Christoph Zürcher, When Governance meets Troubled States in:

Marianne Beisheim, Gunnar Folke Schuppert, Marianne Beisheim, Gunnar Folke Schuppert (Ed.)

Staatszerfall und Governance, page 9 - 27

1. Edition 2007, ISBN print: 978-3-8329-2459-1, ISBN online: 978-3-8452-0100-9,

Series: Schriften zur Governance-Forschung, vol. 7

Bibliographic information
I. Befunde 11 When Governance meets Troubled States Christoph Zürcher A. When Governance meets Troubled States Understanding external state-building, international governance, and the interdependencies between the two, is an extremely challenging task. It is a real life problem, which means that we need to get our empirical evidence from places which are not easily accessible for social research (e.g. from the often messy target states of interventions, and from the often quite abstract realm of the international system). This requires a multidisciplinary approach (political sciences, sociology, international law, social anthropology, and their various sub-disciplines), with its attendant problems and limitations. Nevertheless, if social science wants to be more than just a belated companion of largely incomprehensible social processes going on in places like Afghanistan, Irak, Kosovo, Bosnia, Macedonia, East Timor, DR Congo, Somalia, Somaliland etc., we need to engage in these problems in a more consistent way.1 In order to do so, I argue that we need to embrace in a much more thorough way the empirical realities on the ground. Secondly, we need to refine our concepts. Based on this, we may then be able to develop a better methodological tool kit for assessing the effects of state-building measures – intentional or not – on troubled states. In this paper, I put forward five ideas which may help to achieve this. Firstly, I argue that statehood should best be understood and assessed as four different outputs: domestic authority, security, a measure of material public goods, and institutions for non-violent conflict processing. Secondly, I argue that the state is not any more the only and sole provider of these outputs. Increasingly and especially in the context of post-conflict and/or low-income countries under stress, IOs and INGOs contribute to the provision of these outputs. Thus, it is the state together with external actors that provides statehood (or fails to do so). Thirdly, I propose that the widely used metaphorical concept of a continuum of state capacities ranging from normal to collapsed, should be replaced by the analytical concept of »configuration of statehood«. A specific configuration of statehood is identified by a) assessing of how much domestic authority, security, material public goods, and institutions for non-violent conflict processing are provided, and b) by whom 1 These are only the notorious cases. The number of »troubled states« is obviously a matter of definition resp. a matter of research interest. The World Bank worries about 30 low-income countries at risk. The British DFID identifies 46 fragile states. Foreign Policy and the Fund for Peace list 60 countries in their failed state index, 20 of those red-labeled. The 2005 Peace and Conflict Report lists 10 redlabeled countries and 20 yellow-labeled countries. 12 these outputs are provided. Switching from a continuum to configuration thus enables us to assess the governance input of external actors. Fourthly, I argue that we need to take more into account the social fabric of »troubled states«. Contrary to the first impression, these states usually retain surprisingly high organizational capacities which they invest in regime stability. The mechanisms by which this is achieved are highly informal, but nevertheless accessible to empirical research. Related to this, I introduce the concepts of selective statehood and outsourced statehood. Selective statehood refers to the strategic decision of state elite to only invest in certain core functions of statehood whereas others are consciously outsourced to external actors. Fifthly, building on insights into social fabric of troubled states, I try to illustrate that the interaction between the state, external state-builders and local elites often produces results that are less than successful with regard to state-building. I do this by describing three »games«. The first game is played by IOs and central state representatives; the second game is played by INGOs, local state representatives, and local big men; and a third game is played between and within IOs and INGOs. I hope to demonstrate that, given the preference structure of the actors, these games result in a perpetuation of weak statehood. The effectiveness (or lack thereof) of state-building, seen in this light, is hence also a result of interaction between social actors, and not only a consequence of structural difficulties. B. State-Building External state-building2 usually denotes the well-intended but overambitious attempts of external actors to collate the age-long process of state-formation into a period of a few years, at the end of which the externally re-built states are expected to resemble modern nation states. A more appropriate term for such an endeavour would then rather be »instant statehood« – and as with instant coffee, we should not expect this to be the »Real Thing«. Here, I define external state-building in a broad sense as all intentional measures, aiming explicitly at changing the social and political system of the target state in such a way that core functions of statehood will be executed in a more efficient and less violent way. State-building measures are designed to: a) convince, constrain or pressure elites in a way that leads them to adopt new policies to achieve progress in state-building, b) directly implement institutions at the local, regional and (rarely) central level that enable progress in state-building, c) assume full control over statehood, 2 Jason Brownlee uses the term »foreign-led state-building«. See J. Brownlee, Imperial Designs, Empirical Dilemmas: Why Foreign-Led State-Building Fails, in: CDDRL Working Papers, Center on Democracy, Development, and the Rule of Law, Stanford Institute on International Studies, Stanford 2005. 