INTRODUCTION

High nature value (HNV) farmlands are areas in Europe where agriculture is the dominant use of the land and where agriculture supports, or is associated with, either high species and habitats diversity or the occurrence of species of European conservation concern or both (Andersen et al. 2004). HNV farmland covers over 25% of the European Union’s (EU) agricultural land and contributes to the perseveration of biodiversity, cultural landscapes, territorial cohesion, quality products, and employment (Lomba et al. 2020). However, the area of HNV farmlands is decreasing due to land intensification and abandonment. These two faces of an ongoing techno-economic process, occurring across European agricultural landscapes, have negative impacts on biodiversity and ecosystem services (Plieninger and Bieling 2013, Almeida et al. 2013, Plieninger et al. 2014, European Environment Agency 2013). In a time of globalized food systems characterized by production specialization and concentration, the social-ecological conditions to maintain low-intensity, and thus less productive, HNV farming systems are difficult to meet (O'Rourke and Kramm 2012, Ustaoglu and Collier 2018). HNV farming systems are thus economically and socially marginalized in the wider EU economic space (Jones and Poux 2012).

Halting the loss of HNV farmland requires fostering the socioeconomic viability of HNV farming systems and retention of a management intensity that is compatible with social, cultural, and ecological values. This perspective coincides with the “viable HNV farmlands” scenario described by Lomba et al. (2020), in which the socioeconomic viability of HNV farming systems is increased while maintaining a management intensity that ensures the conservation of farmland biodiversity. Pursuing such viability calls for tailored actions to steer the development of HNV farming systems based on the strength of their local assets. Due to the rapidly changing socioeconomic and ecological contexts, previous studies highlight the need for innovation in HNV farming systems to increase their economic viability. Innovations in HNV farming systems need to balance the transformative character of innovation with the preservation of the practices that generated their nature value initially. Such innovations require a transformational process that extends far beyond the preservation of traditional low-input farming practices (Fischer et al. 2012).

In recent decades, a major shift in theoretical perspectives on agricultural innovation has occurred. The linear paradigm of innovation (Rogers 1962), an approach characterized by the design of technological-based solutions displayed toward passive beneficiaries, is evolving toward an open and dynamic viewpoint of innovation (Beckman and Barry 2007, Berthet et al. 2018). This evolution is characterized by interactive and iterative learning processes supported by multiple social networks (Klerkx et al. 2012, Pittaway et al. 2004). A core element of an open innovation model is stimulating linkages and cooperation opportunities between key actors across sectors, value chains, and territories. Within such complex and multi-actor networks, the essential role of an intermediary, known as an innovation broker, has been highlighted (Howells 2006, Klerkx et al. 2009, EU SCAR AKIS 2019). Previous research on innovation brokerage described the importance of brokers and broader network-bridging organizations for the co-production of knowledge, social learning, and innovation (Van Lenthe et al. 2003, Dedeurwaerdere et al. 2015). Research on design theory has emphasized the importance of the early stages of an innovation process. The negotiation of a shared frame for action is a crucial moment during these early stages, which influences the impact of the entire innovation process (Kilduff et al. 2000, Hey et al. 2007, Callon 1984, Beers et al. 2014). Recent developments in the design of effective innovation co-production systems highlight the role of the innovation broker within agricultural innovation systems (Klerkx et al. 2017). Here, innovation-brokering describes an explicit set of activities to bring together representative actors from different organizations and disciplines to stimulate collaboration toward the co-production of innovation (Klerkx and Nettle 2013).

To date, little research has focused on the design and implementation of innovation-brokering processes and its potential impacts on HNV farmland. Conserving HNV farming systems involves changing territorial dynamics toward better integration of biodiversity at several levels of management (from farm to territorial levels). There is a pressing need to understand how to design innovation-brokering processes so that positive impacts on HNV farmlands social-ecological viability are achieved. In this context, the innovation broker is more than a facilitator who builds bridges between different actors (practitioners and researchers) and organizations. An innovation broker with an environmental goal fosters the co-production of knowledge in a specific context and also plays a strategic role in facilitating certain types of innovation, influencing the wider context in which they operate (Mermet et al. 2013). HNV innovations aim to specifically address HNV challenges and are thus based on an explicit vision of the landscape and HNV attributes in a given territory (Poux and Moran 2017). Thus, we adopt an integrative concept of innovation. HNV innovations are described as a set of new activities and relationships aimed at the conservation of HNV features through socioeconomically-viable farming systems. This approach diverges from standard innovations which mainly focus on economic or efficient resource management regardless of biodiversity conservation.

The aim of this research is to improve the understanding of a range of different approaches of innovation brokers in HNV farmland areas to conserve their environmental characteristics while improving their socioeconomic viability. More specifically, it aims to characterize and compare the agroecological and socioeconomic dynamics occurring in HNV areas, which contribute to HNV conservation in specific territorial contexts. The study focuses on the engagement phase to explore the different dimensions of the design and initiation of an innovation-brokering process and examines the importance of landscape–territorial visioning as an entry point. Finally, our results are discussed, and guidelines are provided for innovation brokers to support the implementation of strategic approaches. These are tailored to the specific territorial contexts of European HNV farmland.

METHODS

Unit of analysis: learning areas

Our research builds on the experience of a Horizon 2020 thematic network titled HNV-Link, which was aimed at improving the profitability of HNV farming without losing its HNV characteristics (EIP-Agri 2016). Overall, the research developed by HNV-Link aimed to identify, foster, and disseminate pertinent innovations and best practices to maintain HNV farmlands by involving key actors in an innovation process tailored to distinct HNV territories.

