Framing a Sustainable Local Food System—How Smaller Cities in Southern Germany Are Facing a New Policy Issue

: Since Agenda 21, the local level has become important in terms of facing global challenges through local action. One of these is ensuring the sustainability of the food system. In German politics, this is a relatively new issue even at the local level. Nevertheless, two smaller cities in southern Germany have decided to change their local food systems towards sustainability. Hence, this paper deals with questions of how local actors are framing the food system and what this means for increasing sustainability. The analysis of qualitative interviews and participant observations based on frame analysis provides deeper insights into understandings of local food systems by actors. This paper aims to explore how framings of problems, solutions and motivations provide or restrict opportunities to increase local food system sustainability. Terms like sustainability or awareness are framed differently. Using the same term to mean different things can have negative effects on the acceptance of policymaking referring to food system transformation. Besides, this paper shows that omissions within the framing counteract the development of sustainable local food policy. Hence, it is important to reﬂect the political implications of absent framings as well to facilitate mutual understanding and consequently, food system change.


Introduction
"By linking sustainable food systems with other policy objectives [ . . . ] food can play an integral role in a city's sustainable development policy" [1]. In this statement, the international organisation ICLEI-Local Governments for Sustainability addresses three key issues. First, food is presented as a relevant policy issue within the context of sustainability. Second, the significance of cities is emphasised. Finally, the statement proposes an integration of the food system into a city-level sustainable development policy.
Since Agenda 21, the value of the local level has taken on greater significance in terms of sustainability [2] (p. 550). There are numerous guiding principles at the local level dealing with sustainability in relation to energy or mobility. In contrast, food and nutrition are underrepresented in local policymaking. This is despite the fact that food accounts for about 13 per cent of per capita carbon emissions [3] (p. 44). Increasingly, different groups of actors stress the importance of transforming local food systems towards sustainability. For some time now, large cities have been the subject of research on local food policies, particularly with regards to the development of food councils or food strategies [4]. Smaller cities, on the other hand, have so far hardly been the subject of food policy research, although many people in Germany live in cities of this size [5] (p. 2). As cities become increasingly important in addressing sustainability issues, it is important not to leave out smaller cities but to focus attention on how they can contribute to the sustainability transformation of local food systems.
Hence, this paper deals with questions of how local actors in smaller cities are framing the food system and what their framing means for increasing food system sustainability. Given that many Germans live in small-and medium-sized cities [5] (p. 2), such as the analysed cases, A-Town and B-Town, it is important to understand framing and policymaking relating to sustainability in these settings (small cities are defined as mostly smaller than 20,000 inhabitants while medium-sized cities are defined as cities with less than 100,000 inhabitants). The underlying assumption is that policymaking is influenced by how actors who are implicitly or explicitly involved in the process frame the issue at hand [6] (p. 182). Frame analysis of local food systems in Germany and Europe has not often been done so far. Using the framing approach, this paper contributes to the scientific debate on the local food system transformation. It will become much clearer what different problem, solution and motivational framings local actors deal with. Thus, the framing approach facilitates an understanding of the difficulties to foster sustainability in local food systems, as actors were framing it differently. A further scientific contribution of this paper is that the analysis of the absence of framings complements the scientific debate. Since this access has rarely been chosen in previous studies on food policy. Focussing on framings used in literature and policy documents but not by the local stakeholders provides a deeper understanding of the prioritisation by local actors referring to food system transformation.
As a result, depicting problem, solution and motivational framings and as well the visualisation of absent framings can foster smaller cities to develop coherent food policies at the local level that integrate different groups of local actors in particular members of city administration and city council, economic actors and the civil society.
This paper starts with an overview of the current state of the research focusing on sustainability and food policy research. In the following sections, the framing is explained as an analytical tool guiding the analysis. Finally, the empirical results are presented, discussed and summarised.

Sustainability and Food Policy Research
This section provides an overview of the current state of research. First, it explains the understanding of sustainability, focusing on the discourse of sustainability within German politics. Second, it explores international and national food policy with particular focus on the academic discourse.

Three Pillars of Sustainability
Sustainability has been defined in various ways and the definition is contested in academic discourse. Originally, Hans Carl von Carlowitz defined sustainability in the forest as the need to not cut more wood than could regrow [7]. His aim was to prevent a deficit of wood because wood meant a guarantee of the welfare and autonomy of future generations [7].
The most common definition of sustainability in German political discourse is drawn from the final report of the Brundtland Commission of the United Nations in 1987 and the conference in Rio de Janeiro in 1992 [8] (p. 16). It comprises three pillars: Environment, economy and social welfare [8] (p. 18). The Study Commission of the German parliament formulated "the guarantee and improvement of ecological, economic and social efficiency" (Translation by the author) as the main aim of sustainability policy [8] (p. 19). The emphasis on efficiency indicates that the protection of nature is not seen as the final goal in itself, but rather that the future stability of the state is a significant benchmark.
In academic discourse, the three-pillar model has been criticised. The model is based on the assumption that the dimensions are of equal rank. Ott et al. criticised this assumption, suggesting that it hides the fact that, depending on the situation, one pillar has been subordinated within the process of modernity [9] (p. 39). The authors concluded that the model should be seen as temporary and, as such, in need of replacement [9] (p. 38). They suggested the categorisation of models of sustainability along the continuum of strong and weak sustainability [9] (p. 103 et sequence). The strongest understanding of sustainability accepts that nature's creatures have a moral intrinsic value, while the weakest understanding is oriented towards increasing the gross national product [9] (p. 103). The concepts of weak sustainability are especially interesting for economists [9] (p. 114) since they mainly focus on the benefits of nature for people [9] (p. 108). The understanding of benefits is mostly based on utilitarian concepts [9] (p. 109). That means that benefits occur if preferences are realized [9] (p. 109). With regard to nature this could imply that parts of nature will be conserved if people for instance like the sight. Following the concepts of weak sustainability, it is often assumed that people adjust their preferences to the existing surroundings [9] (p. 110). Hence, a loss of nature would not be a problem since the preferences of people would change accordingly. Ott et al. emphasis that it is not possible to know about the preferences of future generations [9] (p. 111). Another important aspect of concepts of weak sustainability is technological optimism [9] (p. 116). Following this argumentation, it would be possible to substitute natural resources by technical progress [9] (p. 116). The evaluation of the possibilities of substitution is one of the main differences between weak and strong sustainability [9] (p. 107). The concept of strong sustainability established by Herman Daly, conceptualises the economic system as a subsystem of a geosphere and biosphere [9] (p. 145). Representatives of strong sustainability assume a substantial complementary between nature capital and social capital [9] (p. 146). That means, in contrast to the concept of weak sustainability, that more social capital cannot substitute nature capital but that both types of capital belong together, for instance fish and fishing boat [9] (p. 146). Following Ott et al., the main advantage of the concept of strong sustainability is that it implies more free choice for future generations [9] (p. 167) since it does not suggest any preferences of future generations or completely rely on technical solutions. Finally, Ott et al. conclude that indeed the concept of strong sustainability is environmentally focused, but this is not the only focus [9] (p. 170). The authors' arguments are in line with strong concepts of sustainability while rejecting the concepts of weak sustainability [9] (p. 169). Furthermore, their understanding of sustainability implied that the deterioration of nature should be prohibited and that there should in fact be a requirement to enhance nature [9] (p. 169). In this conceptualisation of sustainability, the three pillars of the Brundtland Commission's model are modified as follows: Efficiency (concerning the economic dimension), sufficiency (referring to the social dimension) and resilience (in relation to the ecological dimension) [9] (p. 170). Aside from such criticisms, Tremmel assumes that academics in general will refer less often to the three-pillar model because of the very broad way in which it is defined by the German parliament's Study Commission [10] (p. 153).
Criticism of the three-pillar model can also be found in food policy literature. Morgan et al. describe "sustainability [as] a highly contested multi-dimensional concept in which trade-offs have to be made between social, economic and ecological values" [11] (p. 222). Smith et al. go one step further, pointing out that food systems "require a multidisciplinary, multi-stakeholder approach that pushes the boundaries beyond economic considerations to include political, social and ecological dimensions" [12] (p. 1). Hence, the definition of sustainability based on the three-pillar model "is being reconsidered as a suitable structure for policy agendas" [12] (p. 1). Nevertheless, it remains the most common model for defining sustainability, also in food policy literature. Hence, this definition is also used in this paper.
In addition to establishing the three-pillar model, the Brundtland Report emphasised the importance of cities for sustainability: "With the publication of the 1987 Brundtland Report, and its crystallisation of the concept of sustainable development, the remit of cities in addressing environmental issues was recast as central to the new agenda" [13] (p. 44). Additionally, van der Heijden pointed out that "[i]n international policymaking, cities are not recognized [sic] as formal actors" [14] (p.81). Though, they "are sites as well as actors when it comes to climate action" [14] (p. 81). This emphasizes the special role of cities within sustainability policymaking. The fact that they are fewer formal actors makes it possible for cities to "experiment with innovative governance instruments" [14] (p. 82). This in turn enables local actors to discuss climate challenges and possible means of addressing urban climate problems [14] (p. 93). In this sense, climate challenges become evident at the local level first and can also be directly addressed at this level. This section has demonstrated the wide variation in the way that sustainability is constructed in academic and political discourses in Germany. Furthermore, it has shown that cities are important actors in the field of sustainability. At the international policy level, it is considered that cities are less formal actors and can thereby operate in different and possibly even more innovative ways. Therefore, the following section aims to take a closer look at the significance of cities as actors in the field of sustainable food policy.

