1The paper starts a discussion with the work of our dear friend and colleague Bernard Jouve who died far too early. For a number of years Bernard critically used European and American debates about governance to develop his own ideas. The paper is an attempt to follow up our debates.
2By contrast to classic views about local government, scholars from various origins have analyzed the political capacity of groups within cities to steer, pilot, change urban society, to adapt to outside pressure, to be transformed by state new policies or by market competition logics (Jouve, Lefèvre, 2003). The urban government/governance debate has proved quite fruitful to contribute to the understanding of the transformation of cities. In the European context, the urban governance/urban regime debate was particularly useful to understand new modes of governance (John, 2001 ; Jouve, 2005) and to contrast different explanatory models of transformation. Opposition between urban governance models based upon transformation of the state in relation to new demands of globalised capitalisms (Brenner, 2004) and models based upon the lesser constraint of the state allowing some cities to develop collective actor strategies in the logic of the Weberian European city have been drawn, sometimes exaggerated (Bagnasco, Le Galès, 2000 ; Kazepov, 2005; Giersig, 2008). Classically, the literature on urban governance (or urban regimes or urban growth coalitions) aimed at pointing towards various mechanisms to create collective capacity to go beyond market and state failures (Stone, 1989 ; Logan, Molotch, 1987).
3Political scientists working on urban governance rightly emphasised its capacity to change the urban society on the one hand, and to raise democratic and participatory issues on the other (Heinelt, Kübler, 2005 ; Denters, Rose, 2005). This proved particularly relevant to the case of European cities that historically represented points of articulation between trade, culture and forms of political autonomy.
4However, as Jessop suggested, and beyond the rhetoric of governance used by political actors, there was no reason to believe that governance failures would not be as spectacular as government failures (Jessop, 2003). In other words, governance and government are not linear, and if analysed as processes are always incomplete. Urban societies are more or less governed over different periods. Studying the limits and discontinuities of government and governance is therefore particularly interesting for urban scholars.
5In this paper, governance is defined as a process of co-ordinating actors, social groups, and institutions to attain particular goals, discussed and defined collectively in fragmented, uncertain environments (Le Galès, 1998). Thus, governance relates to all the institutions, networks, directives, regulations, norms, political and social usages, public and private actors that contribute to the stability of a society and of a political regime, to its orientation, to its capacity to direct, and to its capacity to provide services and to ensure its own legitimacy. In other words, this conceptualisation based upon regulation is useful to answer the following question «who governs when nobody governs ?» On the one hand, the point has been made that governments do not govern all the time. On the other hand, there is rarely no government at all, but more or less strong and codified forms of government. Some sectors of the city can classically be organised and steered according to market logics and actors who may, or may not be dependent upon government resources to develop their project. But markets regulations can be combined with other types of regulation. Risk is one of them.
6Governments govern: but what exactly ? Classically, thinking about government meant looking at its formal apparatus, institutions, or general functions and activities. Governments were defined in terms of: rules of the game, constitutions, organisations and actors, processes of aggregation and segregation, and outputs (Leca, 1995). The governance debate started from the limits of government. This debate led to a dynamic governance research domain, beyond the «who governs» question, organised around the following questions : Can government govern, steer or row (Peters, 1997) ? Could cities be considered as collective actors with governance capacities (Pichierri, 1997 ; Le Galès, 2002) ? Do governments always govern ? What do they govern, and how ? What is not governed ? (Crosta, 1998). Can we identify dysfunctions of governments over time ? Can groups or sectors escape from governments (Mayntz, 1993) ? Who governs when governments do not govern? Can governance replace government or will governance failure replace government failures (Jessop, 2003) ? How does government or governance operate (Lascoumes, Le Galès, 2007) ? What does it mean to govern complex urban societies and networks of cities (Peters, Pierre, 2005 ; Perulli, 2000) ? Do utilities networks govern large cities (Le Galès, Lorrain, 2003) ? What sort of framework is constructed through governance ? (Bevir, 2010).
7Some activities of government take place routinely, such as raising taxes, planning and caring for specific groups. However, most government activities are not continuous (Favre, 2003, p. 165). What is governed is a key question and it may change over time. In periods of war, governments extend their control on a range of activities and sectors; there is massive increase of public policies, taxes and control. By contrast, when times are more peaceful and security threats are reduced, when there is no massive economic crisis or threat to social and political order, governmental activities may be far less important. Similarly, some sectors are heavily governed with dense public policies and laws. By contrast, other sectors are not governed at all or weakly governed. As became obvious in the recent period, loans for housing and subprimes were weakly governed, and more generally weakly regulated. As is well known, many groups and individuals in societies spend considerable time and energy avoiding being governed, leading to the famous question on the ungovernability of societies (Mayntz, 1993).