13 These three types are not mutually exclusive. In the majority of casees, they overlap. State-building measures can be brought along by »state-builders« through positive and negative incentives. Examples of positive incentives are aid for development and reconstruction. Negative incentives are economic sanctions, a weapons embargo, or freezing of aid. Also, state-building can be implemented after a military intervention has cleared the ground for the state-builders, as it is the case in Kosovo, Bosnia, Afghanistan and Iraq. We must set apart the intentional measures of external state-building from the context in which state-building takes place. We even would be well advised to assume that the incentives stemming from the international system very often distort or even contradict the intentional state-building efforts. The strategies of local elites are not only influenced by realities on the ground and by the incentives that the intentional policies of the state-builders create, but no less (and often much more) by incentives that are created within the international system. Of relevance are, firstly, economic flows and dependencies which may make it difficult to overcome the weak, highly informal and often partly illicit economies of troubled states. Secondly, regional power constellations and perceptions of threat which may be supportive of contenders and spoilers can make it difficult to build viable domestic authority. And lastly, the universal norm of legal sovereignty, as coded in international laws and practices, may actually produce negative incentives: In essences, these norms and practices attach a high profit for sovereignty, make good governance extremely expensive, and only weakly punish bad governance. In addition, there is virtually no risk of losing sovereignty for existing states, whereas for would-be states, there are virtually no chances of acquiring full sovereignty.3 Consequently, a rational behaviour for states would be to try enjoying the full sovereignty rent without paying the price for good governance. C. Statehood and its providers If state-building means »all intentional measures, aiming explicitly at changing the social and political system of the target state in such a way that core functions of statehood will be executed in a more efficient and less violent way«, then what is statehood, and what are the core functions of statehood? I argue here that for all practical purposes (not necessarily normative expectations, derived from Max Weber or from international law) statehood is a process at the end of which there are some outputs (»products«). The crucial ones are: Domestic authority e.g. the ability of getting things done: This means that principalagent problems within the bureaucratic structures (or alternative institutions) are (suf- 3 See for a discussion on the – wrong – incentives this state of affairs produces S.D. Krasner, Sovereignty Exhausted. Paper Prepared for Presentation at the International Research Colloquium Berlin and Blankensee, July 16-18, 2004: The Future of Intervention: Intervention, Legitimacy and the Reconstruction of Statehood, 2004. Paul Collier equally advocates to »curb the sovereignty rent«. P. Collier, in a talk at the International Research Colloquium »The Future of Intervention«, Blankensee-Berlin, July 16-18, 2004. 14 ficiently) solved, and that the policies of the state bureaucracies (or alternative institutions) are met with a measure of societal compliance. Security: The physical integrity of the population is guaranteed. Threats of military and common violence are successfully checked. Material public goods: The modern nation state draws a significant proportion of its legitimacy from the fact that it provides the population with a certain measure of material public goods. Which goods these are is certainly a matter of negotiation within society and between society and the state, and expectations will be dependent on more or less realistic expectations of what is possible in the given context. Institutions for non-violent conflict processing: There must be a set of institutions in place that allow the processing of conflict in an embedded, non-violent way for those individuals and groups that share a given territory. Conflict means the struggle over values and claims to scarce status, power and resources in which the aims of the opponents are to neutralize, injure or eliminate their rivals. Conflict itself is not the problem. Conflict is an everyday occurrence in all societies and in most cases it is dealt with in an institutionalized and socially embedded way. The problem is dis-embedded conflict, and conflict that leads to a disruption of inter-group cooperation. All societies have specialised institutions to process conflict in a predictable, institutionalized and non-violent way. These institutions may be formal (e.g. courts, parliaments, elections) or informal (e.g. courts of mediators/elders; markets/bazaars; rules of co-optation, i.e. »buying« contenders and competitors with privileges, or institutionalized forms of corruption). Only the purest Weberians assume that all conflict processing institutions must be formal ones and state-run. In general, it has become a widely shared assumption that societies are more stable and less prone to internal violence when there is a dense network of social institutions in place that allow for conflict processing on various levels and in various issue areas. However, it is usually the state that acts as a supreme rule setter, enabling these informal conflict processing institutions and delimitating areas and issues which are to be regulated by non-state institutions. Whether this function can be executed temporarily or constantly by other agencies (such as IOs or INGOs) remains to be seen. However, this question is one of the most crucial ones for the empirical test of new forms of governance. According to the classical notion of modern statehood in the line of Max Weber, these outputs are seen as being provided by the state (that is, by the bureaucratic structure of the state apparatus). I argue that, against the backdrop of social reality in many, if not most, of today’s states, such a notion rather reflects our normatively based image of how states should be, and less how states today are. To be sure, the tremendously successful self-representation efforts of modern states reinforce this fiction.4 Consequently, I depart form the notion that statehood is provided solely by the state. Instead, I suggest that we think of statehood as a product which is produced by the state in association with other actors. There are examples abundant when states outsource – intentionally or not – the provision of basic functions to external actors. It is 4 On this, see for example J.S. Migdal, Strong Societies and Weak States, State-Society Relations and State Capacities in the Third World, Princeton 1998. 