Enhancing urban autonomy: Towards a new political project for cities

Abstract

Chinese

As the 21st Century world assumes an increasingly urban landscape, the question of how definitive urban spaces are to be governed intensifies. At the heart of this debate lies a question about the degree and type of autonomy that towns and cities might have in shaping their economic, environmental, social and cultural geography. This paper aims to examine this question. Starting with the premise that the degree of autonomy any particular town or city has is inherently an empirical question – one which can only be conceptualised in relational terms vis-à-vis the distributed, networked and territorialised responsibilities and powers of the city and the nation-state and other zones of connection – we examine four different contexts where debates over autonomy have intensified in recent history (Brazil, UK, India and South Africa). Drawing on recent respective histories, we identify key elements and enablers in the making of urban autonomy: a characteristic that exists in a variety of guises and forms and creates a patchwork landscape of differentially powerful fragments. We reveal how, beyond its characteristic as a political ideal, autonomy surfaces as a practice that emerges from within specific sectors of particular societies and through their relationship with national and regional politics. Four alternative forms of urban autonomy are delineated: fragmented, coerced (or enclave), distributed and networked. We contend that the spatial templates for autonomy are not predetermined but can be enhanced in multiple different sites and forms of political space within the city. This enhancement appears essential for the integration and strengthening of capacities for sustainable and just forms of development throughout the urban.

Keywords

autonomy localism political authority scalar governance self-determination

Introduction

What is autonomy and what is its potential in relation to the city? Cities are increasingly being positioned as essential to tackling some of the world’s major challenges, from global environmental issues to economic development and political security. Yet the extent to which cities have the capacity to respond is contested. The capacities of such cities to foster wellbeing, sustainability and justice are intimately related to ways of understanding and practising autonomy. Cities today are variously understood as having either too little autonomy, in their capacities to advance their own development models, address urban inequalities or develop low carbon and sustainable modes of living, or too much – via, for example, financial sectors in London and New York acting beyond regulation. If the level of autonomy that a city has from the national state is often a highly complex and politicised question, the question of how to understand contemporary city autonomy is all the more vexing. Is it a form of autonomy from political regulation by the national state? From the constricted options available in powerfully competitive global capitalist economies? Or from the environmental and other regulations imposed by transnational agreements? As urbanism goes global, at a time when the very idea of the city is being called into question (e.g. Amin, 2013; Brenner and Schmid, 2015; Jacobs, 2012), it may seem odd – outdated, even – to talk of ‘urban autonomy’. Yet, cities remain epistemologically specific places, with specific histories and livelihoods closely linked to the performance and governance of a particular territory, even as that territory is relationally produced (McCann and Ward, 2011). Rather than a redundant category, cities are shaped by – and are central forces reshaping – both global economy and society in ways that complicate autonomy.

In this article we seek to advance our understanding of the role of cities in responding to contemporary challenges by examining the concept of urban autonomy. Initially conceived as part of the Friends of the Earth Big Ideas Change the World project,¹ we undertook to probe the question of how and why urban autonomy can be developed in order to provide the basis for cities to take a role in achieving transformational change for sustainable development. Too often cities are regarded as either subsumed by the powers of national government or as having the free reign to ‘act local’ while thinking global. Neither position has stood the test of time. We argue in favour of a form of ‘enhanced autonomy’ as the basis for improving the social and environmental potential of cities. Following the broad approach of the Big Ideas Change the World project, our collaboration with Friends of the Earth involved drawing on our own fieldwork as well as secondary literature and seeking to learn from a diverse and comparative range of examples of urban autonomy taking place in different contexts and scales, and in relation to a wide array of social, economic and environmental concerns (see Bulkeley et al., 2013). In this article, we further develop these cases and the ideas presented in this policy arena. We develop and extend a set of conceptual arguments concerning the nature and potential of urban autonomy and relate these debates to those concerning the contemporary nature of urban politics. Through this work, our intention is to contribute to the academic debate concerning the nature of contemporary urbanism. Particularly in a context where there is growing momentum to the argument that we are witnessing a new era of planetary urbanism, understanding the political particularities of the urban seems all the more pressing.

We do this by considering four different contexts where debates over urban autonomy have taken place in recent history (Brazil, UK, India and South Africa). Through the use of a historical perspective, we identify key elements and enablers in the making of urban autonomy: a characteristic that exists in a variety of guises and forms and creates a contested patchwork landscape of differentially powerful fragments. Given our focus on promoting social wellbeing and environmental improvement in the city, our focus is on autonomy in relation to the state. We are concerned with how powerful actors of different sorts operate together in the arrangement of power, often in conflictual ways, to address challenges at the city level. There is, of course, another tradition of thinking and practising urban autonomy as a political project away from the state (e.g. traditions of autonomy linked to autonomist Marxist movements that take us away from urban democracies as they are currently constructed, as well as forms of contemporary struggle against neoliberalism such as the Occupy movement; see Katsiaficas, 2006; Merrifield, 2013; Pickerill and Chatterton, 2006; Vasudevan, 2014). While we touch on this disparate tradition at different points in the discussion – in relation to urban social movements – it is not our focus here.

A brief word on the comparative approach we take in the paper. Our discussion examines a number of cities across the Global North–South divide and at different points in time, including Delhi, Mumbai, Durban, Porto Alegre, and various contexts in the UK. Our purpose in doing so is to consider how autonomy operates across distinct political, economic, cultural and historic contexts, with often-different challenges, and to locate general principles that work across urban space globally. The comparative discussion reveals how autonomy is not just a multifaceted process, but how it is differently enabled or inhibited by issues of territory, institutional capacity and political context. Using comparison as a learning strategy (McFarlane, 2011; Robinson, 2016), we show that seeking to enhance autonomy through one dimension or element at a time – be that financial, political or administrative – is unlikely to be successful and may have other unintended consequences.

We also identify four specific forms of autonomy through our comparison: distributed autonomy, where autonomy is purposefully designed as a form of political power sharing; networked autonomy, which is based on the functioning of autonomous units within the city, such as social movements, working in tandem and linking ideas together; fragmented autonomy, based on the presence of autonomous units and processes, but operating in chaotic and fragmented ways, and where objectives and processes are not joined up; and coerced or ‘enclave’ autonomy, a top-down governmental autonomy with strong elements of hierarchical management. The comparison allows us to show that these four forms are not necessarily specific to particular cities, but that cities experience elements of each of them. The comparison also allows us to examine the challenges and opportunities for enhancing meaningful and effective autonomy in the city and to consider what autonomy has been able to achieve.