8In urban terms, this question is particularly interesting. What part, sector, group of the city is really governed ? What is weakly governed ? What is left out ? What is escaping government ?
9Historically, the rise of urban governments was about the institutionalisation of governments against illegal activities, slums, mobile populations, diseases, or rejected poor neighbourhoods. Analysing a city requires focusing not just on governments but also understanding the illegal side of the city, the invisible activities, from undocumented immigrants in clandestine rooms to gangs controlling drug trafficking or private developers financing illegally political activities in order to build new developments. This is not just some dark side of the city which will disappear on the road to rational progress but an irreducible part of any city.
10In her book on «ordinary cities» (2005), J. Robinson calls for more systematic comparison of cities from the North and the South. Governance could well travel in those terms, as suggested by McCartney and Stren (2003). In «The city yet to come», A. Simone‘s characterisation of African cities as «work in progress» (2005, p. 1) is a good reminder that what is not governed in a classic governmental rational way may be more central to understanding what works in a city. Following Robinson’s suggestion, the governance of European cities should also be understood in relation to what is illegal, clientelistic, what is not governed.
11Hence, the question what is governed in the city leads to the question : who is governed in the city ? Government is a two ways process. Accordingly, a lot depends upon the population which is governed, whilst governing a large city is a difficult task because the population is so fluid and diverse.
12Urban riots are now a regular feature of many cities. Reports after the riots usually point to the failures of local schools, social services, discrimination against young people from immigrant background on the labour market, isolation of some ethnic groups, police violence, the rise of illegal trafficking and the absence of legality in some neighbourhoods. In other words, some sections of European cities are not effectively governed. Some inner city and some outer city developments, whilst not completely left out, still have weak infrastructures and public services support. Drug trafficking and informal markets are also part of the urban life. In some neighbourhoods, the police is not welcome and only intervenes when problems emerge and in those cases with significant numbers of policemen. Waves of riots in the UK, France and sometimes Germany, the Netherlands and Italy, point to the development of parallel lives between immigrants and their children and the rest of the population. Analysis point to discrimination, intense segregation in some cities and the lack of policies to integrate some of those groups (Garbaye, 2010; Lagrange, Oberti, 2006; Waddington et al, 2009).
13Are those neighbourhoods governed ? More or less, it depends. In France or Britain, the making of large outer city social housing schemes in the 1960s was not matched by services to the population. In French poor suburbs where some estates concentrated the poorest populations and recent immigrants, local authorities did not have resources, while social services, police, schools and public transports were insufficient. Those places were not governed, or weakly governed. The development of public policies to integrate and incorporate different groups led to some successes and many failures, governance failures.
14In most cities, the question of how to provide services to those populations, how to include them in the democratic debate, how to help children to do well but also how to control those groups and neighbourhood, how to make the law respected, is high on the agenda. In the US, M. Jones Correa and his colleagues show the challenge of ethnic diversity in the process of «Governing the American city» (2001). Similar challenges are very much at stake in European cities and there are numerous cases of problems and lack of governance capacity in relation to ethnic groups in particular (Gagnon, Jouve, 2009).
15The «who is governed question» can also be linked to issues of mobility. In his seminal contribution, G. Martinotti (1993) identified different city users beyond the classic groups of those who work and live in the same place, a minority in most cities. He pointed not only towards the classic regular suburbanite working in the city centre but also the rise of occasional «city users», visitors, with different level of professional constraints or levels of segregation at different scales. A wide literature has developed on «Cities and visitors» (Hoffmann, Fainstein, Judd, 2003). The literature on immigration (Favell, Smith, 2006) has also pointed to the development of transnational networks and the mobility of groups between two places. Who governs transnational networks of retired Moroccan, Pakistanis, Senegalese or Turkish pensioners who spend half the year in their home country where they build a house and half the year in a host country where the children have organised their life ? J. Urry’s «sociology of mobility» (2000) points to the end of a form of a «garden state», when people travel or become more mobile. Census agents have more problems with those students, families or often pensioners, travelling, or organising their life in two different places for a number of years or immigrating for a number of years. Who governs mobile population ? This raises numbers of questions about the provision of services, about tax avoidance, about school population or the provision of housing. The rise of modern governments was related to the development of new technologies of government to make society «legible». One could argue that increased mobility of different sorts make urban society less legible and therefore far more difficult to govern.
