Governing capabilities, not places – how to understand social sustainability implementation in urban development

Social sustainability’s implementation in urban development is a complex endeavour that demands alternative forms of governance. This article draws on the capabilities approach as an evaluative framework to better understand this implementation process. Through an in-depth case comparison of two Dutch urban development projects, the study analyses how collaborative governance situations (i.e. actors, activities and phases) relate to the expansions of resident capabilities in the urban areas. The findings present three principles for a ‘capability-centred governance’ of social sustainability in urban development: (1) integrate human logic into urban governance situations (2) balance strong goal commitment with experimentalist approaches and (3) institutionalise social sustainability implementation. The article concludes that social sustainability’s implementation requires a conceptualisation in which improvements in people’s lives are not seen as the self-evident consequences of a set of place-based policy interventions, but instead as a guiding principle that should continuously be reflected upon and learned from during the different phases of urban development processes.


Introduction: The problem of implementing social sustainability goals
Social goals have recently regained attention in urban policy-making.In global and regional policy programs, terms like inclusiveness and diversity are widely cited (European Commission, 2019;United Nations, 2015).These social policies are responding to globally increasing urban inequalities, manifested through for example, a lack of access to urban services in the global south (Mitlin and Satterthwaite, 2012) and increasing socioeconomic segregation and social exclusion in the global north (Andersen and van Kempen, 2019;Musterd et al., 2017).While cities have been addressing the latter issues via urban policy interventions such as regeneration programmes, improved public spaces and social mixing, their various effects have long formed a centre of critique in urban studies, (see e.g.Davidson, 2011;Lees, 2008;van Gent et al., 2018).Indeed, enduring processes of displacement and gentrification give valid reason to dispute whether social policy interventions are achieving what they intend to achieve.
Complementing enduring urban debates on the effects of different social policies on people's lives, the notion of 'urban sustainable development' offers an alternative conception towards more inclusive cities (Elkington, 1997;Marcuse, 1998;Satterthwaite, 1997).While often understood as one of the conditional dimensions of sustainability, social sustainability is a plural concept containing various value-laden criteria such as social equity, human well-being and quality of life (Dempsey et al., 2011;Rashidfarokhi et al., 2018).Its 'sustainability' part emphasises longitudinality: not only should we care about the needs of people who live in cities now, but also of the ones in the future (Satterthwaite, 1997).In other words, social sustainability does not necessarily strive after immediate policy outcomes, but rather after a positive condition in urban spaces that also very-well develops 'by itself' (Bostro¨m, 2012).
Because of these characteristics, the pursuit of social sustainability in urban development does not align with urban policy goals as static, tangible objectives, but instead as dynamic and ever-changing targets that adapt according to changing generations of residents in urban areas (Dempsey et al., 2011).This conception challenges conventional policy implementation rationales in which policy interventions are designed, realised and evaluated in linear and controllable ways (Hill and Hupe, 2002).Instead, social sustainability 'requires a focus on [...] on-going processes and interactions which continuously constitute the social life in a neighborhood and the relations between the residents' (Langergaard (2019: 468).Social sustainability is therefore also often seen as the process to achieve its desirable condition (Bostro¨m, 2012;McKenzie, 2004).Just as scholars have been arguing that sustainable development requires new ways of collaboration (Kotzebue, 2016;O'Toole, 2004), social sustainability's implementation is a complex endeavour that demands alternative forms of governance with new interdependencies between public, private, volunteering and civic actors (Manzi et al., 2010).
In this article, we draw on Sen's (1999Sen's ( , 2009) ) capabilities approach to offer a conceptualisation of dominant policy-implementation-outcome rationales in urban practices that does not only focus on the realisation of spatial interventions -such as mixed housing or improved public spaces -as desired policy outcomes.Instead, as argued elsewhere (Janssen and Basta, 2022;Janssen et al., 2021), the capabilities approach provides a promising evaluative perspective on social sustainability's implementation as a governance process that is oriented towards the expansion of human aspirations and opportunities (i.e.capabilities) as actual end goals.We specifically zoom in on urban development projects (UDPs) as governance vehicles of urban policy implementation, as such projects can be seen as 'strategic devices ... [that] attempt to settle or to stimulate certain joint courses in individual actions' (Salet, 2007: 3).Thus, we consider UDPs as situations of collaborative governance in which (1) public, private and civic actors work together towards common goals (Fainstein, 2008;Healey, 2006) and (2) in which social sustainability goals undergo a process in which abstract policy ambitions are translated into a set of concrete place-based policy interventions (i.e.place interventions).
Our aim is to better understand social sustainability's implementation by analysing how exactly collaborative governance situations around UDPs relate to expansions of residents' capabilities.Through an interviewbased qualitative case-study analysis, we compare two UDPs with social sustainability goals in the Netherlands, first evaluating how residents were able to convert implemented place interventions into capabilities, and second, analysing the respective governance situations in these UDPs related to these conversions.Subsequently, we reflect on our empirical findings and identify three principles for collaborative governance that advance a 'capability-centred' approach to social sustainability's implementation.
The next section connects theoretical concepts of the capabilities approach with urban development projects and collaborative governance, and results into the empirical question for our case analysis.Section 3 then elaborates on the analytical framework, case selection and methods.Section 4 presents the empirical findings of the case analysis, which are subsequently discussed and interpreted in section 5. Our final conclusions and reflections are collected in section 6.

