The local state’s repertoires of governance strategies for the urban commons: Nuancing current perspectives

Theorists of the commons appear to have rather siloed, ideologically driven understandings of the repertoires of governance strategies enacted by the state in the emergence and development of commons. Neo-institutionalists tend to see the state as a supportive institution, while neo-Marxists generally see the state as either antagonistic and/or co-optative. This article sets its research in the dense and contested urban context, and examines the repertoires of governance strategies adopted by the local state in the emergence and development of the urban commons. By engaging with relational state theory, it aims to help nuance the theory of the urban commons–local state relationship. It does so by carrying out a comparative analysis of two cases of urban commons located in Barcelona. For each case, it delves into the variations, ambiguities and complexities of the repertoires of governance strategies enacted by the local state in its emergence and development. The article concludes that, as suggested by relational state theory, the boundaries between the types of governance strategies that the local state takes on are more nuanced than is perhaps reflected in the literature. These repertoires may well overlap, evolve throughout time and apply selectively to different urban commons.


Introduction
Over recent decades, the commons have re-emerged in the literature of collective action, as selfgoverning initiatives that manage resources or services that work outside the dominant market and state logic (Ostrom, 1990;Hess & Ostrom, 2007;Mattei, 2011;Harvey, 2012;De Angelis, 2017). However, all these initiatives are embedded in market and state relationships (De Angelis, 2017).
This contribution sets its research in the dense and contested Western European urban context, a context packed with people and characterised by different social groups' conflicting imaginaries, interests and uses of space; it aims to examine the repertoires of governance strategies adopted by the local statethe ensemble of municipal government, agencies and institutionsin the emergence and development of the urban commons. The literature on the commons is extremely varied and uses different theoretical approaches, which in turn have different perspectives of the governance strategies enacted by the state. Enright and Rossi (2017) review this diversity and identify two main approaches within it: a neo-institutionalist strand and a neo-Marxist strand. 1 The neo-institutionalist approach and its more recent evolution see the state as a supportive institution (Ostrom, 1990), while the neo-Marxist approach sees the state as an antagonistic and/or co-optative institution (Hardt and Negri, 2009;Mattei, 2011;De Angelis, 2013;Stavrides, 2016). Both perspectives are corroborated by urban studies scholars' theoretical and empirical contributions. Each approach describes a distinct repertoire of governance strategy developed by the state in the emergence and development of the commons. However, they share one main shortcoming: they adopt a relatively siloed understanding of these repertoires, based on ideologically driven perspectives. This shortcoming tends to limit our understanding of the potential scope and limitations of urban commons-led collective action in cities.
This article aims to examine the variations, ambiguities and complexities of the repertoires of governance strategies adopted by the local stateantagonistic, co-optative or supportivein the emergence and development of urban commons, seeking to move beyond the siloed understandings of the commons-state relationship proposed by neo-institutionalist and neo-Marxist scholars. By engaging with relational state theory (Jessop, 2015), the article argues that the boundaries between the different repertoires that the local state may take on are more nuanced than the literature reflects. The article sets its research in the city of Barcelona, and uses a longitudinal cross-case comparison of two different urban commons. It begins by illustrating the neo-institutionalist and neo-Marxist perspectives of the state's repertoires of governance strategies in the emergence and development of the commons, interlinking them with contributions from the urban studies field. It then introduces relational state theory to nuance current theorisations. It continues by presenting and discussing the two cases. It concludes by highlighting how the evidence provided by this study underpins a more nuanced understanding of the commons-state relationship, as suggested by relational state theory.
The role of the state in the emergence and development of the commons: current perspectives According to Enright and Rossi (2017), the two main approaches to studying the commons are the neo-institutionalist and neo-Marxist ones. The neo-institutionalist approach sees the commons as an alternative management model to the state and the market, but one that should exist side by side with them, all within the capitalist system and existing power relations. The neo-Marxist approach also interprets the commons as an alternative management model to the state and the market, but as a model that should be used to overcome both, destroying existing power relations and the capitalist system. Unsurprisingly, each approach adopts a different perspective of the repertoires of governance strategies enacted by the state in the emergence and development of the commons.
Initially, Ostrom appeared to adopt an 'adversarial' attitude towards the state (Van Laerhoven, Schoon and Villamayor-Tomas, 2020). In the design principles that she created to set out the characteristics of long-enduring commons, principle 7 states that 'The rights of appropriators to design their own institutions are not challenged by external government authorities' (Ostrom, 1990, p 90). This approach has often been interpreted as an anti-state message from Ostrom. However, closer attention to her work reveals that this is a serious misinterpretation (Mansbridge, 2014). Design principle 8, aimed at large-scale commons, states that these types of commons need 'nested' governing structures to endure. These structures can provide coercion and resources that make decision-making efficient, and help solve governance problems. This is also the message of another of Ostrom's concepts: 'polycentrism'. Polycentrism implies a form of governance that is formed by multiple, sometimes overlapping, centres of decision-making, each operating with some degree of autonomy, but which are nested at multiple jurisdictional levels (Ostrom, 2010). In subsequent work, Ostrom and other neo-institutionalist scholars have used the term 'polycentric' interchangeably with the 'nested' requirement of design principle 8 (Carlisle and Gruby, 2019). Thus, Ostrom recognises the state as an important player in the development of the commons. However, she sets this aspect aside somewhat, choosing to highlight other aspects of the commons, such as rule-making and enforcement (Anthony and Campbell, 2011).
