Political Reason. Liberalism, Neo-Liberalism, and Rationalities of Government , edited by Barry Andrew , Osborne Thomas , Rose Nikolas , 37 – 64 . Chicago : University of Chicago Press . Rose Nikolas . 1999 . Powers of Freedom. Reframing Political Thought . Cambridge : Cambridge University Press . Rose Nikolas , Miller Peter . 1992 . “ Political Power Beyond the State: Problematics of Government .” The British Journal of Sociology 43 ( 2 ): 173 – 205 . Google Scholar Crossref Search ADS Sabaratnam Meera . 2013 . “ Avatars of Eurocentrism in the Critique of the Liberal Peace .” Security Dialogue 44 ( 3 ): 259 – 78 . Google Scholar Crossref Search ADS Schlichte Klaus . 2012 . “ Der Streit Der Legitimitäten. Der Konflikt Als Grund Einer Historischen Soziologie Des Politischen .” ZeFKo Zeitschrift für Friedens- und Konfliktforschung 1 ( 1 ): 9 – 43 . Google Scholar Crossref Search ADS Schlichte Klaus , Veit Alex . 2007 . “ Coupled Arenas: Why State-Building is So Difficult .” In Working Papers Micropolitics of Armed Groups No. 3 . Berlin : Humbold University . Scott James C. 1998 . Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed . New Haven, CT : Yale University Press . Spittler Gerd . 1980 . “ Abstraktes Wissen Als Herrschaftsbasis: Zur Entstehungsgeschichte Bürokratischer Herrschaft Im Bauernstaat Preußen .” Kölner Zeitschrift für Soziologie und Sozialpsychologie 32 ( 3 ): 574 – 604 . Strunz Sebastian . 2012 . “ Is Conceptual Vagueness an Asset? Arguments from Philosophy of Science Applied to the Concept of Resilience .” Ecological Economics 76 : 112 – 8 . Google Scholar Crossref Search ADS Suhrke Astri . 2007 . “ Reconstruction as Modernisation: The ‘Post-Conflict’ Project in Afghanistan .” Third World Quarterly 28 ( 7 ): 1291 – 308 . Google Scholar Crossref Search ADS Taussig Michael T. 1999 . Defacement: Public Secrecy and the Labor of the Negative . Stanford, CA : Stanford University Press . Trotha Trutz von . 1994 . Koloniale Herrschaft: Zur Soziologischen Theorie der Staatsentstehung am Beispiel des “Schutzgebietes Togo.” Tuebingen, Germany : Mohr Siebeck . Weber Max . 1980 . Wirtschaft Und Gesellschaft. Grundriss der Verstehenden Soziologie . Hrsg. Von Johannes Winckelmann . Tübingen, Germany : J.C.B. Mohr (5. rev. Aufl.) . Williams Paul , Hashi Abdirashid . 2016 . Exit Strategy Challenges for the AU Mission in Somalia . Mogadishu : The Heritage Institute for Policy Studies . Accessed March 13, 2018, http://www.heritageinstitute.org/wp-content/uploads/2016/02/Exit-Strategy-Challenges-for-the-AU-Mission-in-Somalia.pdf . Winckler Joel . 2011 . “ Managing the Complexities of Intervention: United Nations Peace Operations as Organisational Action .” Journal of Peace, Conflict & Development 18 : 83 – 103 . © The Author(s) (2018). Published by Oxford University Press on behalf of the International Studies Association. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

Journal

International Political SociologyOxford University Press

Published: Sep 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off