Finding the pulse of the welfare landscape: reframing green space provision in modernist planning

ABSTRACT

Contemporary planning for urban densification permits the exploitation of the spacious green areas developed for recreation during the welfare planning of the 1960s–70s. Historical studies of welfare planning are needed to better understand the potential values under threat. Answering Colin McFarlane’s call for relational studies of density, this paper offers a complementary examination of the relational geography of green space provision in the 1970s, to reveal what the development of the compact city both silences and (literally) replaces. This relational approach departs from the flat ontology of Actor-network theory. The study captures how ideals of recreation, nature, welfare, planning and the rhythms of life assembled into a geography for recreation in the early 1970s, and how this topology crumbles a decade later. While the green spaces of the 1970s linger on today, their reinterpretation as green structure in the 1980s and 1990s partly veils their former role and potential. The paper interprets the legacy of welfare planning, and provides a base for further examination of the geography of green space provision.

KEYWORDS:

• Actor-network theory
• green structure
• outdoor recreation
• recreational planning
• relational geography of planning
• topology

Introduction

The spacious green areas of the modernist era in Sweden are challenged in contemporary urban planning by the drive for densification in the name of sustainability. Current planning practice argues for the possibility of replacing vast green spaces with compact and programmed parks without compromising on quality, a shift which is frequently described as going from ‘quantity’ to ‘quality’ (Littke 2015; Pries 2019; Stryjan 2020; Åkerlund 2021). This undermines any meaningful criticism based on a quantitative analysis of green space provision, as increased density with less (but more ‘compact’) green space is part of the urban vision. A report of the loss of urban green space could therefore be regarded as an indication of successful densification, and of the reduction of the ‘empty’ and ‘unfriendly’ environments of modernistic projects. As McFarlane (2020, 318) notes, such reinterpretations of urban space are not a matter of ‘a neutral geography of remaking space, but processes of direct and indirect displacement, the valorisation and prioritisation by states and markets of some urban spaces and forms over others (including some forms of density over others)’. This leads McFarlane to ask: ‘what kinds of images, models, approaches and stories of density are being told, and what alternatives ought to be documented or foregrounded?’ (McFarlane 2020, 318). To reveal this, he calls for a relational (and topological) approach to density (McFarlane 2016; 2020). In line with McFarlane, I argue for a relational geography inspired by Actor-network theory to capture the fundamental shifts of green space provision, and to bring to the fore the alternative stories of a green welfare which are being threatened by contemporary urban development. This is captured in a study of the role of green space in relation to recreational planning in Sweden in the 1970s, and the subsequent reconfiguration of this assemblage initiated in the 1980s and 90s. The study examines how the geography of green space is conceptualized in national enquiries and guidelines concerning the interplay between green space provision and recreation/sport, with special focus on two key organizations in respect of this interplay.

The paper offers a relational (and topological) interpretation rather than a rich study of the planning practice that by necessity draws on previous plans and investments, and the local conditions (see, e.g. Turner 1992; Engels 2019; Pries and Qviström 2021). Given the unconventional approach to green space provision, the relational approach has to be introduced before engaging with the historical study.

ANT, topology, and a relational geography for ‘Flat earthers’

‘ … spatiality is not given. It is not fixed, as part of the order of things. Instead it comes in various forms.’ (Law 1999, 6, italics in original)

Relational ontology helps us to unravel the complex geography of an activity, idea or phenomenon on its own terms and without a taken for granted Euclidean framework. This ambition is perhaps most clearly pronounced in the topological discourse within ANT that widened the palette of possible associations making up a geography (Law 1999; Law and Singleton 2005; Murdoch 2006). For this reason, this section uses the topological literature to introduce the ANT-based relational geography of the paper.

The broadening of ANT beyond studies of network space in the early 2000s brought about innovative explorations of topologies of various artefacts and assemblages, from cities to mobile phones, diseases, pumps, mines and forests (Latour and Hermant 1998; Law 1999; De Laet and Mol 2000; Kortelainen 2010; Ek 2012; Law and Singleton 2005; Jensen and Sandström 2020). The examples capture ways of understanding and ordering the world beyond Euclidean space, and the heterogeneity and dynamics of the assemblage in question. Furthermore, with notions of fluid and fire objects, ANT scholars opened up a wider range of understanding of relational space than the previous focus on actor-networks had made possible (Law and Singleton 2005). Such a material-relational understanding has helped to ‘enliven and enlarge human geography's spatial imagination’ (Latham 2011, 315), as in Hoffman and Thatcher’s (2019) notions of creativity, or Latour’s study of a windsurfer’s topology of enjoyment (Latour 2005b).

In the early 2010s topology had become a ‘poststructuralist buzzword’ within human geography, informed by a cocktail of different ontological understandings of the concept (Martin and Secor 2014, 423). While the peak for the buzzword has passed, the diversity and confusion concerning the concept remains. For instance, Martin and Secor (2014) raise concerns about the dichotomous use of the concept as a non-Euclidean space, and as the antipode of topography (see for instance Allen 2016). This has recently led to calls to combine, or even merge topology and topography, for instance in Hönke and Cuesta-Fernandez (2017) concept ‘topolographical’. McFarlane (2016) argues for a similar approach when using topology as an umbrella under which observations of Euclidean, relational, volumetric, experiential and perceptual spatialities should somehow come together. Yet, if topology and topography are indeed so different, then merging them rests on shaky theoretical ground (Lexén and Qviström 2021). The (non-mathematical) topological understanding within ANT offers another approach.

