Skip to navigation – Site map
Dossier : États et territoires du politique
Première partie. Décentralisation et reconfiguration des territoires de l'action publique
1.2 Les territoires dans les représentations d'acteurs

What we talk about when we talk about decentralization? Insights from post-revolution Tunisia

Ce dont nous parlons quand nous parlons de décentralisation : Idées sur la Tunisie après la révolution
Lana Salman
p. 91-108


This article explores the language of decentralization policies in post-revolution Tunisia. It argues that international financial institutions mobilize a particular language about decentralization to create new clients and reinvent themselves in a shifting political landscape. Decentralization policies have gained traction in the Middle East and North Africa region post-Arab revolutions partially as a result of these shifting priorities and their accompanying language. In a region with limited decentralization experiences, it is not enough to look at incentives of politicians to decentralize or the effects of decentralization programs once they are rolled out. Instead, we need to look at the confluence of domestic pressures to pluralize the terrain of politics and development agendas born outside the region’s borders which frames decentralization as an appropriate development framework at this juncture. To make this argument, I analyze the language of World Bank flagship publications that constitute the intellectual backbone for decentralization programs. I then turn to Tunisia and compare the language of these reports with the language that activists use to characterize decentralization in relation to regional inequalities. At stake in this comparison are two discourses of decentralization that call forth diverging development models: one model focuses on the financial, administrative and political autonomy of municipalities from the central government, while the other calls for reparations and a desire for a state that restores past injustices. This language forces us to rethink the mode of politics undergirding decentralization policies, and accordingly to re-theorize decentralization from a particular place on the map, from Tunisia.

Top of page

Full text

  • 1 . The content of this article was developed in conversation with Lamia Zaki whose research and crea (...)

1This article1 explores why decentralization policies have gained traction in the Middle East and North Africa Region (MENA) post-Arab uprisings. In contrast to an international development and political science literature that emphasizes, either the motivations of local politicians to decentralize in response to popular pressure, or assesses the effects of decentralization policies once they are rolled out, I explore the role of international financial institutions (IFIs) in marshalling decentralization policies post Arab uprisings. I argue that decentralization policies are vehicles for IFIs to create new clients and reinvent themselves as partners responsive to post-revolution demands. However, these policies and programs encounter facts on the ground that militate for alternative development models. The facts pertain to how a number of civil society organizations characterize regional inequalities as deliberate marginalization and call for reparations rather than decentralization as a mode of state intervention. Focusing on what different actors talk about when they talk about decentralization is important, particularly in this revolutionary context, because it shows two competing modes of politics: a post-political mode of technocratic intervention that has marked Tunisia’s past, opposed to an adversarial mode of politics promoted by civil society that calls for reparations in the context of transitional justice. To make this argument, I review the above-mentioned literatures on decentralization, and analyze several World Bank publications. I supplement this textual analysis with observations from public forums addressing the problem of regional inequalities today. Although the last section is built on uses empirical observations, the scaffolding of this article builds on a careful review of the grey literature published by international financial organizations as well as my own professional experience at these institutions. I include these empirical observations in the hope to open the debate about questions and topics I continue to research.

2In the first section, I provide a brief review of the international development and political science literature on decentralization, as well as the literature about the limited experiences of political decentralization in the region. In the second section, I analyze the World Development Reports 2008 Agriculture for Development and 2009 Rethinking Economic Geography as well as accompanying publications showing how these reports frame and justify the necessity of decentralization policies. In particular, I highlight the emphasis on an urban centric model of growth and a language that presents the concentration of wealth as simply inevitable. Consequently, the role of local governments becomes restricted to managing the externalities resulting from this unavoidable development model. In the third section, I turn to the language of civil society organizations about regional inequalities and the role of the state in redressing them. I show how this language disrupts the intellectual basis that justifies decentralization as an appropriate policy intervention post-revolution by proposing reparations as an alternative development model. I conclude the article with reflections on what it means to think about decentralization from Tunisia not just as empirical variation, but also as theory that helps us make sense of the current post-revolutionary moment.

Decentralization: political motivations, policy effects and a limited track record in the MENA region

3The political science and international development literature on decentralization can be classified in two categories: (i) the literature about incentives of politicians to decentralize, and (ii) the literature about the effects of decentralization programs once they are rolled out. Motivations to decentralize include efficiency considerations such as overcoming information barriers and heterogeneous preferences, or redistribution concerns such as diffusing social tensions and fostering accountability (Bardhan, 2002). While there is an agreement that the sequencing of decentralization in political, administrative and financial aspects is important (Falleti, 2005), literature about the empirical effects of decentralization on local democracy and accountability as well as development is largely inconclusive (Treisman, 2007). The impacts of decentralization remain country/region specific and usually constrained to particular sectors (Faguet, 2014). Some countries end up recentralizing decision-making (Malesky et al., 2014), while experiences of decentralization in post-conflict situations or countries undergoing transition warn that decentralization can deepen cleavages and increase the likelihood of (sometimes armed) conflict, rather than consolidate democracy at the local level (Eaton 2006 , on the example of Colombia). These various conclusions point out to the difficulty of predicting at the outset the effects of decentralization and to the need for thorough empirical research.

4Decentralization experiences in the Middle East and North African region are limited. Although a new research agenda is emerging (see Harb and Atallah, 2015), earlier research about local governance focused in large part on municipalities rather than decentralization. This research documented the presence of Islamist or Islamist leaning political parties at the municipal level, in countries where they had been (often violently) banished from political life. Because municipalities were considered innocuous and inconsequential spaces of experimentation, the presence of these parties at the local level forced the pluralization of the terrain of politics in otherwise highly centralized regimes. Studying this process of politicization at the local level was an entry point to studying decentralization. At its core was the role of municipalities as sites for the exercise of politics, rather than their potential to be financially, administratively and politically autonomous from the central level.

  • 2 . On the experiences of JDP in local governance, see the article of Coline Mias in this issue addre (...)