Theoretical background: Advancing human capabilities through UDPs
Over the last decade, Sen's (2009) capabilities approach has received a rapidly growing interest in urban studies dealing with social justice and well-being issues (see e.g.Basta, 2017;Fainstein, 2014).Indeed, its individual perspective has proven to be useful for devising evaluative frameworks that aim to centralise human capabilities -i.e., the real opportunities an individual person has to do the things that he/she has reason to value -as a normative direction (Alkire, 2008;Biswas, 2019).Applied to the urban field, this idea resonates in studies defining evaluative spaces as 'spatial capability' (Shin, 2011), 'opportunities to travel' (Cao and Hickman, 2019), or 'housing capability' (Kimhur, 2022).In the same line of thought, we here define social sustainability outcomes as 'urban capabilities', for example, the opportunities an individual person has to perform those functionings that he/she has reason to value for a worthy life in the urban place where he/she lives.
Arguments in favour of the capabilities approach are, despite its essentially critical perspective, of a pragmatic nature in the sense that what happens in 'real life', is what really matters.Hence, 'having the right' to do something, does not say anything about whether a person is actually able to exercise that right (Basta, 2017).The approach incorporates the many diverse individual circumstances of a person, for example, gender, age, cultural background and social environment (Sen, 2009) and is, therefore, sensitive to the complexity in which social injustices occur in real situations in daily life.Accordingly, Fainstein (2014) positions the capabilities approach as a tool to deal with the incompatible core values of the just city (diversity, democracy and equity) and as a useful way 'to devise rules that can govern the evaluation of urban policy' (p.13).
Its acknowledgement of the contextual conditions affecting a person's life is indeed a key feature of the capabilities approach, making it highly relevant for the implementation of social sustainability goals in urban development.These conditions explain, for example, why place interventions do not always lead to intended social outcomes.For example, when a new public park is realised in a neighbourhood, this does not necessarily mean that residents of that area equally experience increased well-being.A young boy who first needs to cross a dangerous road to arrive there, a teenager who sees the playground continuously being occupied by small children, or an adult that simply does not care much about green spaces, may not experience any impact on their lives because of the new intervention.
It is thus not the place intervention itself that should be evaluated, but the conversion into capabilities that a resource like a park evokes for different types of individuals.For this reason, Robeyns (2005) argued that capability studies should not only focus on personal abilities and aspirations, but also pay attention to institutional circumstances, such as social norms and traditions.Next to personal abilities and characteristics, these embedded collective structures greatly affect the way that people convert resources into capabilities (Biggeri and Ferrannini, 2014;Deneulin et al., 2006;Frediani, 2021).If places are understood as productions of not only spatial but also particularly social processes, many new conditions could be discovered inside the political-economic and institutional-historical processes that relate to people's access to and agency in cities (Frediani, 2021).
The governance processes that shape urban development projects -understood as planned and organised place interventions that involve numerous relational efforts (Adams and Tiesdell, 2012) -are full of place-specific institutional and politicaleconomic conditions that function as enabling or disabling factors for people's urban capabilities.Indeed, scholars have shown how places are increasingly shaped through assemblages of activities between the realms of urban politics, planning, real estate and civic engagement (Healey, 2006).These practices may be critically understood to reflect a shift from 'government to governance' in urban practices (Harvey, 1989;Taylor, 2007), but, more recently, as collaborative efforts that pragmatically strive to balance market logics with public policy and local civic objectives through place-based interventions (Kim, 2023).Seeing UDPs as collaborative governance processes (Ansell and Gash, 2008;Emerson et al., 2012) thus implies that urban transformations are not only the result of 'planned' policy interventions, but also shaped by the various activities of and interactions between public, private, and societal actors.
The potential of collaborative governance for social sustainability's implementation lies in the possibility to, through collaborative efforts, create the specific conditions that can function as enabling to human capabilities.Collaboratively delivered UDPs create the opportunity to have more inclusive deliberation processes (Ansell et al., 2020;Kim, 2016) and so, may lead to more inclusive urban outcomes.Although disadvantages of collaborative governance -such as higher transaction costs, the threat of reducing the quality of deliberation (Ansell et al., 2020), or the risk of ignoring an unequal distribution of power between participants (Swyngedouw, 2005) -have been marked in literature, its democratic potential, as well as its instrumental asset to 'substantively better, more widely supported, more robust and innovative processes and solutions' (Ansell et al., 2020: 570) are widely acknowledged (Ansell et al., 2020;Emerson and Gerlak, 2014;Healey, 1998).
Despite its potential, the question remains how precisely collaborative governance can play a role in expanding human capability through urban development.Acknowledging that the more participation is not per se the better (Ansell et al., 2020)