15 sufficient to think of who provides security in Afghanistan or Tajikistan, domestic authority in Kosovo or Bosnia, or public services in Mozambique or Burundi. There are also international institutions and organizations in place to assume these functions – think of the UN transitional administration, the international forces in Afghanistan, or of the World Bank’s suggestion to set up so called ISAs (Independent Service Authorities) in low-income countries under stress (LICUS). ISAs would provide basic services, being independent from government and acting like wholesale contractors with multiple channels for retail provision. In essence, ISA is the outsourcing of basic state services to a private, donor-funded organization.5 The provision of statehood by a multitude of actors is an empirical reality which can be observed in many post-conflict and low-income countries. But what we still critically lack is sound assessment of the implication this has: How sustainable are such arrangements? Are they temporary phenomena, or rather new forms of governance? Is there a way back to classic statehood and classic government? How are these arrangements themselves governed? Where can accountability and legitimacy be localised? Can the bonds between society and the state be replaced by bonds between society and »a coalition between state and IOs«? What does societal compliance mean under such arrangements? Finding answers to these question deserves to be high-up on the agenda of governance research. D. Configurations of Statehood If one wants to assess the effect of state-building, one needs to have a measure of statehood. It is commonplace throughout the vast and growing literature on state failure/state-building to introduce some sort of a typology of the troubled states. States are then classified on a continuum, ranging from strong to weak, failing, collapsed, fragile, and precarious and the like.6 However, there exist a number of substantial problems with most attempts at typologising, and as a result, such typologies are of little analytical value. To start with, these typologies are essentially metaphorical rather than analytical; they are usually not rooted in solid empirical comparative work, and the criteria which were applied to categorize are often unclear and poorly operationalized. Moreover, little attempt is usually made to clearly distinguish between the causes of state failure and the symptoms of state failure. For example, Robert Rotberg claims that internal violence (measured as death in combat) is one of three signals of impeding state failure (the other two being a sudden economic decline and a shift towards a more authoritarian system).7 However, it is not clear whether these signals are treated as the causes for state failure (internal violence leads to state failure) or as a conse- 5 See World Bank Group Work in Low-Income Countries under Stress: A Task Force Report, The World Bank, Washington 2002. 6 See for example R.I. Rotberg, State Failure and State Weakness in a Time of Terror, Cambridge, Ma. Washington, D.C., 2003. 7 Ibid., pp. 21-22. 16 quence (state failure enables internal tensions to escalate). But for designing policy options for presenting state failure, it is essential to know whether internal violence is a cause or a symptom of state failure, because prevention must target the former, not the latter. Finally, it is not clear whether the idea of a continuum ranging from »strong« to »collapsed« is actually best suited to reflect the different empirical realities of troubled states. How to place entities like Afghanistan, Karabakh, Timor and DR Congo on such a continuum? And what good would it do? Clearly, the current states of typologies pose many problems. In the following, I suggest how some of these problems could be overcome. There are three elements to my suggestion: Firstly, typologies should be based primarily on outputs of statehood (security, domestic authority, material public goods, and institutions for non-violent conflict processing). Secondly, typologies should not reflect a linear continuum (or ranking) but rather configurations of statehoods, taking into account that states may do well in one dimension, and bad in another. Such different configurations may then also point to different qualities of states: Some »weak« states may be able to produce security and provide domestic authority, while being locked in a development blockade. Others may provide public goods, but may provide little security. And thirdly, typologies should not only include output of statehood, but also the contributors to statehood, hence taking into account that outputs of statehood are provided no longer solely by the state, but also by NGOs and INGOs. How could a typology reflecting »configurations of statehood« be developed? One way – admittedly not an easy one – to do it would be to develop a composite indicator for each of the four outputs of statehood (domestic authority & societal compliance; security; public material goods; institutions for conflict processing). Each indicator should also show whether this output has been provided mainly by the state, mainly by external actors, or by a cooperation of both. Once such indicators are established for a given state at a given time, it will be possible to judge how much of each output of statehood is currently being provided and by whom it is provided. When such indices are available over time, it will be possible to track changes in the configuration of statehood.8 Take as an illustrative example Tajikistan: The country has suffered from a civil war between 1991 and 1997. A snapshot of 1995 would reveal that there is virtually no domestic authority, (very few) material public goods are provided by humanitarian organizations, institutions for conflict processing are not in place, and security is very low. A snapshot in 2005 would reveal that Tajikistan has considerably increased the output in all four domains. It is one of the few examples of a turn- 8 Consider that most of the data is available. See for example: Indicators on Governance: World Bank’s Country Policy and Institutional Assessment (CPIA); World Bank Governance indicators (; Bertelsmann Transformation Index. Rankings: LICUS; State Failure Index ( 3100). Data on state failure and internal war: Polity IV; Fearon and Laitin Data set; Doyle and Sambanis data set; Collier & Hoeffler Data set; MAR and State Failure Project; Bertelsmann Transformation Index. 17 around, from »weak« to »less weak«. However, we would also see that domestic authority is now mainly provided by the increasingly strong patronage network of the president, whereas security and welfare has been to a large extent outsourced to external actors (Russian troops, and INGOs). In other words, statehood has been recovered, but the Tajik state has changed its social fabric in a fundamental way, and this is reflected in the visualized configuration. Illustrative Example: E. Inside Troubled States: Selective and Outsourced Statehood In understanding how statehood is affected by interventions, one must draw upon an understanding of how statehood de facto works. Most accounts of weak and failed states start and end with a long list of things that do not work: Emphasis is on lack of security, education and health care, poor economic performance, illegal or illicit economies, decay of infrastructure, weak institutions and authoritarian rule. These features may indeed apply to weak and failing states. But only in failed states are they the dominant features. There is only a handful of states that may be best characterized by what is not working anymore. Among these