Framing urban autonomy

Autonomy is a complex idea, encompassing notions of independence and separation, self-sufficiency and self-control, and stretching to regard and respect. Whilst often seen as an interstice between control and regulation (Dahl, 1982), politically it is perhaps the idea of self-determination which has most caught the imagination (Habermas, 1994). In this conceptualisation, autonomy has positive connotations: of a society or community pursuing a common goal with some level of self-determination. But autonomy has also come to be associated with absolute forms of independent rule and strict containment; of boundaries, strong levels of control, and social conservatism (Paddison, 1983). Such a ‘closed’ form of autonomy sits uncomfortably with our idea of a globalising world. Yet, understood as a moral state of independence in which respect for others’ independence is integral, autonomy is naturally bounded or constrained in the interests of the social good – without such constraints, requiring an exercise of hierarchical power. It therefore holds a promise of alternative forms of social organisation and of a politics that could enable radical transformations.

Cities lie at the heart of contemporary questions of autonomy. They are themselves celebrated spaces of cultural, political and economic autonomy, and yet simultaneously cast as increasingly limited in their (autonomous) powers as they become centrally implicated into the flows of global politics, trade and resources. Critically, cities are not to be seen as ‘self-enclosed political [territories] within a nested hierarchy of geographical arenas contained within each other like so many Russian dolls’ (Brenner et al., 2003: 1). In such guise, the question of autonomy would appear straightforward: how much of what kind of autonomy should be allocated to the city, by whom and under what conditions, in order to achieve certain social goals? Here, authority and autonomy are distributed in a ‘zero sum’ fashion, so that as one political territory (the city) gains autonomy a loss of control or determination occurs at another level of political authority.

Yet within the social sciences, there is a growing recognition that to regard political organisation as a ‘nested system’ is to miss the critical ways in which political order and authority have been reconfigured across and within scales, and through new forms of political networks (Bulkeley, 2005; MacLeod and Goodwin, 1999). Others go so far as to interpret political spaces to be a series of rhizomatic, interrelated and interwoven relational becomings or as a pluralist assemblage of human and non-human forces (cf. Amin and Thrift, 2013; McFarlane, 2011). Cities here are symbiotic politically – and economically and culturally – with other spaces at other scales (Massey, 2007). They are not homogenous, but fractured socially, economically and politically, for instance in overlapping forms of power and shifting configurations of urban governance. So we should not (necessarily) expect to find autonomy readily circumscribed within particular boundaries or organised in hierarchical terms. If, from this perspective, ‘the city’ cannot be understood as a homogenous or contiguous political territory, neither can autonomy be seen as a fully formed political condition that can be simply adjusted through the machinations of urban governance.

Autonomy in the city exists in a multiplicity of forms and through various manifestations, creating a patchwork landscape of differentially powerful fragments of autonomy. By viewing urban autonomy from this starting point, we can avoid any assumption that urban autonomy necessarily means ceding power from the (central) state or that it would automatically entail empowering local communities, however defined. Such assumptions have already been challenged by urban scholars who have identified how a collapse in notions of local democracy and local autonomy leads to a false belief that increases in one inevitably lead to increases in the other (Pratchett, 2004), problematising simplistic understandings of the local as an ideal scale configuring a more democratic city that better serves the needs of citizens (Purcell, 2006). Conditional forms of local autonomy (referred to within the UK as ‘earned’ autonomy) create the illusion of greater local influence and power despite an increase in central control (Coaffee and Headlam, 2008; Wilson, 2003). Taking these viewpoints further, but emphasising the dispersed, fragmented and situated nature of local autonomy, we can examine how and why autonomy matters in debates about urban futures, and explore the ways in which autonomy is being realised or could be enhanced within the contemporary urban condition. We argue that the route forward is not to provide blanket calls for greater local autonomy, but in identifying and enabling a portfolio of forms of autonomy – and associated responsibility – that better serve social and environmental agendas.

This does not exclude possibilities for devolution of political power to localities. Indeed in many countries greater fiscal, administrative and political autonomy at the functional city scale may form a desirable or even necessary part of the transformation of fragmented autonomy into some degree of meaningful and effective autonomy from the vicissitudes of global economic competition and financial markets (Janssen-Jansen and Hutton, 2011; Lefèvre, 2010). But it does suggest – as we shall see in the cases outlined later – that in some important ways meaningful autonomy can neither be granted to nor foisted upon cities or their components. Instead, autonomy is relational, ‘always fractured, partial and ongoing’; ‘a temporary and situated social construct’ (Coppola and Vanolo, 2015: 1152, 1155). Practically, it implies that for city leaders, frustrated about constraints on their freedom to implement their visions for change, enhanced autonomy may be found as much through collaboration with communities or businesses within the city, or through networking with other cities, as it is to be found in renegotiated relationships with national authorities.

Critically, autonomy is neither a one-dimensional property of the organisation of the state (in political, administrative or financial terms) nor an attribute that can be readily conferred on a particular territory or form of society; rather, autonomy is a multifaceted political project, achieved relationally and as such subject to political change. But what of the state, long an object of analysis vis-à-vis its own ‘relative autonomy’ from the spectre of economic determination (Jessop, 1990) and increasingly acknowledged to be ‘an important component in the urban development landscape’ (Parnell and Robinson, 2012: 593)? Far from acting as some all-powerful monolith or Leviathan, in substantive terms the state is an institutional ensemble; any power attributed to or distributed through the state only constitutes the power of particular agents and forces which are assembled into its purview (a representational regime), and thereby enabled to act within and through the very institutional ensemble that is the state (Jessop, 2016). It is within such a context that we can envisage how, rather than being devoid of autonomy, cities contain multiple possibilities for autonomy. The particular form and institutional and social base of any urban state can enable a range of forms of autonomy: we have characterised four that appear to be prevalent within contemporary urban conditions – fragmented, coerced (or enclave), distributed and networked. While cities may experience one or more of these forms of autonomy at any one time, they are very different in the extent to which they support the social and environmental outcomes sought from enhanced, effective and meaningful autonomy. Access to the representational regime can enable critical interventions from those with the vision to perhaps transform fragmented and coerced autonomy into distributed and networked forms in order to deliver enhanced social and environmental capacity on the ground.