Method
This study addresses the relations between the governance situations around urban development projects and residents' conversions of place interventions into expanded urban capabilities.We decided to perform an interview-based qualitative case-study analysis of two urban development projects in order to build detailed insights into how the activities of different actors led to these specific urban outcomes.This section elaborates on how the case studies were selected and conducted.

Analysing social sustainability implementation in UDPs in the Netherlands
We conducted our analysis in the Netherlands, which can be seen as an example of a context where urban development projects typically address, among others, social policy goals.Comparing to cities in Anglo-Saxon nations, scholars have exemplified Dutch cities for their governmental commitment to public benefits (Fainstein, 2008) and to issues such as livability, diversity and multiplicity (Healey, 2006).In addition, UDPs in the Netherlands typically emerge as processes of collaborative governance.Since the beginning of the 21st century, regulatory changes and spatial policy reforms diminished the role of the national government in urban development and the responsibilities of regional and municipal governments grew -mirroring a shift in the wider political economy from a social welfare state to a more liberal model (Van Loon et al., 2019).After the financial crisis in 2008, Buitelaar and Bregman (2016) explain how the pillars under the Dutch 'planner's paradise' were trembling, and how a lack of integration of land uses, actors and financial sources changed the practice of UDPs into a landscape of incremental development initiatives with a greater role for individual private initiatives.Currently, a reappearance of state control dominates planning debates in the Netherlands, as new relationships between tiers of government, private actors and third sector organisations are being contemplated (Verdaas, 2019).
The case-study presented in section 4 is analysed through the framework in Table 1.We developed this framework based on a common understanding of UDPs in the Netherlands (Ministry of Infrastructure and Water Management, n.d.) that distinguishes four governance phases to which specific activities and actors belong.These elements form the governance situation around a UDP, which we will analyse in section 4.2.We should note that this framework does not necessary imply linear progress -the four phases may overlap or iterate in practice.In addition, the framework incorporates the typical urban outcomes that are produced per phase.In our evaluative analysis in section 4.1, we focus on the conversions between the outcomes in the realisation and the utilisation phases (i.e.place interventions and urban capabilities).For this study, we thus interpret urban capabilities as people's actual opportunities to utilise implemented place interventions in UDPs.Although we recognise that other interpretations of urban outcomes -such as people's capability to coproduce development plans in the earlier phases -are also highly relevant for achieving more socially sustainable urban developments, we left this interpretation outside the scope of our research.

Buiksloterham&Co and Nieuw-Crooswijk
As cases, we selected two UDPs that had explicit policy goals complying with the concept of social sustainability, to analyse how such goals were translated into specific place interventions.One project is Buiksloterham&Co in Amsterdam, a brownfield development in a former harbour area that is part of a wider transformation zone (i.e.Buiksloterham) and allocated as an experimental area for 'circular urban development' (Gladek et al., 2015).The second project is Nieuw-Crooswijk in Rotterdam, a regeneration area with mostly social housing, where the main goals concerned housing diversification and improvement of safety and livability in the area (Bureau Herstructurering Nieuw Crooswijk, 2003;OCNC, 2005).
Both cases are large-scale UDPs that were carried out in the first two decades of the 21st century.The plan for Buiksloterham&Co Moreover, both projects are examples of 'concessions' in which social housing organisations -in the Netherlands, these are statesupported non-profit companies -developed the land in collaboration with private developers, facilitated by the local government.In Buiksloterham&Co, the social housing organisation acted as the main developer of the land and subcontracted a private real estate developer to develop one-third of the land.In Nieuw-Crooswijk, the social housing organisation formed a consortium with two private real estate developers to develop the area.In 2012, however, the social housing organisation left the consortium and instead took a leading role in managing the renovations.