5In Morocco, a regionalization (rather than a decentralization) policy framed the advent of the Islamist Justice and Development Party, JDP (hizb al-adala wa al-tanmiya) to local power. Banking on their success in parliamentary elections in 2000 and inspired by the Turkish Justice and Development Party AKP, the JDP announced its intent to aggressively compete for local elections. JDP candidates ran for elections in urban areas only and had a few notable victories especially in second-tier cities. Catusse and Zaki (2009) studied the governance practices of the JDP municipal councils in three municipalities where they won the elections. The authors note that far from islamizing politics, the practices of the JDP were couched in technocratic orientations. Their discourse and practices were not without contradictions and oscillated between ‘virginity’ (virginité) in politics because they were outside the political game, and experience accumulated through their sustained presence in charitable associations2. Studying three municipalities controlled by Hezbollah in Lebanon, Harb (2009) shows similar results, noting that Islamist discourse was used as a legitimizing rhetoric post-fact, rather than to guide an electoral program, and that the desire for technical and professionalized management drove the privatization of services even in a context such as Lebanon where the legal framework for municipal action is highly restrictive. This was especially true in large urban municipalities. Comparing trends across various Islamist leaning municipalities both in contexts of weak states such as Lebanon and Pakistan, and strong states such as Morocco and Turkey, Signoles (2009) emphasizes similar results. AKP’s rule at the municipal level, which provided the benchmark for Morocco’s JDP, exhibits the same trends (Massicard, 2009). In Mubarak’s Egypt, a country that lacks a third tier level of municipalities and where elected Local Popular Councils have an advisory role only, some decentralization reforms designed in collaboration with USAID were underway in 2010, but have since been aborted (Boex, 2011).

Decentralization: managing the externalities of an urban-centric growth model

6The literature reviewed above testifies to the hesitant trajectory of decentralization in the region. Reflecting on Tunisia, Beatrice Hibou captures the gist of these experiences under Ben Ali by describing them as highly centralized experiences of decentralization (2015, 118-120). A brief history of the country’s decentralization policies helps illustrate this point.

7Tunisia’s successive post-independence governments have constantly redrawn the map of the territory’s administrative organization with security and development concerns nestled into each other. Geographer Amor Belhedi chronicles the history of these reshufflings, which included the creation of new governorates, municipalities and delegations, or the suppression, amalgamation or division of already existing ones. He shows that this history is a rearrangement of rather than a break with colonial logics of governing (1989, 2-6). Sami Yassin Turki and Eric Verdeil (2015) also emphasize this point. In addition, the government addressed decentralization policies more concretely since 1975 through the stipulations of the Organic Law 75-33 dated May 14 1975 (Loi Organique des Communes OL33-75) and its subsequent modifications in 1981, 1985, 1989, 1993, 1995, 2006, and 2008 (Gilbert & Dafflon 2012)3. The now defunct 1959 Constitution in its Article 71 defined local governments as “municipal councils, regional councils and the structures to which the law grants the status of local authorities to manage local affairs according to conditions provided by law” (UCLG Tunisia Country Profile)4. It was accompanied by four constitutional amendments - the last of which was 2011 –, and pertained to the organization of temporary local governing bodies post-revolution. These temporary governing bodies known as special delegations (délégations spéciales) are expected to head municipalities until the advent of locally elected local governments5.

  • 6 . See Wahid Kraiem 2015, p. 8-10 for more details about the definition of non-municipal areas stipu (...)

8According to Gilbert and Dafflon (ibid), the above legal armature showcases three main points. First, in terms of political autonomy, the country’s institutional architecture has always pitted deconcentration and decentralization. This is evident in two overlapping but incongruent territorial organizations of deconcentrated and decentralized governing bodies. The deconcentrated scheme includes 24 governorates, 264 delegations and 2073 sectors, while the decentralized scheme entails 24 regions, 264 municipalities, and large swaths of non-municipal areas, which have a negative definition only (the absence of local government)6. It is important to note that although their numbers correspond, the spatial boundaries of these two sets of organizational units do not overlap: one delegation can extend across two municipalities, while one municipality may include several delegations. This incongruence helped maintain a confusion as to the attribution of responsibilities for intervention in various jurisdictions. Second, in terms of administrative autonomy, decentralized territorial entities are submitted by law to the power of oversight (tutelle) of deconcentrated entities, especially the governorate in the legal persona of the governor. The power of oversight was as pronounced in municipal and non-municipal areas. In practice, this translates into a priori approval of the governor of all decisions made by municipalities, or by regional and rural councils in non-municipal areas as well as an a posteriori audit of the legality of all implemented decisions. These include decisions related to the dismissal or appointment of council presidents and members, investment decisions in local infrastructure, taxation, and coordination with deconcentrated agencies and concerned ministries for service delivery. Dafflon & Gilbert summarize the reach and extent of the power of oversight as stipulated in the above-mentioned constitutional and organic laws and their amendments in table 1-11 (ibid., p. 41). Third, in terms of financial autonomy, the above legal framework determines the legal provisions of decentralized budgets and expenses. Several amendments were brought to this framework - the most important of which are the 2006 amendments passed in 2007 (ibid). Even with these amendments, the power of oversight remained heavy-handed. Not only did these provisions counter effective decentralization, but they also created confusion in terms of legally stipulated accounting procedures and left municipalities almost chronically dependent on central level discretion. In that perspective, several amendments of the Organic Law sought to redress the structural indebtedness of municipalities by restructuring local finances - for example, “increasing land taxes and user fees” (UCLG Tunisia Country Profile) among other reforms. All these changes however were taken at the central level, and did little to harness the financial autonomy of local governments.