More recently, the important role of the state in neo-institutionalist commons theory has been further reappraised by Ostrom-inspired scholars working in the field of urban governance. Foster and Iaione (2019), studying the application of polycentrism in the urban context, have developed a set of design principles for urban commons. According to their second principle, enduring urban commons need an enabling state (usually a local authority) that facilitates their creation and supports governance arrangements to manage and sustain them (Foster and Iaione, 2019). They see the state's adoption of a supportive repertoire of governance strategy towards the commons as part of a new polycentric approach to city governance, in which a central role is acquired by the provision and management of goods and services by groups of citizens, following principles of subsidiarity and co-operation (Foster and Iaione, 2016). An example of the state adopting this supportive repertoire is the case of housing co-operatives. The literature on the commons has demonstrated extensively how the state is an essential partner in the emergence and development of housing co-operatives, offering policy and planning frameworks that provide the initial capital for the project and that facilitate access to the land required (Huron, 2018).
The neo-Marxist commons literature provides a different and slightly more complex perspective regarding the repertoires of governance strategies adopted by the state in the emergence and development of commons. According to its key proponents, the state is an institution that, in order to accommodate the interests of capital, adopts two repertoires of governance strategies: an antagonistic repertoire through which the state aims to destroy the commons (Hardt and Negri, 2009;Mattei, 2011), or a co-opting repertoire through which the state aims to de-potentiate the commons (De Angelis, 2013;Stavrides, 2016). Both these repertoires are an integral part of the state rationale in neo-Marxist theory, and can alternate over time, or overlap with one other according to sociopolitical circumstances. However, each neo-Marxist scholar tends to focus on the repertoire that they see as predominant.
According to Hardt and Negri, the state is mainly an antagonistic institution, as it is at the forefront of the 'capture and expropriation of the commons' (Hardt and Negri, 2009, p. 256). This repertoire is exerted by privatising and commodifying common resources and services, something that Harvey (2007) has defined 'accumulation by dispossession'. It also acts by suppressing and repressing alternative collective forms of management that could challenge the social-political system. According to Mattei, the centralising force of the state 'shrinks' the commons or 'tends to make them disappear' to benefit the private property-state dichotomy (Mattei, 2011, p. 12). This perspective is reflected in several works by urban commons scholars. Analysing independent cultural spaces in Dublin, Bresnihan and Byrne (2015) show how one of the most relevant constraints to the development of urban commons is the intervention of public authorities, which generally either evict them or shut them down. In his study of urban squats in Rome from 2000 onwards, Di Feliciantonio (2017) highlights how the city council and other state institutions have implemented a series of repressive measures that have led to violent evictions.
Other neo-Marxist scholars see the state mainly as an institution that attempts to co-opt the commons (De Angelis, 2013;Stavrides, 2016). The co-opting repertoire of the state follows two main logics: the economic and the political. The economic logic has been theorised mainly by De Angelis (2013). He argues that the economic feasibility of capitalism has reached an 'impasse' because its growth can no longer come from the social and natural ecosystem on which it currently relies, as this would lead to its self-destruction. The commons are the perfect 'fix' to address this impasse, and to displace the cost of social reproduction onto communities around the world. This logic is implemented in the urban realm through the austerity urbanism paradigm, which has 'rollback redux' as one of its main features (Peck, 2012). According to Peck, if a defining characteristic of early-stage neoliberalism was the rollback of the social state and its redistributive machinery, today's austerity urbanism involves the rollout of a 'shadow' welfare state that relies on community-led, non-profit and faith-based sectors to provide budget-saving welfare services. In her analysis of numerous cases of community-based education initiatives, Martinelli (2013) highlights how they camouflage the rollback of a welfare state that does not provide adequate financial support to its citizens.
The political logic has been theorised mainly by commons scholars who follow the argument of post-Foucauldian studies (Stavrides, 2016). According to these studies, the state acts according to a specific 'governmental rationality' based on the 'presumption that everything can, should, must be managed, administered, regulated by authority' (Allen, 1998, p. 179). This is achieved through the adoption of a set of techniques through which the state regulates, controls and disciplines collectives, especially the most subversive ones, such as the commons (Stavrides, 2016). To do so, it deploys techniques of social-spatial control that aim to tame and normalise them. In her longitudinal analysis of squats in Amsterdam, Dadusc (2019) shows how the local state domesticates the commons, using specific modalities to make them compatible with the economic and political order, and thereby erasing their radical capacity. Frequently, these two co-optation logics work concurrently and overlap. Analysing the Regulation for the urban commons in Bologna, Bologna City Council's framework to institutionalise commons initiatives, shows how the Regulation not only relies on the voluntary work of commoners to carry out activities in the public interest, but also limits their radical demands (Bianchi, 2018).