Data collection and analysis
The case-study material comprises information in the projects' planning documents and semi-structured interviews with relevant actors.Between June 2020 and June 2021, 61 interviews were conducted with professionals involved in the UDPs (18 in Buiksloterham&Co and 16 in Nieuw-Crooswijk) and with residents living in the urban areas of the project (14 in Buiksloterham&Co and 13 in Nieuw-Crooswijk).Most interviews with professionals were held online, while most interviews with residents were held on-site at the homes of participants or in public spaces near to it.Interviewed professionals were selected based on an assumed representation of the different actors involved in the governance processes: among others the municipality, social housing organisations, real estate developers, urban designers, healthcare organisations and civic groups.Interviewed residents were selected based on the sole criterium that they lived in the urban area of the project, thereby leaving room for a diverse representation in terms of age, gender and social, economic and cultural background.Different entry points were used to recruit interview participants, such as through personal references by professionals of the social housing organisation, the 'snow-balling' effect of asking participants to ask their neighbours, addressing persons on the streets, posting a call in a local runners' group on social media, and by putting flyers in mailboxes of houses.
The aim of the interviews with professionals was to bring the governance situation of social sustainability's implementation in the UDP to light by structuring the interviews around questions how professionals defined the project's goals for social sustainability, how these goals were translated into place interventions, and how they perceived collaborations with other actors.In contrast, the aim of the interviews with residents was to analyse how residents experienced the place interventions and to what extent these contributed to residents' expanded capabilities.
The analysis presented in the next section consists of two main steps.The first concerns the evaluations on residents' conversions of place interventions into expanded capabilities.Based on general information in planning documents and the interviews with professionals, we identified four main place interventions for social sustainability in Nieuw-Crooswijk and five in Buiksloterham&Co.Accordingly, while listening in the recordings of the resident interviews, all quotes in which residents referred to those interventions were collected, leading to an overview of the diverse capability outcomes and conversions that residents experienced (presented in section 4.1).
As a second analytical step, we used the interviews with professionals to analyse the governance situations around the UDPs.As the cases were too large to reconstruct a complete picture of the governance situations for all interventions, we decided to focus on realised place interventions as 'community meeting spaces'.We selected these spaces because they well-illustrate the analytical distinction between the place intervention, for example, the realised building or room, and the capabilities that such spaces facilitate, like increased opportunities to meet new neighbours.For this analysis, we transcribed the interviews with professionals and collected quotes about the governance situation around the community spaces in Atlas.TI.This resulted in a narrative description of the implementation processes (section 4.2.1) and an analysis of how the different activities, actors and phases related to residents' conversions of the community spaces into expanded capabilities (section 4.2.2).

Case findings
The findings of the case-study analysis are presented in two parts.The first part focuses on the evaluation of urban outcomes by providing empirical evidence of how residents' conversions of different place interventions into human capabilities took place.The second part discusses the governance situations around a selected type of place interventions, and describes in detail which elements of these situations related to the conversions observed.

Evaluating urban outcomes: Place interventions, conversions and capabilities
Based on the resident interviews, Table 2 lists examples of how residents in the two urban areas were able or unable to convert place interventions into capabilities.Per realised place intervention, the table lists one or multiple capability outcomes -that is, how residents perceived the impact that the place interventions had on their personal livesand the conversions that played a role in these outcomes.
An important finding is that the place interventions led to distinct capability expansions-both positive, neutral and negative -for different groups of residents.In some cases, such as the shared facility rooms in example 7 in Table 2, this diversity relates to individual preferences: whereas some residents saw the realised shared facility rooms as a way to meet neighbours, others mentioned that they simply did not prefer to socialise in such rooms and prefer meeting others in alternative ways.In other cases, however, such as example 4, the distinct capability outcomes more evidently relate to the selective scope of the place intervention.Here, the closed fence around the collective gardens facilitated additional opportunities to recreate for the residents of the housing block, but not for residents of other parts of the neighbourhood.In other words, the design of this place intervention yields an unequal expansion of resident capabilities.
Selective and therefore unequal outcomes were more often observed in Nieuw-Crooswijk (see examples 2, 3, 4, 5) than in Buiksloterham&Co.This can partly be explained by the fact that Buiksloterham&Co was still under construction at the time of data collection, while the transformation of Nieuw-Crooswijk was almost finalised at that time -in later stages, distinct effects might be observed in Buiksloterham&Co too.Nonetheless, it became clear from the interviews that residents in Nieuw-Crooswijk felt affected by the emerging inequalities in the area, -in particular between 'existing' and 'new' residents -due to the design of the masterplan.The decision in this masterplan to demolish a large part of the existing social rental housing stock and to reconstruct the area with owner-occupied dwellings (see example 1), induced strong feelings of displacement amongst residents.Subsequently, the impact of this intervention continued to dominate residents' perceived evaluations about other place interventions, such as the shared gardens (example 4), the cafe´(example 3) and the municipal budgets for resident initiatives (example 5).Since the transformation area in Buiksloterham&Co was previously non-residential, displacement issues here played much less of a role.