9This brief overview of the past forty years of decentralization policies in Tunisia makes it plausible to suggest that decentralization is driven by more than local incentives and agendas. While it may be true that in Tunisia’s post-revolutionary context it is harder for the central government to repress or ignore demands for better infrastructure and services across the territory, it remains difficult to show that state benevolence, accountability mechanisms or electoral incentives alone are driving decentralization policies. This is especially true since local elections have been repeatedly rescheduled since 2011. I am not suggesting that local pressures to decentralize are irrelevant, or that the legal armature detailed above did not enable the groundwork for more meaningful decentralization. Rather, I am pointing out that it is the confluence of donor driven agendas, born outside Tunisia’s borders and domestic pressures, to open up highly centralized regimes that make decentralization policies popular in the current post-revolutionary moment. It is therefore essential to look at the role of IFIs in designing and rolling out decentralization programs in order to fully understand the drivers of decentralization policies in the region. Examining the role of IFIs acquires particular importance since this aspect is understudied in the literature.

10This article argues that in this historical conjuncture, IFIs use decentralization policies as vehicles to create new clients and reinvent themselves as relevant development partners in a shifting political landscape. I focus particularly on the World Bank’s (henceforth WB) role in pushing forward decentralization policies, since apart from the budgetary support of the International Monetary Fund (IMF), the WB is among the largest donors in post-revolution Tunisia. This argument is in-line with a broad literature showing that that the WB has periodically redefined its mission and identity to fend off crisis of legitimacy questioning the benefit and relevance of its interventions.

11The World Bank inaugurated the new millennia in its capacity as a “Knowledge Bank”. On October 1st, 1996, President James D. Wolfenshon spoke of “The New Knowledge Partnership” of the “New Bank” (1996, 14-15). In his address to the board of governors, People and Development, at the World Bank Group’s annual meetings in Washington D.C., Wolfenshon, stressed that “development knowledge is a “global commons” (ibid, 14) and that the Bank must invest to make it available for all. Echoing Wolfenshon, Joseph Stiglitz then his Chief Economist, went on to speak of a “Knowledge Bank” entrusted to generate and disseminate knowledge for development as “a global public good” (Stiglitz, 1999, 319). But Stiglitz was more explicit about “the natural complementarity between [the Knowledge Bank] and the more traditional role of the World Bank in providing capital to less developed countries. Knowledge enhances the productivity of capital” (ibid). However, development scholars studying global World Bank projects have documented how knowledge produced by the institution is also used to generate business, rather than merely as a “global public good”. Whether it is agricultural development in Lesotho (Ferguson, 1994), ecological conservation in Laos (Goldman, 2006), or integrated development and conservation in Indonesia (Murray Li, 2007), the constitutive elements of WB projects usually enable similar outcomes: limiting the role of the state and restricting its interventions to regulate markets so that they can function optimally.

  • 7 . After the appointment of Jim Yong Kim in 2012, the World Bank Group has gone through a restructur (...)
  • 8 . “The International Development Association (IDA) is one of the largest sources of assistance for (...)

12Even when knowledge is produced by The Development and Economic Vice Presidency (DEC), the WB’s research unit as opposed to its operational teams7, it is still meant to disseminate a parsimonious assessment of development problems in view of suggesting solutions that they essentially provide. In this manner, WB managers who develop new financial products, researchers who provide the knowledge, and operation staff who package development solutions using both, all work to further the institution’s main mission: moving money. And they have so far been successful at it, since in the midst of the 2008 financial crisis, the Bank remained AAA rated. This assessment does not aim to vilify the Bank or its business. Rather, the aim is to shed light on an aspect of the institution that is less studied than its development mission of ending poverty; that is its nature as a bank, which cannot remain in business if it does not attract new creditworthy clients to whom it can keep lending money. In an increasingly crowded universe of multilateral aid where more and more countries are graduating out of the International Development Association (IDA) into the International Bank of Development and Reconstruction (IBRD) status8, and with more IBRD countries now able to access credit markets (Kanbur and Summer, 2012), the WB must search for new clients in addition to sustaining its current client base.

13An emphasis on decentralization is one of the many ways the WB can accomplish this mission. A close reading of the World Development Report 2009 Reshaping Economic Geography (WDR2009) illustrates this point. A flagship publication of the institution, development reports are meant to identify the most pressing global development problems and provide state-of-the-art solutions addressing them. They are also trend setters of the World Bank’s thematic focus for development projects pursued across regions, but also for other donors and multilateral development agencies intervening around the globe. The WDR2009 is particularly interesting because it marks the Bank’s definitive break away from large-scale interventions in agriculture/rural areas reorienting all future projects towards an urban-centric development model. The move to an urban-centric development model is not new as Gillian Hart (2001, 2010) shows. But I emphasize the definitive break because what preceded the WDR 2009 Reshaping Economic Geography was the WDR 2008 Agriculture for Development. However, the team who wrote the 2009 report was remarkably entrepreneurial. The differences between the two reports and their subsequent uptake are as fascinating as they are telling.

14Unlike the WDR2008, Reshaping Economic Geography is a modular report. Its main message is that in a globalized world, the economic geography landscape is one of concentrated growth. In the language of the report, this landscape is “bumpy” not “smooth”. The concentration of wealth in densely populated areas connected to world markets is not only natural, it is also inevitable. Dozens of examples the world over and decades of development interventions confirm this conclusion. But while concentrated growth is inevitable, development can be inclusive. Making development inclusive is rarely about targeted interventions in particular regions. Indeed, Reshaping Economic Geography is precisely about moving away from “place-based incentives” which have been the focus of much of the policy debates on urbanization and regional development in a globalized world. Instead, the report invites policy-makers to create institutions for integration that “unify all places and put in place infrastructure that connects some places to others (World Bank 2009, xxiii)”. Achieving inclusive development depends on a diagnosis of the type of regional inequalities policy makers want to redress. Enters the modular nature of the report, which testifies to the entrepreneurialism of its authors.