We focused on the ten territories included in HNV-Link, which are all engaged in an innovation-brokering processes as part of agri-environmental conservation strategies in HNV farmland. Thus, we designate HNV farming areas as potential territories of innovation, as loci of a collective learning process (Kirat and Lung 1999), hereafter designated as “learning areas.” The selection of the ten learning areas was based on the following criteria:

1. occurrence of HNV agroecosystems defined by its land cover and farming systems according to HNV farming typology (Andersen et al. 2004);
2. existence of agroecological management instruments at the territorial level (e.g., Natura 2000 area, UNESCO site, National Park, locally led agro-environment scheme under Common agricultural policy (CAP) rural development programs, etc.);
3. presence of a multi-actor cluster (e.g., private entrepreneurs, professional/farming organizations, local authorities, universities and/or research centers and NGOs) willing to engage in agroecological management to support HNV areas.

The ten learning areas covered ten European countries (Fig. 1) and included territories ranging from 190 to 4200 km², in which agricultural land covered 6–90% of the overall area. The ten targeted learning areas are representative of the diversity of HNV farming areas in Europe, both in terms of their agroecological and social characteristics (Appendix 1 for the full description of the ten learning areas).

Data collection

We focused on the first stages of the innovation-brokering process, namely the engagement process, i.e., the construction of a territorial–landscape vision (Albrechts et al. 2020). The vision was co-elaborated at the scale of the learning area based on a strategic appraisal of the HNV situation. The aim was to produce two alternative scenarios: a “business as usual” scenario, revealing what would take place without the introduction of HNV innovations; and a “HNV vision,” consisting of a social-ecological image for desirable HNV landscapes-territories in the future. To do this, we performed a baseline assessment in each learning area to characterize and compare the agroecological, socioeconomic dynamics, and the territorial governance dimension of each territory. To create a common understanding of the baseline assessment process, a guidance document was provided to set out the main categories of the assessment. It was presented to the local innovation-brokering teams during a methodological workshop, which was discussed and amended before implementation. Four main dimensions were included in the baseline assessment: (1) the agroecosystem dimension, involving the characterization of the HNV farmland area in relation to geology, soils, hydrology, ecological features, land use, and land cover types; (2) the socioeconomic dimension of the farming systems (HNV and non HNV) coexisting within the same area, based on data from agricultural census e.g., crop types, livestock types and densities, farm numbers, size; (3) the wider rural socioeconomic dimension, consisting of the characterization of the population, demography, labor, capital of the area, based on existing available information; and, (4) the institutional dimension: analysis of the governance system and description of the different actor categories, which directly or indirectly influence the development of the agrarian system.

Data collection was undertaken by a local innovation broker team supported by the coordination team. A mid-term review, based on all material gathered in a six-month period, helped to streamline the overall assessment into a common presentation framework of the results across the ten areas. Part of the data collected and organized by each HNV innovation-brokering team was co-produced with local actors of the territory. Local knowledge co-production processes were diverse and specific to each context (i.e., local workshops, focus groups, surveys or in-depth semi-structured interviews or field studies), as well as being highly dependant on the pre-existing knowledge and social capital of the innovation brokers.

In addition, a semi-structured questionnaire employing a blend of closed and open-ended questions (Adams 2015) was administrated to the ten local coordination teams (i.e., innovation brokers) before initiating the baseline assessment. It aimed at providing qualitative and comparable feedback on the general design and initiation of the different engagement processes that occurred within the ten learning areas. The main entry points were formulated according to the four main descriptive dimensions of the facilitation process according to Loeber and Vermulen (2012) (i.e., responsiveness, knowledge, power, and anchoring), complemented with additional questions to capture the nature and anteriority of the local coordination team in the area. Questionnaire items based on these main dimensions of analysis are provided in Table 1.

Data analysis

We first characterized and compared the agroecological and socioeconomic dimensions of each of the ten learning areas. From the baseline assessment, we compared the agri-environmental conservation challenges for each territory through the confrontation of the agroecology dynamics (evolving trends of HNV characteristics within the learning area, from an ecological point of view) with the socioeconomic dynamics of the farming systems (evolving trends of the agricultural sector from a socioeconomic point of view). To capture the latter dimension, we proposed a set of three characterizations of the overall agricultural dynamic within each learning area: (1) gross agricultural product evolution (stable, decline or growth); (2) evolution of the utilized agricultural area (UAA); and (3) farm take-over by young farmers (farm succession). The last indicator was selected as it constitutes a more robust indicator than the evolution of the number of farms or farmers alone as, in some cases, agricultural development has coincided with the replacement of numerous old farmers by a reduced number of younger competitive ones (early-retirement scheme). Together, these indicators enabled the identification of gradients reflecting socioeconomic trends within the ten learning areas (Table 2).

The agroecological dynamics of HNV farmland within each learning area was another dimension described in the baseline assessments. This aimed at characterizing the dynamics occurring in HNV areas from a nature conservation perspective. Overall, such characterization resulted from an interpretative analysis of sections 1 and 2 of the baseline assessments, focusing on the following indicators for each learning area: (1) patterns of HNV farmland distribution; (2) HNV types (Andersen et al. 2004); (3) distribution of HNV farming systems; (4) abandonment of HNV farming areas; (5) intensification of HNV farming areas; (6) share of UAA; and, (7) number of farms. Data illustrating agroecological dynamics of HNV farmland within each learning area are available in Appendix 2. From this, we proposed six distinct categories to qualitatively assess dynamics occurring in the 10 learning areas from a nature conservation perspective. They are summarized in Table 3.