states are the usual suspects (Afghanistan, Haiti, Somalia, DR Congo, Sierra Leone, Liberia, and Iraq). For the great bulk of states which are neither OECD-like nor collapsed, such a list only reveals part of the picture. Surprisingly enough, these states often retain organizational capacities which 18 seem to contradict the image of weak or failing states. Travellers, businessmen or technical advisors are quite aware of how efficient these states are when it comes to: blocking the free movement of people, goods or capital, rulers ensuring regime stability; and regimes nurturing the networks of patronage which are the backbone of governance in such states. Quite differently from their colleagues in the political science departments, area specialists on Africa, Latin America and more recently on post-Soviet periphery and the Balkan have focused on these islands of organizational capacities. Arguably, there is much to learn from the qualitative, empirically rich analysis of weak states in their incarnation as »shadow states«, »network states«, »command states« or »cunning states«. Most accounts of statehood in regions where states are notoriously un-Weberian seem to share common features: they do not depict the differences between Weberian and OECD states as a »deviation«; they try to explain certain features of such states as an outcome of strategic decisions by the state managers rather than as failure; they assume that such arrangements may be rather stable (even though they may present symptoms of permanent crisis), and they assume that the interaction between state and society is not taking place only at the sharp boundaries between the public and the private (as it is assumed for the Weberian state). Rather that these boundaries are constantly reworked, blurred, and contested; and that the borderline between state and society greatly varies. Furthermore, this borderline may significantly differ, depending also on issue areas and on the hierarchical level (central, regional, local). Scholars have coined different terms for what they perceive as a type of state that seems to work in a surprisingly stable fashion, despite the crisis ›symptomology‹. One such term is what the late social anthropologist Georg Elwert calls the ›command state‹. Command states are a widespread phenomenon in Africa, and they are partly reinforced by development aid by western aid agencies.9 Although laws do exist in such a state, they are subordinate to the personal authority of state officials. Economic markets are not primarily regulated by the laws of supply and demand, but by the decision of the patrons who control access to markets. Written laws are left deliberately vague, leaving space for informal bargaining. Every action that is not explicitly allowed might be declared illegal according to the interests of patrons, and permissions are usually obtained through bribes. Such a system is intrinsically corrupt and clientelistic. In the absence of institutionalised venues of settling disputes the risk for violence between competing patrons and networks is high. The »command state« appears to the outsider as overly bureaucratic. But, as Elwert notes, »this is nothing but a façade. The real goal is to block the opponent and to make him invest lots of time, fruitlessly. An entrepreneur needing 70 official signatures for paying a single bill will be ready to bribe one of the 70 civil servants in order to regain the autonomy over his time.«10 9 G. Elwert, The Command State in Africa. State Deficiency, Clientelism and Power-Locked Economies, in: S. Wiggel et al.: (eds.), Entwicklungspolitische Perspektiven im Kontext wachsender Komplexität, 2001, pp. 419-452. 10 G. Elwert, E + Z – Entwicklung und Zusammenarbeit 2001, pp. 284-287. 19 A close relative of the command state is what we have called elsewhere the »network state« in post-Soviet Space.11 In many of the successor polities of the former Soviet Union, surprisingly stable regimes can be found that are at the same time also extremely weak and which often suffer from sustained low-level conflict. We argue that this combination of weak states, conflict and relatively stable regimes is not a paradox, but rather a result of a rational strategy of the elites.12 Elites of newly independent (quasi-) states often find that the best way to govern is to institutionalize networks of patronage. In order to stabilize these networks of patronage, their patrons must satisfy the needs of their clientele. Not surprisingly, patrons in such weak »network states« usually control access to natural resources such as oil, gas or cotton, and positional resources such as tax departments, border guards or police forces. Furthermore, in order to secure his position, the patron must prevent potential challengers from access to resources from which patronage can be derived. Independent economic activities are thus discouraged, and patrons typically try to hinder independent activities outside of their network. As part of this strategy, patrons will also attempt to minimize public goods such as safety, protection, economic opportunities or legal protection; instead they will try to privatize these goods and to make them available only through the network. One means of achieving this aim is to keep the state weak, another means is to tolerate or even promote low-level conflict even within their own state, since this increases insecurity and thus maximizes the dependency of political actors (and the population) on the patron’s good will. State weakness may thus be even a rational choice of leaders, who base their rule on networks of patronage. Weak states may thus emerge by design, and not just by default, as the common wisdom assumes. Governance through networks in weak states may stabilize the regime, but it may also increase the risk of conflict. In contrast to other regions, the »network-state« in the post-Soviet space is highly institutionalized and highly parasitic. Qualitative research in Azerbaijan, Georgia, Kyrgyzstan, Tajikistan and Uzbekistan reveals that buildingup and securing the networks of patronage is done in an institutionalized, socially embedded way and with the help of the state bureaucracy. In other words, here the state (that is the bureaucracy and its coercive power) and the networks (that is the chain of principal and agents, bound by personal loyalty and personal authority) are closely merged – although both retain their specific functional logic. In this respect, the post-socialist network-state shares some similarities with what Will Reno calls the »shadow state« in Western Africa: »The shadow state is the product of personal rule, usually constructed behind the façade of de jure state sovereignty.