Getting acquainted with their landscape: research by design as a tool to understand people's perception of current and future landscapes

In the current policy setting in Western European society, spatial planning tends to emphasise the achievement of policy intentions through the realisation of actual spatial interventions and the growing importance of citizens as spatial actors. The place-based planning approach refers to the fact that the design of integrated interventions must be tailored to places, since it largely depends on the knowledge and preference of people living in it.This paper unpacks the planning practice in Flanders by analysing and describing two landscape cases, as the actual arenas where different social activities compete, many key-actors are gathered and spatial planning processes and interventions take place. The key concepts “governance”, landscape quality” and “research by design” are defined. We investigate if research by design can be used as a tool to explore people's perception of current and future landscapes.The case studies disclose a deliberative process, which makes use of a research by design methodology, is not only able to gain knowledge on people's perception of current and future landscapes but can also improve the mutual understanding of the appreciation of landscapes by various spatial actors.Both cases illustrate the current struggle of authorities with the growing importance and knowledge of citizens and the difficulties the policy makers experience introducing these actors and knowledge in planning processes. The designs and masterplan developed within the cases refer to the place-based approach, as it is proposed by the European authorities.


Introduction
The Western European society is constantly changing.Neoliberalisation and market-friendly policies have been affecting the way cities in Western Europe develop and function since the late 1970s (Tasan-Kok, 2011).In the current policy setting, spatial planning tends to emphasise the achievement of policy intentions through the realisation of actual spatial interventions and the growing importance of citizens as spatial actors (Sager, 2011).This results in a complex dependence of different actors and an institutional and spatial complexity (Albrechts, 2006).The planning authorities are struggling with the growing importance of citizens, and more general the broad range of actors, involved in a planning process.They are developing new forms of cooperation between public authorities and e.g.citizens under the heading "from government to governance".Meanwhile Europe introduced the so-called "place-based approach".The place-based approach (Barca, 2009) refers to the fact that the design of integrated interventions must be tailored to places, since it largely depends on the knowledge and preference of people living in it.Knowledge of people's perception of landscapes to comprehend landscape characteristics and underlying quality is crucial to assure spatial interventions can take them into account and at best enhance them.Determining quality of actual landscapes however is not obvious, let alone finding a consensus on the way this quality can be maintained or enhanced when altering the environment.This paper attempts to unpack the planning practice in Flanders by analysing and describing two landscape cases, as the actual arenas where different social activities compete, many key-actors are gathered and spatial planning processes and interventions take place.The key concepts "governance", landscape quality" and "research by design" are defined in the first section.The second section holds the main hypothesis for the paper, the methodology and research questions.The final ambition of the paper is to develop suggestions for future planning processes in Flanders that aim to change and realise landscapes, making active use of research by design and insights of citizens.