To explore the importance of visioning as an important stage in the engagement of different actors in a multi-actor innovation process, we proceeded with a comparative analysis of the different designs of the innovation-brokering process within the ten learning areas. We built on Loeber and Vermeulen (2012) describing facilitation processes of sustainable agricultural projects. They identified four main dimensions of the facilitation process: responsiveness (identification of an issue for change); knowledge (bringing knowledge that fits with the situation and needs); power (identification of power relations and potential impact of unsustainable drivers); and anchoring (conditions for further elaboration of a dynamic of change). We used these four dimensions as descriptors to account for the design of the ten innovation processes. We compared the initial situation to the situation resulting from the HNV innovation-brokering process during its initiating phase. The description of engagement situations in each learning area (i.e., before initiating the innovation process) builds on the answers obtained through the semi-structured questionnaire administrated to the ten local coordination teams. Full tables with verbatims are available in Appendix 3. The characterization of the engagement situation at the end of the visioning exercise builds on the respective section contained in each baseline assessment. Using the same four descriptive dimensions of the innovation-brokering process, three external readers classified each HNV vision according to a common qualitative assessment grid. A transversal criterion about the participatory process was also added, in order to better qualify the multi-actor dynamic that occurred during the engagement phase. Emerging categories have been proposed to qualify the position of the innovation broker in relation to their external environment: external/consultative, external/co-production, embedded/co-construction (Table 4). Assessments were harmonized at the end of the exercise.

Finally, to explicitly describe the different strategies for initiating an innovation-brokering process, we confronted the prospective dimension of each baseline assessment against their territorial context. In order to represent situations that do not have numerical values, we have chosen to use a qualitative visual interpretation (Coffey and Atkinson 1996, Burke et al. 2005). The aim was to compare and contrast the "business as usual" scenario and the "desirable future" scenario for the HNV areas, as they were elaborated by the multi-actor cluster at the level of each learning area. All ten prospective scenarios were reviewed by three external readers. They qualified the scenarios graphically from their agricultural and conservation dimensions using an arrow, whose length represents the intensity of the change (importance of the change by 2030) and the direction represents the nature of change at territory–landscape level (Albrechts et al. 2020). The judgment of each reader was then validated by the local coordination team, which had developed the scenarios.

RESULTS

A wider typology of HNV farming situations across Europe

Figure 2 represents the agroecological and socioeconomic dynamics resulting from the analysis of the baseline assessment for the ten learning areas. These results illustrate the wide diversity in the evolution of HNV farmland areas throughout Europe and the related conservation challenges in a given macro agri-environmental context. We were able to identify four different evolution patterns in HNV farming situations. The first pattern is characterized by the risk of agricultural intensification, in the context of stable agricultural development such as in Dalsland learning area located south of Sweden (Västra Götaland region); on the Causses karstic plateau in France (Massif Central); and the Montado type landscape in the Sitio de Monfurado learning area in Portugal (Alentejo) (HNV dynamic type a). The second evolution pattern is characterized by the challenge of maintaining HNV farming systems in a general context of eroding farming activities. This is the case for the Western Stara Planina learning area located in the mountain complex of the Balkans in Bulgaria; for the Eastern Hills of Cluj in the Transylvania region of Romania; and for the Cevennes schist valleys in the South of Massif Central (France). Territory–landscape dynamics described in La Vera learning area, located on the piedmont of the Sierra de Gredos mountain range, in Extremadura, Spain, follow the same pattern as Dartmoor, a mountainous region in central Devon, north of Plymouth, England (UK). We see that the erosion of (mostly extensive) farming activities can cause the loss of HNV characteristics in these learning areas (HNV dynamic type b). Two learning areas, both in the Mediterranean, show an evolution pattern characterized by a relict / recomposing farming dynamic associated with medium to good HNV conservation trends (partial conservation or re-composition). This type of evolution is described in the Thessalia learning area, on the east side of the Pindus mountains watershed in Greece, and in the context of the partial conservation of HNV areas in the Dalmatian islands (HNV dynamic type c). Finally, there are situations that articulate stable agricultural development with good conservation of HNV attributes, as in The Burren learning area in County Clare, Ireland (HNV dynamic type d).

Distinctive HNV conservation challenges and associated risks

In order to further describe different strategic configurations for initiating an innovation-brokering process, we analyzed the HNV vision and the BAU scenario against the different types of HNV dynamics observed within the ten learning areas. It builds on our previous typology of HNV dynamics, to which we added the prospective dimension, as elaborated through the visioning exercise (i.e., HNV vision and business as usual scenario). Figure 3 represents HNV conservation challenges and associated risks for a given HNV territorial dynamic.