«13 The Cold War posed an unusual opportunity for rulers in Africa and other parts of the Third World to count on aid from superpower patrons, but in order to receive this aid they had to have control over the internationally recognized state. The rulers used the aid as a resource to bypass popular legitimacy and for buying strong- 11 J. Koehler/C. Zürcher (eds.), Potentials of Dis/Order. Explaining Violence in the Caucasus and in the Former Yugoslavia, Manchester 2003. 12 C. Zürcher/J. Koehler, Berliner Osteuropa Info (2001), pp. 48-52. 13 W. Reno, Shadow States and the Political Economy of Civil Wars, Boulder/Colorado 2000, p. 45. 20 men’s loyalties. Likewise, rather than spending on expensive public services like infrastructure, education or health, ruling elites were buying the obedience and support of local key players in order to strengthen their regime. Ivan Krastev’s »cunning state« is a close relative of the above discussed types of state and shares with them some important features (the most important is that the emergence of this type of state is a result of strategic decision making rather than of »failure«). It differs in so far as the cunning state is also a distinctive product of intentional intervention (which was, ironically, aimed at building up state strength). Analyzing the reasons for weak states in the Balkans, Krastev concludes that state weakness does not only come from the implementation of regulations imposed by international actors like the IMF or the EU, but is also derived from the intention of significant portions of the new elites who use their positions to extract wealth.14 The »cunning state« is a state that is selectively weak and that makes use of its perceived weaknesses: »The fact that most weak states are democracies creates conditions for the strategy of selected weakness. Balancing between the pressures coming from the international community and the pressure from citizens political actors are managing to be weak in areas that are not important for the survival of the regime and at the same time to be strong enough when it comes to the interests of the governing elite.«15 While external constrains ensure policy predictability in the region, they also have a negative effect on relations between politicians and the public, as elites on their quest for the extraction of wealth use external pressure to excuse their lack of social responsibility. As Krastev puts it: »Governments are elected after a love affair with the electorate, but they are married to the international donors.«16 Shalini Randeria applies the concept of the cunning state in the arena of global governance which is characterized by unaccountable international institutions and scattered sovereignties.17 She describes the interaction between localized social actors, international organizations and the cunning state which remains a central actor in selectively transposing neo-liberal policies to the national terrain, and capitalizes on its perceived weakness, in order to render itself unaccountable to its citizens. Randeria cautions against attributing homogeneity to the state since the logic of action on the federal and regional levels might be very different. She also stresses the shifting contours of the boundary between the public and the private as well as the growing entanglement between civil society and state. Against the background of this sketchy insight into the state in its »network«, »cunning«, »shadow« or »command« incarnations, we find that many of the perceived »state failures« are in fact the product of a strategic decision of elites, coming to terms 14 I. Krastev, Journal of Democracy 13 (2002), pp. 39-52. 15 I. Krastev, Bringing the State Up. Paper Delivered at the Confernce »Interethnic Relations in the Western Balkans: Problems, Instruments and Prospects for the Future«, Berlin, September 12 and 13, 2003, p. 13. 16 I. Krastev, In Search of Responsive Government. State-Building and Economic Growth, Policy Studies Series, Center for Policy Studies, Central European University, Budapest 2003, p. 45. 17 S. Randeria, Discussion Paper Nr. SP IV 2003 2005 (Arbeitsgruppe »Zivilgesellschaft: historischsozialwissenschaftliche Perspektiven« des Forschungsschwerpunkts Zivilgesellschaft, Konflikte und Demokratie des Wissenschaftszentrums Berlin für Sozialforschung AGZG, 2003). 21 with the most pressing needs (reconstructing a minimal degree of central authority, keeping potential challengers off, coming to terms with demands from the international community without loosing grip on power). The weakness of many of these states is then often a strategic and selective weakness that enables ruling elites to prey on international support and, if available, on domestic resources, while escaping accountability and responsiveness to domestic and international demands. What is more, such arrangements are often unintentionally sponsored by the international community. To depict these realities, I find it helpful to introduce the concepts of selective statehood and outsourced statehood.18 Selective statehood refers to the strategic decision of state elite to only invest in certain core function of statehood whereas others are consciously left out. Most importantly, elites invest in highly institutionalized, yet informal networks of patronage which become the dominant mode of governance. In order to overcome the principalagent problems inherent to all networks, and to »feed« the clients, elites also invest in control over natural resources, border regimes, the security apparatus and control over staffing procedures for key positions. This will ultimately lead to a high degree of informality, strong vertical ties between elites and local strongmen, frequent cooptation of regional power brokers and constant negotiations between different segments of networks. Patrons, who usually reside in the capital as head of lucrative ministries, constantly need to engage local potentates by granting them admission to privileges and higher levels of autonomies. Direct central control can often only be achieved for a few strategic important regions, whereas the rest of the country is de-fact governed by regional strong men. These regions are often also excluded from the (few) material benefits the state has to deliver. As a rule, INGO then step in and compensate for the states’ lack of commitment. When this becomes a pattern, I speak of outsourced statehood. Certainly, what is most often outsourced to the »internationals« is basic services. In wider parts of lowincome countries, overburdened basic services are only available from the internationals. However, it is not only food security, health care or education that is outsourced. There are also abundant examples of outsourced security (Tajikistan, Bosnia, Kosovo, Afghanistan) and outsourced institution-building on the local and regional level (Mozambique, Afghanistan). Notably, both selective and outsourced statehood produce patterns that may be taken as symptoms of state weakness (weak institutions, high degree of informality and high dependence of foreign donors). However, this is only partly true; if the above presented analysis is correct, then these symptoms represent a weakness which is partly intended (because it helps getting foreign aid, and because rule by patronage depends on insecurity and informality) and partly faked, because vital lines of command remain intact, although they are secured by informal mechanisms. 