Urban autonomy in historical perspective: Identifying elements and enablers

The idea of autonomy is central to the historical development of the city. Autonomy is not only a political idea or set of properties to be implemented, but a practice that emerges from and comes to form part of particular societies. In this section, we consider two different historic contexts – South Africa and Brazil – to illustrate how autonomy is multifaceted and enabled or inhibited by issues of territory, institutional capacity, and political context. This suggests that seeking to enhance autonomy through one dimension or element at a time – be that financial, political or administrative – is unlikely to be successful and may have other unintended consequences.

South Africa: Opportunities and challenges for Local Agenda 21 in Durban

1994 marked the beginning of a post-apartheid era for South Africa. Faced with the need to rapidly democratise society and reinvent government, South Africa embraced pre-existing international frameworks as potential guidelines for future action (Sowman, 2002). Local Agenda 21 (LA21), initiated in Durban in 1994, became more than a tool for the promotion of local autonomy towards environmental action: it was seen as a possibility to incorporate greater local participation towards the identification of priorities for developing strategies and projects (Rossouw and Wiseman, 2004). In the context of the significant transformations of the post-apartheid years, LA21 provided a window of opportunity for the adoption of new local management concepts and practices, including novel forms of understanding urban autonomy. Its success was largely due to the promotion of social and political agendas above and beyond environmental agendas, where community empowerment, health and environmental quality were key drivers in response to jobs and service provision concerns (O’Riordan, 1998).

However, rolling out LA21 in Durban was a learning process marked by challenges (Roberts and Diederichs, 2002), from limited political will and support and limited resources and capacity mismatch to shortcomings on financial autonomy. LA21 was largely seen as a ‘green’ and foreign agenda concerned primarily with environmental issues. In South Africa, given a history of environmental concerns marked by exclusionary strategies for ecosystem conservation, environmental policy issues were often seen ‘as tools for racially based oppression’ (Rossouw and Wiseman, 2004: 131). Resource limitations were both human and financial, impacting the ability of the programme to build its required broad consensus. Capacity building took the shape of environmental education initiatives at different levels, which helped deliver public engagement and support while also broadening the range of political agendas included. Dependency on external donor funding limited the capacity of the municipality to direct investments towards local priorities over those of funders. Finally, there were changes in the city’s territorial configuration created by shifts in administrative boundaries resulting from the transition to democracy. Durban went from 300 km² before 1996 to 2297 km² post-2000, requiring the LA21 process to adjust its operations and expectations to a more functional understanding of the urban area. The challenges experienced highlight the tensions resulting from the deployment of local autonomy initiatives under limited conditions for their realisation. These were characterised by overlapping competences and fragmented environmental functions within and across different scales of government (Rossouw and Wiseman, 2004; Sowman and Brown, 2006). South Africa embraced the principles of LA21 via local environmental programmes, but with a marked absence of ‘mechanisms for translating these policies into practice’ (Sowman, 2002: 185). Despite efforts to the contrary, local environmental issues were incorporated within local and national hierarchical structures primarily in ways that give strategic concerns precedence over local issues (Sowman, 2002).

A broader critical analysis of the LA21 process points to the risks and limitations associated with the institutionalisation of the local as ‘the key scale for environmental action’ (Lawhon and Patel, 2013: 1048). This implies questioning the now common LA21 principle of ‘thinking globally, acting locally’. It demands that we ask what and whose concerns are highlighted and obscured as ‘the local’ becomes the primary space for developing solutions of global reach, and how such emphasis responds to pressing questions on international responsibility and ethics (Lawhon and Patel, 2013). Viewing the local as the privileged site for environmental solutions often overlooks conflicting interests operating at regional, national and international levels. It also fails to account for the wider organisational and political contexts within which such local agendas are necessarily adopted and developed (Marvin and Guy, 1997). Sustainability is somehow to be achieved in tandem with the development of market economies and global competitiveness, despite unanswered questions regarding the compatibility of these two aims (Krueger and Gibbs, 2007). In reality, local environmental initiatives constantly find themselves at odds with the mandates of trade agreements and policy mechanisms operating at national and global levels (Gibbs, 1999). This exemplifies the extent to which the development of urban autonomy via tools such as LA21 needs to ‘make the necessary connections to regulatory processes operating at a variety of spatial scales’ (Gibbs and Jonas, 2000: 2999). Their limited results are as related to the type of challenges and limitations experienced by Durban as much as to the unchallenged ‘acceptance of the merits of the local framing [and] the evasion of questions that this framing silences, particularly questions of responsibility and justice at various scales’ (Lawhon and Patel, 2013: 1048).

However, this analysis does not so much dismiss arguments in favour of a local dimension to sustainable development, as highlight a major obstacle to its likely effectiveness. Such experiences suggest that LA21 initiatives establish neither local freedom, nor responsibility and power. The potential benefits of local environmental and social initiatives for experimentation, innovation and public involvement remain unrealised. The implications are twofold. First, interventions to establish autonomy in single dimensions (such as environmental practice) are doomed to fail – cities need autonomy to negotiate new, less uniform and less constrained relationships with the global economy, rather than greater autonomy to try to be competitive within it as it is currently constituted. Second, ‘local autonomy’ itself emerges as a misleading term, and instead we need to connect power and accountability in multiple dimensions such that the city – constituted by its unique place in these relationships – can become a focal point for the development of livelihoods and lifestyles that deliver and promote wellbeing for all.

Brazil: The political framing of participatory budgeting in Porto Alegre

Some of the most promising experiments with urban autonomy have taken place in Brazil. With the collapse of the military dictatorship and the growth and eventual success of the Partido dos Trabalhadores (PT, the Workers Party) in national elections came decentralisation of power to local levels. A key urban success area for the PT has been in participatory budgeting, with the most notable example being Porto Alegre, where there has been mass participation, elements of redistribution and a balanced budget. Citizens decide and deliberate upon a variety of municipal policies, the cornerstone of which is the much-publicised Orçamento Participativo (Participatory Budget), a neighbourhood-based set of deliberative forums on the city’s budget priorities.