For Law, a topology stands for a ‘family of forms’ (1999, 7), families which play a crucial heuristic role in helping us to widen our understanding of possible ways of thinking and enacting spatialities. As Law (1999, 7) points out this sensibility for complexity is only possible to the extent that we can avoid naturalizing a single form, a single topology (italics in original). Given the radically relational approach of ANT, geography is understood as a matter of relational effects, even if enacted as abstract space. Consequently, Euclideanism is one of the topological families, a geography which need to be performed and maintained as much as any other.

Latour (2005a) offer further clues to how the ‘flat’ ontology and the empirical and myopic approach of ANT differs from other discussions on topology (see also Müller and Schnurr 2016). Latour asks us to follow the actors with an open mind, without adding a ‘context’ or abstract theoretical concepts as shortcuts to an explanation. This flat ontology and empirical methodology bring topology down to earth. Topology, then, is the qualification of the geography and dynamics of an assemblage, not an additional theory that reveals something on its own. Topology, therefore, is not more abstract, fluid or esoteric than the mundane world (see for instance the seminal example of a fluid object in De Laet and Mol 2000). This is not to say that calculations, abstract models, flow charts, ideals or representations cannot be part of the assemblage, only that they have to be drawn in by the actors studied, not by the researchers. Thus, while there is an important heuristic value in concepts such as fluid and fiery objects in opening our eyes for the diversity, dynamics and multiplicity of spatialities, they cannot be used to guide the actual tracing of a topology without adding something new to it. Indeed, such categories (especially if understood as ideal forms) could stand in the way for a rich encounter of the individual case. In this study I refer to topology when aiming to qualify the geography (and dynamics) of an assemblage, and I settle with an interpretation instead of a classification of the topology.

In the topological literature, the notion of actor-network is downgraded from an umbrella concept for any heterogeneous association to a specific topological family, which calls for another overarching concept. Assemblage has often come to play this role, with or without inspiration from Deleuze (see Müller 2015; Müller and Schnurr 2016). In line with the ambition to keep topology (and other potentially abstract concepts) down to earth, assemblage is only used as an umbrella concept in this study, with no additional theoretical belongings.

Since an openness to a multiplicity of relations, and thus of geographies, is key to ANT, the identified assemblages are likely to be heterogeneous, dynamic, or even characterized by internal contestations. The concept boundary object emphasizes that different actors can have diverging agendas and yet collaborate (Star and Griesemer 1989; Star 2010). Boundary objects coordinate action by being ‘plastic enough to adapt to the needs and constraints of the several parties that employ them, yet robust enough to maintain identity’ (Star and Griesemer 1989, 393). Thus, an assemblage can be successful and yet carry internal tensions, which is a balancing act: the interaction comes at a cost, which is acceptable as long as the benefit of being part of it is greater (Latour 1992). To identify such tensions, and the renegotiations they might cause, is then of importance to understand the shift of the topology.

ANT doesn’t offer an alternative to the time-consuming work of tracing heterogeneous relations (Latour 2005a). As Lecomte (2013, 463) argues ‘it is because actor-network theory (ANT) greatly emphasizes the many discontinuities and heterogeneous elements on which any continuous ensemble depends that it constitutes a decisive starting point to engage with the topological complexity of contemporary urban dynamics’. Yet, as Bender (2010, 317) notes, ANT offers an ‘unusually rich heuristic device rather than a formal method’. However, as will be illustrated below, with a paradigmatic shift the actors, intentions and concepts are replaced, which challenges the systematic search or method. Instead, tracing historical topologies requires careful interpretations, otherwise we risk using contemporary topology as a framework for our study, which makes the entire examination rather pointless.

So, where should we start to trace the assemblage? ‘In the middle of things’ or in the controversies suggests Lecomte (2013, 472, see also Latour 2005a). I do both to offer complementary entries into an understanding of the topology of green space for recreation in the welfare society. In the first part, the study takes its point of departure as a matter of central concern identified in previous research: the pressing need to combat escalating lifestyle-related diseases. The second part starts, literally, in the middle of things, poking in the rubble of the former assemblage. This part examines how the shared idea of the topology crumbles from the perspective of two key organizations, and briefly captures the new way of thinking about the city in green structure planning. By doing so, it also uncovers tensions, or the multiplicity/complicity of the former assemblage, and some of its silences.

The two organizations are the Environmental Protection Agency (EPA) and the Swedish Ski and Outdoor Association (SSOA), both of which played a crucial role by bridging an engagement with the outdoors and planning for an increased physical activity. SSOA, EPA and the local govenment focus not only on recreation (or sport or physical activity) as such but also explicitly on its relation to green space. While there are several other actors that could have been studied, the shifting roles of EPA and SSOA in the early 1980s capture the radical shift of green space provision. Primary and secondary sources on their engagements with recreation and Sport For All have been focused upon, especially policy documents and plans. In the case of SSOA, archival studies of their branch for Sport For All (Keep in shape) has also been committed. Yet, this is not a paper on EPA and SSOA, they are merely used as a lens to reveal the shifting geography of green space provision. Therefore, other plans and policy documents are in a few cases brought in to provide a richer picture of the planning for green space. Particular attention has been payed to the geography of green space provision expressed in the documents, and (to a lesser extent) to the assemblage of actors brought in to realize the planning intentions.