Both the neo-institutionalist and the neo-Marxist perspectives, with different degrees of sophistication, describe various repertoires of governance strategies enacted by the state in the emergence and development of the commons. Neo-institutionalists see collaboration as the only repertoire enacted by the state, adopting a tendentially monolithic and static perspective of its governance strategies. Conversely, neo-Marxists adopt a more dynamic and plural perspective, seeing the state as enacting antagonistic and/or co-optative repertoires, either simultaneously or alternatively, depending on the socio-political context. However, both sets of scholars display one main shortcoming: they tend to avoid entering into dialogue with each other, thus creating an epistemological separation in the discipline that seems more ideological than empirically based. To bring together and nuance these perspectives, the article will now draw on insights from recent development in relational state theory.

Bringing in relational state theory
Within relational state theory, the state is considered as a complex assemblage of institutions, organisations and interactions that are far from acting and exercising power as a coherent and unitary entity (Jessop, 2015). This means that coherence and unity are not a given, but have to be built by mobilising the state's different capacities and resources in the move to pursue certain state projects and objectives (Baker and McGuirk, 2019). In doing so, the state assemblage does not act uniformly throughout time and space, but adopts different repertoires of governance strategies according to the socio-economic context, the political system and scalar geographies within which it is embedded (Brenner, 2004). When defining strategies, this assemblage engages in an intricate networking relationship with other actors to multiply its effectiveness, gathering them together to become part of the state assemblage, and thus blurring state-society boundaries (Jessop, 2015). However, these networking relationships are established in a selective manner, that is, only including actors that are identified as willing to co-operate and can enhance political consensus, while tending to exclude the more problematic and subversive ones, in order to build what the literature has defined as a post-political configuration, a configuration that marginalises conflictual voices and demands from the process of governance (Swyngedouw, 2009).
This relational understanding of the state allows us to move beyond commons' scholars' somewhat siloed understanding of repertoires of governance strategiessupportive, antagonist, cooptativeto bring them into a coherent theoretical framework that can be used to build less ideologically driven knowledge regarding the commons-state relationship. Such a framework is based on three main propositions. Firstly, the diverse, heterogeneous and fragmentary nature of the state elucidated in relational state theory implies that a local state's repertoires of governance strategies adopted in the emergence and development of the commons do not have a clear-cut interpretation and cannot be ascribed to a unique rationality (either co-operative, antagonist or cooptative in our cases). Instead, they are characterised by contradictions, compromises and incompleteness, with such repertoires blending into each other and sometimes overlapping. Secondly, the fact that, according to relational state theorists, state actions are not carried out evenly through time and space implies that the local state's repertoires of governance strategies enacted in the emergence and development of the commons are not static and pre-defined, conversely to what is supposed in the neo-institutionalist account that sees the state as exclusively supportive, but are changing and evolving, as rightly suggested by neo-Marxist theorists. However, for neo-Marxists, such changes and evolutions take place in the continuum between antagonism and co-optation and do not include co-operation, which should also be considered as a possible repertoire. Finally, relational state theory's idea of the selective nature of the state in its engagement with different actors implies that even in cases where the local state might appear to adopt a supportive repertoire in the emergence and development of the commons, this repertoire is not applied consistently to all social groups that develop commons, but only to the more moderate, co-operative ones. To elucidate this, in the following sections, the article shows how these three propositions play out in the urban environment.

Methodology
This article sets its research in the context of Barcelona, and adopts a qualitative-interpretative approach based on a longitudinal cross-case comparison of two urban commons, both established in the same turbulent post-crisis period and with the same political forces in office, but that had different trajectories and outcomes: i) Puigcerdà, a housing squat set up in 2011 by a heterogeneous group of sub-Saharan immigrants, who were eventually evicted in 2013; ii) Can Batlló, a cultural centre set up in 2011 by neighbourhood residents and activists that is now one of the most acclaimed self-managed spaces in the city. The cases were selected by following two main criteria. Firstly, they began to be self-managed at around the same time. Secondly, they were carried out by different social groups. These two case studies were projects that started in the same period, but were carried out by two different social groupsone made up of marginalised sub-Saharan immigrants, the other low-and middle-income active citizens. The projects gave shape to different commons: one was an informal collective space to cope with a kind of poverty usually present in the Global South but that is also becoming increasingly widespread in Global North cities (Montagna and Grazioli, 2019), while the other was a traditional Western European self-managed cultural space (Martínez López, 2013). This allowed us to examine how the repertoires of governance strategies adopted by the same local state, that is, one governed by same political forces, may differ in relation to different social groups, practices and claims. This comparison permits us to work towards a more comprehensive understanding of the commons-state relationship at the city level. This understanding is furthered by an analysis of the changing political context in Barcelona: the demands and evolution of social movements, power shifts in the local government, and changes in the urban agenda. The research was carried out from March 2016 to June 2017. Data were collected by analysing documents and carrying out 18 in-depth semi-structured interviews with members of the urban commons' social groups, as well as public officers, politicians, activists and professionals (architects, lawyers) that were directly involved in the case. Questions were posed to understand the interviewee's perspective on the local government's strategy towards the urban commons. The interviews were recorded and manually transcribed. They were analysed using Nvivo10 software that facilitates data coding and cross-case analysis. The cases are presented and discussed along a temporal narrative that follows the power shifts in local government. This structure allows us to better embed the cases within Barcelona's changing political context.