Governance, landscape quality and research by design
The policy-making in Europe is dispersed across a broad range of actors, including politicians, officers, interest groups but also citizens.This demands specific ways of cooperation to achieve a common goal, often referred to as governance.Governance has become an important concept with the decreasing role of the welfare state since the 1980s (Tasan- Kok & Vranken, 2011).Güntner (2011) distinguishes two competing interpretations of the term governance.The first uses the term in a neutral, analytical way, distinguishing different modes of governance: command, competition, cooperation, negotiation and others.A second perspective considers governance as a relatively new concept describing new forms of cooperation between public authorities and e.g.citizens under the heading "from government to governance".Within the ESPON TANGO Project territorial governance is defined as (ESPON & Politecno di Torino, 2014, p. 5) "the formulation and implementation of public policies, programmes and projects for the development of a place/territory by: coordinating, actions of actors and institutions; integrating policy sectors; mobilising stakeholder participation; being adaptive to changing contexts; realising place-based/ territorial specificities and impacts".Therefore territorial governance is seen as an extension of multi-level governance, adding explicitly territorial and knowledge related elements, thus focusing on a place-based and territorially sensitive approach.The participation and mobilisation of stakeholders is decisive within the concepts of territorial and multi-level governance.Consequently knowledge of people's perception is crucial for the success of cooperation between spatial actors, institutions and policy sectors and the achievement of policy goals.In recent documents of the Flemish Authorities the concepts of territorial multilevel and multi-actor governance are considered as two of the main challenges for the spatial planning in Flanders (Vlaamse Overheid, 2014).Landscape can be defined as "an area, as perceived by people, whose character is the natural and human factors and their interaction" (Council of Europe, 2000) This broad definition which is used in the European Landscape Convention, the main framework for policy and management of European landscapes, shows a clear holistic approach and refers to all landscapes: natural, rural and urban, exceptional, mediocre or even degraded.Further deepening and application of the definition indicates that landscape largely deals with the quality and identity of an area.Landscape quality is a broad concept that implies personal values and subjectivity, it is constructed by a complex process of dialogue between spatial actors who try to establish a mutual understanding of the concept given a specific place, specific time and specific context (Reijndorp, Truijens, Nio, Visser, & Kompier, 1998;Hajer, Sijmons, & Feddes, 2006).Hajer et al. (2006) therefore point out the need to establish a shared set of terms for landscape quality so actors can truly understand each other's perspectives during the deliberative process.
In this paper we explore if design can be used as a tool to explore people's perception of current and future landscapes.Cross (2006) describes several capacities of design.Design can help to understand the nature of a problem by the production of alternatives, can focus on possibilities and synthesis rather than analysis, can explore abstract concepts through concrete objects, is constructive and holistic, is able to cope with ill-defined problems and can cross bridges between experts and laymen.Schreurs and Martens emphasise the interactive and process-character of research by design."Research by design is not focusing on the final product, the design.Research by design is dealing with the "production" of an environment by testing sites, programs, perceptions, intention, goodwill, feasibility…" (Schreurs & Martens, 2005, p. 4).Research by design has an explorative dimension and a new vision on forecasting and backcasting (Dreborg, 1996).Most research by design is done to explore, spatialise and visualize new concepts, to make new transitions in urban planning matters understandable (Zaman et al., 2014).Within this theoretical framework research by design can be used in a territorial governance context in order to mobilise stakeholders and to be adaptive to changing place based conditions.

Research Methodology
Since design is capable of tackling similar issues (Cross, 2006), our research hypothesis suggests research by design is a suitable tool to explore people's perception of current and future landscapes.Furthermore our research also hypotheses the improvement of mutual understanding of the appreciation by various spatial actors through a deliberative process guided by research by design enhances the success for actual realisation.
Through case study analysis this research examines to which extent this hypotheses is observed and explores conditions for success.Therefore following research questions are formulated: To which extent was research by design able to gain knowledge regarding people's perception of landscapes?To which extent and how was this knowledge explicit during the research?To which extent did the deliberative process contribute to the mobilisation of the stakeholders, the adaptiveness to changing contexts and the realisation of place-based/territorial specificities and impacts?What were critical factors for success?
Given the objectives, research questions, resources and time the case study is based on two cases and is conducted through a qualitative data-analysis of research reporting (Goethals, Moulaert, & Schreurs, 2011), government documents (Stad Antwerpen, 2006;Stad Turnhout, 2008;Stramien, 2008;Stad Antwerpen, 2012) and other written sources (Vlaamse Vereniging voor Ruimte en Planning, 2010; Claes, Coppens, De Wever, Pittillion, & Schoeters, 2011;Segers et al., 2013;Van Damme, 2013).The quality of the research results is being assured through the usage of a wider range of different sources, data types and independent authors.This methodology provides a certain level of inter-subjectivity which diminishes the risks of unilateral or unbalanced insights.Diagrams of both cases are used as analytical instruments to examine and compare both cases in a structured way.