The pulse of welfare planning

‘A Swedish male heart ticks 5.3 years longer than its American counterpart. Why?’ (Brattnäs 1973, II)

This question, and a photo of the inside of a heart (covering two pages), introduces the book ‘Fit for Fun, a Swedish Message’. It was a timely question. As part of the societal progress of the post-war society a new phenomenon had emerged: lifestyle-related diseases. The risk of a heart attack became an especially emblematic threat, not least for men in the middle of their career (McKenzie 2013). In medical science, this would lead to a paradigm shift in the understanding of the importance of everyday exercise in the 1950s (Bolling 2005; Svensson 2013; see, e.g. Åstrand 1968; SOU 1969:29), a shift that came to affect the Swedish urban and suburban landscape primarily in the 1960s and 70s. The book ‘Fit for Fun – a Swedish Message’ not only sums up the relatively novel understanding of the need for everyday exercise but captures its key role in the emergent Swedish welfare society, and the pride of Sweden as a pilot country in this work (Brattnäs 1973). The author was the sports attaché of the Swedish Consulate General in New York, and part of her duty was to market everyday Swedish fitness exercises - and Sweden - to a North American audience. The book (which was also translated into German and Swedish) underlines how a pragmatic and moderate form of recreational exercise had become part and parcel of everyday life in Sweden, for all ages. It pictures an elderly woman taking part in a workout at an urban square, an elderly male member of parliament at an aerobic class, children having fun in an indoor swimming pool, women working out in a modern office landscape, a family enjoying a ‘health break’ with exercises at the roadside during a car journey, and so on. The heart was a thread running through the book, with a heart symbol introducing every new theme. The book includes a longer interview with professor Per-Olof Åstrand, expert in exercise physiology and cardiology, on health and fitness, which had made the Swedes aware of the importance of exercising the heart (see Åstrand 1968). Åstrand, who held a central role in the public debate on health and fitness, also acted as a technical advisor for the book. As Åstrand had been involved in launching recreational running as a remedy for welfare diseases in the late 1950s (Qviström 2013), it is not surprising that this is what features on the cover of the book, nor that Brattnäs introduces recreational exercise with running, due to its cardiological effects:

‘The thing about running – or jogging – is that it stimulates the heart and muscles, as well as the legs. It gets the blood moving.’ (Brattnäs 1973, 9).

With its emphasis on the heart and the rhythms of everyday life in a modern, urban society, the book captures a particular profile of the welfare society that would come to characterize how it mobilized the landscape as a welfare resource. As Svensson notices, ‘[r]ational training for everyone and rationalisation in general were at the core of the Swedish welfare state’ (2013, 897). This rational exercise, with its special focus on combating coronary diseases, would enter planning in the 1960s. Yet, the concern for health and wellbeing was not new as such, nor was public investment in facilities for recreation, albeit with very limited resources compared to the recreational planning in the 1970s. As Åquist (2001) illustrates in her close reading of a comprehensive plan from the mid 1950s, the way of structuring planning interventions based on the rhythms of everyday life was already part of the functionalist rationale of modernist urban planning. However, recreational exercise is not one of the rhythms that Åquist identifies. What was new, then, was the focus on weekly exercise for all and the key role of the pulse, but also the places and actors associated with these new practices (Qviström 2013).

What doesn’t come to the fore in the English edition of ‘Fit for Fun’ is how these ideas took shape in and through spatial planning. Leisure planning was high on the agenda in post-war planning in Sweden, as in many other Western countries (United Nations, Economic Commission for Europe 1969; Gold 1973; Clark, Jokela, and Saarikivi 2011; Alon-Mozes 2020; Pries and Qviström 2021). When the association of local authorities in 1975 launched a new journal, ‘On Leisure’ as a forum for municipal politicians and staff, it was due to the rapid increase of municipal spending, from 100 million SEK in 1961–1,500 million SEK in 1975, with higher figures expected in the following years (Järdler 1975). These investments were however brought to public attention in the introduction of the Swedish publication of ‘Fit for Fun’, which argued that Sweden (in contrast to North America) was well equipped for such activities:

‘Swimming pools and sports grounds are available to everybody. Municipal recreational centres and floodlight trails for running and skiing are practically everywhere. The gymnastic halls of schools are open and accessible to the public after school-hours, and it is usually just a stone’s throw to vast, green areas for walking and hiking when we want to relax. On top of this, associations and clubs engage in sport for all in a manner which is entirely unfamiliar in the USA’. (Bedrup 1973, 4, own translation)

Green space would become a key element for recreational planning, as a resource as such but also as a site for locating infrastructures for recreation. A significant example of this conceptualization of green space is found in a report which sums up the norms for planning in the City of Stockholm in 1970 (Stockholm stad 1970). These standards only address green space through play, sport and outdoor recreation. With these norms, recreational rhythms become an ‘obligatory point of passage’, or the compulsory lens through which the establishment of new public parks or green spaces had to be justified and understood (see Callon 1986).

With Swedish corporatism as a general model for the distribution of governance, planning involved a number of actors at local, regional and national level, which makes any attempt to offer an overview very difficult. Even though state funding played a crucial role for the rapid development of a varied infrastructure for recreation, the support served partially different purposes, it was not entirely coordinated, and it was dependent on the initiatives of local actors. A national inquiry in 1979 reveals the complex structure, with 38 out of 39 national bodies offering financial support to grassroots organizations, adding to more than 60 different kinds of grants from the state. State support for children’s activities and recreation is portrayed as especially scattered (SOU 1979:60). Thus, we shouldn’t expect to find one blueprint model or one set of norms for the welfare landscape: the governance of this landscape was complex and only loosely coordinated with recommendations, financial incentives, moral statements, and allusions to landscape ideals and imageries (see Pries and Qviström 2021). Yet, the Environmental Protection Agency (EPA) had a national responsibility for outdoor recreation in the 1970s, and had a major influence on what kinds of places were assembled for recreation, which makes it an important actor to study.