Building further on the innovation brokers interviews, we were able to better describe and compare the strategic nature of HNV conservation challenges within each group of learning areas sharing the same broad HNV dynamics. For learning areas showing HNV evolution patterns characterized by an intensification risk to HNV systems in a favorable farming development context (HNV dynamic type a), most of the HNV challenges are formulated to recover HNV ecological characteristics (horizontal arrows) within the existing farming systems. Adaptation or change in production practices is seen as the appropriate way to address the HNV challenge, targeting the agricultural productive sector mainly. Farmers are then at the center of the engagement process (farm level). In these types of situations, the level of risk is moderated. A failure of the innovation-brokering process would lead to a status quo situation (circular doted arrows). For territories where the HNV strategic challenge is to maintain HNV farming systems in a general context of eroding farming activities (HNV dynamic type b), HNV visions follow the same visual pattern (diagonal arrow) showing that the HNV conservation objective aims both at energizing the agricultural sector and increasing its environmental performance. We observe that the innovation-brokering strategies are aimed at the agricultural sector (more specifically the extensive livestock sector). They target mainly public schemes and regulation (i.e., rebalancing policy schemes in favor of extensive farming systems through agri-environmental payments and others available CAP measures). They also focus their strategy on market dynamics using quality schemes to better valorize the agricultural production (cheese and meat products) and public regulations (on hygiene rules). The risk is that a decline of farming activities entails a decline of HNV attributes (dotted arrows). In the learning areas where the HNV conservation challenge is to increase the share of HNV agriculture in a context of agricultural reconquest (HNV dynamic type c), we can observe very different HNV conservation strategies (different directional arrows). The first HNV vision proposes an important change both in the agricultural dynamic as well as in the HNV conservation trends (diagonal arrow). This ambitious scenario builds on a territorial strategy of re-articulation of production systems between plain and mountain, mainly through pastoral farming (i.e., Learning area of Thessaly, Greece). It targets both the agricultural sector as well as other territorial actors. Alternatively, the second HNV vision is mainly articulated toward the development of agricultural projects linked with tourism activities, taking place on fallow recovery areas and former agricultural plains (i.e., Learning area of Dalmatian islands, Croatia). The aim being to reinforce HNV ecological features (mosaic agricultural landscapes) targeting rather small-scale agricultural activities and niche markets (i.e., orchards, aromatic plants, honey, small-scale breeding). But both visions are confronted by the same risk, which is to favor the agricultural reclaim dynamics with adverse impacts from an environmental view point. Finally, in territorial–landscape situations where the overall territorial context is characterized by stable agricultural development with good conservation of HNV attributes, as in The Burren learning area, Ireland (HNV dynamic type d), the HNV strategic challenge is to maintain an existing situation, balancing nature conservation with the development of farming activities. The HNV vision proposes to better anchor biodiversity conservation in individual farming practices thanks to locally led agro-environment schemes. The risk is to observe a progressive disengagement of farmers, and an intensification of practice, leading to the loss of the environmental value of HNV features (e.g., semi-natural vegetation maintained by extensive winter grazing) in this particular case.

Different design for engaging an innovation-brokering process in favor of HNV areas

In addition to better characterization of HNV dynamics, related conservation challenges and strategies in each learning area, we sought to explore the different designs of engagement within the HNV innovation process. Building on our comparative assessment grid (cf. Table 4), we were able to characterize different types of HNV vision from a procedural view point. These were characterized according to the four constitutive dimensions of facilitation processes in the field of agricultural innovation, as proposed by Loeber and Vermeulen (2012). Our results show three types of HNV vision-building processes. In some cases, HNV visions result in a rather general territorial narrative based on experts’ work and a participatory consultative process. This type of visioning exercise delivers a partial analysis of the production systems as well as a view of its impacts on HNV conservation dynamics (responsiveness). Even if the HNV vision is based on descriptive elements involving different disciplines, it does not deliver an actionable understanding at a territorial–landscape level. This kind of partial and scattered narrative does not provide a strategic understanding of the system of actors and key levers for transformative collective actions in favor of HNV territories (power). This production is a good start but proved to be too general to inform the local debate with an actionable ideal perspective and will not be a major tool for stakeholder engagement in the long run (anchoring). Some visions prove to deliver a detailed and explicit HNV vision in terms of landscape and ecological dynamics based on experts’ work and a co-production participatory process with some actors of the territory. This type of exercise makes explicit the functioning of agricultural production systems and related ecological issues (responsiveness). It builds on multidisciplinary material describing the socioeconomic functioning of farming systems and ecological conservation trends at the territory–landscape level. It displays a synthetic understanding of major HNV conservation issues through a visualization component (knowledge). Key strategic actors for the maintenance of the HNV farming system in the area are identified, as well as socioeconomic and/or technical lock-ins (power). This type of HNV vision might be useful for engagement but lacks being anchored within the current system of actors. It does not provide an actionable roadmap for potential ambassadors of HNV innovations (anchoring). Finally, a third type of HNV vision convincingly combines landscape and ecological strategic perspectives based on a co-construction participatory process with key actors having an impact on HNV dynamics in the area (responsiveness). As in the previous situation, this type of visioning exercise builds on rich multidisciplinary material to deliver a good understanding of the socioeconomic trends in farming systems, as well as a clear understanding of ecological conservation dynamics and respective challenges, into an actionable assessment (knowledge). This type of HNV vision is co-constructed with strategic actors for the maintenance of HNV farming. Possible options for change at the landscape–territorial level are identified. It articulates a multi-actor strategy in the medium- and long-term. It proves to be useful for engagement and for selection of transformative HNV innovations (anchoring).