Governance situations: Phases, activities and actors
The evaluations above show how implemented place interventions led to both successful and failed conversions for residents.It is therefore crucial to assess what specific conditions in the UDPs affect one or the other.
To do so, we reconstruct the governance situations of both cases and analyse how -not observed yet -Residents doubtfully waiting how the integration between residents with a lower and higher incomes will work out 7 Shared facility rooms including laundry machines in social housing blocks The central locations at the entrance of the buildings and the multiple functions such as a laundry machine, coffee machine, library, and bike repair, facilitate physical encounter among residents.
-Residents experiencing increased opportunities to interact and engage with neighbours -Some residents not experiencing any social effects by the intervention 8 Shared gardens for social housing blocks and selfmanaged semi-public inner gardens for homeowners The semi-public gardens compensate for the lack of public green spaces in the neighbourhood; however, they have a closed character.
-Residents not feeling welcome in semi-public gardens and worrying about the lack of green space to let their children play outside 9 Coaches allocated in facility rooms in social housing blocks as communitybuilders The intensive presence of the coaches in the shared facility rooms (7 days per week, 10-12 hours per day), make them directly visible and approachable in residents' living environment.
-Residents feeling supported in their own initiatives and questions -Residents appreciating that they have a central contact point to approach and an external party that looks after the neighbourhood cohesion -Some residents not experiencing any effects by the coaches these situations related to the capability outcomes.We focus on a selective set of place interventions (examples 2, 3, 7 and 9 in Table 2), which are forms of 'community meeting spaces'.First, section 4.2.1 describes how these community spaces were implemented throughout the different phases of the project, and which actors were involved.Accordingly, section 4.2.2 assesses how precisely the configuration of governance activities, actors and phases in both projects related to the conversions observed around the community spaces.
Implementing community spaces.In Buiksloterham&Co, the eventual realisation of the shared facility rooms in the social housing blocks where coaches are located (examples 7 and 9 in Table 2) can be traced back to the design of several 'urban activators' in the operationalisation phase by the urban designer.These activators were envisioned as buildings that provide clustered office spaces for societal, non-commercial service providers (Studio Ninedots, Delva Landscape Architects and De Alliantie, 2015).In the transition from operationalisation to realisation phase, however, the idea turned out not to be financially feasible for the social housing organisation.As an alternative, during the realisation phase, the social housing organisation started to collaborate with a healthcare organisation that was selected to rent a number of apartments in the new social housing blocks.They negotiated that, if the social housing organisation created community meeting rooms on the ground floors of the buildings, the health care organisation would operate these rooms in the utilisation phase and allocate coaches to them who perform community-building support for all residents, including other residents in the neighbourhood.In return, the healthcare organisation was allowed to use the rooms as a daycare facility for clients who did not necessarily live in the neighbourhood.The shared facility rooms were realised and operated as such.
In Nieuw-Crooswijk, the school building and the cafe´in the urban area (examples 2 and 3 in Table 2) were the results of different implementation processes.The school building can be traced back to the initiation phase in which early planning documents already included an idea of a centrally-located, 'multifunctional' primary school building that serves multiple educational, pedagogical and communal functions (Bureau Herstructurering Nieuw Crooswijk, 2003).Accordingly, this idea was integrated into the masterplan during the operationalisation phase and was successfully constructed in the realisation phase.Because of a delay in housing construction due to the global financial crisis in 2008 and subsequent economic recession in the Netherlands, the school operators changed in the utilisation phase.Few new families had arrived to the area, for which the first operator faced a lack of children and the municipal education department decided to change the operator to one that was in more need of space.
Parallel to this, the social housing organisation came up with an idea during the realisation phase, after the social housing organisation had stepped out of the developer's consortium and wished to strengthen the dialogue with residents.The social housing organisation commissioned a resident architectural research consultant to conduct a study on local needs in the area, which eventually led to a cafe´run by three residents.During the utilisation phase, the social housing organisation facilitated this cafe´by offering below-market rents in the first years of operation.