Case study: Ruggeveld-Boterlaar-Silsburg (Antwerp) and Schorvoort (Turnhout)
The study consists of two cases situated in the North of Belgium.Both cases make use of research by design within a deliberative spatial planning process.They are of different scale, process structure and aim.The cases are both set in urban fringe landscapes which are characterised by a high degree of complexity.
In order to examine and compare these cases their process structure was unravelled and presented both in a textual description and as a diagram (figures 5 and 7), furthermore a focus was laid on the specific role of research by design as a means to gain knowledge regarding people's perception of landscapes and to contribute to the deliberative process.The first case, Ruggeveld-Boterlaar-Silsburg, is situated in Deurne, a district in the east of Antwerp.The area of 83 ha is part of the valley of the river Schijn which is considered an important green infrastructure connecting the city centre with the eastern outskirts and the surrounding hinterland (Stad Antwerpen, 2012).At the other hand it is also a patchwork of various functions such as allotments, soccer pitches, a ski slope and ice rink, an athletics track, basketball courts, a campsite, a cemetery and is in use by several small associations.Despite these many features and functions the landscape is highly deteriorated and has a desolate character.Indeed the hodgepodge of both unplanned developments and rather randomly juxtaposed functions in existing zoning plans results in a very fragmented area.Although it represents five percent of the total green and recreational space of Antwerp, the area is only been used by the present sport clubs and associations and not by the general public.To enhance the landscape quality in order to make the area more appealing for recreation the local government wants to reorganise the area and develop it as a large public park with sports and community facilities (Claes et al., 2011).Therefore in the year 2000 the local government took the initiative to develop a zoning plan which linked the Ruggeveld-Boterlaar-Silsburg area to the adjacent park Rivierenhof.The associations and sports clubs in the area were not consulted nor involved in the conception and design process.The local government did not expect disagreement, because at first sight, both land users and local authorities had similar interests, however the conceived zoning plan proved to be highly contested and a committee was formed to resist the reorganisation of "their" area.Clearly both parties had a different interpretation of the landscape quality.
In 2006 the local government restarted a spatial planning process, based on the newly adopted strategic spatial structure plan for Antwerp (Stad Antwerpen, 2006), addressing the issues that led to the failure of the previous plans.Nevertheless protest resurged and the zoning plan was not adopted.In spite of this contestation the associations and sports clubs gradually started to realise the existent fragmented use of the site was problematic for the overall development of the area.Furthermore in order to legally expand the existing activities a new zoning plan was mandatory (Claes et al., 2011).
To break this repetitious failure, a different approach was conceived to establish a masterplan (figure 4) by means of a co-production between stakeholders and local authorities, based on a mutual understanding of landscape quality.Research by design was used within this coproduction process to explore the differences in interpretation between experts and local actors, to investigate feasibility of possible solutions in different scenarios and to establish a common vision for the development of the area.The research tried to establish a mutual understanding of the landscape quality through a deliberative process as described by Hajer et al. (2006) and Reijndorp et al. (1998), but also used the research by design process to test the feasibility of Belgeo, 3 | 2015 possible solutions for different stakeholders as pointed out by Schreurs and Martens (2005) in order to improve the chances for actual realisation.Before engaging in a research by design process, the committee of stakeholders and the local authorities signed a participatory charter, (indicated in dark grey on figure 5) which stated a mutual commitment for the coproduction of a masterplan.It defined the itinerary of the process, it cleared out expectations, it identified delegates to participate and debate and established a distinct structure for the co-production process.In accordance with the charter the planning officials of the local authorities and the stakeholders appointed an independent research group, a consortium of Belgian and Dutch designers: Maxwan Architects, 1010, Karres & Brands landscape architects and Goudappel Coffeng.At first researchers made an inventory and survey concerning needs and expectations of all stakeholders.Based on this first exploration interviews with the involved stakeholders and field surveys were conducted to map existing use and scan possible future development.This mapping process provided insight in the spatial distribution of the issues.This preliminary research provided a basis for the development of a masterplan (figure 4) through research by design, which started simultaneously at different scales with distinct groups of stakeholders (indicated in light grey on figure 5).The most important task and eventually result of the research was to combine spatial claims or visions into scenarios for multifunctional land use, because space was scarce and the added sum of development acreage was vast.The sketches, some examples are shown on figure 3, show how these combinations of diverse land uses might look like, making the ideas less abstract and more clear to imagine for the layman among the stakeholders.The designs gave all the parties more insight in the spatial possibilities or restraints, moreover it provided a common language for discussion and negotiations (Claes et al., 2011;Van Damme, 2013).