Environmental welfare planning

The EPA was established in 1967, with outdoor recreation as one of its responsibilities. In the early years of its existence, key issues concerned the need for an ordered exploitation of natural resources, with a strong emphasis on social and societal needs. The nature protection would primarily be solved by technology and planning as part of the modern society (Mårald and Nordlund 2020). The strong belief in planning would also characterize their engagement with outdoor recreation, a theme that in effect concerned physical activity in the outdoors more generally. With the Nature Conservation Act as the legal framework for their responsibility for outdoor recreation, the EPA would aim for an environmental planning in which landscape and nature conservation merged with the welfare ambition of sport for all.

The national inquiry ‘Sport For All’ (SOU 1969:29) reframed the EPA’s involvement in outdoor recreation. The inquiry has been described as a milestone in Swedish sport policy (e.g. Österlind and Wright 2014), and would play a fundamental role for recreational planning in the 1970s. The inquiry would result in significant increase of financial support for sport, not least due to its important role in a welfare society. It would also redefine sport in the widest possible way, thus including almost any kind of physical activity. In accordance with this radically inclusive definition, the inquiry emphasizes that a divide between sports and outdoor recreation is redundant, as the former includes the latter. The report justifies the wide definition with the societal challenges caused by an increasingly sedentary society, which requires a change in focus from primarily youth to the middle-aged and elderly as a target group for sport activities. Thus, physical activity, and the need to raise the pulse, is in focus rather than competitive sports. This becomes especially evident in the appendices of the inquiry, one of which sums up the health benefits of physical activity with a strong focus on coronary disease, and an appendix calculating the financial benefits of reducing the number of heart attacks. The main research presented as a background for the inquiry is on coronary diseases, together with an analysis of the existing organization, finances and ongoing activities.

As the EPA held the main responsibility for outdoor recreation, they would become responsible for funding recreational facilities within the conceptual framework of ‘Sport for All’ in which physical activity was put at the forefront and a division between sport and outdoor recreation was left behind. With sport for all as the main objective, national support for developing ‘conventional sports arenas’ was removed in 1970 (with a few exceptions), leaving the responsibility for financing such arenas to the municipalities. Instead, the EPA would fund a number of different kinds of facilities for outdoor recreation, such as outdoor swimming pools, camping grounds, harbours for leisure boats, infrastructure for fishing, slalom slopes, parking spaces for outdoor recreation/sport, running tracks and outdoor recreational centres (Statens naturvårdsverk 1970).

The EPA structured their work and policies on certain scales, which reflected the rhythms of work and leisure in the welfare society. One example is given in a background report for ‘National Spatial Planning’ (Guinchard 1971). The report presents a model structured around different kinds of ‘leisure types’, that is, the different rhythms of leisure, and their different reach in terms of places for recreation (see Table 1).

Table 1. Norms for accessibility to areas for outdoor recreation.

Part of a day	Areas in the vicinity, class A	walking, cycling, wheelchair	1 (1½)km, 15 min
Part of a day	Areas in the vicinity, class B	bus, or cycling, wheelchair	5 km, ½ hour
One day	Areas for a daytrip	car, bus/railway, bicycle	5 (−100) km, ½ (−1½) hour
2–3 days	Areas for a weekend trip	car, bus/railway, possibly flight	100 (−200) km, 1½ (−3) hours
More than 3 days	Areas for a holiday	car, bus/railway, flight	Unlimited

Note: Adjusted and translated Guinchard (1971, 18).

Different activities, facilities and places could then be classified according to the scales and rhythms of leisure; part of a work-day (activities supported within neighbourhood), day-based activities (supported on a municipal level), activities for the weekend (supported by regional planning), holiday (supported by national planning). A strikingly similar model was presented the following year by Norrbom (1970), representing Kommunikationsdepartementet, the governmental body in charge of the National spatial planning. The similarities with the scalar thinking of the modernist road planning developed in the late 1960s are hardly a coincidence, as the road network provided the backbone for connecting recreational spaces. This hierarchical ‘tree-model’ (Alexander 2015) with its scalar divide and its particular nature ideal (e.g. favouring forests as places for outdoor recreation) would also be adopted in ‘Planning for Exercise’, a guide targeting local and regional authorities as well as sport and outdoor recreation organizations involved in planning for exercise for all (Centrala motionsrådet 1976, see also Qviström 2016). The guide was published by the Central Council for Exercise, a joint venture of the Swedish Sports Confederation, the Swedish Association of Inter-company Athletics, SSOA, the National Board of Health and Welfare’s committee for health information, the Association of County Councils and the Association of Local Authorities. This collaboration illustrates as such the importance of planning for recreation within the welfare society – and the multiple actors involved in its realization.

The EPA would, with this model as a base, develop a hybrid spatial planning for recreation within the jurisdiction of environmental protection, in which a specific form of everyday recreation, a specific kind of nature, and the rhythms of the welfare society were cornerstones.

On the local level, the EPA would support the municipalities financially (Statens naturvårdsverk 1970). A radical merger of town and countryside municipalities into much larger municipalities offered an arena for developing the welfare state (Erlingsson, Ödalen, and Wångmar 2015). The ambition had been to ensure that each municipality was a functional region (SOU 1961:9; Wikman 2019), and thus an efficient unit for developing spatial planning and for supporting local social service and welfare facilities. Importantly, with a focus on functional regions, each municipality would gain equal administrative and legal status, which had not been the case with the former division between cities, townships and countryside municipalities. Also, this would play down a divide between town and country, a division which was seen as arcane by the investigators (SOU 1961:9). In line with ideas of spatial justice, municipal planning would take on the entire municipality, at least in strategic planning documents, with a focus on developing a community rather than building a town and preserving countryside. In this new geography, the role of the EPA was not restricted to the countryside but could follow the rhythms of everyday life, from neighbourhood facilities to weekend recreational centres.