Scholars have outlined various characteristics supporting the success of this initiative (Baiocchi, 2001; Sintomer et al., 2008): clear procedures for organisation, representation and participation; engagement of both individual and collective actors, such that participation is two-tiered involving both individuals and community organisations; provision of spaces for debate, information distribution and activism (with participatory forums functioning as spaces for airing local demands and problems and for sharing information about government functioning); and the provision of opportunities for people to learn about content-related issues (e.g. budgets) as well as the technical criteria involved in budgeting and its ramifications for planning urban operations. Once people become involved in participatory budgeting, they are more likely to take part in participatory forums in other sectors, including education, health, infrastructure services and sports facilities, amongst others (Abers, 2000; Baiocchi, 2001). Success was aided by a combination of different modes of democracy: participatory (e.g. regular regional debates and learning initiatives) and representative (e.g. through the Municipal Council of the Budget).

The process resulted in ‘a reversal of priorities: primary health care was set up in the living areas of the poor, the number of schools and nursery schools was extended, and in the meantime the streets were asphalted and most of the households have access to water supply and waste water systems’ (Sintomer et al., 2008: 166–167). However, none of this means, of course, that the views of the poor and the better-off register in equal measure, and there is no reason to believe that participatory forums somehow undo existing inequalities: indeed unequal access to service provision was a high-level stake in protests across many Brazilian cities before the 2014 World Cup. And the success of urban participatory forums in Porto Alegre is, to be sure, partly a function of the city’s relative wealth compared with other cities in Brazil. Yet the experience of Porto Alegre suggests that successful autonomy in urban governance requires political will, social demand, an innovative idea and a sense of genuine meaningful group participation, and emerges from a process of negotiation between the state and society. It is, moreover, not a question of more or less state involvement.

The South African and Brazilian examples contain key lessons for thinking about urban autonomy, such as the need to attend to the realistic potential of formal devolution of powers when set against economic constraints and imperatives. Cities need autonomy that extends beyond just one sphere (environmental, political or economic) if autonomy as a principle is to succeed; otherwise, they will be unable to assert their own issues and will be corralled by the structures of competition within the global economy. Successful autonomy requires political will, but also more than this: social demand, innovative ideas that energise people, and a sense of genuine rather than tokenistic participation (combining, for example, forms of representative and participatory democracy). ‘Local autonomy’ itself is therefore a misleading term; the challenge is more accurately to link power and accountability in multiple dimensions and geographies that allow the city to exert as much capacity as it can to promote the wellbeing of its inhabitants.

Realising autonomy: Learning from urban realities

The history of urban politics has been shaped by the need to determine its autonomy in relation to other levels of the state (Cox, 1993). But rather than being entirely subsumed by other levels of the state, or the global economy, the city is rich with possibilities for greater autonomy. At any given moment, it contains a multiplicity of sites where different forms of autonomy can be realised. Everyday life is made, in large part, through activities that are locally or self-organised. Communities and municipalities work to develop and deploy different forms of autonomy and self-determination, from the organisation of social services to efforts to address environmental challenges. The extent and limitations of urban autonomy are subject to continual renegotiation between urban authorities, nation-states, international agencies and economically powerful actors, creating new possibilities for greater autonomy. Yet not all forms of autonomy are the same. Some are fleeting, others embedded, some disempowering and others a force for change.

Building on the analysis developed in the previous section, here we argue for the need to recognise four alternative forms of urban autonomy – fragmented, coerced (or enclave), distributed and networked – that could help provide the architecture for building alternative social and environmental futures. Cities normally experience elements of each of these four forms of autonomy. By focusing on four different urban contexts (London, Mumbai, Delhi and the contemporary localism initiatives promoted by UK governments), this section explores how these different understandings of autonomy are currently being played out. From this starting point, we examine the challenges and opportunities for enhancing meaningful and effective autonomy in the contemporary city, what autonomy has been able to achieve, and how issues of responsibilities have been addressed in such attempts to fashion autonomy.

London: Creating distributed autonomy towards urban sustainability from a space of partial autonomy?

London, like many metropolitan areas, has a complex governance structure. As part of the Thatcher Government’s culling of local socialism, the Greater London Council was abolished in 1986, and in 2000 a Greater London Authority (GLA) was reinstated alongside 33 borough councils responsible for particular districts in the city. The GLA combines elements of administrative autonomy in relation to planning, political autonomy in the form of a Mayor and an elected body with decision-making powers, and a small degree of financial autonomy achieved through an additional ‘council tax’ payment by London residents. This combination of partial administrative, political and financial autonomy has been critical in the development of an urban sustainability agenda, particularly in terms of transport, energy and climate change. Examples of this are the introduction of the ‘congestion charge’ (2003), investment in alternative modes of transportation, the development of one of the first urban climate adaptation plans, a long-term commitment to reducing GHG emissions by 2050 and a policy goal of 25% of energy needs to be met by decentralised generation (Bulkeley and Schroeder, 2008, 2012). The actions of the Mayor and the GLA take inspiration and learning from the pioneering work of several London borough councils, and rely on the alignment of this agenda with private-sector and non-profit interests. In addition, the engagement of communities and local organisations has been critical both to the development of London’s environmental programmes and their achievements.

There are several examples of how the GLA has taken advantage of forms of distributed autonomy to advance an environmental agenda. For example, many of the GLA’s sustainable energy initiatives draw on the earlier work of the London Borough of Merton and what is known as the Merton Rule: a planning provision which requires new urban developments to incorporate on-site renewable energy generation to meet 20% of its energy requirements. Organisations with varying degrees of autonomy such as London First, the City of London Corporation and the Climate Group have been closely involved in the development of a ‘low carbon transition’ narrative (Bulkeley and Schroeder, 2012; see also Thornley et al., 2005). Particular parts of the city have had a catalytic role as well. In Brixton, the Brixton Transition Town movement (BTT) has been a crucial partner in the delivery of the city’s Low Carbon Zone established by the GLA. BTT operates locally as a community interest company raising awareness on climate change and peak oil issues, promoting a change in behaviours and engaging the community in a low-energy future.

All of these actors draw on their own forms of autonomy for their work, highlighting the links between distributed and networked autonomy in the city. BTT, for example, experiments with innovative forms of community management, including participative design and democratic resource allocation. Although a territorially demarcated community-based organisation and social movement, BTT’s autonomy is not clearly demarcated by territory but rather is fluid and continually renegotiated, and it enjoys some autonomy from mainstream political culture, operating on the principles of ‘direct democracy’ rather than representative democracy. From an administrative perspective, it selects and runs its own projects, though these are often done in partnership with other community groups and local government. Financially, they have established a successful local currency, the Brixton Pound, which has fostered a new local economic space albeit one that is limited in extent and in comparison with wider financial flows in the area.