The emergence of two urban commons in post-crisis Barcelona
Barcelona has a tradition of self-organisation that is over a century long, and is founded on its anarcho-collectivist culture. This tradition, although eclipsed during the Franco dictatorship, reemerged in the late 1980s with the growth of the squatting and co-operative movement, and exploded with the 2007/8 crisis when urban commons proliferated once again (Cruz et al., 2017). Barcelona was severely affected by the economic recession. From a political point of view, in 2011 this led to the birth of the 15M Movement (15M), a set of mass social protests linked to previous urban mobilisations related to anti-eviction movements, that spread through the public squares of major Spanish cities, including Plaça Catalunya in Barcelona. At the local electoral level, in May 2011 it led to the victory of a liberal-conservative coalition after more than 32 years of left-wing city councils in power . During and after the mobilisation, different commoning initiatives emerged, some of them directly or indirectly linked to the re-territorialisation of the 15M from squares to neighbourhoodssuch as Can Batlló - (Varvarousis et al., 2020), others born of marginalised social groups' attempts to survivesuch as Puigcerdà (Pradel-Miquel, 2017). Within this turbulent socio-political context, marked by the spread of different modalities of collective action, from marches and square occupations to urban commons that contested and challenged the economic and political order, the local government adopted a repressive governance strategy towards the most disruptive forms of political action (occupations of the city squares), but it did not prevent, and at times even apparently facilitated, the emergence of commoning initiatives. However, this apparently enabling governance strategy adopted by the local government towards less disruptive forms of contentious politics could also be interpreted as economic and political co-optation.

The emergence of Can Batlló: Between support and co-optation
According to the Barcelona General Metropolitan Plan approved in 1976, the privately owned 14 hectares of the former Can Batlló industrial site located in La Bordeta neighbourhood were earmarked for transformation into public facilities and spaces. However, the left-wing coalitions that governed the city never began the necessary public works. Over the decades, the residents of La Bordeta repeated their demands for the public facilities and spaces that had been promised to them. In 2009, these demands coalesced into the 'Can Batlló is for the neighbourhood' Platform (CB Platform), stemming from newly mobilised groups created from residents' associations, along with activists and social organisations. With the onset of the financial crisis, it became clear to the residents that the building work needed was not going to be carried out, and they decided to push their claims through with a structured protest. They set a specific date, 11 June 2011, close to the local municipal electionswhich were to be held on 22 Maythreatening to squat in the factory if the work had not started by then. During the campaign, the demand for public facilities and spaces evolved into a demand for self-management.
In May 2011, the 15M movement erupted, and Plaça Catalunya became the scenario of antiausterity protests. On 22 May, the liberal-conservative coalition won the elections; it was set to take power on 11 June. The centre-left administration decided to establish negotiations with the CB Platform and the owner of the building, and on 11 June, an agreement was reached. This transferred one of the Can Batlló warehouses from the owner to the city council, which would in turn temporarily grant it to the CB Platform under an agreement to be renewed yearly. Thus, on 12 June, the CB Platform triumphantly entered the Can Batlló site to begin self-managing it. Indeed, the decision to temporarily transfer the warehouse to the CB Platform was crucial for the birth of Can Batlló. Without this space, it is likely that Can Batlló would never have been set up. However, the transfer of the property could also be interpreted as the adoption of a political co-optation strategy: it could be argued that the council's actions were moved by a need to prevent the CB Platform from engaging in squatting-based forms of disruptive politics, and that they used the transfer of the space as a tactic to quell protests in a jeopardised socio-political context (Rossini and Bianchi, 2020).

The emergence of Puigcerdà: Between (indirect) support and co-optation
In August 2011, a few months after Barcelona City Council granted one of the Can Batlló buildings to the CB Platform, a privately owned estate on the other side of the city was squatted by a group of 20-30 sub-Saharan immigrants (Cameroonian, Nigerian and Senegalese immigrants, among others). They had found a vacant factory in Puigcerdà Street, a vast space made up of numerous warehouses, and had moved in. This began when the liberal-conservative coalition was already in office, but the emergence of the urban commons was linked to the outbreak of the financial crisis and to the approach that the left-wing coalition had been adopting towards sub-Saharan immigrants. During the crisis, this social group, previously employed in the construction and agricultural sectors, had found itself ejected from the formal labour and housing markets. However, the left-wing coalition had not developed any policies to tackle the problem because, as one resident said, 'sub-Saharan immigrants were an invisible group whose condition never reached the public debate'. Thus, they had found a viable survival strategy by practising informal housing solutions and economic activities.
Within the walls of the Puigcerdà squat, the immigrants self-organised and managed to subsist: they built informal domestic spaces, so that their housing needs were met; they carried out informal economic activities, mainly waste-picking, but they also set up a restaurant, a second-hand shop and a bicycle workshop; they engaged in forms of mutual support, helping each other with their daily tasks. Moreover, within the walls of the factory, they began to become a visible and organised political body, understanding that they needed to take their social, administrative and political demands to the city council. Living all together in the factory was not easy: co-habitation conflicts proliferated because of the allocation of sleeping places, and because of inter-ethnic and intergenerational rivalries. However, to deal with the daily problems and conflicts that arose, they managed to build an organisational structure, based on national sub-groups where each had a representative.