The Role of Cities in Sustainable Food Policy
At the global level, the constellation of actors influencing the food system has changed over time. As noted by Lang et al., food policy has in the past, "tended to be seen as the preserve of governments" [15] (p. 11). From this perspective, the government played the role of a supplier who needed to provide food for the population. "Today, however, the world of food is straddled by giant food and drink corporations who are equally, if not more, significant and de facto formulate their own food policies" [15] (p. 11). In this way, the factors influencing food policy have changed due to conglomerates as powerful new actors. Significantly, such conglomerates are not democratically legitimised and this creates the general conditions in which food policy is embedded, even at the local level.
Since World War II, food policy in Germany has been embedded in agricultural policy and has become an important topic in light of agriculture-related occurrences, such as food scandals. The relative prosperity of the country has made it possible for German politicians to deal less with questions of security of supply than with the question of whether food enables the population to lead a healthy life. Hence, food policy in Germany has primarily been related to health and consumer protection. This is reflected, for instance, in the naming of the ministries responsible for nutrition. For example, the section of the ministry for food and agriculture's homepage that deals especially with nutrition is entitled "healthy nutrition, safe food" (translation by the author) [16]. Barlösius identified the development after the Second World War as the third phase of food policy in Europe. Whereas in the first phase the policy was focused on the prevention of famine, food policy in the second phase aimed to enhance industrial and military efficiency [17] (p. 245). In the third phase, health was the major focus of food policy [17] (p. 246). Thus, food policy transitioned from economic governance to health policy [17] (p. 246). In contrast, sustainability is a relatively new topic in food policy discourse. According to the classification by Barlösius, the concept of sustainability has the potential to initiate the fourth phase.
Following Lang et al., the topic of food security was of central importance during the 20th century while "the core 21st-century task is to create a sustainable food system" [18] (p. 313). Topics, such as food security or food sovereignty, which have always been central to food policy discourse, are also part of the understanding of sustainability. Moreover, at the same time there are different efforts to force a linkage between food systems and sustainability at different levels. Since Agenda 21 and the development of food policy councils in the Global North, the local level has taken on greater importance. However, smaller cities have rarely been the subject of academic studies on this topic.
There are many general differences between larger and smaller cities, and differences relating specifically to food systems can also be identified. Morgan points out that "[w]hile small municipalities have been the real pioneers in getting high quality food into schools and hospitals for example, the larger city authorities have recently produced urban food strategies under the banners of public health, social justice or sustainability" [19] (p. 344). This means that the activities of smaller municipalities have so far focused mainly on supply within public institutions. Urban food strategies concerning sustainability have been established primarily in larger cities. Accordingly, food policy research located in the Global North has mainly focused on big cities, such as Vancouver [20], London, New York [11] or Cleveland [21].
The focus on big cities regarding food system transformation towards sustainability becomes obvious when looking at food policy councils. The first of these was founded in Knoxville, Tennessee in 1982. Referring to the analysis of food strategies from Canada, the USA and the United Kingdom, Sonnino suggests that food strategies in general contain "'notions of 'freshness' and 'healthiness'" that "are central in the narratives of urban food strategies. Significantly, however, they are never discussed in isolation from other sustainability objectives" [4] (p. 194). In Germany, local food initiatives are also mainly located in big cities, such as Cologne and Berlin, where the first German food councils were established in 2016. In this kind of initiative, different groups of actors, such as civil society, economic actors and politicians are involved, even if these are bottom-up initiatives that have emerged through civil society involvement. The establishment of local food councils in Germany is part of the most recent history. One output of these can be the development of a food policy strategy. Germany and other European governments "have developed or are discussing the development of national food strategies" [22] (p. 90). While there is currently no official national food strategy in Germany, such an initiative is under consideration. In contrast, analysing cases in the Netherlands and the United Kingdom, Hebinck et al. show that "Urban Food Strategies have become regarded as a pivotal point of change in addressing food systems failings and transitioning to sustainable food systems at the local level" [23] (p. 1). Given that Germany does not have a national food strategy, it is hardly surprising that the body of literature regarding sustainable local food systems in Germany is rather small.