The case of London illustrates how different degrees of partial autonomy can be combined to leverage significant levels of response to some of the key sustainability and social justice issues facing cities. The case also shows that autonomy does not arise purely from legal, institutional or financial grounds. Rather, it points to the importance of a distributed form of autonomy, garnered at the intersections of multiple organisations. Instead of being conferred on institutions or organisations, autonomy is generated through the relation between them. Recognising autonomy in this way means that we need to also acknowledge that, as a political project, autonomy is usually not a means to achieve separation, but is rather an attempt to move towards plural self-determination or the possibility of engaging other actors on their own terms to work towards common goals.

Mumbai: Building networked autonomy from forced autonomy?

Many informal neighbourhoods in the Global South are forced, as a result of long-entrenched and growing urban inequalities, into a condition of forced autonomy, where a disconnection from some of the city’s networks occurs not by will or desire but rather in the context of exclusion. Yet, sometimes these autonomously organised practices can form templates for the emergence of more formal organisation of different sorts, from social activism and resistance to incorporation (not necessarily co-option) within government programmes. The incremental processes through which residents organise everyday life, for example, can provide a platform for social activism. This is the case of Slum/Shack Dwellers International (SDI), where the learning processes, knowledge and partnerships that emerge from informal practices give rise to forms of urban networked autonomy.

Starting through organising ‘pavement dwellers’ on the streets of central Mumbai in the early 1980s, SDI has grown into a global movement operating in over 20 countries. The premise of the movement is to take the autonomous practices through which people organise everyday life – such as residents’ knowledge attained through incrementally building houses and infrastructures of sanitation and water – and use that knowledge to negotiate with municipal and state authorities. For example, a key strategy in SDI’s work is to build full-size model houses and public toilets and put these on display to government officials. By doing so, the aim – sometimes successful, often not – is to draw officials into discussions that lead to land and resources being provided to the poor so that they can build their own developments (McFarlane, 2011). In this process, the movement shifts the autonomous work of informal urban residents from incrementalism to a ‘radical incrementalism’ (Pieterse, 2008), where ordinary practices become a template for social change and people’s empowerment.

Given SDI’s global reach, the movement has become an example of networked autonomy. The local members of SDI regularly meet to exchange information and ideas on organising. In such way, they produce an informal peer learning network that encourages other groups to use strategies such as self-built house modelling to negotiate with their local authorities. House modelling is one of a set of central ideas that circulate in the movement, producing a fine balance between the ideas that come from SDI leaders at national and international levels and the autonomy of the local SDI member groups interested in working out their own priorities. Nonetheless, this is a tension that does not always work successfully, and some groups – notably in South Africa – have felt their own priorities being marginalised by those of the global network. SDI’s story therefore provides two lessons for thinking about urban autonomy: first, the potential of using people’s existing practices to build more formal autonomous organisations; and, second, the difficult balance of linking autonomous groups through larger networks, and the risk that the larger network might erode that local autonomy through commitment to particular agendas, ideas and directions, sometimes via endeavours to ‘mainstream’ across wider territories. This is not to argue that autonomy should be imposed or forced upon groups or localities, but to illustrate how capacities can be built to convert forced autonomy into something more effective and meaningful.

Delhi: Fragmented autonomy out of neighbourhood autonomy?

Our analysis of autonomy in Delhi explores a different element: the role of neighbourhood associations and the ‘new middle classes’ of the Global South in the make-up of a fragmented autonomous city. The emphasis here is on the political risks associated with an increase in neighbourhood autonomy – and influence – in a city characterised by the emergence of marked class identities (Harriss, 2006). Since 2000 the Delhi government has implemented the Bhagidari programme, which seeks to institutionalise citizen participation through a collaborative form of government aimed at involving the public in problem-solving activities and the management of public assets. The main focus of the programme is the city’s Resident Welfare Associations, or RWAs, essentially neighbourhood management committees in the city’s formally planned neighbourhoods, where members pay regular charges towards security and the maintenance of common resources (Chakrabarti, 2007). The programme has resulted in improved urban services in those areas where it has been implemented, with RWAs actively involved in services such as waste management, neighbourhood security and the maintenance of parks, street lighting and roads (Kundu, 2009). ‘Most of the activities of the Delhi government are now conducted under the rubric of Bhagidari’ (Chakrabarti, 2007: 98). Through umbrella organisations, RWAs have become a powerful force in the city, as exemplified by their successful opposition to increases in electricity prices. However, the Bhagidari programme has been criticised for having an exclusionary character and for creating urban fragmentation, as a result of its focus on elite and middle class neighbourhoods associated to formal land tenure (property) modes, and its exclusion of informal neighbourhoods where inhabitants have no tenure rights.

The exclusionary and fragmented character of this form of autonomy rests on the ambiguity of government interventions in informal settlements: whilst government agencies continue to provide some basic services in slums, such as water, street lighting and pavement, they ‘avoid negotiations that may result in the provision of land titles’ (Chakrabarti, 2007: 99). In such cities land reform is likely to be a necessary condition for effective autonomy. The scope for low income urban dwellers to influence is further constrained as the RWAs, through the Bhagidari programme, have a relatively direct access to city administrators. The RWAs have become an effective mode of local representation, albeit accessible only to specific sectors of society (Ghertner, 2011). The result has been a reduction in the influence of locally elected representatives and an increase in the power of middle and upper classes in determining the priorities of the city, moving away from a pro-poor agenda (Baud and Nainan, 2008). This represents the deepening of class identities as an active driver for urban politics: in a context of historic religious and caste identities – alongside their associated exclusions – and recent economic transformations that have failed to incorporate the needs of the poor, the middle classes are emerging as a strong political actor driving urban agendas through novel forms of autonomy (Fernandes and Heller, 2006).

The case of Delhi highlights the extent to which neighbourhood autonomy is charged with conflicting local politics and identities, and the tensions and contradictions associated with increases in autonomy for certain groups at the expense of others. In this case, ‘programs designed to increase citizen–government partnership can have the surprising effect of making this space shallower and narrower, reducing the avenues of political participation open to the poor’ (Ghertner, 2011: 526).

Coalition localism in the UK: Enclave autonomy?