At the beginning, the occupation was tolerated by the owner and by the liberal-conservative government, which decided not to send in the police. This tolerant approach was crucial in establishing this urban commons. An approach such as this cannot be framed wholly within the vision of the enabling state, as it did not provide active support to the urban commons. However, if the conservative government had not been so tolerant, Puigcerdà would most likely never have been established. This means that merely tolerating the birth and development of an urban commons can also be interpreted as a passive yet supportive governance strategy. However, it can also be interpreted as political and economic co-optation. On the one hand, the tolerant approach towards the sub-Saharan immigrants' occupation of the factory can be claimed to be a governing technique that aims to control this social group by preventing them from both bringing their social, administrative and political demands to the local government and from swelling the ranks of the city-wide political mobilisations. On the other, the sub-Saharan immigrants' occupation was used by local government to shift the social costs onto the immigrants themselves, and avoid providing any public policies to address the issue.
The development of the two urban commons under the liberal-conservative government A liberal-conservative coalition led the City of Barcelona from 2011 to 2015. It was fully aligned with the urban austerity measures imposed by the European Union and by the Spanish government, which imposed new fiscal measures and spending controls on the municipal budget (Davies and Blanco, 2017). In parallel, the disruptive political force of the 15M movement progressively faded; however, commoning initiatives that had begun to spring up during and after the mobilisation continued to expand, constantly challenging government decisions. Thus, the development of the two urban commons must be framed in this urban austerity context in which the disruption of 15M had almost disappeared, but had left modalities of collective action that were willing to challenge the economic and political order. Within this still turbulent social-political context, the conservative government adopted an apparently accommodating governance strategy towards the more moderate and lawful commoning initiativestime banks, urban gardens and self-managed cultural centres, including Can Batllóseeking to recognise them administratively and support them financially, while hindering the development of more controversial and legally ambiguous ones, such as housing and social squats, including Puigcerdà. However, also in this case, the city council's governance strategies can be interpreted in several ways.

The development of Can Batlló: Between support and co-optation
The liberal-conservative coalition decided to facilitate the development of Can Batlló and grant it an important degree of decision-making autonomy. In this way, Can Batlló became a fully functioning, self-managed cultural centre, with a meeting space, café, auditorium and library. During its mandate, the city council carried out structural work on the building to make it safe, and began paying the water and electricity bills. Furthermore, a relatively constructive relationship between the government and the CB platform grew up. Through bi-weekly meetings, representatives of the CB Platform discussed a variety of issues with the local government, including improvements to the surrounding area, and administrative and financial requests. During this mandate, additional warehouses and spaces were temporally entrusted to the CB Platform, with further spaces set up: a collaborative carpentry workshop, an urban vegetable garden, a collective brewery, a bicycle workshop, a collective printing service and a family space. Moreover, land to set up the La Borda cooperative housing project was ceded to a Can Batlló-related group on the Can Batlló site.
The Can Batlló initiative combined many different activities and projects. Each activity and project had its own governing body, but all formed part of Can Batlló, participating in its general assemblies, where representatives autonomously decided on urban commons issues together. In relation to the possible contradiction between public support and their independence, one activist said: 'what they [the city council] do is to invest in their assets, and what we do is use them and manage them as public facilities for the neighbourhood. Our use and management mean that we decide on our cultural programme, the activities and our economic model'.
The liberal-conservative government's facilitating approach favoured the development of Can Batlló, and one could claim that it represents a supportive governance strategy as advocated by Ostrom-inspired urban commons scholars. However, it could also be interpreted as a both political and economic co-optation governance strategy. While the liberal-conservative government was squeezing municipal spending, it implemented dedicated policy arrangements to deal with the extraordinary increase in self-managing initiatives in the city. Two key arrangements were Pla Buits (Vacant Spaces Plan), adopted in 2012, and Gestió Cívica (Civic Management), adopted in 2015. The former was used to temporarily grant vacant spaces owned by the city council to citizen-based organisations to carry out 'activities of public interest' (Ajuntament de Barcelona, 2013). The latter was used to transfer the management of public facilities to community organisations (Ajuntament de Barcelona, 2015). Many activists have upheld that these arrangements have twofold objectives, in line with the political and economic logic of co-optation. On the one hand, they aim to institutionalise and thus control the more moderate commoning initiatives, reducing their capacity to challenge the established order; on the other, they support citizen-provided services, thus reducing public-service delivery expenditure. To maintain decision-making independence, Can Batlló decided not to sign up to either of these arrangements. However, the 'informal' support received by the liberal-conservative government can be considered to have some of the same institutional logic lying behind it.

The development of Puigcerdà: Towards destruction
While the liberal-conservative government apparently decided to facilitate the development of Can Batlló, a very different governance strategy was adopted for Puigcerdà. Due to a combination of factors, such as the size of the space and the worsening conditions many sub-Saharan immigrants found themselves in, Puigcerdà quickly grew very large. According to one of the people who lived there, the initial number of 20-30 people over the first days of the occupation grew to 200-300 in just 1 month. Once Puigcerdà had grown to those proportions, informants say that it began to be perceived as a threat by the city council, which saw eviction as the only solution to the occupation. As one neighbourhood resident said: '[Puigcerdà] was considered a problem because it became an African town within the city of Barcelona and the administration did not like this kind of self-organisation'.