DISCUSSION

Reconsidering the understanding of HNV conservation challenges

Our results show an important diversity of territorial situations, both in terms of HNV conservation patterns and their related strategic challenges. Indeed, the founding texts of HNV farming insist on the fact that low-input farming is a necessary condition for HNV conservation (Beaufoy et al. 1994, Bignal and McCracken 1996). Thus, when this low-input farming is lost, HNV attributes are also lost. However, we illustrate that this prevailing perception of HNV conservation challenges is only one among other possibilities. The loss of HNV attributes results from a gradient of land abandonment and of intensification (Bignal and McCracken 1996) (HNV dynamics b and c). In some cases, the risk of intensification prevails over the risk of abandonment, even though it does not fit the archetypal image of HNV areas (e.g., HNV dynamics a and d). This consideration changes the strategic perspective of the HNV broker and thus the relevant approach for engagement. Therefore, agricultural development in HNV farmland areas does not necessarily correspond to the conservation of HNV farming (HNV scenarios in Lomba et al. 2020). For instance, the innovation of branding local cheese produced by HNV farmers to attract a higher price illustrates this aspect. It might appear to be an appealing innovation per se as it may combine biodiversity conservation and farm viability, regardless of the context. However, its environmental benefits are not guaranteed and depend on many other factors. If such an innovation takes place in areas with potential for intensification and a weak socio-institutional setting, there is a high risk that this brand will encourage land intensification and farm concentration. The weak institutional setting is unable to monitor the market development on the basis of environmental and social criteria. It may correspond to increased viability of (former) HNV farms in a (former) HNV area. Indeed, such trends have been described for some areas producing cheese under protected denomination of origin, with adverse impacts on biodiversity (Quetier et al. 2005, Duvaleix-Treguer et al. 2018). Therefore, we argue that it is crucial for a HNV innovation broker to understand the territorial context and HNV dynamics when engaging. They need to consider the two axes: 1) the dynamics in place in terms of the development of farming systems (under the main indicator of overall land use: does agriculture maintain its role in land use or is it losing importance?); and 2) the dynamics in terms of HNV conservation (Lomba et al. 2014). An accurate analysis is needed to ensure that biodiversity conservation and agricultural development go side-by-side.

HNV conservation challenges: the importance of strategically framing the innovation-brokering process through a landscape–territorial approach

Our results led us to revisit the classic diffusionist approach of innovation-brokering (OECD 2005) and better consider its strategic and landscape–territorial dimensions. Along with Klerkx et al. (2012:458), we highlight the social dimensions of innovation. This involves not only a change of technology but rather a holistic vision of what the future should look like and the changes that are needed to achieve it. In that sense, our results show that for a given type of HNV agri-environemental dynamic, HNV conservation challenges call for different strategies and involve different levels of risk from an agri-environmental perspective. Thus, the framing of the HNV conservation challenges (i.e., what is at stake with HNV farming within a given landscape–territory?) is of crucial importance, as well as the degree to which the framing is shared within the learning area. If we go back to the definition of Schön and Rein, frames are implicit values, goals, or underlying assumptions that designate or make visible specific issues and how they should be best addressed (Schön and Rein 1994). The selection of a desired end state implicitly includes the identification of a problem or a focused narrative of collective action. This step then determines the constitution of a coalition of actors, the possibility of creating a “collective becoming” (Metzger 2013) or a “vision.” It would help to formulate a set of values, norms, cognitive and shared conception of reality and meaning, as well as related habits and routines, in a new and coherent organizational system. In that sense, HNV innovation brokers are displaying a wide range of activities to negotiate shared frames. The objective is to build a game-changing coalition of actors to achieve an agri-environmental objective: the conservation of HNV attributes in a given territory. This calls for a more reflexive approach to engaging successful territorial coalitions and leads to the question of how to tactically equip HNV innovation brokers to achieve successful engagement (Mizrahi and Rosenthal 2001, Aldrich and Marsden 1989, Ospina Peralta and Hollenstein 2015). Our exploration of the different HNV vision-building processes builds on a strategic landscape–territorial approach and leads to different levels of achievement. This raises a fundamental question: is it preferable to build a HNV desirable vision under the form of a long-term perspective of what constitutes a sustainable HNV landscape–territory (Albrechts et al. 2020, Metzger 2013), or is it more efficient to focus on specific immediate needs to deliver socioeconomic viability of the farming systems in HNV areas? That is assuming the latter coincides with an overall biodiversity conservation goal.

Exploring the strategic dimension of innovation process design: the ideal versus practical engagement for the HNV innovation broker

In their analysis of the facilitation of sustainable agricultural projects, Loeber and Vermeulen (2012) distinguished two different types of brokering approaches. The first approach aims at developing an understanding of, and responsiveness to, participants’ needs as a basis for pushing ideas and visions beyond their immediate reach (henceforth “ideal”). The facilitator designs the project as a co-learning process to induce reflexivity from the participants. The second approach of brokering is “designing implementable structures” (henceforth “practical”). It “strives for an understanding of and responsiveness to the participants’ needs and motives to gain their mandate for steering the process toward a shared vision and an implementable design.” The aim is to steer toward tangible and rapid results, building on intermediate milestones. The two approaches do not have to be in opposition but are meant to be complementary. However, this complementarity cannot be taken for granted as each approach insists on different dimensions and entry points for innovation and HNV farming coalition building. There is a need for a HNV broker to make a strategic choice when engaging with other actors. When developing innovation for HNV conservation, they need to decide on what aspects they insist on introducing into the (already existing) local process: as an ideal or a practical broker. Although it should be made clear that the words themselves should not be literally used as such. In our research, the engagement process explicitly included HNV agri-environmental dynamics as the basis for a landscape–territory vision. This placed the exercise in an “ideal” perspective. The enrollment of actors was based on the common understanding of what is at stake for the long-term biodiversity management of the area (HNV conservation challenges). It insists on the social process component of innovation, highlighting the role of strategic narratives in territorial transformations in favor of HNV conservation. In this perspective, the HNV innovation broker’s role is to bring an explicit HNV conservation goal into the actors’ interactions and to facilitate their understanding and sharing across a growing community of actors. Our results show that, in some situations, the visioning exercise proved to be a successful tool to frame the engagement process and strategically build a coalition promoting HNV innovations, but not always. The issue of the social capital of the innovation broker, the nature of the boundary organization from which they operate (Leith et al. 2016), and, therefore, the financial and human resources they can mobilize are central to success. Though seemingly obvious, this is frequently underestimated. The alternative to this first option (i.e., building a HNV vision) is to engage in an innovation process from a practical standpoint by focusing on practical innovations aimed at improving farm viability. Our results show that it could be a choice with higher risk from an agroecological perspective. This is especially true in territorial situations where the risk of agricultural intensification is prevailing. Therefore, we claim that it is important to design any innovation-brokering strategy based on some inputs from an “ideal” strategy. We argue that there is a need to clearly articulate the risk. This is of particular importance where it is rarely accepted as such; in contexts where a certain level of intensification might be judged as normal from a farming perspective and too high from a biodiversity conservation perspective. Developing a landscape–territory vision helps to justify HNV conservation goals and equip a collective learning process with evaluative agri-environmental landmarks. Otherwise, a practical approach alone may result in the risk of engaging with actors on the basis of actions which respond to their expressed needs (i.e., to achieve socioeconomic viability of agriculture) without visibility of the impact on the environment of the HNV landscape–territory in the medium-term.