Shortly after entering power in 2010, the UK’s Conservative–Liberal Democrat coalition government asserted the significance of localism as ‘laying the foundations’ for its vision of a ‘Big Society’. Coalition localism was presented as a counterweight to decades of central government control, and the then Prime Minister David Cameron envisaged this process being enabled through a ‘catalysing state’ designed to ‘unleash community engagement’. Amid a period of savage public spending cuts, Cameron envisaged this move would enable a crucial role for voluntary agencies, social entrepreneurs and community activists to combine effective public service reform with active citizenship (Cameron, 2010). In 2011, the government introduced the Localism Bill, arguably designed to transform central–local relations and ‘strengthen local democracy’ by devolving more power and freedom to councils and neighbourhoods while also revolutionising the planning system in favour of local communities (Townsend, 2010). However, coalition localism can also be interpreted as an expression of top-down governmental hierarchy which, in the context of such austere cuts in public expenditure, forces the creation of enclaves where freedom to act is constrained not by a lack of power but a lack of resources, with increasing levels of responsibility under financial constraint. The localism legislation heralds potentially far-reaching changes in the relationship between central government, local government, communities and individuals and a reconfiguration of the geographies of governmental autonomy, welfare and citizenship across Britain (Clarke and Cochrane, 2013).

The Localism Act (HM Government, 2011) heralded several notable changes in policy and planning. Some of these are, first, ‘new freedoms and flexibility for local government’, including the ‘general power of competence’ to undertake any activities that do not contravene law (Department of Communities and Local Government (CLG), 2011: 1). Second, ‘new rights and powers for communities’, particularly the right for community associations to ‘challenge’ local authorities and ‘buy’ and assume control over and delivery of local services such as libraries and sports centres (CLG, 2011: 1). Third, a reform of the planning system – effectively a rescaling of planning responsibility – based on the abolition of Regional Development Agencies and the encouragement of Neighbourhood Plans designed to offer local communities ‘genuine opportunities’ to influence the future of their places, enabling a community’s right to build as well as deciding where new houses and commercial ventures are to be located (CLG, 2010). Finally, ‘reform to ensure that decisions about housing are taken locally’, whereby social landlords are ‘freed up’ to offer flexible (as opposed to ‘lifetime’) tenures and local authorities are given greater freedom to set their own policies on who qualifies for social housing (CLG, 2011: 1).

This localism approach, amid a zealous fiscal austerity in local government and state finances, might be viewed to operate as a mode of coerced and strictly geographically limited autonomy. Such policy measures actively hamper the development of networked and distributed autonomy with mutual respect between different actors at different scales. Instead, such coerced autonomy can be used to support the replacement of a state-supported menu of community-oriented initiatives by one more markedly invoking (neo-) liberalising ‘freedoms’ for individuals and communities to design and implement their own local ventures (Rose, 1999), with all the widening social inequalities such a transition risks. It marks a moment where ‘civil society actors begin increasingly to inherit public policy-making and service delivery functions in the past exercised largely by government’ (Deas, 2013: 65). While this could in principle offer some scope for local democracy, it may also become a smokescreen for government to off-load responsibilities onto communities (Purcell, 2006), without the financial or other capacities needed to deliver them fairly. Such coerced autonomy could compromise nationwide scope for service delivery and social citizenship, and perhaps even see services and major elements of the built environment privatised in favour of powerful economic and corporate interests (Featherstone et al., 2012; Painter et al., 2011).

These brief examples from the UK and India show that autonomy cannot be forced, nor thoughtlessly devolved onto already unequal communities without risking the exclusion of new groups from city spaces and city economies. They illustrate risks and potentials in the relationship between city authorities and already existing forms of local and networked autonomy taking place within and between cities. Alongside the other examples examined, they illustrate how autonomy, beyond a political ideal, is a practice that emerges from within specific sectors of particular societies and through their relationship with national and regional politics.

Conclusion: Renewing autonomy?

Whilst ideas of autonomy travel easily from site to site and country to country, the specific environment where such autonomy is to be realised plays a key role in determining what it can really achieve. Autonomy, usually considered in terms of its financial, political or administrative dimensions, raises a multiplicity of issues at stake. As illustrated by the examples used in this paper, geographical and historical context are critical, and notions of autonomy that have been successful in certain countries at the time of particular political regimes can fail when implemented elsewhere. National and regional politics do not just sit in the background; they are the actual arenas and sites of everyday practice around and through which possibilities for local autonomy open or close. The nation state, as a site of struggle where different interests compete, has historically played a primary role in promoting or limiting autonomy (Jessop, 2016). Where urban autonomy emerges, this is often the result of a potent combination of political will and public demand. In turn, autonomy creates new kinds of politics.

We identified four alternative forms of urban autonomy, arguing that the spatial templates for autonomy are not predetermined, but can be enhanced in multiple different sites and forms of political space within the city. The first is distributed autonomy, where autonomy is purposefully designed as a form of political power sharing. This can enable alternative forms of political expressions and action, and is sometimes managed hierarchically. Second, networked autonomy, which is based on the functioning of autonomous units within the city, such as social movements, working in tandem and linking ideas together. The third is fragmented autonomy, based on the presence of autonomous units and processes, but operating in chaotic and fragmented ways, and where objectives and processes are not joined up. Finally coerced or ‘enclave’ autonomy, a top-down governmental autonomy with strong elements of hierarchical management. We argue that distributed and networked autonomy cannot function without enhanced participation for all, and that urban autonomy itself is a generative and relational process made through the autonomy of different institutions in and beyond the city, within the context of unequal and often changing power and social relations. Enhanced autonomy, transcending forms of ‘fragmented’ and ‘coerced autonomy’ – especially in the economic realm – appears essential for the integration and strengthening of capacities for sustainable and just development of and in our cities.

Forms of forced autonomy are common in many cities, as illustrated by the examples of Mumbai, Delhi and localism in the UK; and although these can provide the basis for new forms of social organisation and networked autonomy that might provide empowerment across different urban contexts, they also risk creating new forms of social exclusion, and can also be seen as a result of asymmetries of power and influence, or more meaningfully, where those charged with shaping the institutional ensemble of the state engage in distributing resources selectively, unfairly and often in the face of democracy. Challenging these – and the interests such inequalities serve – is a critical task for a new political project of enhanced autonomy. This is all the more important as autonomy can be a means through which existing inequalities are maintained (via fragmented autonomy) or where new forms of hierarchical control are imposed on communities who have little capacity to respond to the challenges they face (via coerced autonomy); this reminds us that autonomy is not necessarily progressive, and that it matters a great deal who gets to participate in the political project and who is excluded.