16Regulating is not governing. However, the economic sociology of regulation is helpful to understand regulations. In analytical terms, cities are more or less organised around markets or governments, in more or less conflictual or combined ways. Regulations can be seen as mechanisms of governance and defined on the basis of three dimensions : (i) the mode of co-ordinating diverse activities or relationships among actors, (ii) the allocation of resources in relation to these activities or these actors, and (iii) the structuring of conflicts (prevented or resolved). Consequently, the word «regulation» can be used, for example, when highlighting relatively stabilized relationships between actors or social groups, relationships which allow the distribution of resources according to explicit or implicit norms and rules (Lange, Regini, 1989).Three ideal types are usually defined : (i) State regulation (frequently identified with hierarchical or political regulation), where the state structures conflicts, distributes resources and co-ordinates activities and groups. This type of regulation implies domination and control as well as the capacity to sanction. This description can also fit certain large, hierarchized organizations where authority is the principal moving force, even if only informally.(ii) Market regulation : since the emergence of capitalism, this type of regulation has played a growing role in organizing exchanges between supply and demand, adjusted through prices (or sometimes through volumes...).(iii) Co-operative/reciprocal regulation (sometimes called regulation through social or political exchange) based on values and norms, on a single identity, and on the trust that expresses forms of exchange and/or solidarity between the members of a community, a clan, a family, or a district (Crouch, Le Galès, Trigilia, Voelzkow, 2001).
17These three regulations are mixed in the governance of cities. Governments never completely govern a city because they have to deal with market forces (private developers in urban growth coalitions), along with religious groups, familial interest, social movements or NGOs.
18A classic form of regulation in the third category (exchange reciprocity) is clientelism, patronage, corruption. An underdeveloped research agenda has in particular suggested taking seriously cooperation/reciprocity regulation beyond the «enchanted» view of governance through traditional regulation. Patronage is defined as : «…the complex social arrangements known as patron-clients relations denote, in their fullest expression, a distinct mode of regulating crucial aspects of institutional order : the structuring of the flow of resources, exchange and power relations and their legitimation in society» (Eisenstadt, Roniger, 1984, p. 209). Corruption is classically defined as a system of exchange where public officials obtain financial resources in exchange for decisions. The «community» or «cooperative/reciprocal» regulation is also a broad church where one can at least distinguish between the logic of reciprocity and the logic of social/political exchange. As far as urban politics is concerned, the latter is more relevant. There is a long tradition of research to examine patronage and corruption as one type of social or political regulation in anthropology in particular and later in sociology and political science, taking into account the bureaucracy in particular or factions in urban political machines. In urban politics, classics include in the US, Banfield and Wilson’s «City Politics» (1963) or P. Allum’s monograph on Naples (1973). In the last two decades in many European cities, some forms of patronage, clientelism or corruption have been identified for instance with the «mani pulite» operation in Italy seeing the demise of the socialist domination of Milan; in the Paris case where several leaders of the Chirac post-Gaullist party were condemned by the Courts; but also in the North East of England or Liège in Belgium. In other words, parts of the city can be regulated according to non classic governmental principles but with the participation of governmental actors.
19Who governs when nobody governs ? Corrupt elite networks sometimes. Illegal organisations are also a classic case where they can «run», organise, possibly govern, some sectors, some neighbourhoods or parts of the city. The implication of the Camorra in Napoli has become a classic case but more limited examples could be found in other settings.
20Finally, urban governance and most public policies are part of the world of overlapping powers within the global and regional (such as European) governance in the making : municipalities, metropolitan authorities, regions sometimes, federal states or Autonomies, the nation state, the EU and sometimes the OECD urban group, the UN (Habitat Summit) international rules comprising environmental norms, can play a role in urban policies. There are endless cases of urban policies where the norm is now for the overlapping funding and influence of different levels of government, for the better i.e. more targeted and coordinated effort, or the worse, more piecemeal fragmented actions. In most countries, the territorial organisation of the nation state has been facing serious reshaping, an ongoing process which leads to the pluralisation of territorial interests within the state. Associations, voluntary sector organisations, from neighbourhood group to giant utility firms have a say and some power in urban policies. Urban policy therefore covers a wide range of actors from different sector of societies, with different status, acting at different levels. Emerging problems raise questions which cross horizontally over bureaucracies and sectors, and vertically over different levels of government. In that world, one wonders if governance is more than the aggregation of various incremental choices, and random developments.