15Regional inequalities imply that the inevitable concentration of wealth results in a “portfolio of places” some of which are “leading” and others that are “lagging”. But not all “lagging regions” are born the same, they vary across three dimensions: density (human geography, although what is meant is economic density but population concentration is considered an acceptable proxy), distance (physical geography, although this is not Euclidian distance only), and division (artificial barriers to the free movement of people and goods). Matching these 3 Ds (density, distance, and division) are 3 Is (institutions, infrastructure, and interventions). Institutions are spatially blind in the sense that they are applied uniformly across space and include for example alienable property rights and regulations that ensure efficacious transition between land uses (for example from agriculture to industrial use). Infrastructure is connective infrastructure such as roads and highways but also sanitation and electricity, and water networks, whereas interventions are spatially targeted; they are place-based incentives that should only be used in extreme cases. For instance, achieving inclusive development in a lagging region that is sparsely populated, distant and disconnected from world markets, that is a lagging region characterized by all the 3 Ds, requires all three Is: institutions to make land markets function better, connective infrastructure to level the playing field in terms of service delivery and targeted interventions to overcome divisions or upgrade slums. This is then a modular report in the sense that it provides a grid of adaptable solutions: achieving inclusive development is choosing among various combinations of Ds and Is. The formula is attractive and simple, and the report provides ample examples, “best practices” as well as failures, to show where and how such combinations can be deployed. Figure 0.1 A navigational aid for the reader on page 43 of the report illustrates these various policy choices. The figure encapsulates the universe of development solutions to regional inequalities in a 3X3 table.

16In 2013, a number of authors of the WDR2009 developed a related knowledge product that is a distilled version of Reshaping Economic Geography and targets city managers more explicitly. Planning, Connecting and Financing Cities – Now: What City Leaders Need to Know (World Bank, 2013) is the structuring framework of a series of Urbanization Reviews that WB staff are writing about countries across the world. Since it is a spin-off of the WDR2009, it is no surprise that it too is organized in three modules: planning, connecting and financing. In Planning, Connecting and Financing Cities – Now, the emphasis on decentralization is more explicit, especially in the financing module. According to this document, decentralization ultimately enables municipalities to become creditworthy. Creditworthiness in turn means that municipalities could engage in public-private partnerships with the private sector, and eventually take on loans to finance their investments without sovereign guarantees. Thus, decentralization is the means to create a market of potential subnational lenders where cities and not countries constitute a new frontier of subnational lending. Since 2011, World Bank experts have produced 16 Urbanization Reviews across the developing world that constitute a basis for lending operations across and between sectors (transport, urban development, governance, etc.)9. No development report has been followed by as explicit an entrepreneurial effort to market its findings and make them operational. This shows that besides its mission of ending world poverty, the World Bank is a heterogeneous institution where different individuals are each pursuing their own careers by doing what banks must do to survive: gaining new clients to keep moving money around.

17I end this section by returning to Reshaping Economic Geography to suggest that unlike what it painstakingly wants its readers to believe, the report’s language and main policy recommendations are not neutral. What they leave us with is a development model that is attentive to space as a structuring factor, but the implementation of which presupposes a post-political mode of politics (Mouffe 2011, Swyngedouw, 2010). In this post-political world, inclusive development is achievable without disagreements or contention, without recalling past policies which may have been unjust, without naming and shaming friend and foe. A reminder about the main messages of Reshaping Economic Geography illustrates this point: (i) although, in a globalized world and in the short to medium terms, we observe a non-linear growth trajectory that leads to a bumpy rather than smooth economic geography landscape, in the long term the spatial distribution of resources and people tends towards equilibrium,(ii) concentrated growth is thus inevitable but inclusive development remains achievable, (iii) by shifting the policy debate about regional development from place-based targeted interventions to integration. It is worth quoting from the report to demonstrate the post-political undertone of this language.

“Many policy makers perceive cities as constructs of the state – to be managed and manipulated to serve some social objective. In reality, cities and towns, just like firms and farms, are creatures of the market. Just as firms and farms deliver final and intermediate goods and services, towns and cities deliver agglomeration economies to producers and workers. So city administrators are better advised to learn what their city does, and to help it do this well, rather than try to abruptly change the course of their city’s destiny. Planners and policy makers should see their role as prudent managers of a portfolio of places, to get the most from agglomeration economies.” (World Bank 2009, 128)

18In addition to a language of lagging and leading regions, Reshaping Economic Geography asserts that it is futile to intervene in cities, and futile to try and redress regional inequalities, except within the framework of 3Ds and 3Is discussed earlier. These facts are presented as self-evident and universal. Franco Moretti and Domonique Pestre (2015) conducted a quantitative linguistic analysis of 50 years’ worth of World Bank publications, including all World Development Reports since 1958. Their analysis sheds light on why and how such universal conclusions emerge. The authors point out to a gradual shift in these publications from a concrete language of loans, infrastructure, investments, and equipment to a diluted language of governance, finance, management, initiatives, frameworks and platforms. They refer to this language as “Bankspeak”. What accompanies the shift to “Banspeak”, the authors argue, is a blurred temporality which enables “’freeing’ a statement from ‘all determinants of place and time, and all references to its producers” (Moretti & Pestre, 2015, p. 96). Statements that at first seem contentious and subject to criticism become universal truths which can be presented as inevitable. Although Reshaping Economic Geography is filled with dozens of case studies grounding the authoritative points of the report in concrete examples situated in specific times and locations, the conclusions abstracted from these examples are evacuated from any historical and geographic specificity. The universalism of these conclusions disavows that these examples are always already situated (see Chakrabarty 2000), and therefore must be questioned within their contexts and times rather than framed as “principles true in every country” (Mitchell, 2002, p. 54-79).

“Marginalized” vs. “lagging” regions: the role of local governments

  • 10 . The FTDES is a Tunisian non-governmental organization created in 2011 with the objective of defen (...)

19Against the language and analysis of World Bank reports that present “concentrated growth” as both natural and inevitable, a language about “marginalized regions” has emerged within the framework of Transitional Justice. In a conference organized by the Tunisian Forum of Economic and Social Right (Forum Tunisien des Droits Economiques et Sociaux FTDES)10 in September 2016, panelists including academics, members of the FTDES and syndicalists of the General Tunisian Labor Union (Union Générale Tunisienne de Travail UGTT) referred to a crisis of the existing development model. They dissected regional inequalities within the framework of marginalization and reparations “atahmysh w jabr alarar”. In these interventions, “lagging regions” such as Kasserine are not simply “lagging” in relation to other “leading” coastal areas as predicted by the WB’s growth model. Instead, they are marginalized regions “manatiq muhamasha”. The relevance of this language is first and foremost its ability to eviscerate a passive voice in the description of development phenomena. While framing some regions as lagging, either because they are remotely located, sparsely populated, or lack an endowment of natural resources is an objective qualification of an unfortunate fate, identifying a region as marginalized recalls the agency of particular actors whose policy choices determined its fate.