Using comparison as a learning strategy, the examples in this paper suggest that effective distributed autonomy cannot be achieved through coercion, but this does not preclude efforts to enable and incentivise it through careful policy design. Nor can genuine autonomy even be simply ‘granted’ – it must be asserted and negotiated by those who would wield it with responsibility. Distributed autonomy can provide a means through which to generate new spaces for politics in which environmental and social justice goals can be pursued; but raise challenges of co-ordination, co-option and the remaining presence of hierarchy – in London, Delhi and Mumbai for example, new forms of autonomy are being generated at the intersections of the partial distributed autonomy and social networks present in the city, which could provide a means through which to leverage new spaces for politics in the city. A vital challenge here is responding to the fact the city is already fragmented. Consider, for instance, the powerful inequalities in housing in all three cities with the competing agendas of social justice and vested real estate. Any attempt to incentivise autonomy in such a context will always be up against the challenges established by the conflicting positions of the city’s different actors, and the socio-economic positions that they inherit and are forced to confront. The associated development and redevelopment of urban space, with ever more enclosure and privatisation of previously public or common spaces, as documented by Harvey (2012), responds directly to the interests of financial capital, thinly shrouded in the notion of ‘competitiveness in the modern knowledge economy’. Such economic relational autonomy is a powerful influence – perhaps the dominant one in many cities – but it is not what we consider meaningful or responsible autonomy.

Yet, the challenges facing declining industrial conurbations and growing cities in developing countries are very different. How can distributed autonomy provide – if not ‘autonomy from’ then ‘insulation against’ – the negative impacts of global competition or of limited institutional capacities? In practice, the pursuit of autonomy – however conceived – will remain shaped by financial and institutional capacity, contestation of the scope, function and boundaries of autonomy, and the nature of public participation. To strengthen and knit together forms of enhanced autonomy will take a mix of actions, in some respects inevitably specific. In each state, and potentially each city, different (state and non-state) bodies will need different new powers and particularly new capacities. To focus on enhanced fiscal, political and administrative powers for city authorities misses the importance of other capacities and other actors. The examples from London, Durban, Porto Alegre and Mumbai suggest that NGOs, community organisations, social processes and city authorities all need enhanced capacities to collaborate in creating new political spaces, as well as greater financial autonomy from the centre. Inevitably autonomy, as a political project, must involve significant challenges of addressing social and environmental justice, both internally and in relation to wider global concerns.

Footnotes

Acknowledgements

The authors would like to acknowledge Duncan MacLaren, former Chief Executive of Friends of the Earth Scotland, for his input into the development of the ideas and argument here and Mike Child, Friends of the Earth, for initiating and supporting the collaboration under the Big Ideas project. They are also grateful to the three Urban Studies referees for helpful and instructive comments.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Abers

(2000) Inventing Local Democracy: Grassroots Politics in Brazil. Boulder, CO and London: Lynne Rienner Publishers.

Amin

(2013) The urban condition: A challenge to social science. Public Culture 25(2–70): 201–208.

Amin

Thrift

(2013) Arts of the political: New openings for the left. Durham, NC: Duke University Press.

Baiocchi

(2001) The Porto Alegre experiment and deliberative democratic theory. Politics & Society 29: 43–72.

Baud

Nainan

(2008) ‘Negotiated spaces’ for representation in Mumbai: Ward committees, advanced locality management and the politics of middle-class activism. Environment and Urbanization 20: 483–499.

Brenner

Schmid

(2015) Towards a new epistemology of the urban? City: Analysis of Urban Trends, Culture, Theory, Policy, Action 19(2–3): 151–182.

Brenner

Jessop

Jones

et al . (2003) State/Space. Malden, MA: Blackwell.

Bulkeley

(2005) Reconfiguring environmental governance: Towards a politics of scales and networks. Political Geography 24: 875–902.

Bulkeley

Schroeder

(2008) Governing Climate Change Post-2012: The Role of Global Cities – London. Tyndal Centre Working Paper 123. Norwich: Tyndall Center for Climate Change Research. Available at: http://www.tyndall.ac.uk/content/governing-climate-change-post-2012-role-global-cities-london (accessed 20 August 2016).

10.

Bulkeley

Schroeder

(2012) Beyond state/non-state divides: Global cities and the governing of climate change. European Journal of International Relations 18: 743–766.

11.

Bulkeley

Luque

McFarlane

et al . (2013) Enhancing Urban Autonomy: Towards a new political project for cities. Discussion paper for Friends of the Earth’s Big Ideas Change the World project. Durham University. Available at: https://www.foe.co.uk/sites/default/files/downloads/autonomy_briefing.pdf (accessed 20 August 2016).

12.

Cameron

(2010) Big Society Speech – 19th July. Liverpool.

13.

Chakrabarti

(2007) Inclusion or exclusion? Emerging effects of middle-class citizen participation on Delhi’s urban poor. IDS Bulletin 38: 96–104.

14.

Clarke

Cochrane

(2013) Geographies and politics of localism: The localism of the United Kingdom’s coalition government. Political Geography 34: 10–23.

15.

Coaffee

Headlam

(2008) Pragmatic localism uncovered: The search for locally contingent solutions to national reform agendas. Geoforum 39: 1585–1599.

16.

Coppola

Vanolo

(2015) Normalising autonomous spaces: Ongoing transformations in Christiania, Copenhagen. Urban Studies 52(6): 1152–1168.

17.

Cox

(1993) The local and the global in the new urban politics: A critical view. Environment and Planning D: Society and Space 11: 433–448

18.

Dahl

(1982) Dilemmas of Pluralist Democracy: Autonomy vs. Control: New Haven, CT and London: Yale University Press.

19.

Deas

(2013) Towards post-political consensus in urban policy? Localism and the emerging agenda for regeneration under the Cameron government. Planning Practice and Research 28(1): 65–82.

20.

Department of Communities and Local Government (CLG) (2010) Decentralisation and the Localism Bill: An Essential Guide. London: Department of Communities and Local Government.

21.

Department of Communities and Local Government (CLG) (ed.) (2011) A Plain English Guide to the Localism Act. London: Department of Communities and Local Government.

22.

Featherstone

Ince

Mackinnon

et al . (2012) Progressive localism and the construction of political alternatives. Transactions of the Institute of British Geographers 37: 177–182.

23.