21The questions «what is governed ?» also finds a partial answer in a recent trend, which sees cities as places exposed to a multiplicity of risks, which urban governments are expected to manage with different instruments, standards and approaches.
22In many respects, the risks most often cited have always been familiar to modern cities : contagious diseases, protest movements, social disruptions, crime and delinquency, poor quality housing, natural and man-made disasters. Cities were considered vulnerable to these risks due to: the way they were built and organized, the concentration of large numbers of individuals in a limited (and often crowded) space, the ensuing collective problems (hygiene, waste, …), the exacerbation and proximity of social and economic distinctions, the attraction of populations looking for work, shelter or food, the siting of dangerous and polluting activities, the presence of institutions of economic and political power offering a target for attacks or social disruption, the multiple and often interdependent networks (information, finance, transports, energy, …), etc. But until recently, these problems were not framed as risks. They were collective problems that called for institutions such as the police (historically in charge of maintaining both order and hygiene) and technical solutions such as collective networks to collect used waters and waste (de Swaan, 1988).
23Many of these problems are now defined as risks. Although this can be seen as part of a wider extension of the notion of risk to policy issues, risk holds some special features in urban settings, and these relate to the questions addressed in the previous section.
24For the most part, as we just noted, problems framed as risks are not new. Nor can it be said that cities today are more vulnerable than they were before. True, they are more densely populated and more dependent upon a whole set of complex and interdependent networks. Large cities are also at the crossroads of multiple flows of population, information, goods and services. But the middle age city could easily be characterized in similar terms, when set against their wider political and economic environment.
25There are of course some exceptions, notably in the US where scholars have analyzed how the development of large cities systematically denied the importance of environmental factors, even reducing natural defences and introducing new vulnerabilities that proved in time to be catastrophic (Davis, 1998; Freudenburg et al, 2008). There are no such studies in Europe, due for the most part to the way cities were built and developed over the centuries. Vulnerabilities and hazards are well known, but they have for the most part always been there. The development of urbanization may have increased the risk of flooding in cities like Paris, London and Prague, for example, but one can hardly state that these cities are more at risk than they were a hundred years ago.
26If risk, per se, is not a new feature of the European city, risk as a notion has entered the realm of urban governance. This can be see as an effort to depoliticize public problems, to suggest technical solutions to often complex issues, to redefine rules of accountability, and finally to promote a modern approach to the resolution of urban problems – in line with a general trend that characterizes organizations, public and private alike.
27Indeed, risk management or governance suggests that, through the application of scientific or technical approaches, a problem initially fraught with uncertainty can be understood, measured, made predictable, in other words controlled and governed. Applying a risk framework to flooding or earthquakes, industrial accidents, crime, drug trafficking or a pandemic thus reveals causal factors, patterns, recurrences, which can be used to find solutions. These will be based on planning or building codes, monitoring devices, alert systems, protection (such as levees or floodwalls), or contingency plans to mitigate the consequences of a catastrophic event. Experts, academics, consultants are important actors in providing authorities with data to define risks, along with solutions to manage them.
28Strongly linked to the notion of risk is the notion of vulnerability. This notion first arrived with seismologists but was quickly picked up by geographers and later international organizations working on risks and disasters. It suggests that to manage risks, local authorities need not only define the dangers or hazards that threaten their city, but must also identify the populations or neighbourhoods that are most exposed. Although vulnerability is often defined in technical terms, such as the capacity of buildings to resist an earthquake or of a population to adapt to a pandemic, defining vulnerable groups and parts of the city is in fact a highly political decision, for it determines directly who is responsible for individuals or groups that are deemed less capable of protecting themselves in the face of a danger or disaster. Factors such as age, class, ethnicity, gender are considered important in defining vulnerable populations (Tierney, 1999). But contextual elements also play a determining role in explaining how these factors actually play out (Klinenberg, 2002).