In the city council's discourse, eviction was justified because of the health and safety risks in the factory. According to a public officer at Puigcerdà 'sanitation was embarrassing as there was no water, no electricity, and there was rubbish all over the place'. This risk was perceived differently by Puigcerdà's inhabitants, who did not see it as problematic; they merely asked the city council to recognise it and help them by setting up water and electricity services. However, the city council refused to do this, and were bent on eviction. As an opposition politician said: 'the liberal-conservative government never had a personal and human attitude and never empathised with the possibility that something could have been developed there [within Puigcerdà]'. Nevertheless, the city council could not enforce the eviction, as it did not own the property. Eventually, at the end of 2011, the owner undertook legal action. Press articles gave the impression that the owner wanted to regain possession of the property to carry out an urban redevelopment project (La Vanguardia, 2012). However, various sources have confirmed that the city council put pressure on the owner to instigate the proceedings. This was confirmed by one activist who said: 'the city council pressured the owner to undertake the eviction with the excuse that it would be accountable for any accident at Puigcerdà, but the owner didn't care. It was the city council itself that truly aimed to end the occupation'.
When legal action was initiated, local social movements organisations gave their support to the immigrants' struggle. Together, they organised collective marches and events to defend the occupation but also to claim social rights (in terms of housing and working conditions), administrative rights (residency and work permits) and political (voting) rights for the whole African community in Barcelona. In the meantime, the city council set up a dedicated social care service called the Oficina del Plan de Asentamientos Irregulares -Irregular Settlements Plan -(OPAI), to deal with the humanitarian crisis that was likely to explode with the eviction. The OPAI aims to 'provide support to all people living in informal settlements in Barcelona along their path towards social, economic and administrative inclusion, guaranteeing decent living conditions, in terms of housing, employment and administrative status' (Ajuntament de Barcelona, 2012). According to the activists who supported the occupation, the city council's objective in implementing the OPAI was not to provide for the sub-Saharan immigrants' subsistence but 'to disintegrate them, disperse them'. In other words, according to them, the city council's objective was to disarticulate the sub-Saharan political subject by eradicating the urban commons.
Eventually, the eviction order was executed on 24 July 2011. Some immigrants used the OPAI. They were housed in hostels but they were not guaranteed either employment or a path to becoming legal residents. (Ajuntament de Barcelona, 2014). Some immigrants, especially the undocumented ones, tried to find other informal subsistence solutions on their own. The result of the Puigcerdà eviction and the implementation of the OPAI was that the sub-Saharan community was dynamited. They lost the possibility of collectively self-providing a wide-reaching form of subsistence and collectively building a political body to demand their social, administrative and political rights.
It is apparent that the local government crushed the Puigcerdà project, adopting the antagonist governance strategy towards the urban commons that neo-Marxist scholars usually attribute to it. This strategy seems to be partly due to the differing perceptions of risk held by the Puigcerdà's inhabitants and by the council. The council prevented the development of sub-Saharan immigrants' urban commons by insisting on reducing a health and safety risk that it perceived as serious. It seems obvious that recognising and supporting a space such as Puigcerdà would have implied both a major shift in the rationale of the local government that, like many others in the Global North, tends to identify informal practices as practices to be eradicated (when they cease to be tolerated as a matter of convenience, the initial case of Puigcerdà) (Pradel-Miquel, 2017), and a shift in the legaladministrative practice that has a rigid expert-based codification of risk (Morrow, 2019). However, the decision to dismantle Puigcerdà was also partly due to the threat that this group was posing in terms of their demands to the local government for social, administrative and political rights. The needs of the marginalised sub-Saharan population had never emerged in the public debate, and politicising them then might have entailed political risks (losing legitimacy and consensus) for the ruling coalition. Can Batlló did not need to be dismantled in the same way, since it posed no such political threat to the local government, with its members mainly asking the city council to improve the surrounding environment and to support the urban commons. These were more politically neutral demands that could be easily addressed by the local government, without needing to shift its rationale and legal/administrative practices or taking any major political risks.

The development of two urban commons under the Barcelona en comú progressive coalition
One of the long-term impacts of the 15M mobilisation in Spain was the emergence of a series of municipalist coalitions made up of existing left-wing political parties and activists that aimed to transform local politics, moving beyond the imposition of the urban neoliberal austerity paradigm. Many of these coalitions gained power in the 2015 local elections, among them Barcelona en Comú (BComú), led by Ada Colau, the former leader of the anti-eviction movement. In this way, many activists 'entered' public institutions with the idea of increasing the protective role of the state, but still maintained close relationships with the social movement organisations from which they came . The new government agenda was characterised by a twofold approach: on the one hand, an approach that recalls traditional municipal socialism through reversing previously implemented austerity policies and strengthening state-provided local public welfare; and, on the other, an approach that follows the principle of new municipalism that sees promotion and support of self-governing initiatives as a fundamental feature of city governance (Blanco and Gomà, 2020). Thus, the development of the two urban commons must be framed in this changing scenario of local politics, which was characterised by an important power shift in the local government with a consequent implementation of a new urban agenda, and the adoption of an apparently supportive governance strategy by the city council towards commoning initiatives. However, this strategy could not always be framed as supportive, and above all, it was not always applied uniformly. Even BComú shies away from facilitating the development of controversial urban commons that do not have a solid legal framework. During the first BComú mandate, some notable squats, such as the Banc Expropiat, were evicted, and a similar approach was applied to the sub-Saharan commoning initiatives.