Localisation of Food
Against the background of globalised food production, processing, trade and disposal, food system localisation becomes an important counterpoint in the political and academic discussions about sustainable food [12] (p. 1). Different sociologists have considered food-related actions, such as preparation and consumption as significant practices for socialisation in general [17,24,25]. Their research provides various insights into the constituting role of food-related practices in different societies. Elias describes the process of civilisation as a development of new concepts of shame that are manifested, for instance, in the analysis of changing table manners [24]. Bourdieu focuses on processes of distinguishing between different social classes based on eating habits [25]. Barlösius deals with both the phenomena of communitization and socialisation, as well as with processes of making distinctions [17]. Thus, food and eating habits are always related to questions of identity, whether it be personal or group identity, such as in the case of a city. This link between food and identity is an important political issue that is often associated with the localisation of food systems.
Obviously, where cities are seeking to increase the sustainability of local food systems, localisation becomes an important part of the policy discussion. However, there is no clear definition of what is meant by localisation. In general, the aim of localisation is that all food-related processes ideally take place at the local level and with the participation of local actors. With reference to the three-pillar model of sustainability, there are three basic assumptions as to why food system localisation may be an adequate response to problems related to globalisation processes. With regard to the ecological pillar, it is assumed that the food system localisation implies, for instance, less transport and correspondingly reduced carbon emissions. The social pillar implies a closer relationship of trust between producers and consumers that leads to an increasing local added value. This has the effect that "food system localization is often assumed to be a good, progressive and desirable process" [26] (p. 33). However, the assumption that food system localisation is inherently good is a fallacy. Morgan et al. point out that even if localisation is a necessary basis for a sustainable food system, localisation and sustainability are not necessarily one and the same thing [11] (p. 212). A sharp criticism of the concept of localisation is expressed by Born et al. They conclude that a scale cannot be inherently good or bad but rather that the quality of a food system depends on its content, including "the actors and agendas that are empowered by the particular social relations in a given food system" [27] (p. 196). Hence, they recommend that urban planners in particular avoid the local trap. Even researchers sometimes become ensnared in the local trap [27] (p. 200). The empirical part of this paper shows that food system localisation, and consequently also the local trap, plays an important role in the analysed cases.

Sustainable Local Food Policy
This research focuses on the framing of sustainable local food policy. The second part of the paper is split into the underlying concept of sustainability used, the role of cities in sustainable food policy and the importance of the local level. It was shown above that food has always been an important issue for policymakers. However, the current discourse in the countries of the Global North has only just started to deal with this issue in relation to sustainability. While the operationalisation of sustainability with reference to the three-pillar model is contested, it is nevertheless the most common model in food policy literature and is thus also used in this paper. Since Agenda 21, cities have become important actors in terms of sustainable local policy. However, the focus on the local level can be accompanied by an overemphasis of local identity. To date, sustainable food policy research has been much more focused on larger cities than on smaller ones. This has created a research gap that is addressed in this paper. It has been shown that discussion around the role of smaller cities is an important contribution to sustainable food policy literature. Referring to the research question, the next section sheds light on the framing approach and why it is useful in this context.

The Framing Approach
This paper discusses framing within the context of local food system changes. It is assumed that the concept of framing is a useful approach since it enables access to an interactive and intersubjective process [28] (p. 93) in which problems, solutions and motivations-which in turn are the basis for policymaking-are constructed.
There is no consistent definition of what is meant by 'frame' or 'framing'. Hence, it is important to clarify the understanding applied in this paper. One of the first definitions of framing stems from Erving Goffman who understands the concept as principles of organisation that help to define a social situation [29] (p. 19). In this sense, framing is a process in which people structure incidents and experience. In contrast, "'frames' are often treated as objects people possess in their heads and develop for explicitly strategic purposes" [28] (p. 93). Van Hulst et al. provide an approach that is useful for analysing framing within policymaking, assuming that, "[i]n policy-making, framing is a process in and through which policy-relevant actors intersubjectively construct the meanings of the policy-relevant situations with which they are involved, whether directly or as onlookers and stakeholders" [28] (p. 97). Since food policy is a new topic for municipalities, this paper refers to the approach of Benford et al. This approach does not delve as deeply as the elaborated framing approach by van Hulst et al. Nevertheless, the framing approach of Benford et al. is well suited to this research as they organise different framings into three main categories: Problem framing, solution framing and motivational framing. These categories are very useful for staking out sustainable food as a new issue in policymaking at the local level. The construction of meanings stressed by van Hulst et al. remains a basic component of the framing approach used herein.
Benford et al. focus on social movements [30]. Accordingly, they divide their understanding of framing into three categories as noted above. The first is diagnostic framing, also referred to as problem framing. It is connected to the identification of an entity that can be blamed for, or addressed as, being responsible for a problem [31] (p. 110). Moreover, for an issue to be added to the political agenda it has to have become a problem beforehand [32] (p. 231). "It is broadly acknowledged that the way in which problems are framed and how knowledge is produced has significant implications for policy development and societal outcomes" [33] (p. 35). In that sense, this paper assumes that the framing of problems is a fundamental aspect of policymaking.
The second category of framing refers to solutions, whereas the third category is motivational framing [30] (p. 615). Based on the understanding of Benford et al., this paper focuses on analysing the framing of problems, solutions and motivations within two smaller cities in southern Germany. Through these two cases, it is possible to examine current framings of the food system at a local level. In contrast to Benford et al., this analysis is not based on social movements as these are not present in the cases examined.
However, the focus on social movements provides the opportunity to analyse framing in light of the research question of this paper: How is the local food system framed by local actors in smaller cities in southern Germany with regard to sustainability? Within food policy research in Germany, analysis of local food systems in small cities has so far been limited despite the fact that most people live in cities of this size. A further research gap relates to the methodological approach. In contrast to the usual approach, which involves analysing articles or official documents, this research involved qualitative interviews with local actors, as well as participant observations. This use of a qualitative approach to focus on the local food systems in two smaller cities in Germany addresses the identified gaps in current research.
In contrast, Moschitz uses a slightly different approach. In her paper she aims "to explore the ways in which food is framed in official policies in Switzerland and thereby gain a better understanding of the potential for the development of urban food policies" [6] (p. 180). Using the frame approach for the analysis of documents, Moschitz focuses on the status quo. In comparison, this paper is based on qualitative interviews and participant observations using the framing approach. Hence, it refers to the process of policymaking.
The process of framing includes different kinds of articulation, whether it be verbal or in the form of policy documents [30] (p. 616). This focus on articulation neglects all those things that are left unsaid even though the unspoken things are of great political importance. As such, the framing approach is not useful for understanding the framing of verbal exclusion. This gap can be closed by using situational analysis which covers the whole situation, including underlying discourses or the role of nonhuman actors and actants [34]. There can be different reasons for the fact that specific things remain unsaid. Two of these are taboo subjects that people are not justified to talk about and self-evident things that people do not need to talk about.
The following analysis will show that the framing of problems, solutions and motivations together with the framing of the unsaid can illuminate how local actors in smaller cities are framing the food system and what their framing means for increasing food system sustainability.

Empirical Design
Framing understood as the construction of meanings [28] (p. 97) or the definition of a situation [29] (p. 19), as discussed above, calls for a qualitative approach. Qualitative social research focuses on either subjective perspectives or collective patterns of orientation [35] (p. 240). Therefore, semi-structured interviews with local actors are necessary. Firstly, semi-structured interviews are open enough for the interviewees to develop their own viewpoints and priorities. Secondly, the focus on local actors provides insights into their construction of reality, which affects policy change and cannot be garnered only through analysing policy documents.