On a regional level, the EPA supported regional centres for recreation, and developed methods for analysing landscape qualities in a McHargian fashion (e.g. Statens naturvårdsverk 1975; compare with McHarg 1992, originally published 1967). On a national level, the agency introduced and provided financial support for the development of 25 ‘primary recreation areas’ for active holidays (Proposition 1975:46; Statens naturvårdsverk 1975; Sandell 2008). To make this available for everyone affordable accommodation and journeys to the sites were part of the strategy. The active engagement of the agency in regional and national planning was done within the framework of the ongoing National spatial planning, a nationwide attempt to designate areas of national and regional importance for a number of public interests, including recreation and nature conservation (Statens naturvårdsverk 1975; Forsberg 1992).

Establishing places for outdoor recreation in a welfare society

The new infrastructure for recreation was established in close collaboration with a number of actors, spanning governmental bodies and insurance companies to grass-root organizations and local sport clubs. One of these organizations was the Swedish Ski and Outdoor Association (SSOA), the main organization for outdoor recreation in Sweden. The SSOA took an active role in developing an infrastructure for ‘sport for all’ as part of the welfare project. They developed, in close collaboration with medical doctors and cardiologists (notably P-O Åstrand), a model for outdoor recreational centres, which would be implemented by almost every Swedish municipality (Qviström 2017). As Qviström (2013) illustrates, the model clearly combines two ideologies: the cardiological focus on the heart and physical exercise with the therapeutic focus of the outdoor association on the importance of being (active) in nature. Furthermore, SSOA’s division Keep in Shape was one of the important initiatives taken by grassroot organizations, with ample financial support from private insurance companies and the state (Grauers and Sundbom 1972; Qviström 2013). Keep in Shape aimed nationwide campaigns at the public, with skiing, brisk walks and running, arranged ‘active holidays’ for families, provided courses in jogging and skiing, running campaigns for school children, and so on (Grauers and Sundbom 1972). In the early 1970s up to 200 local clubs offered at least some of these summer activities (Friluftsfrämjandet, B1c vol. 2). The SSOA was also involved in developing the first national guidelines for facilities for outdoor recreation (Statens Naturvårdsverk 1971). Finally, the regions of Värmland and Sörmland collaborated with the SSOA to develop regional plans for wellness (Friskvårdsplan), with the recreation centres as key (Friluftsfrämjandet, E2, vol. 4). These plans offer insights into the geography they aimed to establish.

The wellness-plan for the region of Sörmland was developed in 1972, a plan which was supposed to cover the needs for the coming 5–10 years and thus limit the escalating costs for medical care. The plan was developed in dialogue with a large number of actors, including the regional branches of the Swedish Sports Confederation and the Swedish Association of Inter-company Athletics, the county architect, regional authorities for rural development, leisure and nature protection, the Association of Local Authorities and Land Owners. One part of the strategy was to clarify the responsibilities of these actors, which resulted in highlighting of the key role not only of the municipalities and the regions, but also of private firms, among others.

The report primarily aimed to encourage exercise for all, arguing for running, brisk walks, swimming and cross-country skiing as the most accessible and suitable forms of exercise (Frohm 1972). To support this, a specific infrastructure was suggested (see Table 2).

Table 2. A regional infrastructure for well-being suggested by SSOA.

Type of facility	Location and a rough estimation of the need
Walking and running paths	2 metres/inhabitant in the neighbourhood area
Floodlit running/skiing trail	Not more than 2 km from residential areas and workplaces
Other trails	Hiking trails more than 50 km in length, and canoe trails more than 50 km in length, aiming at specific destinations for excursions.
Recreational centre	1 per 5,000-10,000 inhabitants, with sauna, floodlit running trails, changing rooms, fitness test room and relaxation room.
Wellness centre	For health checks, with a nurse and medical doctor: 1 per 50,000 inhabitants.
Local activity areas	With tracks, trails, open grass fields and facilities, 200–300 m^2 per inhabitant, located within the neighbourhood or its immediate surroundings.
Activity area, for excursions	10–100 km from the built-up areas to support outdoor recreation during the weekend or holiday periods. With rest shelters, a basic café and basic camping facilities.
Indoor gyms	Basic gyms in some housing complexes, medium versions at workplaces, and advanced gyms at schools and large workplaces.

Note: Adjusted and translated from Frohm (1972, 19–20).

The plan stipulates a rhythm with daily exercise (for the heart and other muscles) of 20-30 min, and weekly exercise in the forest ‘preferably on a foot-friendly path in or adjacent to a recreational centre’. However, the report also recommends

‘meditation and relaxation … preferably with the family … in forest and land (which can be combined with the weekly exercise), by listening to the sounds of the forest, studying animals and nature, or in anther silent and calm environment’. (Frohm 1972, 16, own translation)

In addition, fitness tests 2–3 times a year, and a health check (especially for the middle-aged and elderly) are recommended. Notice how the model emphasizes societal and bodily rhythms (and the recreational and therapeutic role of the forest) as its point of departure.

The report also offers a characterization of the landscape of the region, describing waterways for canoeing and water sports, wilderness in need of protection, and the archipelago as a site for an active holiday, etc.

Similar plans can be found in other regions, though with a more prominent focus on sports in some cases. Yet, the rhythms are there, as is the importance of basic forms of recreation in which the contact with nature is an important component (c.f. Lillienhöök 1971a, 1971b; Qviström 2016).

A topological interpretation

The green welfare planning draws to some extent on previous infrastructures, investments and institutions when assembling places, actors and activities to support physical activity and wellbeing for all, even though the level of ambition (in terms of, e.g. state funding) was clearly scaled up during the welfare era. Yet, in comparison with previous planning it differs in its emphasis and understanding of bodily and societal rhythms, with an emphasis on the week instead of holiday periods, the pulse instead of just recreational leisure, and with the rhythm of the car and office hours as important factors. It might be even more accurate to locate the specific character in the alignment of these rhythms with a specific idea of recreation and nature, and its given role in planning for a welfare society. This alignment seems to work as an elusive boundary object, as a story which can be told from several different points of entry and with different emphasis on particular aspects, coordinated through various forms of collaborations and state interventions.