The development of Can Batlló: Genuine support?
When BComú took power, it continued to support Can Batlló, just as their liberal-conservative forerunners had done. However, they changed the structure that this support took. Firstly, more spaces were transferred to the residents, to develop co-operative projects that had not prospered under the liberal-conservative government, such as Arcadia, a co-operative school. And very importantly, in 2019, the city council decided to modify the agreement, and grant all the spaces for 30 years, with the possibility of a 20-year extension. The BComú government justified this decision by underlining the public services and facilities being offered by the urban commons and the savings that this entailed for the municipal budget. Estimating the cost of equivalent management carried out by the public administration, they indicated that: 'if the city council were to take on the services offered by the organisation [CB Platform], it would involve a municipal investment of 1.43 million euros per year […] while a total of 1.1 million euros of public investment has been invested in Can Batlló since 2011' (Ajuntament de Barcelona, 2018).
This statement would appear to imply that the BComú government is adopting a co-optative governance strategy towards Can Batlló by implementing an austerity urbanism approach. However, the idea of transferring the spaces on the Can Batlló site to the CB platform for thirty consecutive years does not appear to arise from an expenditure-saving initiative, but from the political need to strengthen the self-managing capacity of Can Batlló. This need is part of a more general vision held by a city council that aims to support the various urban commons in the city. In 2017 the BComú government adopted a new legal framework to achieve this objective: Patrimoni Ciutadà (Citizen Assets). This aims to support self-managed initiatives (urban gardens, cultural centre, time banks, etc.) by promoting them but allowing them to maintain their diversity and independence (Ayuntamiento de Barcelona, 2017b). Moreover, this approach has been supported by a substantial increase in public spending: in 2011, the city's public expenditure per inhabitant was €1383, while in 2018, it was €1691 (Gobierto, 2017). Thus, the BComú government's support of Can Batlló seems to fall more into the supportive governance strategy than into the co-opting one. The rationale is not actually part of an austerity-driven model of urban governance, but a model that rejects the austerity measures implemented by the previous administration. However, even in this case, interpreting such a governance strategy as co-optative cannot be entirely dismissed, as it did shift the cost of social reproduction onto a community.
The evolution of Puigcerdà: An ongoing lack of support? While many commoning initiatives have met with the BComú government's apparent support, this cannot be said for the most controversial commoning initiatives with ambiguous legal status. During the first BComú mandate, the city's sub-Saharan immigrants' living conditions remained inadequate, and they had to find informal forms to subsist. To do so, they stopped squatting in big industrial factories in the Poblenou area, and began to occupy single flats in the Barcelona periphery. Some informal settlements persist in Poblenou, but with significantly fewer residents (ranging from 6 to 10 people). The BComú government's programme addresses the needs of vulnerable groups, trying to deal with the informal activities of these groups in a less repressive way and offering social rather than legalistic solutions (Ajuntament de Barcelona, 2017a). However, the BComú government does not appear to have acted very differently from its forerunner in relation to the sub-Saharan immigrants' urban commons. Both neighbourhood residents and activists claim that if Ada Colau's government had had to deal with an occupation such as that of Puigcerdà, it would not have been able to stop the eviction. In fact, as a neighbourhood resident said: 'the BComú government had already had to resolve some cases of informal sub-Saharan settlements, albeit smaller ones, such as in Calle Pamplona (Poblenou) and, in the end, it did so in the same way as its predecessors'.
It can be seen that while the BComú government was apparently enacting a supportive governance strategy towards many urban commons in the city, this strategy was not adopted in the case of the urban commons that were most controversial and that lacked a legal framework, such as those of sub-Saharan immigrants. Whilst it is politically and administratively simple to support legally clear-cut commoning initiatives run by moderate active citizens, such as Can Batlló, it is far more complex, even for BComú, to support controversial urban commons whose status is legally ambiguous, especially those made up of marginalised immigrants. The support of the council was crucial in improving the material and safety conditions of a potentially risky commons, but this support would have involved implementing major changes in legal and administrative practices to deal with the legal-administrative limitations (health and safety regulations) that need to be adequately addressed for urban commons to thrive. Another obstacle was that, just like for the previous council, this support might pose a political risk for BComú, since it could be heavily criticised by the opposition, discrediting the work of its government in general. The fact is that, despite their possibly good intentions or their apparently different rationale, BComú adopted an antagonist governance strategy towards the sub-Saharan immigrants' urban commons, similar to that of the former liberal-conservative government.
Understanding the urban commons-local state relationship through relational state theory The comparative analysis carried out in Barcelona of the two urban commons and the repertoires of governance strategies adopted by the local government regarding each of them permits the corroboration of the three propositions on the urban commons-local state relationship developed previously based on relational state theory.