Governance
situations affecting capability conversions.As shown in Table 2, the implemented community spaces in the UDPs led to both positive and negative capability outcomes.In this section, we assess which elements of the governance situations precisely related to the successful or failed conversions observed around these spaces.Following our framework in Table 1, we draw out the specific configurations of governance activities, actors, and phases related to these conversions.
A first general observation is that particularly negotiations between developing actors and utilising actors about the design and operation of place interventions were fruitful for achieving resident capability conversions.In Nieuw-Crooswijk, the collaboration between the social housing organisation (as a developing actor) and a resident architectural research consultant and local residents (as utilising actors) led to new possibilities for residents to participate in the project and to realise their own ideas for place interventions (namely, the cafe´).In Buiksloterham&Co, the negotiation between the social housing organisation as a developing actor and the healthcare organisation as a utilising actor led to an intensive management of the facility rooms through the availability of coaches 7 days per week, 10-12 hours per day.This created the opportunity for them to observe locally-emerging needs and to provide tailored-made support that facilitated capability conversions (example 9 in Table 2).Such support consisted of practical things such as a laptop that enabled a resident initiative to start language lessons in the facility rooms, or personal conversations to motivate residents to realise their ideas.From a capability perspective, our findings thus show that not only large policy efforts, but small-scaled interventions also particularly make a difference in people's daily lives.
Second, we observe that sociallyprincipled collaborations stretching over the realisation and utilisation phase were advantageous for capability expansions.In such collaborations, the intended social principles behind a place intervention are warranted in the transition from its realisation to its actual utilisation.This was for example, the case in Buiksloterham&Co, where the health care organisation was enabled to perform activities after the spaces were realised, and in Nieuw-Crooswijk where a lower rent during utilisation enabled residents to run the cafe´.Moreover, an observed advantage of stretching socially-principled collaborations over the realisation and utilisation phases, was that this created the possibility for developing actors to adjust design decisions according to the perspectives of actors who later utilised the intervention, such as the health care organisation that helped to design the interior of the shared facility rooms.Opposed to those examples, the school building in Nieuw-Crooswijk provides evidence of a collaboration between actors in the realisation and utilising phase where such intended social principles were not warranted.After the school's construction, the developer's consortium delivered and transferred it to the municipal education department.The subsequent decision by this department to change the school operator was fully reasoned from this actor's perspective, who did not have a specific interest into the goals of the urban development project.As a consequence, the new operator was a school with a religious profile that attracted children from all over the city -not necessarily from the neighbourhood -and the school building was run as a regular school without the intended 'multifunctionality' for the neighbourhood (see example 2 in Table 2).
Our final observation is that goal-setting activities in the initiation phase influence whether place interventions lead to resident capability expansions or not.This became clear from the differently-phrased ambitions and the eventually-realised place interventions in the two projects.In Buiksloterham&Co for example, interviewees emphasised that they were only able to realise the facility rooms with the coaches because the team was kept small and the project was kept 'under the radar' of the larger project administration.While the professionals thus feared the initiative to fail if it had to be justified by 'higher-level' decision-makers, an earlier manifest in the initiation phase helped them in this justification.Because it was signed by 24 organisations, the manifest created a commitment among project actors to innovate (and experiment) on sustainability and circularity goals.On the contrary, Nieuw-Crooswijk showed that too narrow or not commonly shared definitions of social sustainability goals in the early project phases withheld developing actors to implement interventions that could support capability outcomes.As the project goals resonated with a municipal policy to attract higher income households to the city by improving the livability of urban areas and diversifying the housing stock, interviewed developers defended their place interventions by arguing that improving individual life situations had never been a goal of the project.

Discussion: Towards principles for capability-centred governance
The case comparison presented above provides a number of insights about the relations between governance situations in Dutch urban development projects and residents' conversions of place interventions into capabilities.The findings reveal that for pursuing urban outcomes as human capabilities, -thus not only as delivering place interventions -, several collaborative governance activities are pivotal, namely: negotiating about the design and operation of place interventions between developing and utilising actors, upholding social principles in the transition from realisation to utilisation phase, and setting jointly shared human-oriented goals during the initiation phase of projects.These empirical insights allow us to reason more generally on how the implementation of social sustainability in urban development can be effectively governed.By discussing our findings in light of wider urban debates, we here present three principles of a 'capability-centred governance' for social sustainability in urban development.