Fernandes

Heller

(2006) Hegemonic aspirations: New middle class politics and India’s democracy in comparative perspective. Critical Asian Studies 38: 495–522.

24.

Ghertner

(2011) Gentrifying the state, gentrifying participation: Elite governance programs in Delhi. International Journal of Urban and Regional Research 35: 504–532.

25.

Gibbs

Jonas

(2000) Governance and regulation in local environmental policy: The utility of a regime approach. Geoforum 31: 299–313.

26.

Gibbs

(1999) Sustainable cities in Europe. European Urban and Regional Studies 6: 265–268.

27.

Habermas

(1994) Three normative models of democracy. Constellations 1: 1–10.

28.

Harriss

(2006) Middle-class activism and the politics of the informal working class: A perspective on class relations and civil society in Indian cities. Critical Asian Studies 38: 445–465.

29.

Harvey

(2012) Rebel cities: From the right to the city to the urban revolution. London and New York: Verso Books.

30.

HM Government (2011) The Localism Act. London: The Stationery Office.

31.

Jacobs

(2012) Urban geographies: Still thinking cities relationally. Progress in Human Geography 36(3): 412–422.

32.

Janssen-Jansen

Hutton

(2011) Rethinking the Metropolis: Reconfiguring the governance structures of the twenty-first-century city-region. International Planning Studies 16: 201–215.

33.

Jessop

(1990) State Theory: Putting Capitalist States in their Place. Cambridge: Polity.

34.

Jessop

(2016) The State: Past Present, Future. Cambridge: Polity.

35.

Katsiaficas

(2006) The Subversion of Politics: European Autonomous Social Movements and the Decolonization of Everyday Life. Edinburgh: AK Press.

36.

Krueger

Gibbs

(2007) Introduction: Problematizing the Politics of Sustainability. In: Krueger

Gibbs

(eds) The Sustainable Development Paradox: Urban Political Economy in the United States and Europe. New York: Guilford Press.

37.

Kundu

(2009) Elite Capture and Marginalisation of the Poor in Participatory Urban Governance: A Case of Resident Welfare Associations in Metro Cities. In: India Urban Poverty Report 2009. Ministry of Housing and Urban Poverty Alleviation, Government of India / UNDP. New Delhi: Oxford University Press.

38.

Lawhon

Patel

(2013) Scalar politics and local sustainability: Rethinking governance and justice in an era of political and environmental change. Environment and Planning C 31(6): 1048–1062.

39.

Lefèvre

(2010) The improbable metropolis: Decentralization, local democracy and metropolitan areas in the Western world. Analise Social 45(197): 623–637.

40.

McCann

Ward

(2011) Introduction: Urban assemblages: Territories, relations, practices, and power. In: McCann

Ward

(eds) Mobile Urbanism: Cities and Policymaking in the Global Age. Minneapolis, MN: Minnesota University Press, pp. xiii–xxxv.

41.

McFarlane

(2011) Learning the City: Knowledge and Translocal Assemblage. Oxford: Wiley-Blackwell.

42.

MacLeod

Goodwin

(1999) Space, scale and state strategy: Rethinking urban and regional governance. Progress in Human Geography 23: 503–527.

43.

Marvin

Guy

(1997) Creating myths rather than sustainability: The transition fallacies of the new localism. Local Environment 2: 311–318.

44.

Massey

(2007) World City. Cambridge: Polity.

45.

Merrifield

(2013) The urban question under planetary urbanization. International Journal of Urban and Regional Research 37(3): 909–922.

46.

O’Riordan

(1998) Sustainability for survival in South Africa. Global Environmental Change 8: 99–108.

47.

Paddison

(1983) The fragmented state: The political geography of power. Oxford: Blackwell.

48.

Painter

Orton

Macleod

et al . (2011) Connecting Localism and Community Empowerment: Research Review and Critical Synthesis for the AHRC Connected Community Programme. Durham: Department of Geography, Durham University.

49.

Parnell

Robinson

(2012) (Re)theorizing cities from the Global South: Looking beyond neoliberalism. Urban Geography 33: 593–617.

50.

Pickerill

Chatterton

(2006) Notes towards autonomous geographies: Creation, resistance and self-management as survival tactics. Progress in Human Geography 30(6): 1–17.

51.

Pieterse

(2008) City Futures: Confronting the Crisis of Urban Development. London: Zed Books.

52.

Pratchett

(2004) Local autonomy, local democracy and the ‘new localism’. Political Studies 52: 358–375.

53.

Purcell

(2006) Urban democracy and the local trap. Urban Studies 43: 1921–1941.

54.

Roberts

Diederichs

(2002) Durban’s Local Agenda 21 programme: Tackling sustainable development in a post-apartheid city. Environment and Urbanization 14: 189–201.

55.

Robinson

(2016) Comparative urbanism: New geographies and cultures of theorizing the urban. International Journal of Urban and Regional Research 40(1): 187–199.

56.

Rose

(1999) Powers of Freedom: Reframing Political Thought. Cambridge: Cambridge University Press.

57.

Rossouw

Wiseman

(2004) Learning from the implementation of environmental public policy instruments after the first ten years of democracy in South Africa. Impact Assessment and Project Appraisal 22: 131–140.

58.

Sintomer

Herzberg

Röcke

(2008) Participatory budgeting in Europe: Potentials and challenges. International Journal of Urban and Regional Research 32: 164–178.

59.

Sowman

(2002) Integrating environmental sustainability issues into local government planning and decision-making processes. In: Parnell

Pieterse

Swilling

et al . (eds) Democratising Local Government: The South African Experiment. Cape Town: University of Cape Town Press, pp. 181–203.

60.

Sowman

Brown

(2006) Mainstreaming environmental sustainability into South Africa’s integrated development planning process. Journal of Environmental Planning and Management 49: 695–712.

61.

Thornley

Rydin

Scanlon

et al . (2005) Business privilege and the strategic planning agenda of the Greater London Authority. Urban Studies 42: 1947–1968.

62.

Townsend

(2010) Pickles: Localism is my top priority. Available at: http://www.regen.net/news/1008529/Pickles-Localism-top-priority?DCMP=EMC-RegenDailyBulletin.

63.

Vasudevan

(2014) The autonomous city: Towards a critical geography of occupation. Progress in Human Geography 39(3): 316–337.

64.

Wilson

(2003) Unravelling control freakery: Redefining central–local government relations. The British Journal of Politics & International Relations 5: 317–346.