The consecutive altered sketches were used to test whether certain spatial solutions could be feasible and desirable for all stakeholders and to make sure the resulting masterplan would truly address the underlying landscape quality issues.This way the researchers distilled a common understanding of the existing landscape quality features by all stakeholders through envisioning and discussing possible future landscapes.The insights in the existing landscape quality were not always made explicit, since the aim was to create shared vision on the future landscape quality, rather than evaluating the present landscape perception.However Belgeo, 3 | 2015 the knowledge was incorporated in the successive designs (X, Y, Z on figure 5) which can be seen as a design resume of the insights.Successive discussion and design, both plenary and in smaller groups, incrementally changed the plans and ultimately resulted in a masterplan The independence of the researchers as an interface between local authorities and the other stakeholders proved to be a formula for success to guide the discussion, to establish a common vision based on existing qualities and to provide a wide coalition to implement the masterplan.The second case, the neighbourhood Schorvoort, is situated in the south of Turnhout.In this area the local government intends to increase the housing density by developping three projectzones for urban housing, according to the zoning plan the regional government adopted (Stramien, 2008).In addition the city's local policy plan aspires preserving of enhancing spatial quality and encouraging public involvement within spatial developments (Stad Turnhout, 2008).Private owners undertook several attempts to develop their land in the projectzones, however not consistent with the zoningplans and exiting policy intentions (Segers et al., 2013).The local government therefore intends to adopt a masterplan for the neighbourhood as a basis for urban and landscape development (Goethals et al., 2011).Schorvoort used to be a small agricultural village, but during time it devoloped as a part of the adjacent town of Turnhout.It is characterised by ribbon development along the former agricultural roads built up with large introvert housing projects on the former meadows and fields, both social and market-oriented.The neighbourhood of Schorvoort is situated in the valley of the river Aa which connects a public park with the surrounding agricultural area and has a history of flooding problems.Within the area several facilities are located such as a primary school, a church, a grocery store, a bakery and a soccer club.Because of the lack of spatial legibility, dwellers have difficulties orientating in the area and different parts of the neighbourhood are rather experienced as separate enclaves.Furthermore as a result of the relatively high amount of social housing Schorvoort also has to cope with problems of social segregation.Hence the neighbourhood lacks coherence from both spatial and social perspective, with a difficult understanding of the landscape and its qualities as a result (Goethals et al., 2011).
Before commissioning a masterplan for urban and landscape development the local government wanted to establish a clear project definition in which key qualities are addressed, a range of spatial issues are identified and an ambition for design is formulated.Therefore the Department of Architecture, Urbanism and Planning (unit Planning & Development) of the University of Leuven was commissioned to conduct the required research and to synthesise their findings.Because of the mentioned spatial and social issues researchers opted for the use of "research by design" being a methodology they trusted to be more suitable to gain knowledge on existing and future quality both in the spatial as in the social fabric (Goethals et al., 2011).Roughly three groups of stakeholders were involved in the participative research by design process: the landowners of the projectzones, the local government and the dwellers.
The government officials made a selection for representation of dwellers with attention to the variety of social groups.Within the group of landowners the private owners were represented by a local architect who had already proposed the contested masterplans for development in the projectzones.The bulk of remaining land is owned by a social housing company, which was also invited to participate in the planning process.Firstly the researchers made a quickscan of the area through field survey, consulting policy documents and spatial analysis, which led to a preliminary evaluation of the existing spatial quality.Based on these findings interviews were conducted with landowners, politicians, policy makers and experts which resulted in schematic designs for potential development.
In contrast to the first case the research process was organised in a strictly consecutive manner (in light grey on figure 7).The first designs were adjusted on the basis of input from dwellers in a design workshop.A second workshop with representatives of all stakeholders refined the designs.Goethals et al., 2011 Two final workshops and related designs attempted to find a shared opinion on future developments and spatial quality ambition between different landowners and the local government.To guide the deliberative process and to facilitate the quest for establishing shared terms for spatial quality (Hajer et al., 2006) throughout the deliberative process the researchers used a framework for spatial quality based on a scientific concept.Remarks, suggestions, ideas and discussion during the process were structured according this framework to facilitate and clarify the debate.
The framework composes spatial quality by five dimensions: 'vitality', 'sense', 'fit', 'access' and 'control' (Lynch, 1984).This framework enables the assessment of landscape quality, but can also serve as a guide ex ante to ensure landscape quality within a transformation process (Vervoort, 2012).For landscape quality the dimension "sense" is particularly of importance, since it is the degree to which the appearance of a landscape is capable to enhance the perception of the underlying interaction between natural and human factors (Pisman, Vervoort & Appermont, 2013), therefore ensuring a broader comprehension and appreciation of the landscape values.