18 C. Zürcher, Internationale Politik 60 (2005), pp. 13-14. 22 F. When Governance meets Troubled States: Games actors play What happens when external state-building meets with selective statehood – and why is the outcome often less than satisfying? In the following, I am concerned in the first place with intentional external state-building. This, as a rule, takes place in the context of an UN mission and/or massive development and reconstruction aid. Most of these measures are implemented by INGOs or NGOs on behalf of IOs or nation states. For the sake of simplicity, let us assume that we can depict this process as three simultaneously ongoing games. The first one is between the »internationals«, represented by an International Organisation (IO) and the central state representatives. The second one takes place on the ground between state representatives, INGOs and local big men. The third one takes place within and between INGOS and IOs. I intend these »games« to be illustrative examples of why state-building measures often produce suboptimal results. Let us also assume the stage is an ordinary low-income post-conflict state with weak institutions, the central elite’s grip on the regions is weak, and networks of patronage are the dominant mode of social interaction within the polity. I. The first game is played by IOs and central state representatives: The key interest of the state representatives is to preserve their regime. This requires a minimal measure of domestic authority. Domestic authority in turn depends on the capacity to incorporate important parts of the elite into the dominant network of patronage (compliance), and to block societal mobilization. Both objectives in turn cause certain costs: networks must be nurtured and monitored. Co-opted potential contenders want their share, the population expects a certain minimal provision of public goods, and security services (»muscles«) must be paid. State representatives will thus reject, or at least hinder, policies that weaken their networks of patronage and that could promote mobilization of the population. They will also reject policies that threaten to undermine their ways to secure financing. An effective way to gauge the potential for cooperation between IO and state representatives is then to look into the strategies of how state representatives maximize profits and minimize costs. Typically, state representatives of such countries do not depend on taxes. Usually the poverty of the population and the difficulties of the terrain make tax collecting unrewarding. Instead, state elites typically focus on the control over natural resources and on export and import licences. In order to minimize the cost associated with statehood, they allow state bureaucracies to be highly corrupt in order to give them the opportunity for self-financing. At the same time, corrupted officials are easy to control by the central state, because they live under the constant threat that the state (that is, the executive, or the president) may actually apply the law and sack them. A further means of minimizing the costs of statehood is to attract foreign donors, especially by encouraging INGOs to provide basic services to the population. Another means is to encourage labour migration of the population. 23 Given this state of affairs, what is there to be hoped for the international organization that steps in? The IO has on its broader agenda »state-building« and or »peace-building«19. This huge task is broken down into manageable pieces, such as humanitarian assistance, reconstruction of infrastructure, reconciliation, fostering institutional growth and finally arriving at good governance. However, there is a widely shared and well grounded consensus, that implementing any of these requires security in the first place. And this in turn requires a minimal degree of domestic authority. Thus, the IO will be anxious not to develop policies that may undermine the shaky little amount of domestic authority that there is. And that is where it meets the expectation of the state representatives. In short, the field for cooperation is largest when it comes to minimizing the costs of statehood, and it is smallest where key mechanisms of ruling and financing the network-state are concerned. This explains why prolonged efforts at state-building often get stuck after a short initial success in improving the living conditions of the population. When state-building meets selective statehood, there is an inbuilt tendency for the status-quo to be perpetuated – meaning: networks of patronage and high degrees of informality remain strong, notwithstanding the fact that services for the basic needs of the population may be improved.20 II. A second game is played by INGO, local state representatives, and local big men: Let us now consider a second »game« that takes place on the ground, in any province of the above-mentioned typical post-conflict, low-income country. After civil war or internal unrest, there usually emerges a type of actor that exercises considerable influence in the village or in the region. These local big men used to be field commanders of the rebel armies, or police officers that helped the state to win the war. They control the lucrative parts of the formal and, perhaps more frequently, the illicit economy. They drive the big SUVs, and they support their community with donations. They reside in the provincial capital, and they have protection from somebody in the capital. They own large amounts of land or run the only industries in the region. These local big men are an unavoidable by-product of state-weakness and internal war, and thus they are a social reality in most of the 190 or so nation-states. A second player are local representatives of the state, such as regional and provincial governors. They are tasked with advocating the claims of the state towards society, meaning that they try to ensure a measure of societal compliance. The third player is the locally active INGO, most often an implementing agency of state-building programmes of IO (such as the UNDP) and national governments (USAID, GTZ). For the representatives of the state and for the locally active INGOs, local big men are a problem: On the one hand, their autonomy defies meaningful efforts at state-build- 19 This is a benevolent assumption. If state-building is only second priority, whereas hunting terrorists is the first, then the game is played differently. This also is a clear hint that state-building and battling rebels or terrorists on the ground (directly or by proxy) can be conflicting objectives. 