29Both the introduction of risk and vulnerability thus shape the «what is governed» issue in a very specific way. While some groups of the population will be considered as being at risk, others may be defined as risk factors (by their behaviour, for instance minorities accused of drug trafficking or carrying diseases). Maps help decide which neighbourhoods are exposed to dangers, be they flooding or criminal activities, and this will have an (indirect) impact on property values. Technical devices, such as cameras and once again maps, are used as instruments destined to change the behaviour of individuals; but more often then not they will force them to move to other parts of the city either to avoid a danger or to avoid being watched. City officials have lists of vulnerable populations they must provide assistance to in case of a pandemic, heatwave or other major catastrophe. But these lists rarely match from one city to another and once again reveal distinctive preferences.
30The extension of risk to different urban issues also contributes to a redefinition of boundaries between the state and different levels of government. Preparations for a potential H5N1 pandemic and management of the recent H1N1 pandemic both offer interesting illustrations. While in some countries, the state relies on local government to take part in the management of the pandemic, in others it organizes the response with its own resources. For instance, French authorities still consider that major public health problems are a state prerogative, with little room for local authorities. On the contrary, British authorities deem that it is the role of local authorities to take part in the management of a major health event, since they have a better knowledge of their populations.
31Other risk issues reveal the political capacities of metropolitan governments, compared to smaller urban governments. For example, on issues related to water or waste, metropolitan governments possess not only the resources but also the policy style needed to address these often complex technical and social problems. For not only do these problems require technical skills and financial capacities, which they possess, they also require decision-making procedures that can achieve some form of consensus among the different stakeholders.
32This last feature may be observed, not only on more traditional risks such as waste management, but also on emerging risks. This notion refers to issues that present multiple uncertainties, both scientific and social, are characterized by a high level of controversy, and present an important political risk for public authorities. Good examples are radiofrequencies and nanotechnologies. Faced with emerging risks, local authorities are initially at pains to manage the scientific debates on whether or not these technologies present a risk for health or the environment. But once this dimension of the problem has been delegated to national authorities and their experts, local authorities demonstrate a capacity to invent solutions to reduce the social uncertainties, by engaging in negotiations with the different stakeholders and producing new sets of «local» rules. In many instances, these contribute actively to a reduction in the general level of controversy and political risk. Once again, metropolitan governments are more prone to these inventions than municipal governments.
33But the questions raised by these local solutions reveal two ambiguities of local risk governance.
34First, faced with multiple demands for protection against different risks by their populations, local authorities tend to favour precautionary approaches. This is all the more noteworthy in decentralized countries where urban governments are competing with both central and regional governments to demonstrate their capacity to protect their populations. On radiofrequencies, for example, this can be observed in Spain, Italy and Belgium, in contrast with France and the UK where central government maintains its prerogatives on public health issues (Borraz, Salomon, 2007). In turn, this competition may fuel the general controversy and help maintain some risk issues high on the public agenda – whatever the actual risks for the population may be.
35Second, local solutions often entail either externalities or disparities between places : this can imply a risk transfer. Such a transfer calls for renewed institutional capacities, either at the level of an urban region or within the state. Yet, many risk issues reveal that such capacities are still lacking. Hence, cities tend to become risk factors for their neighbouring territories, and in turn this fuels multiple conflicts around activities or facilities that are considered by suburban or rural zones as negative externalities from the large city that present a risk for the health or environment of their populations. More and more, these conflicts are framed as risk issues.
36In line with the last comment, what is governed and what is not governed must also be considered in relation to local government boundaries. The city is proving more elusive, populations more diverse, governments are being rescaled and new modes of governance are being structured. The classic European city comprised within city walls disappeared long ago. Although the city in a classical sense has remained relatively robust, suburbanisation has developed all over eastern and western European cities. In territorial terms, what is governed is anything but an obvious question.