The heterogeneous constellation was dependent on certain agreements being treated as a matter of fact, certain given absences and presences in the coalition. For instance, outdoor recreation was by definition a matter of environmental management and planning, and indoor recreation had a secondary position if included at all. Outdoor recreation was clearly related to nature, but nature did not stand in opposition to built up areas. Competitive sport, on the other hand, was not part of the equation. Furthermore, the territory of the municipality was the arena for planning, which marginalized the importance of a division between town and country. This, in turn, can be related to ideas of spatial (or distributional) justice, with the same planning standard and planning procedure for all places, related to population density only. Health was dependent on an alliance of two ideas: the beneficial values of being outdoors in nature, and the specific focus on exercise physiology - particularly the pulse. The planning was based on a scalar idea of everyday life in the welfare society. In combination with the new municipalities and the other state funds etc., a rather loose set of standards would develop, working on the level of the state down to the sandpit (see, e.g. Statens naturvårdsverk 1970; Statens naturvårdsverk and Statens institut för byggnadsforskning 1971; Statens planverk 1977; Mack 2019; Pries and Qviström 2021). This complex assemblage alone gives a sufficient explanation of the type of recreational landscape which would materialize in the 1970s.

The planning trend described above does not capture the entire greening of the welfare cities, for two reasons. First, history and the local topography play a key role in planning. Already established parks and urban forests were either reinterpreted as part of the new topology, or lingered as anomalies. Secondly, the topology described does not capture the green buffer zones which were also part of welfare planning. Wide corridors were set aside to separate conflicting land-uses, especially arterial roads and housing. These buffer zones were not intended (or counted as) green assets, but would nevertheless characterize the welfare landscape and contribute to its spacious greenery. While these buffer zones probably play a minor role for recreation (and therefore are less important to examine in this paper), this nevertheless reminds us of the complexity of planning in which several topologies are at work. Yet, it is the alignment of green space provision, an outdoor recreation with a focus on the pulse, welfare and nature that would be challenged in the coming years, which is why this topology deserves special attention.

Back to nature

In a recent study on ‘environmental dissonances’ in modern Swedish environmental planning, Mårald and Nordlund (2020) describe a polarized debate between proponents of ecological modernization in the name of societal progress and the welfare state, and an environmental concern on the other hand. As Sandell (2008) illustrates, these tensions also concerned 1970s planning for outdoor recreation. Yet, the conflicts between progress and the environment was also internalized by actors engaged in welfare planning. The writings of Bertil Lillienhöök, agronomist and consultant at K-Konsult, capture this internal tension. Lillienhöök delivers a devastating critique of the inhumane town planning, and of a society with escalating costs for medical health care. As a consolation for the alienated modern city and its lurking lifestyle-related diseases, he proposes the development of a ‘leisure landscape’ as a vast green belt. To compensate for the urban environment, these landscapes should be as extensive as the cities and towns taken together, in order to provide sufficient assets for running, skiing, hiking and other recreational activities that bring the urban population closer to nature (Lillienhöök 1971a, 1971b). Thus, in the writings of Lillienhöök, a very harsh criticism forms the basis of his engagement with the welfare project. The engagement of SSOA reflects a similar critique of modern society. These tensions materialized as a modern suburban and urban nature, embodied through new practices such as recreational running.

A topology with such built-in frictions is frail. In the early 1980s, welfare planning would shift in character and in some parts be challenged, and another green topology would slowly emerge. It is not my intention to capture this process or the new topology in detail, only to use the shift to gain a better understanding of the welfare topology, not the least by hinting at the fundamental differences compared to the emerging topology.

One tension concerned the inclusive idea of sport for all (with a focus on recreation), another one the relation between sport for all and nature. When the concept ‘sport for all’ was introduced, two institutions had raised concerns over the idea of bridging sport and recreation: the SSOA and the EPA. The SSOA made the argument for a qualitative difference between outdoor recreation and the inevitable focus on competition within sport (Kungl. Maj:t 1970). Another reasonable concern was that sport involved a wider range of activities and places beyond the nature-based idea of recreation which they had invested in. While the ‘sport for all’ concept would strengthen the recreation-nature assemblage in the 1970s, a renegotiation of ‘sport for all’ would challenge it in the 1980s.

The 1980s saw an increasing sportification and commercialization of outdoor recreation, but also a shift from its given setting in nature, not least the introduction of an American version of jogging in 1979 with its close connection to both sport and an urban environment (Olson 1993; Qviström 2017). This development would cause trouble for the SSOA, as some of the novel trends were adopted by local clubs for their Keep in Shape activities. The minutes from the Keep in Shape committee in the early 1980s offer an insight into how the tensions between outdoor recreation and sport developed (Friluftsfrämjandet, A23d, vol. 20–27). In 1984, an internal report offers scathing criticism of Keep in Shape and describes it as the most ‘debated’ division of the organization which needs to ‘sanitise’ its assortment of activities, i.e. avoid competition and commercialization, in order to protect the ‘good reputation’ of the organization. Instead, focus should be put on facilitating the wellbeing of SSOA’s own members, a shift not only from targeting the general public, but also towards a more limited understanding of recreation (Friluftsfrämjandet, F 3:18). This reframing caused an immediate reaction from one of the main actors, the insurance company Trygg-Hansa who had provided funding for the campaign for over two decades, which stepped back from its active role on the Keep in Shape board and withdrew its funding since the campaign was no longer ‘for all’. With new guidelines for Keep in Shape in 1984, the SSOA abandoned the welfare project and all its funding, networking and marketing opportunities. Instead it returned to its roots: a brochure promoting a romantic idea of the forest, and of family-based leisurely recreation, far away not only from competition but also from the outdoor gyms and other facilities which the organization had developed in the past (Friluftsfrämjandet 1984).