20In other words, the language of marginalization delineates the responsibility of the central government that at best was negligent and at worst violent and dis-possessive. In that sense, it is no surprise that the dual focus of this conference was marginalization and reparations, since the latter represents a call for action, to demand that what was deliberately taken away from these regions be restored to them. The concept of marginalization is multi-faceted but the dearth of data at the appropriate scale makes it difficult to quantify it. In an effort to compensate for this lack of quantitative indicators, panelists provided a qualitative understanding of marginalization, which stressed several features. The first feature is a structural absence of the welfare state apparent in lacking or degrading services and physical infrastructure (roads, hospitals but also healthcare provision, schools but also quality of education). The second is a precarious labor force and its distribution across sectors, which is inappropriately captured in the figures of the National Institute of Statistics (INS). Another related feature of a lagging region is unemployment, especially of women and youth, compounded by the degrading quality of education, which reinforces the perception that education is no longer a vehicle for social mobility. This perception is pervasive among youth who go to school to satisfy a parent’s wish, and end up dropping out, both because of degraded physical infrastructure such as the absence of means of transportation to go to school, and the poor quality of education. Young people in marginalized regions perceive education as useless, since they will end up unemployed with or without it. Non-governmental welfare provision is prevalent in marginalized regions. It either entrenches already existing clientelistic networks now emboldened by the presence of political parties, or provides a cover for a laissez-faire state policy that looks the other way when rampant informal/black markets sustain local economies. Finally, marginalized regions experience an angry and shameful self-characterization of being marginalized (hogra) that transcends age groups and ignites a deep sense of accumulated historical injustices that were never acknowledged.

21Based on this assessment of marginalized regions, the FTDES has organized various social movements “from the margins” (men al hamish) to ask the government for reparations in view of mending these historical and ongoing territorial injustices. Reparations represent an alternative development model tied to the creation of employment and investment in quality infrastructure and services. These movements are calling for the restauration of what they perceive as the right to development that has been denied to these regions, rather than just decentralization as the financial, administrative and political autonomy to make decisions at the local level. Of course, there is no reason why reparations cannot be included in the design of decentralization policies, especially financial decentralization. However, this is not the current framework of decentralization in Tunisia. In an article published in Nawaat e-zine, journalist Yassine Bellamine criticizes government approval of the design of a WB project on local governance and urban development currently under implementation. Specifically, Bellamine argues that the formula for central to local subsidies was revised to favor larger and already developed municipalities. Instead of favoring weaker and smaller local governments, these transfers promote “an even more unequal decentralization” (une décentralisation encore plus inégalitaire)11.

22Three initiatives underway are giving visibility for a regional development agenda focused on reparations. Taken together, these initiatives disrupt the attempt to eschew the demand for regional development, which was at the heart of the revolution, by offering decentralization policies as a viable alternative. The first initiative was launched in June 2015 when the FTDES in collaboration with Lawyers Without Borders (Avocats Sans Frontières – ASF)12 submitted a dossier to Tunisia’s Truth and Dignity Commission (Instance Vérité et Dignité – IVD)13 demanding that the governorate of Kasserine be designated a “victimized region” (région victime) (ASF & FTDES, 2015)14. The dossier raises the issue of wide disparities between coastal and inner regions at the heart of the grievances of the Tunisian revolution. Its legal provisions are in line with Article 10, al.3 of Organic Law number 2013-53 on Transitional Justice. It mobilizes a set of definitions and indictors that serve as evidence documenting the “systemic and organized exclusion and marginalization” that lead to important gaps in human and economic development between the coast and its hinterlands. Kasserine was one example of such marginalized inner regions. By default, and since the law associates regions with governorates, this model dossier was meant to be replicated for other governorates deemed to have endured similar processes of marginalization and exclusion. Based on other transitional justice frameworks around the globe, the dossier adopts definitions of marginalization and exclusion that take into account the specificities of Tunisia. It defines marginalization as “(i) a form of distinction, discrimination or persistent and pronounced disadvantage, (ii) which compromises life chances of groups of people, (iii) and which is a result of particular economic, social and political processes. These processes do not necessarily produce discrimination or disadvantage, it suffices that they induce it, and the resultant discrimination be pronounced and persistent over time” (ibid, 24). The dossier defines exclusion as “a process which results in eliminating access of individuals to certain rights and services, including access to employment opportunities and other sources of revenue, to education, and to participation in decisions that have the potential to affect their lives” (ibid)15. The definition also emphasizes that in legal terms, and in order for a region to be considered “victimized”, evidence must show that the marginalization and exclusion in question are organized efforts – meaning systematic and targeted policies rather than conjectural circumstances tied to external factors, such as a declining global economy, or a climate crisis affecting a particular region of the world. It is important to note that this dossier draws heavily on many World Bank reports assessing Tunisia’s political economy, including the question of regional disparities. However, it notes, that “unlike other countries, unequal growth in Tunisia did not lead to inclusive development, the benefits of economic growth did not spill into inner regions, and did not lead to improved opportunities in marginalized areas.” (ibid, 10). In other words, in its language, this dossier disagrees with one of the most important conclusions the WDR 2009: that although growth is concentrated, development can be inclusive.

23It is important to understand what defines Kasserine as a “victimized region”, and to stop at the minutia of definitions of marginalization and exclusion because what is at stake is not only a disagreement about definitions, but a fundamentally different narrative and argumentation about regional development. At their core, these definitions cast doubt on the ability of decentralization defined as political, administrative and financial autonomy of local governments to achieve equitable regional development. This narrative about “victimized regions” is repeated in public forums discussing regional inequalities to counter discourses that present these outcomes as natural. It is also used to chart solutions that do not consist of decentralization policies only. For the FTDES and their partners, the next step is to prepare similar dossiers for other regions similarly victimized under the previous regime.