Sampling and Methods
As mentioned above, most research on food policy and local food systems has been carried out in big cities. In contrast, this paper is interested in smaller cities because these represent the reality for a larger proportion of citizens [36]. Thus, the essential criterion was the size of the cities. Embedded in the project "KERNiG" two concrete cases, A-Town and B-Town, were selected. KERNiG, a German acronym for "Municipal food systems as the key to comprehensive integrative sustainability governance" is a collaborative research project founded by the Federal Ministry of Education and Research. It started in September 2016 and will run over three years. The overarching goal is to initiate and conduct research in two smaller cities on a locally sustainable transformation involving active shaping of the food system. Both cities have already made efforts regarding sustainability and climate protection. So far, one of the cities has focused on sustainable energy and mobility while the other one is part of a food-related movement. Both cities have previous experience in implementing issues related to sustainability, and partly even related to food. That means that dealing with these issues is not completely new for them. Hence, they were selected as suitable cases for combining both issues at the local level. A third important selection criterion was that both cities should be different with regard to the surroundings to find out if this would have an influence on the framing. While one of the cities is very close to a big city, the other one is located in a rural area. Since production is a very important part of the food system, the two cities were selected against the background of their dominating agriculture as well. While agriculture in the rural area is especially based on dairy farming, which is often done organically, agriculture in the other city is especially based on special crops, mostly done conventionally. The decision to analyse two cities and not just one was based on the question of whether or not differences between cities could be identified.
After the case selection, the sampling structure was the main determinant within the empirical design [37] (p. 178). Focusing on policy change, this research sample was based on actors relevant to the processes of policy change concerning the topic of food at the local level ( Table 1). The food policy literature defines three groups of relevant actors significant for the food system: "[S]tate, supply chain and civil society" [15] (p. 11). Transferred to the local level, it was useful to split the group of state actors into local politicians and city administration. The first predetermined sampling criterion [37] (p. 182) was to cover these four predefined groups. Since the aim was to gain a comprehensive overview of the framing of the food system in the two cases, the second criterion was that the sample includes actors from different policy sectors, such as, for instance, tourism, education, nature conservation or urban planning. Obviously one actor can be part of more than one sector. The third sampling criterion was the importance of actors concerning the process of policy change. The importance of an actor can be assessed on the basis of their position (e.g., faction head, head of a department, director of a local company, member of a citizens' initiative, etc.) or based on the fact that the actor was identified by other interviewees as being relevant for local decision-making processes. Whether an actor was referred to by other actors was not a predefined criterion but rather a result of a snowball procedure [37] (p. 184). In this case, they also had to fit the predetermined sampling criteria. In addition, the sampling strategy took into account different genders and a range of ages.
In addition to the semi-structured interviews, participatory observations focusing on collective patterns of orientation were undertaken. The participant observation took place in ten stakeholder workshops, initiated by the KERNiG project, and dealing with the potential development of the local food system in the two cases. Actors identified as relevant based on the sampling criteria were invited to participate in these workshops. In addition, council meetings were observed in which the aims and measures concerning the local food system were resolved. Furthermore, meetings of citizen initiatives dealing with the topic of food or sustainability were also observed.

Analysis Procedure
This paper presents empirical results based on 26 semi-structured interviews. The survey period started in February 2017 and ended in December 2017. The interviews were transcribed, anonymised and analysed using qualitative text analysis with reference to Mayring [38]. Although this approach is used quite often, the qualitative text analysis is described as weak in terms of the analysis of sense-making processes like framing [37] (p. 189). Hence, the qualitative text analysis was complemented with an in-depth analysis of sequences when a deeper analysis appeared to be useful with regard to the framing process. The coding was carried out using computer-assisted MAXQDA (VERBI Software GmbH, Berlin, Germany). Based on the analysis scheme provided by the framing theory of Benford et al., the main deductive codes were problem, solution and motivation. The characteristics of each code were identified as deductively and inductively (see Table A1 in Appendix A). Framing is a process of sense-making. The following is a simple example of how the different framings were identified. If an interviewee pointed out that children eat too much fast food and that this causes malnutrition, I defined this as a framing of a problem on the basis of negative valuations, such as "too much" and "malnutrition". The superordinate framing is the framing of the problem. Furthermore, such a framing as a framing of health was defined, since malnutrition causes health problems. A similar but slightly different example is the following: An interviewee pointed out that children eat too much fast food and that this could only be prevented by implementing a school subject dealing with nutrition. In this case, a problem framing concerning eating habits of children that are valued in a negative way was identified. Moreover, this sentence provides a framing of a solution.
Since the proposed solution is a school subject, education as a solution framing was identified. These examples illustrate the analysis process on which the following results are based.
In addition to the framing, the research analysed which group of actors (politicians/ administration, economic actors and civil society) operates with what kind of framing and if there are differences regarding the two cases and/or the groups of actors. According to the concept of problem framing by Benford et al., this approach provides information about how blame is assigned. Moreover, the additional focus on actors provides insights into different or similar motivations decisive for policy change.
In the following section, the empirical findings of the framing process in the two cases in southern Germany are presented and analysed. The results are structured in the following way: Following Benford et al., the empirical results are divided into the three superordinate categories: Problem framing, solution framing and motivational framing. The framing of problems is divided into four components. First, the problem is described. Second, and with reference to Bagdonis et al. [31] (p. 110), it is explained who is blamed for the existence of the problem. In the third step, the cited solution framings are outlined. Last but not least, the groups of actors using this kind of problem framing are identified.
Based on Benford et al., the second important aspect of framing is the framing of solutions. This section is not structured in a particular way given that only one main solution was identified in the interviews. Likewise, the framing of motivations does not need a special structure. In this section, different motivations of the actor groups are explained.

Empirical Results
The first important finding is that there are no substantial differences between the two cases with regard to framing. That said, if it is not the settings within individual cities that are decisive for framing, it is important to more closely examine what other factors may constitute the framing. To this end, the presentation of the empirical results starts with the framing of problems, followed by the framing of solutions and motivations.

Problem
In the two cities, food security was not apparent as a relevant problem framing because the food supply is seen as guaranteed. Expressions like "we are principally well-positioned" illustrate the absence of food security as a problem framing. (Interviewee BS15, cf. Interviewee AS6; AS12; AS15) (All interviewees were German speaking. Direct quotations were translated by the author of this paper).
The only identified problem framing concerning security is related to small-scale business in agriculture, food service and the crafts-based food industry. These actors describe their economic existence as being directly threatened. They gave two reasons for this. The first is regarding new laws passed by the European Union that specifically concern hygiene requirements and documentation obligations, for instance, in the context of certification processes. A restaurateur, for example, pointed out that the regulation processes of the European Union have not improved the situation for the restaurants (cf. Interviewee AS16). A second issue identified as being relevant to the problem of existential security is the competitive situation between discount structures and small-scale business. Against the background that big companies are able to offer their products at a lower price, small-scale businesses assume that they cannot keep pace. In this context, one farmer explained that there is strong competition and that this is particularly related to large companies (cf. Interviewee BS9). Furthermore, she noted that "we increasingly have to consider how to stay on the ball" (Interviewee BS9).