Belgeo, 3 | 2015
In this case the researchers created twodimensional design schemes to unravel underlying structures in order to examine the key qualities of the landscape and to show possibilities for enhancement (an example is shown on figure 6).The design was mostly used to visualise the discussion and to distil elements that could be used in the eventual project definition.In the Schorvoort case "research by design" did not lead to a masterplan, but the designs and conclusions of the various workshops were used as a way to clarify the perceptions of different stakeholders on the existing qualities of the neighbourhood.It mainly made use of twodimensional design schemes to structure and distil elements that could be used in the project definition.Although the process had the ambition to investigate possibilities for a common concept for spatial development with enhancement of landscape quality, it rather clarified the different point of views of stakeholders than unifying them in a clear and generally supported project definition.The project definition (as shown in dark grey on figure 7) in Schorvoort and the charter of Ruggeveld-Silsburg-Boterlaar (indicated in dark grey on figure 5) are somehow comparable, in terms of mutual understanding of landscape quality and cooperation between spatial actors, institutions and policy sectors and the achievement of policy goals.

Results and discussion
In both cases the interactive and process-character of research by design proved to be a good way to incrementally gain knowledge regarding people's perception of the landscape qualities.For instance comparison of survey responses on landscape quality in the Turnhout case before and after the research by design process shows a large increase in landscape quality expertise among dwellers, responses being clearer, more accurate and more detailed (Segers et al., 2013).The use of the scientific based framework in the Turnhout case has to be pointed out as a way to clarify and structure the research results, but also to guide capacity building and to find common terms for landscape quality (Goethals et al., 2011;Segers et al., 2013).On the other hand researchers indicate little fundamental knowledge on the perceptions of the (private) landowners was gained, and explain the latter due to a lack of cooperation during the design process (Goethals et al., 2011;Segers et al., 2013).Perhaps the visualisation of the forecasting in the Turnhout case in two-dimensional plans made it more difficult for laymen to participate in the debate.In the Antwerp case the eventual masterplan was able to formulate Belgeo, 3 | 2015 a clear design solution in consensus with all stakeholders.Although not verified explicitly, this could indicate most of the different views on existing and future landscape quality were successful identified and integrated.
The cases differ in the way the knowledge was recognized, explicit and consolidated.In the Antwerp case the different designs were used as preliminary synthesis documents, which incrementally were adjusted during the various discussion sessions.Although the researchers made record of the proceedings the evolving design itself was the main method to capture knowledge with a clear focus on future landscape quality.In contrast in the Turnhout case the workshops resulted in various maps and designs, which were useful to gain insight in the landscape perception, but establishing a comprehensive scheme of the area was no goal.The gained knowledge, captured through the use of the scientific based framework, was explicit and synthesized in a mere textual document explaining the current perceived qualities by stakeholders and pointing out issues to be addressed in further design.
The deliberative process in both cases led to an increased expertise in place based/territorial specificities of local government and stakeholders.However using design as a medium for structuring and synthesising the debate, resulted in different levels of success, referring to the realization of landscapes.In the Antwerp case the process resulted in a mutual understanding with new allies and a grown involvement of stakeholders in the spatial interventions and the enhancement of landscape quality (Vlaamse Vereniging voor Ruimte en Planning, 2010; Claes et al., 2011;Segers et al., 2013;Van Damme, 2013).In the Turnhout case the latter is less clear.The deliberative process caused a distinct increase in awareness of landscape quality among dwellers, which resulted in a new neighbourhood committee "Schorvoort 2020", that unites different social groups and aims at ensuring the future landscape quality (Goethals et al., 2011;Segers et al., 2013).But the process was not able to fully engage the (private) landowners in the debate.Researchers indicated this might had been caused due to a lack of trust in the independent role of the research group and a resultant perceived competition with the proposed plans of their representing architect.Also they acknowledge a possible flaw of the research approach, which had less regard for market oriented development logics, that led to poor cooperation (Segers et al., 2013).Therefore the research did not result in a mutual understanding or an integrated vision.On the contrary, no allies for implementation with key stakeholders (i.e.landowners) were established, which makes the success for further implementation and realisation questionable.
To conclude some possible factors for success were identified by comparison of both cases.The use of a charter in the Antwerp case resulted in mutual commitment of local government and stakeholders and clarified expectations with respect of possible outcome before engaging in the research by design process.The Turnhout case did not explicit these specifics in a similar way, therefore expectations and incentives for commitment by stakeholders were less clear, which possibly led to a lack of cooperation by the (private) landowners (Segers et al., 2013).Accordingly the doubt on the independence of the researchers clearly was a stumbling stone in the Turnhout case (Segers et al., 2013).In the Antwerp case the independent status of the researcher on the contrary ensured the validity of the research by design process among participants, moreover being a neutral mediator, the researchers were able to propel and monitor the deliberative process (Claes et al., 2011).
The scope of the research did not allow more profound investigation through supplemental interviews of key witnesses such as involved spatial actors, researchers and government officials and longitudinal research.It is strongly recommended to discuss the results with the key actors involved in the two cases and to monitor the realisation of the two cases in the future.