24The second initiative is to rethink the definition and concept of the region, since the word “region” is not an appropriate translation of the Arabic word jiha. The subtleties of translation were echoed in the question and answer session and were tied both to historical belonging to territory, as well to the most appropriate scale at which statistics should be collected. A panelist member of the IVD argued that in order to build convincing cases of reparations for past territorial injustices, reports need to refer to indicators at the smallest possible geographical scale that could be aggregated up to the governorate level if needed. But most development indicators exist only at the level of the governorate, a scale too large to account for important variations within areas of interest. Beyond this technical requirement of appropriate scales, the discussion about statistics pushed the concept of marginalization – as opposed to lagging regions – even further, casting a serious critique on the presumed role of the state. A panelist who has been processing requests for reparations from various neighborhoods and cities, argued that an analysis of statistical figures over the period 1955-2013 shows a methodical process of marginalization at work in specific regions through rigged statistics. The figures collected by the National Institute of Statistics (Institut National des Statistiques INS) showed that mortality rates, educational attainment, and unemployment rates in certain regions were well below national averages but the INS revised their sampling and data collection strategies to deliberately reflect a different situation. This article is not the appropriate place to explore such claims but existing literature supports a similar narrative about the use of statistics to manufacture Tunisia as a growing national economy because of exemplary reforms implemented under Ben Ali’s rule (see Hibou 2006a, 2006b, 2012). The third initiative is what panelists referred to as preserving memory (hofth al thakira). Its aim is two-fold; first to re-write the history of these regions from the periphery, to displace a history written by the center, as way of documenting the myriad injustices they have been subjected to, and second to break the cycle of intergenerational reproduction of marginalization in its various forms.

25Why document the language activists use to describe and discuss regional inequalities in the context of Transitional Justice? Why stop at the subtleties of translating the Arabic word jiha into region? And why provide an account of a vision of reparations to mend the injustices of territorial marginalization in an article about decentralization? I want to suggest that what is at stake here is a politicized understanding of decentralization that ultimately reinstitutes regional development at the heart of its prerogatives. In the language used by civil society organizations, with its subtleties and the vision it conveys is a notion of politics that redefines what development models are appropriate for a post-revolutionary moment. As Wendy Brown reminds us “[n]aming, of course, is never innocent” (210, 684) since whoever gets to designate and qualify a certain situation as a problem ultimately proceeds to identify and put forward the solutions to rectify it. The solutions put forward are path dependent, in the sense that they are predicated on the ways the problem was qualified and framed in the first place. To rehearse a point I made earlier, when the World Bank presents an urban centric model of “concentrated growth” as inevitable, regional inequalities become simply a matter of externalities that should be addressed to achieve “inclusive development”. The role of local governments is restricted to the management of these externalities arising from this inevitable growth model. In other words, local governments become technical actors among a multitude of others who are striving to “smoothen a bumpy economic geography” (World Bank, 2009, p. 62). The endpoint of this political model is an equilibrium that emerges out of the converging rational interests of all actors. Reasoning along these lines calls forth a language of lagging rather than marginalized or victimized regions to evacuate any historically constituted relations of power between center and periphery that have created the current development landscape. Put simply, this is a language that eschews politics as disagreement and contestation, it is the technical language of the post-political (Mouffe, 2013).

26Although Wendy Brown (2005) discusses the role of critical theorists rather than that of activists and community organizers, she provides insights about the importance of this language that I suggest we understand as critique. When Tunisian activists use the language of deliberate marginalization, they are critiquing the fact that lagging regions are naturally existing geographies of a given territory. The author reflects on the etymology of the word – critique which originally comes from the Greek word krisis in its juridical meaning. She writes “[p]rocedurally, juridical krisis thus consisted of recognizing an objective crisis and convening subjective critics who then passed a critical judgement and provided a formula for restorative action […] krisis thus comes close to what we would today call deliberation, and its connotations are quite removed from either negativity or scholasticism” (Brown 2005, 7). Although the current disappointment with the empty promises of the revolution could be characterized as a moment of crisis, I am not suggesting that these activists are in a process of “deliberating” with the World Bank or the central government for that matter about the most appropriate development model. Deliberation requires a level playing field and power differentials in this case simply prohibit it. In all cases, Wendy Brown is concerned with the role of critical theorists and not activists.

  • 16 . On the top-down process of political decentralization, see the article of Eric Gobe in this issue (...)

27But I bring up this notion of critique to highlight the importance of language, and the possibility it opens up for an adversarial mode of politics that does not evacuate local governments from their political role, and simply renders them managers of a pre-determined spatial concentration of wealth. In that sense, focusing on what we talk about when we talk about decentralization is an avenue to rethink how a particular register re-politicizes decentralization from the margins, imbuing it with a strong orientation towards regional development especially as Tunisia embarks of the creation of municipalities across the entirety of its national territory. This is no easy task given that political decentralization (municiplaisaiton) was a top-down process16: the central government decided on the sequencing of political decentralization chronologically and procedurally including which existing jurisdictions will be divided into more than one municipality, which jurisdictions will be amalgamated and which jurisdictions will be created from scratch. The central government will also be responsible for the initial allocation of resources to set-up these municipalities. In the current post-revolution context, these practices will not go uncontested, especially that they are a reminder of a not so distant autocratic past of manipulating Tunisia’s territorial organization as mentioned in the introduction to this article. A political decentralization process which does not include the peripheral voices of those who have self-identified as dwellers of “victimized regions” will give birth to politically atrophied municipalities, shiny buildings in otherwise desolate and impoverished territories.