Principle 1: Integrate human logic into urban governance situations
On the most fundamental note, our empirical findings reveal a disparity between the logic of the governance situation around a UDP on the one hand, and the logic of what is needed for a person's quality of life on the other hand.By identifying the shortcomings in how implemented place interventions in urban projects impact the perceived opportunities of residents, our analysis proves the relevance of a capability-centred evaluative perspective to urban outcomes.Although our analysis reveals examples of how such a perspective can be pursued through smallscale, tailored support additional to planned place interventions, it also showcases the formal barriers in the governance processes to actually provide such support.When reflecting on our interview questions, interviewed professionals brought forward various procedural, financial and administrative obstacles to realising interventions that were not formally planned for in the UDP.For example, when residents asked for a small budget to create a small self-managed garden in public space, the municipality found it difficult to do so since it was not formally owning the space during the UDP.Moreover, whereas our analysis points out how especially utilising actors who manage or operate places, such as schools, healthcare professionals, social workers or residents themselves, play a crucial role for providing such tailored support, these actors were not formally part of the governance situations around the UDPs studied.
Fundamental questions about whether cities cater the actual needs of people (Cardoso et al., 2022) or whether the locus of social policy should be shifted towards more people-centred objectives (Lees, 2008) keep returning in urban debates.The position that we take here is that, if human needs are effectively addressed through the activities of actors that actually utilise the urban spaces developed -that is, those who act according to 'human logic' -, these activities should become an integrated part of the various political, economic and juridical governance activities that shape an UDP.Indeed, building on critical urban theorists who showed how structural forces of capitalism and neoliberal politics use UDPs as conduits of capital accumulation and urban consumerism (Swyngedouw et al., 2002), it can be argued that the collaborative model is overestimated as long as such forces 'produce endemic social conflict and domination by the powerful' (Fainstein, 2000: 455 in Kim, 2016: 3548).Certainly, the governance situations we studied also show clear power imbalances among actors.It is particularly for that reason that take a critical-pragmatic position and pursue the question how a capability perspective can improve the existing political-economic play of prevailing actors in urban governance situations.
Principle 2: Balance strong goal commitment with experimentalist governance Second, our empirical results signal a tension between, on the one hand, encouraging flexibility in the governance situation around a UDP in order to adapt place interventions according to emerging human needs, and on the other hand, having a strong commitment to pre-defined goals.Due to the diverse ways in which people convert place interventions into capabilities, it is a misconception that one can 'predict' urban capabilities in the early phases of a UDP.As often mentioned in studies on collaborative approaches (Gaete Cruz et al., 2022;Kim, 2016) incrementally integrating knowledge about local needs into decisionmaking processes is key to serving humanor community-based goals.While it may therefore be tempting to assume that 'the more flexibility, the better', our empirical findings also revealed the relevance of defining clear goals at the early stages of a UDP.The cases showed how developing actors in fact searched for ways to embed their innovative practices in the already-established governance structure around the UDP in order to justify unplanned investments.Similar dynamics are identified as a 'double bind' dilemma of social innovations in urban governance by Bartels (2017): while innovations typically intend to challenge conventional systems, they often have little chance to survive if they do not give in with the prevailing order.
The tension between flexibility and room for innovation on the one hand, and the necessity to embedded incremental changes in existing structures on the other hand, call for reflexive governance that leaves room for experimentation.Without substantive changes in the relational dynamics of governance, Bartels (2017) argues, the experimental learning that is so typical of innovative practices fades away in attempts to embed it in the daily practices signified by 'hierarchical policy making, competition over funding and status, and engrained routines and knowledge' (p.3800).Indeed, as some argue, the adoption of an 'experimentalist governance process' (Sabel and Zeitlin, 2012) remains politically challenging if public bodies in institutionally weak environments do not have the capacity to act accordingly (Morgan, 2018).More than organising one-off reflection moments, an experimentalist approach to the governance of UDPs thus requires specific institutional capacities.According to Frediani (2021), embedding critical urban learning in urban development would imply to deepen urban diagnoses, promote collective analytical exercises, re-examine the nature of planning practice, co-produce situated strategies, and even, to facilitate an 'reflexive and empowering ethos of city-making practice and collaboration ' (p. 143).This latter point touches upon the empirically observed importance of strong commitment to pre-defined goals in UDPs.In fact, the capabilities approach carries the potential to provide a solid ethical ground in urban development processes to justify the flexible, experimentalist approaches that come along with the highly value-laden and interpretive character of social sustainability's implementation.