A parallel shift would occur at the EPA, who redefined their remit in relation to outdoor recreation. The tensions between recreation and nature conservation surfaces in an EPA report of the late 1970s (Statens naturvårdsverk and Statens planverk 1979), and the rationale behind funding local facilities for sports and recreation was officially questioned in the year-book of EPA in 1978, in which it is suggested this is a matter for urban development rather than environmental protection (Anonymous 1978). Moreover, the definition of recreation would be redefined, from a matter of ‘[outdoor] exercise, relaxation and leisure activities of various kinds’ (SOU 1973: 52, 21) to ‘[outdoor] relaxing and stimulating activities done during leisure-time’ (SOU 1981:28, 15), noticing that sport and wellness primarily belonged to other spheres than recreation. This captures a rift in the wide approach to sport for all – and how the pulse went down in areas related to outdoor recreation. In the year-book of EPA in 1984 Zettersten states:

‘Today we describe the agency’s responsibility with the keywords: Outdoor recreation in nature. This means that the agency is responsible for that part of outdoor recreation which is based on the right to roam – to be able to freely stay and travel in nature. … the agency does not have the responsibility for sport or for tourism as a business.’ (Zettersten 1984, 7, own translation)

With such narrow definitions of outdoor recreation and nature, the urban environment and recreational centres are left behind. The marginalization of the EPA in relation to ‘sport for all’ is evident in the EPA’s financial reports. In 1980/81 the EPA supported facilities for sport and outdoor organizations with more than 7 million SEK. The following year, the sum was around 400,000 SEK, and only 340,000 SEK in 1982/83 (Hjelm and Jäderberg 1984). This budget was used to, for instance, support recreation facilities in remote mountain regions, far away from the arena of everyday recreation.

With national spatial planning coming to an end in the 1980s, the active role of the EPA would be left to the municipalities, especially with the new Planning and Building Act in 1987. The EPA’s responsibility for outdoor recreation would end in 1991, without being transferred to any other agency, which in practice meant it was decentralized to the municipalities. Petersson-Forsberg (2014) portrays this as an all-time low for Swedish outdoor recreation politics. Furthermore, the geography of recreation became a local matter governed primarily by planning legislation which had lost its formal norms concerning green space provision (SOU 1994:36, appendix 3).

A green topology beyond welfare planning

In planning, a criticism of modernistic ideas and a renewed interest in the urban would emerge in the early 1980s (e.g. in SOU 1981:99), before gaining greater prominence in the 1990s (e.g. Boverket 1994b; see Kristersson 2003; Tunström 2009). This would call for other studies of green space, in which its role was redefined (or defended) as part of the urban fabric (see Lundgren Alm 2001 for a more detailed description).

In 1993, a national inquiry was instigated to ‘develop strategies for the treatment of green areas within densely populated areas’ within the legal framework of the Planning and Building Act (SOU 1994:36, appendix 3). This inquiry would affect how green space provision was to be framed in the 1996 revision of the Planning and Building Act (Lundgren Alm 2001).

The inquiry clearly acknowledges the threat of densification, and notes a significant loss of green space in the cities over the past 20 years. This calls, they argue, for giving green structure a stronger position in planning. Thus, the main aim of the inquiry was to re-conceptualise the idea of green space as green structure, in an urban context. The authors are explicit about leaving recreation, exercise and play behind, since these aspects, they argue, had already been acknowledged within planning. What is missing, they argue, is its cultural heritage (which the modernist paradigm had left behind), but more importantly: the ecological function of the green structure and its role for biodiversity. From an ecological perspective, green structure is defined not as a flat surface, but with the ground, the soil, the hydrological system and open water as part and parcel of the structure. Furthermore, the definition defies any jurisdictional boundaries and includes private properties as well as public space. This (landscape) ecological approach is combined with a focus on everyday environments in cities, which leads to a geographical limitation to the densely built-up areas and their immediate surroundings. The increased urbanization is taken as an argument for focusing on the urban sphere in contrast to weekend or holiday facilities further away (SOU 1994:36, appendix 3, 40). Finally, the inquiry clearly highlights ‘nature areas’ and ‘parks’ in the city, whereas facilities for sport and recreation are almost entirely silenced.

With the reframing of green space planning as green structure, the former idea of accessibility and spatial justice comes into question. As the inquiry notes:

‘A majority of the municipalities strive for equal access for the inhabitants to usable greenspace within the settlement. [Yet] the green corridors which are presented on a principal level for the three largest cities … are in practice not connected. The documentation of the green structure should therefore also include an analysis of the real accessibility’ (SOU 1994:36, appendix 3, 25, own translation).

In this way, connectivity, which is of importance from a landscape ecological point of departure, also frames the understanding of accessibility (and equal access), as if green space could only be accessed through another green space. Further on, the inquiry turns this into an ontological claim arguing that ‘The green areas of cities and settlements are comprised of connected structures from the perspective of human use as well as in terms of ecological functions.’ (SOU 1994:36, appendix 3, 41).

The inquiry was soon to be followed up by a publication from the National Board of Housing, Building and Planning (Boverket), to provide knowledge on the green structure and to guide in how to analyse and plan for the green structure. In the foreword, the general director of the National Board of Housing, Building and Planning states:

‘The planning of green areas in our cities has been defective and short-sighted over the past decades. … But a shift in attitude is about to occur.’ (Gösta Blücher in Boverket 1994a, 5, own translation).