28To understand the role of decentralization policies in post-revolution contexts, scholars must go beyond the incentives of politicians to decentralize, or the effects of decentralization policies per se. They must pay serious attention to the emerging language that frames decentralization. In post-revolution Tunisia, activists and international financial organizations talk differently about decentralization. Among activists, decentralization talk is wedded to the question of regional development and the fate of marginalized regions, and demands reparations to acknowledge and mend past territorial injustices. The World Bank too ties decentralization to the question of territorial inequalities. However, compared to the language of local activists, World Bank reports use a post-political register of lagging vs. leading regions in a globalized world where spatially concentrated growth is inevitable, but development can be inclusive. At stake in these different registers are competing models not only of development, but also of politics that intersect at the level of decentralization policies. It should not be surprising then to note that the model of politics local activists evoke in their language is almost antithetical to that proposed in decentralization policies promoted by IFIs. To borrow a characterization from Nitzan and Bichler, activists are using a language that critiques “a liberal theory of politics […] where the ideal political system is one that intervenes least, and the best way to guarantee minimal intervention it to make politics itself operate as a free market” (2009, 27). In doing so, they don’t point to decentralization as the political, financial and administrative autonomy from the central government, but rather call for a different kind of state presence, one incarnated in a development model that rescripts the territorial relation of the center to its peripheries.

Top of page


Bardhan Pranab, 2002, “Decentralization of Governance and Development,” The Journal of Economic Perspectives 16 (4), p. 185–205.

Belhedi Amor, 1989, “Le Découpage Administratif En Tunisie,” Revue de Géographie Du Maroc 13 (2), p. 3–25.

Bellamine Yassine, 2015, “Gouvernance Locale : Quand La Banque Mondiale S’immisce Dans La Gestion de Nos Municipalités !,” Nawaat, February 3.

Boex Jameson, 2011, “Democratization in Egypt: The Potential Role of Decentralization,” Urban Institute Policy Brief.

Bono Irene, Meddeb Hamza, Hibou Beatrice, 2015, L’État d’Injustice au Maghreb. Maroc et Tunisie, Paris, Karthala.

Brown Wendy, 2009, Edgework: Critical Essays on Knowledge and Politics, Princeton, Princeton University Press.

Brown Wendy, 2010, “Political Theory Is Not a Luxury: A Response to Timothy Kaufman-Osborn’s ‘Political Theory as a Profession,’” Political Research Quarterly 63 (3), p. 680–685.

Catusse Myriam, Zaki Lamia, 2009, “Gestion communale et clientélisme moral au Maroc : les politiques du Parti de la justice et du développement,” Critique internationale n° 42 (1), p. 73–91.

Dafflon Bernard, Gilbert Guy, 2013, “L’économie politique et institutionnelle de la décentralisation en Tunisie: État des lieux,” Université de Fribourg Suisse, École Normale Supérieure Cachan France, Banque Mondiale Washington. .

Eaton Kent, 2006, “The Downside of Decentralization: Armed Clientelism in Colombia,” Security Studies 15 (4), p. 533-562.

Faguet Jean-Paul, 2014, “Decentralization and Governance,” World Development, Decentralization and Governance, 53 (January), p. 2-13. doi:10.1016/j.worlddev.2013.01.002.

Falleti Tulia G., 2005, “A Sequential Theory of Decentralization: Latin American Cases in Comparative Perspective,” The American Political Science Review 99 (3), p. 327–346.

Ferguson James, 1994, The Anti-Politics Machine: Development, Depoliticization, and Bureaucratic Power in Lesotho, Minneapolis, University of Minnesota Press.

Forum Tunisien pour les Droits Économiques et Sociaux, 2015, “Demande relative à l’établissement du statut de « Région-Victime » de Kasserine (Art.10, Al. 3 de La Loi Organique 2013- 53),” Report submitted to the Instance de la Vérité et de Dignité, Tunis, Tunisia.

Goldman Michael, 2006, Imperial Nature: The World Bank and Struggles for Social Justice in the Age of Globalization, 59806th edition, New Haven, Conn. ; London, Yale University Press.

Harb Mona, 2009, “La gestion du local par les maires du Hezbollah au Liban,” Critique internationale n° 42 (1), p. 57–72.

Harb Mona, Atallah Sami, eds., 2015, Local Governments and Public Goods: Assessing Decentralization in the Arab World, 1st edition, Beirut, Lebanon, The Lebanese Center for Policy Studies.

Hart Gillian, 2001, “Development Critiques in the 1990s: Culs de Sacand Promising Paths,” Progress in Human Geography 25 (4), p. 649–658.

Hart Gillian, 2010, “D/Developments after the Meltdown,” Antipode 41 (January), p. 117–141.

Hibou Beatrice, 2006a, La force de l’obéissance; économie politique de la répression en tunisie, Paris, La Découverte.

Hibou Beatrice, 2006b, “Le libéralisme réformiste, ou comment perpétuer l’étatisme tunisien,” L’Économie politique no 32 (4), p. 9–28.

Hibou Beatrice, 2012, “Macroéconomie et domination politique en Tunisie : du « miracle économique » benaliste aux enjeux socio-économiques du moment révolutionnaire,” Politique africaine, no. 124 (November), p. 127–154.

Kanbur Ravi, Sumner Andy, 2012, “Poor Countries or Poor People? Development Assistance and the New Geography of Global Poverty,” Journal of International Development 24 (6), p. 686–695.

Kraiem Wahid, 2015, “Local Governance in Non-Municipal Areas,” Unpublished background paper for the World Bank Local Governance & Development Program.

Li Tania Murray, 2007, The Will to Improve: Governmentality, Development, and the Practice of Politics, Duke University Press.

Lund Christian, Boone Catherine, 2013, “Introduction: Land Politics in Africa – Constituting Authority over Territory, Property and Persons,” Africa 83 (Special Issue 01), p. 1–13.

Malesky Edmund J., Cuong Viet Nguyen, Tran Anh, 2014, “The Impact of Recentralization on Public Services: A Difference-in-Differences Analysis of the Abolition of Elected Councils in Vietnam,” American Political Science Review 108 (01), p. 144–168.

Massicard Élise, 2009, “L’islamisme turc à l’épreuve du pouvoir municipal,” Critique internationale n° 42 (1), p. 21–38.