CONCLUSION

This study argues for a strategic approach to HNV conservation at the landscape–territory level. Building on existing literature on innovation-brokering and coalition building. It explores distinct designs of the engagement process for HNV innovation within different territorial settings. It goes beyond the classical definition of innovation-brokering and challenges the neutrality of innovation brokers. As strategic actors, they endorse an active commitment in favor of HNV area conservation. They act as coalition builders, reconciling biodiversity conservation and agricultural development needs. The issue then turns to how to engage with other actors to have an impact on the wider environment? What does the broker promise to be credible? We demonstrate the role of landscape–territory visions and visioning processes as guiding principles which are a prominent characteristic of the brokering process. The process of engagement and the overall framing have to be considered carefully within the overall territorial context of the area. In this perspective, the intention of this study was to show that there is an issue for such engagement and a need to improve understanding of the context in which a HNV broker might engage. Building on Loeber and Vermeulen’s (2012) work on approaches and attitudes developed by facilitators in the field of agricultural innovation, we discussed two different designs of coalition building for HNV innovations. The practical approach proposes a set of innovation solutions as an entry point in the social process. This might be appealing in a context where biodiversity conservation is rarely a priority within the mainstream agricultural agenda. In this situation, the needs of biodiversity conservation may disappear from the agenda of the coalition in the medium- and long-term. Another more indirect way to engage is to collectively frame the HNV conservation issue, building on a long-term territorial–landscape vision (i.e., the ideal approach). This strategic approach to innovation brokerage requires being explicit about what is at stake for biodiversity conservation for a given territory and what needs to change in the course of agricultural development (framing). Such understanding of the innovation-brokering process is addressing two blind spots of the classic diffusionist approach of innovation. It questions the nature of the innovation itself in a territorial–landscape perspective and it also raises the issue of the broker’s “assets” (social, scientific, and financial capital) in a given organizational field. The challenge is to design an engagement process that articulates quick wins (implementable structures) with sufficient strategic reflexivity (landscape–territory vision) so that they effectively answer HNV conservation challenges. This can pave the way for an incremental strategy which seeks to secure agri-environmental gains for agricultural systems with high nature value in the long-term.

LITERATURE CITED

Adams, W. C. 2015. Conducting Semi-Structured Interviews. Pages 492-505 in K. E. Newcomer, H. P. Hatry, and J. S. Wholey, editors. Handbook of Practical Program Evaluation. John Wiley & Sons, Inc., Hoboken, New jersey, USA. https://doi.org/10.1002/9781119171386.ch19

Albrechts, L., A. Barbanente, and V. Monno. 2020. Practicing transformative planning: the territory-landscape plan as a catalyst for change. City, Territory and Architecture 7(1):1. http://doi.org/10.1186/s40410-019-0111-2

Aldrich, H. E., and P. V. Marsden. 1989. Environments and organizations. Pages 361-392 in N. J. Smelser, editor. Handbook of Sociology. Sage Publications, Newbury Park, CA, USA. https://doi.org/10.1146/annurev.so.02.080176.000455

Almeida, M., C. Guerra, and T. Pinto-Correia. 2013. Unfolding relations between land cover and farm management: high nature value assessment in complex silvo-pastoral systems. Geografisk Tidsskrift-Danish Journal of Geography 113(2):97-108 https://doi.org/10.1080/00167223.2013.848611

Andersen, E., D. Baldock, H. Bennett, G. Beaufoy, E. Bignal, F. Bourwer, B. Elbersen, G. Eiden, F. Godeshchalk, G. Jones, D. Mac Cracken, W. Neiuwenhuizen, M. Van Eupen, S. Hennekensand, and G. Zervas. 2004. Developing a high nature value farming area indicator: final report. Report to the European Environment Agency, Copenhaguen, Denmark. https://ieep.eu/publications/developing-a-high-nature-value-farming-area-indicator

Beaufoy, G., D. Baldock, and J. Clarke. 1994. The nature of farming - low intensity farming systems in nine European countries. Institute for European Environmental Policy, London, UK.