Firstly, as suggested by relational state theory, the repertoires of governance strategies that the local state takes on in the emergence and development of the urban commons may sometimes seem to overlap; they do not have a clear-cut interpretation and cannot be ascribed to a unique rationale. Support and co-optation, for instance, cannot always be understood as two wholly different and mutually exclusive repertoires of the local state, but as two overlapping ones. Both the Can Batlló and Puigcerdà cases demonstrate this. The emergence of both urban commons may be interpreted as a directly or indirectly supportive repertoire of governance strategy or as a co-optative one. This relativism does not mean that both interpretations are always possible, but when they do seem to overlap, different interpretations are possible depending on the analytical focus used. If we focus on understanding how each urban commons emerged, we see that neither of them would have been established without the initial direct or indirect support of Barcelona City Council. If we focus more on critically understanding why each urban commons emerged, Barcelona City Council's initial support can be interpreted as a form of co-optation in both the economic and political sense. It is true that examining the state's approach within a broader analysis of the city's governance can help us understand which repertoire of governance strategy is predominant, but it will not necessarily exclude the other.
Secondly, the repertoires of governance strategies that the local state takes on in the emergence and development of the urban commons evolve considerably over time according to the challenges posed by the socio-political context present and the need to pursue specific state projects. This evolution sometimes, but not always, depends on changes in the local political forces. Some might be committed to pursuing neoliberal austerity programmes; others, some kind of socialist project. If we interpret the development of Can Batlló as a predominantly political and economic co-optation repertoire of governance strategy adopted by the liberal-conservative government, aiming to appease social conflict in the city while pursuing a neoliberal austerity agenda, then with the arrival of the progressive government, the repertoire enacted by the new local government can be interpreted as a mainly supportive one. This is due to BComú's close relationship with commoning initiatives and its commitment to implementing progressive city policy. Nevertheless, this evolution also depends on changes in the contingent economic and political needs of the local state, independently of who holds the majority. The Puigcerdà urban commons can be interpreted as the adoption of a of co-optative repertoire of governance by the liberal-conservative government: the initial development of the commons was politically and economically convenient for the council, but once it became a political threat, that is, once the Puigcerdà informal space grew too large and its residents began demanding social, administrative and political and rights, the same administration decided to destroy it. This governance strategy did not change even with the arrival of progressive forces to the council.
Thirdly, the repertoires of governance strategies that the local state takes on in the emergence and development of the urban commons are applied selectively. The local state does not act in a uniform way with all the urban commons in the city, but appears to adopt one repertoire or another depending on each social group's composition, political demands, activities, whether they fit into the local state's rationale and legal/administrative practice, and if they represent a political risk. Our two cases show this very clearly. Facilitating the development of Can Batlló did not involve any major changes in the local state's rationale. Dealing with low-and middle-income active citizens who demanded to self-manage a cultural centre and providing recognition and (limited) economic and administrative support for this project fitted in nicely with both the conservative government's neoliberal austerity agenda and the progressive government's collaborative one. Nor did it involve any major challenges to the local state's current legal/administrative practices. Conversely, Puigcerdà, an informal space constituted by a marginalised social group, was never directly supported by any local government, since such support would have involved an important change in both the rationale and the legal/administrative practices of the local state, which would have had to deal with ambiguous legal structures in a way that is alien to how Western governments usually act. Moreover, dealing with socially, administrative and politically marginalised people that might, as happened in this case, increase their demands to gain rights could be extremely risky, even for a progressive government like BComú, especially in these current times in which the integration of immigrants has become a thorny political issue rolled out in the propaganda of reactionary rightwing parties to discredit progressive parties and governments. This risk is not posed by an urban commons such as Can Batlló, support for which is not a hot political issue for parties, and is something that can even help broaden the consensus of both right-wing and left-wing parties within the centre-left electorate.
The empirical evidence emerging from the case studies thus confirms that the limits between the repertoires of governance strategies adopted by the local state in the emergence and development of the urban commons are more nuanced than what is reflected in either neo-institutionalist or neo-Marxist literature.

Conclusion
According to the literature, the state can take on different repertoires of governance strategies towards the commons: antagonistic, co-optative and/or supportive. The theorisation of these repertoires tends to correspond to the ideological positions of those who uphold them, with neo-Marxists seeing the state as an antagonistic and/or co-optative institution, and neo-institutionalists seeing the state as supportive. As these perspectives tend not to enter into dialogue with each other, the aim of this article is to delve into the varieties, ambiguities and complexities of the repertoires of governance strategies adopted by the local state in the emergence and development of urban commons. It does so by engaging with relational state theory and analysing two different types of urban commons in the city of Barcelonathe Can Batlló self-managed cultural centre and the Puigcerdà housing squat, all the while embedding this analysis in Barcelona's ever-changing political context. Following the theoretical contributions made by relational state theory that has helped us to nuance the boundaries of commons' scholars' siloed theorisations of commons-state relationship, we have tried to underpin such contributions with the empirical evidence provided by the two cases. This empirical evidence confirms the suggestions provided by relational state theory, that is, that the boundaries between the different repertoires of governance strategies that the local state may take onantagonistic, co-optative or supportiveare far more nuanced than implied by the somewhat siloed nature of commons literature. These repertoires may overlap, evolve throughout time and be applied selectively to different urban commons. Considering the important role that the local state plays in the emergence and development of urban commons in the dense and contested urban space, it is necessary to continue to analyse such nuances in different urban commons, over varying historical periods and in diverse urban contexts, privileging crosscase comparisons. Only through an empirically informed knowledge of urban commons-local state relationships articulated adequately with state theory will we be able to advance our understanding of the potential scope and limitations of urban commons-led collective action in cities.