This echoes a general historiography of a revival of green planning after a few lost decades of modernist planning (see, e.g. Lundgren Alm 2001).

The publication from the National Board of Housing, Building and Planning reiterated the thematic and geographical conceptualization of green structure from the inquiry in detail. The report suggests a map-overlay with six themes to capture the green structure, including the theme ‘Recreation, Work-day, Leisure and Health’. The theme brings to the fore the importance for children playing, the pedagogical aspects of the parks and urban nature etc., but the main part of the chapter is dominated by an environmental psychological perspective on the importance of parks. Following this research, a key challenge, according to the chapter, is to experience nature to reduce stress: this is done by walking or just experiencing the park, not through physical activities (Boverket 1994a). Thus, the pulse and the body didn’t vanish from the planning, but with a shift from coronary heart diseases to mental stress, the pulse has gone down and sport and active recreation (at least for others than children) are largely put aside.

With the concept of green structure, the contribution to green space provision made during the welfare era was reinterpreted and revalued. A multi-scale system with a focus on the rhythms of life was replaced with a morphological (and ecological) focus on urban land-use. With the EPA and the SSOA being out of the picture, and less emphasis on sport for all, few interfaces remained between sport provision, green space provision and environmental planning. Furthermore, the very definition of green structure, with its focus on landscape ecological conditions, marginalized the role of facilities for recreation. Thus, planning for recreation would be divided between (and to some extent fall between the stools of) park planning, planning for sports, and nature conservation. This division would marginalize recreational planning in the coming years.

Conclusions

This paper provides an example of how a relational approach facilitates novel insights about taken-for-granted geographical configurations within planning. Green planning is a case in point, where paradigmatic changes in the conception of green space provision have occurred over the past decades. A map based study could easily overlook these shifts. Furthermore, a relational approach can help us understand the different rationales that has governed green planning. It brings to the fore the shifting places, actors and ideals assembled as green space. By doing so, it opens up for a wider discourse on the future of urban greenery, and a better understanding of the green legacy that is currently being challenged.

The modernist legacy, and its geography, have been questioned ever since Alexander’s devastating (and very influential) critique of the geography of modernist planning (Alexander 2015, originally published 1965). Alexander revealed a hierarchical order of functions in modern city planning, forming the mathematical model of a tree, a model that, according to Alexander creates isolation and reduces the liveliness and complexity of the ‘natural’ city. This is one of the seminal studies that needs to be fleshed out with studies going beyond the morphology of the built environment. The green welfare topology detailed in this paper is based on the tree-model which Alexander (2015) captures, but it isn’t just any tree: the scalar model is based on the pulse and the rhythms of everyday life, especially the rhythms of leisure. This tree is indeed not a city (to refer to the main argument, and title, of Alexander’s paper), but a foundation for a welfare society beyond a rural-urban divide. Furthermore, its geography was constituted by certain ideas of nature, the body and the society as much as of a scalar (or hierarchical) approach to space, which makes the heuristic value of the tree-model rather limiting. Yet, the tree-structure was important as it echoed and could easily be combined with welfare planning at large, with obvious links and similarities to, for instance, traffic planning. This welfare geography resulted in a type of planning where the logic is more likely to be seen in a flow chart than on a map.

The loose (and frail) alliances which formed this assemblage are key, I would argue, to understanding why this particular landscape materialized primarily in the 1970s, only to be challenged and crumble a decade later. By recognizing these alliances, a historically informed critique of the contemporary planning is possible. This critique goes beyond a discussion of green space per inhabitant. Instead, it directs attention to how green space have lost its pulse and its societal rhythms, and thus become marginalized as a societal resource. Furthermore, it hints at the potential to reassemble green spaces with, for instance, the EPA and, on the local level, the leisure department, rejoining a conversation on the planning of everyday landscapes for health and wellbeing. Such an alternative future cannot just rely on historical studies but requires us to, among other things, rethink what rhythms and scales matter in contemporary society. Even so, a historical relational study can open up for such a discussion on other futures.

The description of green structure planning hints at an important relationship between the old and the new topology, as the latter is explicitly framed as a complement to previous planning. The green planning emerging in the 1980s and 90s relied on and aimed to complement an existing landscape, shaped by previous green space provision regimes. This explains why it differs in almost every respect – except for the green places which it acknowledges as part of the green structure. The two topologies would therefore complement each other – at least initially. However, the shift in focus led soon enough to disinvestment and decay of the previous welfare geography and its facilities.

The topology of green space provision alone does not suffice to capture the situated and complex character of the urban landscape, not least due to the crucial importance of the local topography and history. Furthermore, planning practice is affected by numerous topologies, partly materialized in the landscape, partly in planning documents and regulations. This results, as Pries and Qviström (2021) illustrate, in a landscape shaped by a patchwork of plans rather than by one strong idea of time, speed, space and place. Yet, precisely because new ideas and trends only slowly materialize in the landscape, and could take time to make a difference in the planning practice too, it is vital to examine shifts in thinking, measuring or acting in relation to the landscape. A relational approach, I argue, is necessary to provide general interpretations of how shifts in the planning regime will affect the landscape. More specifically, this paper argues for the necessity of better understanding the topology of welfare planning, before its assets in the landscape are too heavily reinterpreted or transformed.

Acknowledgement

I am grateful to Mia Ågren, Amalia Engström, Sued Ferreira da Silva, Märit Jansson, Johan Pries and Jack Richold for useful comments on earlier versions of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Swedish Research Council Formas: [grant number 2016-00264].