Beaufoy, G. (editor), HNV-Link Partners. 2017. THE HNV-LINK COMPENDIUM. Comparative collection of High Nature Value innovations, experiences, needs and lessons, from 10 European “Learning Areas.” Cuacos (Spain): EFNCP & Montpellier (France): CIHEAM-IAMM, (HNV-Link H2020 Project). http://www.hnvlink.eu/download/D2.6_HNVLinkCOMPENDIUM.pdf

Beckman, S. L., and M. Barry. 2007. Innovation as a learning process: embedding design thinking. California Management Review 50(1):25-56. https://doi.org/10.2307/41166415

Beers, P. J., F. Hermans, T. Veldkamp, and J. Hinssen. 2014. Social learning inside and outside transition projects: playing free jazz for a heavy metal audience. NJAS - Wageningen Journal of Life Sciences 69:5-13. https://doi.org/10.1016/j.njas.2013.10.001

Berthet, E. T., G. M. Hickey, and L. Klerkx. 2018. Opening design and innovation processes in agriculture: Insights from design and management sciences and future directions. Agricultural Systems 165:111-115. https://doi.org/10.1016/j.agsy.2018.06.004

Bignal, E. M., and D. I. Mccracken. 1996. Low-Intensity farming systems in the conservation of the countryside. Journal of Applied Ecology 33(3):413-424. https://doi.org/10.2307/2404973

Burke, J. G., P. O'campo, G. L. Peak, A. C. Gielen, K. A. Mcdonnell, and W. M. Trochim. 2005. An introduction to concept mapping as a participatory public health research method. Quality Health Research 15(10):1392-1410. https://doi.org/10.1177/1049732305278876

Callon, M. 1984. Some elements of a sociology of translation: domestication of the scallops and the fishermen of St Brieuc Bay. The sociological review 32(1_suppl):196-233. https://doi.org/10.1111/j.1467-954X.1984.tb00113.x

Coffey, A., and P. Atkinson. 1996. Making Sense of Qualitative Data. Complementary Research Strategies. Sage Publications, Thousand Oaks, CA, USA.

Dedeurwaerdere, T., A. Polard, and P. Melindi-Ghidi. 2015. The role of network bridging organisations in compensation payments for agri-environmental services under the EU Common Agricultural Policy. Ecological Economics 119:24-38. https://doi.org/10.1016/j.ecolecon.2015.07.025

Duvaleix-Treguer, S., C. Emlinger, C. Gaigné, and K. Latouche. 2018. On the competitiveness effects of quality labels: evidence from the French cheese industry, working paper. CEPII. http://www.cepii.fr/PDF_PUB/wp/2018/wp2018-17.pdf

Eip-Agri (European Innovation Partnership). 2016. How to make HNV farming more profitable without losing the HNV characteristics? Report of the EIP Focus Group on HNV farming profitability. https://ec.europa.eu/eip/agriculture/en/focus-groups/high-nature-value-hnv-farming-profitability

Eu Scar Akis (Standing Committee on Agricultural Research: Agricultural Knowledge and innovation Systems). 2019. Preparing for Future AKIS in Europe, European Commission, Brussels. https://scar-europe.org/index.php/akis-documents

European Environment Agency. 2013. The European Grassland Butterfly Indicator: 1990-201, EEA Technical report n°11/2013. Publications Office of the European Union, Luxembourg. https://www.eea.europa.eu/publications/the-european-grassland-butterfly-indicator-19902011/at_download/file

Fischer, J., T. Hartel, and T. Kuemmerle. 2012. Conservation policy in traditional farming landscapes. Conservation Letters 5(3):167-175. https://doi.org/10.1111/j.1755-263X.2012.00227.x

Hey, J., C. Joyce, and S. Beckman. 2007. Framing innovation: negotiating shared frames during early design phases. Journal of Design Research 6(1-2):79-99. https://doi.org/10.1504/JDR.2007.015564

Howells, J. 2006. Intermediation and the role of intermediaries in innovation. Research Policy 35(5):715-728. https://doi.org/10.1016/j.respol.2006.03.005

Jones, G., and X. Poux. 2012. Socio-economic perspective on HNV farming. Pages 451-458 in R. Opperman, G. Beaufoy, and G. Jones, editors. High Nature Farming in Europe, Verlag Regionalkultur, Ubstadt-Weiher, Basel.

Kilduff, M., R. Angelmar, and A. Mehra. 2000. Top management-team diversity and firm performance: examining the role of cognitions. Organization Science 11(1):21-34. https://doi.org/10.1287/orsc.11.1.21.12569

Kirat, T., and Y. Lung. 1999. Innovation and proximity: territories as loci of collective learning processes. European Urban and Regional Studies 6(1):27-38. https://doi.org/10.1177/096977649900600103

Klerkx, L., A. Hall, and C. Leeuwis. 2009. Strengthening agricultural innovation capacity: are innovation brokers the answer? International Journal of Agricultural Resources, Governance and Ecology 8(5-6):409-438. https://doi.org/10.1504/IJARGE.2009.032643

Klerkx, L., and R. Nettle. 2013. Achievements and challenges of innovation co-production support initiatives in the Australian and Dutch dairy sectors: A comparative study. Food Policy 40:74-89. https://doi.org/10.1016/j.foodpol.2013.02.004

Klerkx, L., E. Petter Stræte, G.-T. Kvam, E. Ystad, and R. M. Butli Hårstad. 2017. Achieving best-fit configurations through advisory subsystems in AKIS: case studies of advisory service provisioning for diverse types of farmers in Norway. The Journal of Agricultural Education and Extension 23(3):213-229. https://doi.org/10.1080/1389224X.2017.1320640