Responsibility
In the case of threatened existential security, the European Union and large companies were cited as being responsible.

Solution
Accordingly, interviewees demanded a reduction of bureaucratic red tape to solve the problem of increasing regulations. For instance, a representative of the agricultural sector said: "[ . . . ] yes, definitely streamlining bureaucracy" (Interviewee BS9). Concerning the problem of economic competition, the suggested solution was the creation of awareness at an individual level. This is discussed in more detail in the next section.

Actors
The cited examples of existential security as a problem came from a group of economic actors. Nevertheless, it was also observed that politicians and members of the administration supported this concept. The special connection between politicians, administration and the economy can be illustrated by, for instance, one head of the city administration who talked about "our producers", "our restaurateurs" and "the citizens' initiative" (Interviewee BS13). Use of the word "our" instead of "the" underlines the significance attributed to local economic actors for the city as a whole. In contrast, active groups of civil society are devalued as "civically engaged members of the city" (In the original transcript, BS13 said, "bürgerengagierte Mitglieder der Stadt". The context indicates a clearly negative connotation) (Interviewee BS13) or "professional participation citizens" (In the original transcript, the term "Berufsbeteiligungsbürger" is used. In German, this term invokes a negative connotation) (Observation BS_LG_2).

Lack of Agency at the Administrative Level
Problem A second problem framing identified in the analysis was agency, or rather a lack of agency. The security framing shows that economic actors attribute agency to the European Union or conglomerates whereas the related problem of threatened security is not framed as a problem of agency. In contrast, the city council and the local politicians were viewed as having no agency concerning the food system and its transformation. This lack of agency is regarded as a serious problem.
The only exception is school nutrition. In this sector, the city council seems to have agency and sees itself as being capable of acting. A representative of the city administration pointed out that the administration can influence the food system in the field of school meals because of a department that is responsible for schools (cf. Interviewee BS1). It is noteworthy that the school sector is the only one in the city that is directly linked to the food system. Regarding the city as a whole, the interviewees consider that the administration has only little influence.

Responsibility
There was no entity directly blamed for the lack of agency but rather it was seen as a result of political structures. On the one hand, this could mean that these structures are more or less accepted as given. On the other, this acceptance could also be an excuse since a lack of agency makes it possible to externalise the need to act. This becomes obvious in formulations, such as "one would have to" (Interviewee AS15). Such statements leave out the concrete actor and make clear that the interviewees do not see themselves as being responsible.

Solution
Since structures are perceived as being responsible for the lack of agency and it is assumed that these cannot be addressed, the proposed solution focuses on the individual level of consumers. One head of the city administration explained that the potential influence of the administration is "obviously very low" (Interviewee AS17); yet, he considered that there is an opportunity to support the development of consumer awareness (cf. Interviewee AS17). The idea is that increasing consumer awareness would guide appropriate individual decision-making processes relating to food consumption. This would in turn negate the need for changing political structures and addressing the lack of agency.

Actors
All actors agreed that the lack of agency of the city administration and local policy concerning the food system is a significant problem. This assumption was expressed by representatives of the city council itself, as well as by economic actors and the active civil society.

Problem
The alienation of production and consumption is described as an important problem framing. Alienation concerning the food system was defined as a lack of knowledge about food; a loss of cooking practices; a lack of visible production processes due to mechanisation and the relocation of production; and a lack of interest in general. These circumstances seem to be reasons for why people cannot value locally grown products and the process of production. As a result, people buy their food in discount shops and do not or cannot afford to care about the consequences for local companies or people and their living conditions all over the world. This again causes the problem of threatened existence and sustainability. For instance, one interviewee assumed that most people think that "[ . . . ] if I want bananas, I am going to buy bananas and, in that moment, I don't care where they came from" (Interviewee AS1). The assumption behind this is that people connected to local producers are more likely to value locally grown products and are also willing to pay more money for these (cf. Interviewee AS16). It is noteworthy that it is always the consumer who is considered to be alienated from production whereas production or producers are not described as being alienated from the consumers.

Responsibility
There was no clear attribution of responsibility in this problem framing. No entity was explicitly blamed and actors instead made reference to the transformation of society in terms of job requirements; family structures; and processes of globalisation, such as, for instance, mechanization. However, the attribution of blame was generally vague.

Solution
The only solution proposed to address the problem of alienation was awareness-raising.

Actors
All groups of actors agreed that the alienation of consumers from production is the most significant problem for the transformation of the local food system.

Carbon Emissions as Climate Problem
Problem By participating in the project, representatives of the city council demonstrated their willingness to look at their food system from a sustainability perspective. In this context, the influence of nutrition in relation to the carbon footprint emerged as an important issue in the stakeholder workshops and the interviews. For instance, one interviewee pointed out that "[ . . . ] it is really fundamental, our nutrition is at least a third of our carbon emissions" (Interviewee BS7_2). The analysis shows that for the interviewees, carbon emissions-especially relating to production, transport and-consumption seem to be the most significant climate problem in relation to food.

Responsibility
Again, in this problem framing, the attribution of blame remains vague. Thus, the solution framing focuses on the individual level as this is considered to be addressable.

Solution
The interviewees consider that it is necessary to reduce carbon dioxide emissions. The analysis found that it is considered useful to start at the individual level where the raising of awareness is again seen as the only solution. One member of the city administration explained that "[ . . . ] the reduction of emissions of carbon dioxide has primarily been a topic of traffic avoidance [ . . . ] now that nutrition constitutes thirty per cent of the carbon footprint is in general, from my point of view, hardly fixed in people's minds [ . . . ]" (Interviewee BS1).

Actors
The problem framing in relation to climate is primarily articulated by active civil society. This group of actors engages in initiatives defending or protecting the environment, sustainability and climate. Furthermore, there are people from the administration, as well as politicians who highlighted that the linkage between climate and nutrition is something new but nevertheless important for them. The economic actors framed climate aspects as a problem if their companies were dealing with these and if they were using environmental aspects as a selling point.
Summarising the presented framings, four main problems are evident: (i) Existential security; (ii) the agency of the city administration; (iii) alienation of consumers from production; and (iv) climate impacts. In all cases, awareness raising was mentioned as a significantly valuable solution. Hence, the following approaches the question of how the framing of solution proceeds and how awareness is defined.