Conclusion
We would like to conclude the paper with some suggestions for future planning processes that aim to change and realise landscapes, making active use of research by design and insights of citizens.
The case studies disclose that a deliberative process, which makes use of a research by design methodology, is not only able to gain knowledge on people's perception of current and future landscapes but can also improve the mutual understanding of the appreciation of landscapes by various spatial actors.The research indicates the latter can result in new allies among spatial actors with an increased involvement in the spatial interventions and the enhancement of landscape quality, but also points out possible flaws and conditions which could complicate further actual realisation.Establishing a participatory charter can guide the process to realisation since the itinerary of the process is defined, expectations are cleared out, delegates to participate and debate are appointed and a distinct structure for the co-production process is agreed on.The independence of the researcher is important both for a unconditional commitment in the research by design process, and for propelling and monitoring the deliberative process.Also the usage of concrete, easy to understand graphics, like for instance three-dimensional sketches, seems to be essential to profoundly gain insights in the perception of laymen and to synthesise in a holistic way the mutual understanding of the landscape quality.
Both cases illustrate the current struggle of authorities with the growing importance and knowledge of citizens and the difficulties the policy makers experience introducing these actors and knowledge in planning processes.The designs and masterplan developed within the cases refer to the place-based approach, as it is proposed by the European authorities.

Résumés
In the current policy setting in Western European society, spatial planning tends to emphasise the achievement of policy intentions through the realisation of actual spatial interventions and the growing importance of citizens as spatial actors.The place-based planning approach refers to the fact that the design of integrated interventions must be tailored to places, since it largely depends on the knowledge and preference of people living in it.This paper unpacks the planning practice in Flanders by analysing and describing two landscape cases, as the actual arenas where different social activities compete, many key-actors are gathered and spatial planning processes and interventions take place.The key concepts "governance", landscape quality" and "research by design" are defined.We investigate if research by design can be used as a tool to explore people's perception of current and future landscapes.
The case studies disclose a deliberative process, which makes use of a research by design methodology, is not only able to gain knowledge on people's perception of current and future landscapes but can also improve the mutual understanding of the appreciation of landscapes by various spatial actors.Both cases illustrate the current struggle of authorities with the growing importance and knowledge of citizens and the difficulties the policy makers experience introducing these actors and knowledge in planning processes.The designs and masterplan developed within the cases refer to the place-based approach, as it is proposed by the European authorities.

Figure 6 .
Figure 6.Research by design showing the possible development of public space as a way to clarify landscape structures, KU Leuven-R&D.

Figure 7 .
Figure 7. Diagram research by design process Turnhout case.