Klerkx, L., B. Van Mierlo, and C. Leeuwis. 2012. Pages 457-483 in Evolution of systems approaches to agricultural innovation: concepts, analysis and interventions. Farming Systems Research into the 21st century: The new dynamic. Springer, Netherlands. https://doi.org/10.1007/978-94-007-4503-2_20

Leith, P., M. Haward, C. Rees, and E. Ogier. 2016. Success and evolution of a boundary organization. Science, Technology & Human Values 41(3):375-401. https://doi.org/10.1177/0162243915601900

Loeber, A., and T. Vermeulen. 2012. The art of ‘doing’ sustainable agricultural innovation: approaches and attitudes to facilitating transitional projects. Pages 102-117 in M. Barbier, and B. Elzen, editors. System Innovations, Knowledge Regimes, and Design Practices towards Transitions for Sustainable Agriculture. INRA, France.

Lomba, A., C. Guerra, J. Alonso, J. Pradinho Honrado, R. Jongman, and D. Mccracken. 2014. Mapping and monitoring High Nature Value farmlands: challenges in European Landscapes. Journal of Environmental Management 143:140-150. https://doi.org/10.1016/j.jenvman.2014.04.029

Lomba, A., F. Moreira, S. Klimek, R. H. Jongman, C. Sullivan, J. Moran, X. Poux, J. P. Honrado, T. Pinto-Correia, T. Plieninger, and D. I. Mccracken. 2020. Back to the future: rethinking socioecological systems underlying high nature value farmlands. Frontiers in Ecology and the Environment 18(1):36-42. https://doi.org/10.1002/fee.2116

Mermet, L., K. Homewood, A. Dobson, and R. Billé. 2013. Five paradigms of collective action underlying the human dimension of conservation. Pages 42-58 in D. W. Macdonald, and K. J. Willis, editors. Key Topics in Conservation Biology. John Wiley & Sons, Ltd., Hoboken, New Jersey, USA.https://doi.org/10.1002/9781118520178.ch3

Metzger, J. 2013. Placing the Stakes: The enactment of territorial stakeholders in planning processes. Environment and Planning A: Economy and Space 45(4):781-796. https://doi.org/10.1068/a45116

Mizrahi, T., and B. B. Rosenthal. 2001. Complexities of coalition building: Leaders' successes, strategies, struggles, and solutions. Social work 46(1):63-78. https://doi.org/10.1093/sw/46.1.63

O'Rourke, E., and N. Kramm. 2012. High nature value (HNV) farming and the management of upland diversity. A review. European Countryside 4(2):116-133. https://sciendo.com/it/article/10.2478/v10091-012-0018-3

OECD 2005. Oslo Manual: Guidelines for Collecting and Interpreting Innovation Data, Eurostat, Paris.

Ospina Peralta, P., and P. Hollenstein. 2015. Territorial coalitions and rural dynamics in Ecuador. Why history matters. World Development 73:85-95. https://doi.org/10.1016/j.worlddev.2014.10.026

Pittaway, L., M. Robertson, K. Munir, D. Denyer, and A. Neely. 2004. Networking and innovation: a systematic review of the evidence. International journal of management reviews 5(3‐4):137-168. https://doi.org/10.1111/j.1460-8545.2004.00101.x

Plieninger, T., and C. Bieling. 2013. Resilience-based perspectives to guiding high-nature-value farmland through socioeconomic change. Ecology and Society 18(4):20. http://dx.doi.org/10.5751/ES-05877-180420

Plieninger, T., C. Hui, M. Gaertner, and L. Huntsinger. 2014. The impact of land abandonment on species richness and abundance in the Mediterranean Basin: a meta-analysis. PloS one 9(5):e98355. https://doi.org/10.1371/journal.pone.0098355

Poux, X., C. Bernard-Mongin, and F. Lerin. 2017. The HNV-Link Atlas-Crossed perspectives on 10 learning areas. Understanding the field of play for High Nature Value innovation projects. HNV-Link, EU H2020 project, Paris, France. http://www.hnvlink.eu/download/atlas-HNV-Link-D1.4.3.pdf

Poux, X., and J. Moran. 2017. Assessment of the baseline situation. Guideline for LA coordinators. HNV-Link, EU H2020 project, Paris, France. http://www.hnvlink.eu/download/D1.2.2WP1guidancenotefinal.compressed.pdf

Quetier, F., P. Marty, and J. Lepart. 2005. Farmers’ management strategies and land use in an agropastoral landscape: roquefort cheese production rules as a driver of change. Agricultural Systems 84(2):171-193. https://doi.org/10.1016/j.agsy.2004.05.005

Rogers, E. M. 1962. Diffusion of Innovations. Free Press, New York, USA. https://doi.org/10.1002/jps.2600520633

Schön, D. A., and M. Rein 1994. Frame Reflection: Towards the Resolution of Intractable Policy Controversies. Basic Books, New York, NY, USA.

Ustaoglu, E., and M. J. Collier. 2018. Farmland abandonment in Europe: an overview of drivers, consequences, and assessment of the sustainability implications. Environmental Reviews 26(4):396-416. https://doi.org/10.1139/er-2018-0001

Van Lenthe, H., M. Hekkert, S. S., and B. Van Waveren. 2003. Roles of systemics intermediaries in transition processes. International Journal of Innovation Management 7(03):247-279. https://doi.org/10.1142/S1363919603000817