Raising Awareness as Main Solution
The problem framing shows that awareness raising is considered to be the general solution in relation to a more sustainable food system. Education, closeness and visibility were described by interviewees as key conditions for increased awareness. Interviewees argued that if people receive the message that sustainable produce and locally grown products are better and if this information is linked to visual impressions of production, processing, supply and disposal, they are likely to attribute greater value to the local food system as a whole. Consequently, they would be willing to pay more for their food. As mentioned above, this would address the problems of existential threat and increasing carbon dioxide emissions. From the actors' perspective, the logical conclusion is to promote education programmes, especially for children.
However, the proposed solution generates new problem framings. If the interviewees were asked to think of opportunities for raising awareness, they appeared uncertain. Many provided similar responses, such as, "I do not know myself how to approach it but one ought to try it somehow" (Interviewee AS4). Another interviewee asked: "[A]nd how do I get this into the minds of the population?" (Interviewee AS9). Interviewees who work in the field of education focusing on food were also unsure: "[W]e do not get them no matter with which entry [ . . . ]. That is a very serious problem" (Interviewee AS7).
The second problem is that awareness remains more or less an empty word, meaning everything and nothing. The analysis shows that awareness is an extensively covered but nevertheless unspecific topic. The sequence analysis provides a more precise picture, which shows that from the interviewees' point of view, awareness means that people act sustainably in relation to nature, climate and regional value creation. The importance of awareness as a solution is justified by the fact that the political and administrative actors distance themselves from the possibility of regulating the dietary or consumer behaviour of citizens. For instance, one politician argued that: "[ . . . ] one cannot and one is not allowed to dictate to anyone how to behave" (Interviewee BS22).
The analysis of the problem framing shows a very broad and unspecific definition of problems and the same mechanism was observed concerning the framing of solutions. Awareness is very important but hard to achieve. The following section outlines the framing of motivations concerning the local food system.

Personal Motivation Is Not Enough for a Movement
The findings regarding problem and solution framing show that the definitions of these are unspecific. This makes it more complicated to identify motivations for action.
Among the different groups of relevant actors, the active civil society had the clearest motivational framing. This group describes knowledge as the starting point for action. More specifically, they refer to lectures (cf. Interviewee BS7_1) or films (cf. Interviewee BS7; AS14) about agriculture, consequences of globalisation processes or alternative lifestyles as the starting point of their active engagement. It is noteworthy that these films are also mentioned by politicians, as well as the administration but that they appear to motivate these actors to moderate their actions only in their private life (cf. Interviewee AS9). These framings motivate the interviewees to deal with problems of climate protection or lack of sustainability in relation to the food system.
Moreover, the individual motivation for increasing awareness of nutrition is frequently traced to personal experiences, such as pregnancy (cf. Interviewee BS11) or diseases (cf. Interviewee BS8). This kind of motivation does not focus on sustainability but on personal health and it is usually not perceived as a feasible motivation for a group of people.
Among economic actors, financial benefit is significant motivational framing. Companies, restaurants or farms are willing to act sustainably, but only if this is expected to garner economic benefits (cf. Interviewee AS4; BS9). In their professional lives, politicians and representatives of the administration connect to this economic argumentation. Focusing especially on school meals, this argument becomes important. One member of the city council pointed out that "[ . . . ] at the moment, one has decided for another one [supplier], because of the price, but that one discusses it already once, is good" (Interviewee AS9). Further, he expressed his hope that in the next public tender process "maybe the regional supplier is even so attractively priced" (Interviewee AS9) that the decision in his or her favour could be justified.
The third often mentioned aspect motivating all groups of actors is the expectation that the city council act as a role model with regard to sustainability. A representative of the city administration highlighted its exemplary function (cf. Interviewee AS2). The administration has to force the topic and to show how policymaking regarding the food system should be carried out. The interviewees argued that by functioning as a role model, the city council could motivate its citizens to move in the same direction either by influencing changes in their behaviour or by founding new initiatives.
In summary, four motivations for acting were apparent: Knowledge concerning climate change, personal experiences related to health, economic benefits and authorities as role models. The analysis shows that knowledge, personal experience and health are the most important individual motivations necessary for one person to act; however, these do not appear to be sufficient for the emergence of movements towards a sustainable local food system. Even though the authority as a role model focusing on the food system does not exist in our two cases, this is seen as a possible solution and accordingly as a sufficient condition for prompting action.