Mookherjee Dilip, 2015, “Political Decentralization,” Annual Review of Economics 7 (2), p. 231–249.

Moretti Franco, Pestre Dominique, 2015, “Bankspeak,” New Left Review, II, no. 92, p. 75–99.

Mouffe Chantal, 2005, On the Political, 1st edition, London, New York, Routledge.

Mouffe Chantal, 2013, Agonistics: Thinking The World Politically, 1st edition. London, New York, Verso.

Nitzan Jonathan, Bichler Shimshon, 2009, Capital as Power: A Study of Order and Creorder, 1st edition, London, New York, Routledge.

Signoles Aude, 2009, “L’islamisme municipal, enjeu et garant de la modernisation des pratiques politiques ?,” Critique internationale n° 42 (1), p. 9–19.

Stiglitz Joseph E., 1999, “Knowledge as a Public Good,” In Global Public Goods, edited by Kaul Inge, Isabelle Grunberg, and Marc A. Stern, New York, NY, Oxford University Press, Inc., p. 208–325.

Swyngedouw Erik, 2010, “Apocalypse Forever? Post-Political Populism and the Spectre of Climate Change,” Theory, Culture & Society 27 (2–3), p. 213–32.

The World Bank, 2008, Agriculture for Development, World Development Report 2008, World Bank.

The World Bank, 2009, Reshaping Economic Geography, World Development Report 2009, World Bank.

Treisman Daniel, 2007, The Architecture of Government: Rethinking Political Decentralization, Cambridge, Cambridge University Press.

Turki Sami Yassine, and Eric Verdei, 2015, “Chapter 1: Tunisie: La constitution (du printemps) ouvre le débat sur la décentralisation,” in Local Governments and Public Goods: Assessing Decentralization in the Arab World, edited by Mona Harb and Sami Atallah, 1st edition, Beirut, Lebanon: The Lebanese Center for Policy Studies, p. 12–43.

“UCLG Country Profiles: Tunisian Republic.” n.d. United Cities and Local Governments.

Wolfensohn James D., 1996, “People and Development, Address to the Board of Governors, Washington D.C.,” The World Bank.

Top of page


1 . The content of this article was developed in conversation with Lamia Zaki whose research and creativity have inspired me. I am grateful for participants at the workshop ‘Etat et Territoires du Politique. La décentralisation en débat dans le monde Arabe’ for their engagements with an earlier draft, for two anonymous reviewers who provided helpful feedback as well to Chris Mizes and Ananya Roy for their comments on earlier drafts. This research is part of a larger dissertation project on experiences of the state at the local level in post-revolution Tunisia. Fieldwork for this ongoing research is funded by the Zeit-Stiftung Trajectories of Change Research Program as well as the Center for Middle Eastern Studies, Institute of International Studies and Global Metropolitan Studies at the University of California Berkeley.

2 . On the experiences of JDP in local governance, see the article of Coline Mias in this issue addressing the case of Salé-Tabriquet.

3 .

4 .

5 . On the role of Special Delegations, see the article of Sami Yassine Turki and Chiara Loschi in this issue.

6 . See Wahid Kraiem 2015, p. 8-10 for more details about the definition of non-municipal areas stipulated in the 1989 amendments of OL33-75

7 . After the appointment of Jim Yong Kim in 2012, the World Bank Group has gone through a restructuring process away from the geography-based organization that characterized it. See for example a Financial Times article on the topic ( This latest restructuring provides further evidence to the point made earlier that the World Bank Group as an institution is constantly reinventing itself to remain relevant in a constantly shift global economy.

8 . “The International Development Association (IDA) is one of the largest sources of assistance for the world’s 77 poorest countries, 39 of which are in Africa, and is the single largest source of donor funds for basic services in these countries. IDA lends money on concessional terms. This means that IDA credits have a zero or very low interest charge […] IDA complements the World Bank’s original lending arm – the International Bank for Reconstruction and Development (IBRD) which was established to function as a self-sustaining business and provides loans and advice to middle-income and credit-worthy poor countries.” see

9 . The 16 completed Urbanization Reviews can be consulted online here .

10 . The FTDES is a Tunisian non-governmental organization created in 2011 with the objective of defending basic economic and social rights of peoples at the national and international levels. It aims to: (i) reinforce the role of civil society in the elaboration of alternatives of economic development at the national and regional levels, (ii) reconsider the role of the state in the redistribution policies and investment especially in providing affordable and quality public services across the territory, (iii) to contribute to social and economic development within institution, (iv) to pay particular attention to the social security system to improve its coverage and services, and (v) to fight corruption, protect public goods, and ensure the sound management of public finances

11 .

12 . ASF is an international NGO created in 1992 with the objective of putting in place institutions and mechanisms that enable access to an impartial and just adjudication, capable of ensuring juridical security and guaranteeing the protection of effectivity of basic rights .

13 . The IVD is an independent state institution with legal personality and financial and administrative autonomy. It was established by Organic Law No. 2013-53 of 24 December 2013 related to transitional justice. The aim of the IVD is to dismantle the authoritarian system and to facilitate the transition to the rule of law by revealing the truth about past violations, determining the State’s responsibility in these violations, asking the perpetrators of these violations to admit to their acts, to restoring victims to their rights and dignity, preserving their memory and facilitating national reconciliation .

14 . .

15 . The aforementioned references are available in French and Arabic languages only. All
translations to English are mine.

16 . On the top-down process of political decentralization, see the article of Eric Gobe in this issue.

Top of page


Bibliographical reference

Lana Salman, « What we talk about when we talk about decentralization? Insights from post-revolution Tunisia », L’Année du Maghreb, 16 | 2017, 91-108.

Electronic reference

Lana Salman, « What we talk about when we talk about decentralization? Insights from post-revolution Tunisia », L’Année du Maghreb [Online], 16 | 2017, Online since 05 July 2017, connection on 21 September 2017. URL : ; DOI : 10.4000/anneemaghreb.2975

Top of page


Lana Salman

Department of City and Regional Planning, University of California at

Top of page