20 This is consistent with econometric studies of aid effectiveness that find that aid effectiveness sharply drops after initial successes. See P. Collier/A. Hoeffler, Aid, Policy and Growth in Post-Conflict Societies. World Bank Policy Research Working Paper 2902, October 2002, World Bank, 2002. 24 ing, and their influence and prestige often greatly exceeds that of local state representatives. State-building efforts are a threat to the autonomy of local big men, which they have achieved during times of civil unrest and war. On the other hand, given their considerable influence on local politics, it is usually quite impossible to get things done, faced by the resistance or opposition, of the local big man. Both representatives of the state and locally active INGOs must then define their strategies. They can even try to circumvent local big men, hoping to marginalize them. Or they can co-opt local big men, hoping to ›civilize‹ them. Empirical evidence shows that the first option is risky and rarely works. Furthermore, both local state representatives and INGOs fear that they themselves would be marginalized if they choose to circumvent and the other actor chooses to co-opt. Consequently, both local representatives of the state and locally active INGOs tend to choose option two, trying to co-opt local big men. What then follows is an absurd but predictable competition between INGO and state representatives around the co-opting of local big men. Whatever the outcome in the long run is (meaning the prospect for civilization), there is a high probability, that in the short run, the autonomy of local big men will only gradually diminish, and that the cooperation between the state and the INGO will remain difficult (or will not take place at all). 21 Both results are not favourable to state-building efforts. III. A third game«: The NGO scramble22 The first two games were based on the benevolent assumptions that IO and implementing INGOs were not only honest brokers of state-building, but also capable of acting in a goal-oriented, coordinated way. The first may be true, the second never is. Imagine a state-building intervention which is mandated by the UN, supported by an overwhelming majority of UN member states, generously funded by the OECD states, and implemented by various INGOs and local branches of IOs: It will soon become clear that the quality, and the eventual success of the intervention, depends not only on the commitment and the strategy of the interveners, but also to great extent on the coordination of the various actors in the field. Not only the quality of the strategy matters, but also the quality of the implementation and this in turn crucially depends on how efficiently the various agents resolve the coordination problem. Viewed from below, the neat typology of actors – states, INGOs and IOs – and their different, clear cut spheres of action get blurred. In the reality of state-building measures, various states (resp. ministries charged with overseeing intervention measures), IOs, implementing INGOs (operating mainly on national level), implementing local NGOs (operating mainly on local level), implementing local branches of IOs (for example the UNDP office in Tajikistan) form a complex structure which can be 21 This is consistent with the explicit warnings that have recently been forwarded by development practitioners about the dangers of »circumventing the state«, see A. Mehler, Oligopolies of Violence in Africa South of the Sahara, Institut für Afrikakunde, 2004. 22 See A. Cooley/J. Ron, International Security 27 (2002), pp. 5-39. 25 best understood as multiple principal-agent relationships. As in all principal-agent relationships, the problem of coordination within the principal-agent structure and between various and often competing principal-agent structures is of utmost importance. The quality of implementation depends on how efficient state actors, IOs, local branches of IOs, INGOs and local NGOs coordinate their efforts, and this, as organizational theory predicts, depends in turn on external incentives and internal organizational structure. While only few authors fully question the positive contribution and motivation of IOs and NGOs, critics now increasingly explain certain deficiencies of peace and statebuilding processes by applying a sociological institutionalist approach of organizational behaviour to IOs and NGOs involved in such interventions. Instead of being mere tools executing the goals for which they were founded, IOs and NGOs are first and foremost institutions/bureaucracies in the Weberian sense that develop a life of their own and become actors in their own right. Viewing IOs and NGOs involved in state-building interventions as agents rather than structures opens up new explanatory frameworks. In this venue bureaucratic culture emerges as a factor significantly forming IO and NGO behaviour: »Ethnographic studies of IOs describe a world in which organizational goals are strongly shaped by norms of the profession that dominate the bureaucracy and in which interests themselves are varied, often in flux, debated, and worked out through interactions between the staff of the bureaucracy and the world in which they are embedded«.23 In the extreme, this can lead to »pathological« behaviour typically associated with bureaucracies, i.e. an obsession with fulfilling internal rules thereby creating a distorted organisation-internal vision of reality and remaining insulated from feedback from outside.24 Looking at IOs and NGOs as agents following their own institutionally defined goals also draws attention to their interest in self-preservation. This has been criticized widely as leading to suboptimal outcomes with regard to the stated goals of these organizations. The Berlin-based think tank, the European Stability Initiative (ESI),25 strongly criticises the large Balkan peace building missions in Bosnia and Kosovo of »inventing« new fields of activities to justify their continued presence in the Balkans. In this vein the Office of the High Representative in Bosnia, the main IO implementing the Dayton Agreement in Bosnia, after accomplishing initially set goals such as establishing the political framework defined in the Dayton Agreement in 1995, supporting the return of refugees, etc., is now involved in a large scale restructuring of the judicial system and conducting an anti-corruption campaign. Cooley and Ron (2002) highlight another aspect of IO and NGO inefficiency. By applying a political economy approach, the authors find that competition for funds in a highly competitive »aid business«-environment can create »powerful institutional imperatives [that] can subvert IO and INGO efforts, prolong inappropriate aid 23 M. Barnett/M. Finnemore, International Organizations 53 (1999), p. 706. 24 Ibid. 25 See G. Knaus/M. Felix, Journal of Democracy 14 (2003). 