37This is a classic theme in the urban literature. Those writing on mega-cities, giga-cities or the rise of global urban regions (Scott, 2001) point to the rise of networks and governance failures related to obsolete governmental boundaries. Another way to think along the same line relates to the idea of the end of cities and the triumph of urban sprawl, in other words the suburbanisation of cities and the urbanisation of suburbs (Dear, 2000). In that line of analysis, the dissolution of the city is taking place within a large fragmented, chaotic, unstable urban world which is not governed. Is the urban world becoming ungovernable ? A classic argument dismisses this view because the relatively stable core of Europe’s urban system is made up of medium-sized and reasonably large cities, which are fairly close to one another, and a few metropolises. This importance of regional capital cities, of medium sized cities (200 000 to 2 million inhabitants) remains a major feature of contemporary European societies (Le Galès, 2002). However, there is also suburbanization (Phelps, 2006) based upon the departure of population from both the city and the metropolitan area because of deindustrialization for instance and the rise of the metropolitan area (London, Paris, Brussels, Milan, Marseille or Lisbon) (Jouve, Lefèvre, 2004). Increasing urban concentration has been accompanied by apparently inescapable, unlimited dispersal into conurbations and urban regions with fluctuating outlines. Cities have expanded, fragmented, and sometimes organized into networks like those in Northern Italy or the Netherlands, and this is said to be rendering traditional urban governments obsolete. Europe is made of a few declining cities, many dynamic medium size and large cities, and two dynamic large «global cities», whatever that means. European cities make a fairly general category of urban space, relatively original forms of compromise, aggregation of interest and culture that bring together local social groups, associations, organized interests, private firms and urban governments. The pressures created by property developers, major groups in the urban services sector, and cultural and economic globalization processes, provoke reactions and adaptation processes of actors within European cities, defending the idea of a particular type of city that is not yet in terminal decline. The modernized myth of the European city remains a very strongly mobilized resource, and is strengthened by growing political autonomy and transverse mobilizations.
38Despite sprawling movements in most European cities, the resistance of the old city centres epitomizes their peculiarity. Lévy (1997) takes the example of large public collective transport (in particular the tramway) together with pedestrian areas and cycling paths to demonstrate the remaining strength of the idea of European city. There is a continuing representation of the city as a whole : Crouch (1999) suggests a «Durkheimian» view of the city that still exists in Europe. The increased legitimacy of political urban elites sustains and re-invents this presentation. European cities are still strongly regulated by public authorities and complex arrangement of public and private actors. European cities appear to be relatively robust, despite pressures from economic actors, individuals, and states (including welfare states) being reshaped within the European Union. Processes of exclusion, strengthening and transformation of inequalities, segregation, and domination are also unfolding in these cities. The development of residential suburbs separated from the city and of polycentric cities, the isolation of disadvantaged districts, the development of cultural complexes, leisure facilities and shopping centres, as well as diverse cultural models and migrations, all clearly demonstrate the pressures exerted on the traditional medium-sized city. The urban regions of Milan and the Randstadt are good examples of more polycentric structure and interdependent dynamics between the city centre and other cities.
39These developments have led in most countries towards never ending debates about the rescaling of metropolitan governments, the making and unmaking of metropolitan governments (Lefèvre, 1998), and the developments of all sorts of collaborative form of governance including private-public partnerships, charters, plans, contracts or joined up governments.
40In most places in Europe, the reorganisation of urban governments has given salience to the particular status of (big) cities, now comprised in terms of metropolitan areas and often organised under state pressure (Brenner, 2004). Firstly, in comparison with fifty or a hundred years ago, the larger European cities have more autonomy and much more vigorous local leadership (Borraz, John, 2004). The major capital cities of Europe have only in the last years or decades had an elected unified government and mayor (Paris 1977, London 2000). Despite some resistance, a metropolitan government is being recreated in Britain, and slowly in Italy. In France too, the restructuring of local government based on a mix of direct constraints and strong financial incentives is creating an original and powerful structure of inter-municipal urban government benefiting from strategic and public policy delivery powers together with important financial and human resources. Metropolitan government emerged in the Stockholm area in the l970s and has developed in the other Nordic capital regions too. In Eastern Europe, reforms of local government in the 1990s led to differentiated sets of legal statuses, in particular for the capital. During the negotiation to join the EU, a particular emphasis was put on decentralisation reforms supposed to undermine existing bureaucracies and to reinforce the democratisation of the political regimes. Within that decentralisation trend, cities did particularly well in terms of new powers. In the Hungarian two tier system, the capital has been granted a special legal status with specific powers given to the district government of Budapest and the urban mayors are directly elected. The same applies to the Czech Republic where the 2002 restructuring of local government grants a special status to Prague and 19 other cities. In Poland too, 65 cities were given county status. Relatively high levels of devolution were also granted to Baltic state cities.
41There is a good deal of urban governance going on in European cities but not all the time, not for all the groups, not for all the neighbourhoods and not so much for the peripheries of the city. This feeds the rise for new policy instruments to increase governing capacity (Pinson, 2009 ; Lascoumes, Le Galès, 2007). Governance discontinuities should therefore be analysed more systematically.