Principle 3: Institutionalise social sustainability implementation
Finally, our empirical findings point out that the pursuit of inclusive policy outcomes in cities demand a certain institutionalisation of social sustainability goals in the governance situations around UDPs.In both cases presented in this article, the most innovative place interventions emerged as unplanned, self-initiated ideas by small groups of professional actors within the UDPs.While such emergent initiatives may be expected because they form a flexible response to locally emerging needs -they also lead to rather vulnerable urban outcomes.In both cases, these place interventions only addressed a selected area or group of people, resulting in unequal capability expansions.Although it may be inherent to implementation processes that broad, idealistic ambitions need to be bought down to practical proportions, it is crucial that the impacts that these interventions generate are being critically assessed.On whose lives do new place interventions have no impact?Do some groups gain more benefits than others?And: at what point does this become unjust?Following Sen's (1999) original intention, the capabilities approach sets out a normative principle for comparative exercises to spotlight the most severe inequalities in society, and to determine where to start to make them more just.It is particularly the connection between this critical outlook on local urban outcomes and the wider structural drivers that relate to them (Biggeri and Ferrannini, 2014;Frediani, 2021) that we think is central in a capability-centred governance of social sustainability in urban development.Indeed, the often-heard critique of trusting in local collaborative initiatives is that this ignores systematic issues and in fact creates a 'belief in better urban futures while perpetuating spatial and social injustices' (Cardoso et al., 2022(Cardoso et al., : 2644)).In line with scholars arguing for the embeddedness of civil society in multi-scalar democratic governance (Gerometta et al., 2005) or for 'bottom-link-approaches' (Eizaguirre et al., 2012), we agree that discussion on institutional design is crucial to not let collaborative governance be overshadowed by issues such as opaque systems of representation, asymmetrical power distributions or favouritism (Kim, 2016), but instead, let its many proven advantages thrive.

Final conclusions
The overarching aim of this article was to better understand the implementation of social sustainability ambitions in contemporary urban development.By analysing the relations between the governance situations of two Dutch urban development projects and residents' conversions of place interventions into capabilities, our casestudy analysis has provided empirical evidence about the specific conditions in collaborative governance processes that help to achieve more socially sustainable urban outcomes, that is, outcomes meeting the standards of the capabilities approach (Sen, 2009).These conditions enabled us to propose three general principles for a 'capability-centred governance' of social sustainability in urban development: (1) integrate human logic into urban governance situations, (2) balance strong goal commitment with experimentalist governance and (3) institutionalise social sustainability implementation.
We conclude that social sustainability's implementation requires a conceptualisation in which improvements in people's lives are not seen as the self-evident consequences of a set of place interventions, but instead as a guiding principle that should continuously be reflected upon and learned from during the different governance phases in urban development processes.Whereas our study showed that the utilisation of place interventions is not necessarily governed in urban development projects, it is particularly in this 'utilisation phase' that residents' actual needs for social sustainability can be properly understood.Connecting the activities of utilising actors who manage or operate places such as schools, healthcare organisations, local companies, and residents themselves, with the activities of planners, designers and property developers is therefore crucial to embed social sustainability goals in urban development.More than technical sustainable urban development goals such as energy-efficiency, social sustainability is concerned with a high diversity of personal values and needs, and therefore yields uncertainty about 'the right' interventions an urban place needs.It is this normative and dynamic character of social sustainability, as well as the meager institutionalisation of its goals in the governance situations around urban development projects, that makes the concept currently vulnerable in implementation processes.We therefore argue that serious attempts to further institutionalise social sustainability in existing governance structures are needed, if socially sustainable urban development is to empirically emerge.
How exactly such institutionalisation should take place can differ from place to place.In institutional contexts where public actors play strong roles in urban development, developing participatory or intermediary practices as part of planning procedures may be a way to go, whereas in private sector-led development contexts regulations that favour, for example, corporate social investment can be more impactful.Moreover, enlarging the financial and organisational capacities of self-organised community initiatives is an evident and ofteninvestigated way relevant to many urban contexts.Finally, in light of our empirical findings, developing procedures to continuously monitor the effects of collaborative activities throughout the different phases in urban development projects may very well function as a way of institutionalising social sustainability goals in governance practices.After all, designing and implementing the place-based policy interventions for communities is one thing, but to make them appropriate for local residents in a socially sustainable manner requires a much more reflexive and on-going commitment among all actors involved.thank de Alliantie, Philadelphia and Woonstad Rotterdam for their support in this case study by welcoming us as researchers and by sharing relevant contacts and documents.We also particularly thank Co Verdaas without whose endless support and guidance this article would not have been possible.We thank the anonymous reviewers who provided constructive comments on how to improve this article.Last but not least, we thank Sultan Cetin-Ozturk and Daniel van Staveren for their valuable comments on the development of this article.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was conducted at TU Delft's chair Urban Area Development, that is supported by the Foundation for Knowledge on Area Development (Stichting Kennis Gebiedsontwikkeling).

Table 1 .
Framework for analysing social sustainability's implementation in UDPs.

Table 2 .
case examples of conversions from place interventions to capability outcomes.