26 projects, or promote destructive competition among well-meaning transnational actors. Attempts by IOs and INGOs to reconcile material pressures with normative motivations often produce outcomes dramatically at odds with liberal expectations«.26 In summary, by depicting the »in-the-field« actors of international state-building interventions, IOs and NGOs, as actors in their own right, strongly interested, among other issues, in their own survival and growth, sheds fresh light on the difficult realities of external state-building. On the one hand, IOs and NGOs take significant responsibility in implementing the main aspects of international state-building interventions. On the other hand, derived from their nature as organizations, there appears to be a tendency of these IOs and NGOs to (over-)emphasize their roles and thus to take on responsibilities that should, at least, be critically reviewed. From the perspective of sustainable state-building, this tendency is highly worrying as it can impede the emergence of a local ownership of the state-building process and inhibit local political groupings taking responsibility for the political processes within their country. In this paper, I have suggested that statehood could be »measured« along four dimensions: domestic authority, security, provision of public goods, and institutional arrangements for non-violent conflict processing. In many developing and/or postconflict states, which are usually the prime target of international governance assistance, these outputs are provided not only and not solely by the state, but also by international actors. A meaningful typology of state »strengths« or »weakness« then should reflect the overall level of these outputs, but also the contribution of external actors to it. Based on such data, it would then be possible to identify different »configurations« of statehood. In assessing the outputs of statehood, I argue that it is of crucial importance to acknowledge that many of these outputs, but especially domestic authority, are built on informal, but nevertheless surprisingly effective mechanisms, whereas many »weaknesses« of states are actually a result of deliberately outsourcing some of the costs of statehood. Finally, I have argued that the interaction between external state-builders and local elites often reinforces the specific arrangements and configuration in these »troubled« states, which may explain the limited success of external state-building. G. References Barnett, Michael/Finnemore, Martha: The Politics, Power, and Pathologies of International Organizations, International Organizations 53 (1999), pp. 699-732. Brownlee, Jason: Imperial Designs, Empirical Dilemmas: Why Foreign-Led State-Building Fails, in: CDDRL Working Papers, Stanford: Center on Democracy, Development, and the Rule of Law, Stanford Institute on International Studies, Stanford 2005. Collier, Paul/Hoeffler, Anke: Aid, Policy and Growth in Post-Conflict Societies. World Bank Policy Research Working Paper 2902, October 2002, World Bank, 2002. 26 A. Cooley/J. Ron, International Security 27 (2002). 27 Cooley, Alexander/James Ron: The NGO Scramble. Organizational Insecurity and the Political Economy of Transnational Action, International Security 27 (2002), pp. 5-39. Elwert, Georg: The Command State in Africa. State Deficiency, Clientelism and Power-Locked Economies, in: Steffen Wippel et al. (eds.), Entwicklungspolitische Perspektiven im Kontext wachsender Komplexität. Festschrift für Dieter Weiss, 2001, pp. 419-452. Elwert, Georg: Für eine neue deutsche Alpha-Politik, E+Z – Entwicklung und Zusammenarbeit, 2001, pp. 284-287. Knaus, Gerald/Felix, Martin: A European Raj?, Journal of Democracy 14 (2003), pp. 23-33. Koehler, Jan/Zürcher, Christoph (eds.): Potentials of Dis/Order. Explaining Violence in the Caucasus and in the Former Yugoslavia, Manchester 2003. Krasner, Stephen D.: Sovereignty Exhausted. Paper Prepared for Presentation at the International Research Colloquium Berlin and Blankensee, July 16-18, 2004: The Future of Intervention: Intervention, Legitimacy and the Reconstruction of Statehood, 2004. Krastev, Ivan: The Balkans: Democracy without Choices, Journal of Democracy 13 (2002), pp. 39-53. –: Bringing the State Up. Paper Delivered at the Conference »Interethnic Relations in the Western Balkans: Problems, Instruments and Prospects for the Future«, Berlin, September 12 and 13, 2003. –: In Search of Responsive Government. State-Building and Economic Growth, Policy Studies Series, Center for Policy Studies, Central European University, Budapest 2003. Mehler, Andreas: Oligopolies of Violence in Africa South of the Sahara, Institut für Afrikakunde, 2004. Migdal, Joe S.: Strong Societies and Weak States, State-Society Relations and State Capacities in the Third World, Princeton 1998. Randeria, Shalini: Between Cunning States and Unaccountable International Institutions: Social Movements and Rights of Local Communities to Common Property Resources, in: Discussion Paper Nr. SP IV 2003-502: Arbeitsgruppe »Zivilgesellschaft: historisch-sozialwissenschaftliche Perspektiven« des Forschungsschwerpunkts Zivilgesellschaft, Konflikte und Demokratie des Wissenschaftszentrums Berlin für Sozialforschung AGZG, 2003. Reno, William: Shadow States and the Political Economy of Civil Wars, in: Mats Berdal/David M. Malone (eds.), Greed and Grievance: Economic Agendas in Civil Wars, Boulder/Colorado 2000, pp. 43-68. Rotberg, Robert I.: State Failure and State Weakness in a Time of Terror. Cambridge, Ma./ Washington, D.C. 2003. Roy, Olivier: Development and Politial Legitimacy: The Case of Iraq and Afghanistan, Conflict, Security & Development 4 (2004), pp. 167-180. World Bank Group Work in Low-Income Countries under Stress: A Task Force Report, The World Bank, Washington 2002. Zürcher, Christoph: Gewollte Schwäche, Internationale Politik 60 (2005), pp. 13-24. Zürcher, Christoph/Koehler, Jan: Institutions & Organizing Violence in Post-Socialist Societies, Berliner Osteuropa Info (2001), pp. 48-52.

Chapter Preview



Zu erforschende Governance-Probleme stellen sich in den unterschiedlichsten Räumen von Staatlichkeit: Zum einen in den - und darauf war bisher das Hauptaugenmerk der Governance-Forschung gerichtet - vergleichsweise wohlgeordneten Staaten der OECD-Welt wie in postnationalen Herrschaftsverbünden wie der Europäischen Union (Stichwort: European Governance), ferner auf der globalen Ebene (Stichwort: Global Governance) sowie schließlich in Räumen, die man als Räume begrenzter Staatlichkeit bezeichnen kann. Governance in Räumen begrenzter Staatlichkeit, d.h. in so genannten »weak« oder »failing states«, war Gegenstand einer Ende September 2005 in Berlin von der Alfred-Herrhausen-Gesellschaft und dem Wissenschaftszentrum Berlin veranstalteten Konferenz.

Die dort gehaltenen Vorträge sowie einige ergänzende Beiträge werden im vorliegenden Sammelband dokumentiert.