Discussion
This paper started with the consideration that it is important to focus on small cities in Germany and their efforts to move towards sustainability. The first underlying assumption was that cities as designers of sustainability are of great importance since they represent the living conditions of many people in Germany. Second, it was assumed that sustainable food is becoming an increasingly important policy issue in Germany. The KERNiG project connects these assumptions by analysing efforts to transform the local food systems in smaller cities towards sustainability. This paper focuses on the framing of the local food systems because framing plays a significant role in policymaking. According to Benford et al., framing can be divided into the framing of problems, solutions and motivations. In line with this, four main problem framings were identified in this study. The first is the framing of security, which is generally an important framing in food policy literature [18]. In the two analysed cases, this problem framing does not focus on the security of food and nutrition but rather on the existential security of small-scale businesses. The connection between ecological agriculture and the food sector has been elaborated in various studies on regional development [39,40]. In that sense, regional value creation emerged as an important topic. The second problem framing focuses on how local actors could influence processes within the food system. School nutrition seems to be the only sector in which the local administration attributes agency to itself. The third problem framing refers to the alienation of consumers from food production. This is a very basic problem framing since it is related to and even causes other problem framings. Without framing alienation as a problem, external security would not have been described in the way it was. In 1999 Tappeser et al. found out that especially children in industrial nations were not familiar with how food production in the agricultural sector looks [41] (p. 8). These findings were confirmed by Astleithner et al. who argued that the increasing alienation of consumers from production leads to the longing for regionality [42] (p. 210). Moreover, the lack of awareness or rejection of climate effects linked to nutrition were related to the problem of alienation. Hence, the last problem framing deals with the climate aspect of food. Here, the significant influence of food systems on climate seems to be increased carbon dioxide emissions. Bönisch et al. for instance pointed out, that the current food system has a significant influence on the emission of greenhouse gases [43] (p. 32). In the two cities there were very engaged economic actors sometimes they were also members of the city council. As a result, the administration and local politicians assume the focus on regional value creation. The problem of alienation is part of it as well. Hence, the problem of existential security seems to be the most important one. In contrast, referring to the literature on food system transformation towards sustainability climate issues are especially important. Hence, it is used by all stakeholders. The problem of agency is a problem of structure. To sum up, it can be said that the problems mentioned address different levels. Since the problem referring to the existential security of small-scale businesses addresses a direct concern and everyday experience, it is a crucial problem for them. Referring to the literature on food policy, the problem of climate issues is regarded as much more important.
There is only one solution proposed for the four different problems. The raising of awareness at the individual level is expected to lead to changed behaviour, ideally including sustainable, healthy, regional and seasonal consumption. The solution framing does not start at the structural level but at the individual one. Awareness is operationalised as the link between knowledge, visibility and closeness. However, the problem remains as to how greater awareness is to be achieved. In this study, the interviewees did not identify concrete means of increasing awareness.
The analysis shows that sustainability is not a frame in itself. Nevertheless, it is part of each framing. For instance, the support of regional small-scale business has been inter alia justified by the argument that regional value creation is associated with shorter transport routes and thereby fewer carbon dioxide emissions. The quintessence of the presented problem and solution framing is the idea that if consumers buy locally grown products they will enhance the local infrastructure, learn more about the processes from production to waste disposal, and eat fresh, healthy and ultimately sustainable food products. This understanding of sustainability coincides with the finding of Hinrichs [26] and Morgan [19] who observed that localisation of the food system is mostly seen as a synonym for sustainability. Because of this, localisation is always framed as a positive and desirable process which becomes obvious by the expression of 'Heimatgefühle'. In this context, it is assumed that sustainability does not need to be framed explicitly as a problem within the local food system because it will automatically be addressed through regionality or localisation which are more concrete and more visible.
The understanding of sustainability in Germany concentrates on the economic aspect of the concepts and is thus defined by Ott et al. as a rather weak form of sustainability [9] (p. 39). In this study, the primarily economic understanding of sustainability was confirmed with regard to the group of economic actors as well for the group of politicians and the administration. In contrast, the economic-based understanding of sustainability does not appear to be particularly important for the active civil society group. This group used an ecological understanding of sustainability. Thus, the analysis shows that there is an awareness of sustainability as a problem but that this awareness is not fundamentally the same for all groups of actors.
These different groups of actors also become important in relation to the motivational framing. Currently, this motivation does not exist in the two analysed cases. The analysis shows that the active civil society was motivated by media reporting, literature and lectures about climate change and consequences of globalisation. This personal experience also motivates actors from other groups but it only leads to changes in their own behaviour and does not provide the basis for wider structural change. This contradicts the idea that important actors, such as the leaders of the city council functioning as role models for a sustainable food system, could motivate people to follow the same course.
The analysis shows that the inclusion of aspects that remain unsaid is also important for understanding sustainable food policy in smaller cities. The analysis of the problem framing showed that there is no entity explicitly attributed with responsibility for the lack of agency, the alienation or the climate issues. The creation of awareness framed as a potential solution also remains broad and unspecific. These two aspects lead to the assumption that the articulated framing cannot exclusively provide a comprehensive analysis of sustainable food policymaking in smaller cities. In this respect, it is useful to shed light on the unsaid, as well as the articulated. For instance, one very useful approach for investigating underlying but silent discourses is situational analysis [34]. This paper is based on the framing approach and not on situational analysis. Nevertheless, aspects of situational analysis help "to also address and analyse salient discourses within the situation of inquiry" [34] (p. 14). Following this approach, it was observed that the elephant in the room was globalisation.
Regional production and regional value creation were either implicitly or explicitly part of all the different framings. At first glance, this is hardly remarkable given the project's focus on the local level and the transformation of local food systems. However, the local and the regional only exist in contrast to the global. In the literature, the global and local as a contrasting pair are often emphasised. Hinrichs, for instance, describes food system localisation as "an apparent counterpoint to globalization" [26] (p. 33) while Baker talks about a "local-global binary" [44] (p. 10). In contrast, a quantitative list of terms used by all 55 interviewees (interviews were conducted with a total of 55 people. Twenty-six interviews were included in the qualitative evaluation) in this study shows that words like "regional", "regionality", "local" and "home" are mentioned more than 200 times in the analysed data whereas words like "global", "globalisation" and "international trade" are mentioned approximately 20 times. It is conspicuous that the global is hardly mentioned although it is the counter term of the local. This raises the question of what the exclusion of the global could mean with regard to the food system. Proven explanations, such as self-evident aspects or taboos are not relevant here. Globalisation is not taboo but rather an accepted concept especially when focusing on the economy. Hence, it is in a way self-evident that the food system is globally embedded. It would be expected that the interviewees would stress the significance of the local by referring to the negative effects they associate with globalisation. However, when they did so they did not explicitly refer to globalisation, for instance, when stating that price developments are linked to world trade (cf. Interviewee AS3). On this basis, an assumption is made that the missing naming of globalisation as a counterpoint of the local is of great importance.
The framing of globalisation as a problem or even the explicit mention of globalisation as one aspect of framing [28] (p. 92) would have two effects. First, referring to globalisation in terms of dissociation from the global is as politically undesirable as the overemphasis of the local which tends towards protectionism. Thus, the declaration of the G20 Meeting of Agriculture Ministers in 2018 underlines that they "are concerned about the increasing use of protectionist non-tariff measures" [45] (p. 7). Smith et al. confirm that there have been concerns "expressed about the pursuit of the 'local' as a hidden form of protectionism in the name of sustainability" [12] (p. 2). Therefore, a framing of globalisation as a problem linked to an emphasis of the local could be associated with a protectionist policy.
The second effect of naming globalisation explicitly could be that framing globalisation as a problematic counterpoint of the local means that globalisation processes have to be addressed. Klein points out that globalisation sounds like a mystical power that paralyses politicians and prevents them from acting [46] (p. 8). This leads directly to the lack of agency that was identified as a problem framing in this study. In this context, the study found that politicians and the administration consider that they have little influence on the food system. The fact that globalisation was not explicitly mentioned here shows that globalisation could not be addressed as an entity given that all levels and actors influence globalisation processes in one way or another [46]. The explicit framing of globalisation as a problem would require an assignment of responsibility and consequently the development of a solution. However, since this is not the case, the transformation of the local food system towards sustainability can be negotiated exclusively at the local level. Hence, there is no perceived need to frame solutions at other levels. In other words: The framing of globalisation as a problem of local food systems or the blaming of globalisation as the responsible entity would entail a criticism of the existing system. The missing framings as an important part of policymaking could only be touched upon in this paper. Nevertheless, it seems to be an important aspect that requires further investigation in future research dealing with the framing of rather new policy issues.
Returning to the main research question of how local actors in smaller cities frame the local food system and what this framing means for transforming the local food system towards sustainability, the analysis shows that in both cases the existential security problem of small-scale business is more tangible in contrast to agency, alienation and climate. There are three reasons identified for the fact that the existential security problem of small-scale business seems to be more tangible. First, this problem is visible. For instance, people can walk the streets recognising vacant stores where a bakery has been before. Moreover, terms like "dying of small farms" are well known particularly in affected regions. The third reason is, that the topic has strong advocates either in the form of lobby associations or in the form of the respective affected persons. According to Benford et al. this problem has a supposed clear cause-and-effect relationship, thus different entities could be directly blamed for it. On the one hand, the EU is blamed for their political decisions with regard to hygiene and control, on the other hand consumers are blamed for their wrong purchasing behaviour. People are aware that this problem cannot be solved at the local level. For the politicians this is beneficial since they can support the local economy confirming their efforts to forward this problem to higher political levels. This type of single-issue problems in food policy are particularly interesting for policymakers since they "find it hard to address the inter-relatedness of the whole food chain and the whole food cycle" [18] (p. 318). In contrast, agency, alienation and climate issues are not directly visible. The problem of agency refers to the structure of polity and administration. Thus, it is beyond the realm of everyday life. Alienation is rather causing the problem of existential security than being a separate problem itself since it leads to supposed wrong decisions of consumers. It is not tangible since not all people are equally alienated, but this alienation is attributed only to certain groups of people who are not aware of their supposed alienation. In contrast to the vacant stores you cannot show alienation. Climate issues in contrast are very complex. There are no clear cause-and-effect relationships and cause and effect are often disconnected from time and space. Thus, climate impacts are rarely visible. Moreover, this problem can be described as "a complex moral problem that our current political system is not well suited to address" [47] (p. 475). Besides, the climate often lacks influential advocates, as dealing with climate issues sometimes thwarts economic interests. Just like the problem of existential security can be visible, the success can be visible as well. And just like the problem of climate remains invisible, the success would not be directly apparent. Thus, the existential security problem is more tangible for people, for politicians it is easier to communicate and in sum, it is less complex especially in contrast to climate issues related to the food system. Local authorities use the focus on awareness as a proposed solution at the individual level, thereby externalising the responsibility to change policymaking.

Conclusions
Dealing with the local food system is a relatively new topic for actors in smaller cities in Germany. It also seems to be a policy issue that is credited with little importance. There are other issues that draw more attention. The administration and the politicians try to link the more prominent topics with sustainability in the food system. Efforts towards regional value creation are one example of this. Moreover, except amongst the group of civil society actors, the sustainable food system is not a topic of heated debates. Rather, the other groups of actors, such as administration, politicians and the economy aim to find consensus and political harmony.
On the one hand, transforming the local food system towards sustainability requires a clear framing. Currently, different groups of actors use the same framing but mean different things. This facilitates the building of consensus but at the same time complicates the initiation of a transformation process. In the two analysed cases, there is a lack of clear framing and the framing in general remains vague.
On the other hand, creating a more sustainable local food system also requires powerful actors. The analysis shows that politicians and the administration adopt and support the economic framing. In doing so, they run the danger of falling into the local trap of using local food system change as a unique selling point for the city marketing.
Powerful actors using a strong sustainability framing would be able to push forward food system change. This could be civil society actors, as well as economic actors. In the analysed cases, the economic actors are much stronger than civil society. The important role of municipalities concerning sustainability is stressed by politicians at all policy levels. Nevertheless, for the analysed cases it can be stated that the city administrations do not feel responsible and do not perceive their agency. Hence, the transformation of the local food system in cities, such as A-Town and B-Town would ultimately be more feasible if there was an obligation from a higher policy level.
This paper provides three main policy implications for local food system transformation: All involved actors, in particular local politicians and administrative staff should be aware that the fact that everybody uses terms like sustainability and awareness rising does not necessarily mean that people are framing it in the same way. Hence, it is necessary to clarify the contextual significance. Moreover, it would facilitate food system transformation at the local level if higher political levels commission the cities to shape local food system transformation towards sustainability, since the topic is new for cities and they are currently struggling how to deal with it. That is why they adapt their argumentation to the local economy. This leads to the third implication. Food system transformation at the local level needs a mutual understanding and respect. It is a comprehensive topic that needs all groups of actors. This aspect is especially relevant for small and medium-sized cities where people know each other from different contexts.
Smaller cities are particularly interesting for local food system transformation since they have the possibility of bringing together all relevant actors involved in the food system. Additionally, and perhaps most important, smaller cities have long-standing traditions of local societies taking the initiatives, since people identify with their respective city.
This analysis shows that insights into the framing of the food system in smaller German cities provide an important contribution to the broader food policy literature given that the qualitative approach reveals new results. The qualitative approach provides the opportunity to follow the argumentative process. In contrast to official documents, this analysis shows that local actors struggle to answer the question of how awareness raising could be achieved and are sceptical with regard to the agency of local actors. However, this paper shows that the framing approach could not account for significant aspects of policymaking, such as underlying discourses. One academic conclusion provided by this paper is that food policy research requires a comprehensive analysis using different methods, analysing diverse data surveyed at different policy levels, and including different groups of actors. A further suggestion would be to extend food policy research to small and medium-sized cities since what applies to bigger cities is not necessarily transferable to smaller cities because of different socio-spatial conditions.
Referring to the role of smaller cities within the process of increasing sustainability in local food systems, it would be interesting to analyse the framing of sustainable local food systems in different regions of Germany to find out if these findings can be generalised or if A-Town and B-Town represent specific and unique cases that differ from smaller cities in other parts of the country. In any case, the debate on sustainable nutrition at the local level supports the suggestion that sustainability may become an elementary component of the fourth phase of European food policy in the sense of Barlösius. Acknowledgments: My special thanks go to Daniela Kleinschmit and Sylvia Kruse for the discussions and critical comments that led to this paper. Moreover, I am grateful to all interviewees for their time and openness and to the stakeholders who allowed me to observe the workshops. Last but not least, I thank the two anonymous reviewers, whose comments helped me to improve the paper.

Conflicts of Interest:
The author declares no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. Table A1. Tabular overview of coding.

Codes Operationalisation Examples
Deductive Coding-Framing approach Problem The framing of problems is a fundamental part of the framing concept by Benford et al. [30]. Hence, both problems and the absence of problems were coded. A problem is something that is either explicitly labelled as a problem or something that is described in a negative way and therefore is obviously framed as a problem.
"[ . . . ] food has up to now never been a problem for us in our history, rather that we have none." (Interviewee AS5) "I believe, that often these small marketers are struggling with positioning themselves" (Interviewee BS14) Deductive Coding-Framing approach Responsibility Bagdonis et al. [31] pointed out that a problem definition includes the attribution of responsibility within a framing process. The coding includes persons or entities seen as responsible for the emergence of a problem or its resolution. "[ . . . ] but regarding high politics, now you're talking about Berlin-more should be done from above." (Interviewee BS22) Deductive Coding-Framing approach Solution According to Benford et al. [30], one deductive code is a "solution". This code includes all aspects mentioned as being possible or existing solutions to a problem.

Codes Operationalisation Examples
Deductive Coding-Food policy research Regionality According to the concept of "localization" [26], the code regionality was developed to analyse the importance and the meaning of the region within the food system. The coding of agency [48] includes the attribution of one's own agency and the agency of others.

Inductive Coding Alienation
The code of alienation was developed inductively based on the data. It is associated with the problem code. This code includes a description of how people are connected to or disconnected from their surroundings.
"[ . . . ] but as already mentioned, the alienation between production and those who are buying it then, is just big [ . . . ]" (Interviewee AS4) "[ . . . ] we still know that milk comes out of the udder and not out of the tetra pack" (Interviewee AS18)

Inductive Coding Climate
The code of climate was developed because of the interviewees' focus on climate aspects. This code includes explications in the context of climate. "[ . . . ] regarding the co2-print that this simply reduces and thus also makes a significant contribution to climate protection [ . . . ]" (Interviewee BS1)

Inductive Coding Knowledge
The code of knowledge was developed because knowledge is often mentioned as a solution while ignorance is defined as a problem. The code includes knowledge and ignorance equally.
"[ . . . ] and so the people get to know a little more what grows. Because, many do not even know that anymore." (Interviewee AS14) Table A1. Cont.

Inductive Coding Awareness
Awareness is an inductively developed in vivo code because the label adopts the wording of the interviewees. The code contains all parts of the interviews in which "awareness" is explicitly mentioned, as well as descriptions of awareness without mentioning the word. Awareness is associated with the coding of solutions while a lack of awareness is also coded as a problem.