Food (di)lemmas: disentangling the Italian Local Food Policy narratives

. Recently, food-related policy initiatives have proliferated, such as food strategies, food plans, food councils, food districts and food communities, just to name the more relevant ones. Far from being systematically defined and logically systematized, these concepts often overlap or are used as synonyms. The paper has systematically traced the current trends showing how these concepts are used in the current debate, the theoretical background on which they are grounded and the public policies they call for, following a threefold approach: (1) a literature analysis to establish the state of academic research on food systems in its multidimensionality; (2) a review of the existing national legislation to detect the utilization of food policy-related lemmas in the normative; (3) a computational linguistic analysis applied on institutional documentation to explore how cities and territories are using concepts and definitions in the grey literature. The results show that the construction of narratives around the topic of food systems planning is experiencing a momentum, with particular emphasis on principles, background premises and governance aspects. In this context, the risk of marginality for the agricultural sector in such discourses and narratives is highlighted.


HIGHLIGHTS
• The construction of narratives around food policies is experiencing a momentum in Italy, especially with regard to governance structures and principles.• Local food policy initiatives are identified more for their best-practices than their results or effectiveness.• Agricultural sector is the weakest element in local food policy narratives, which mostly concentrate on post-production stages.• Local governments have acknowledged the political reach of food systems, fuelling a narrative that relies mostly on stakeholder participation and representativeness.

INTRODUCTION
The lemma "food" recalls a plurality of semantic values, due to the cultural and symbolic significance it evokes.Food takes on environmental, cultural, social, legal, economic, historical, geographical and symbolic meanings and values originating from territories (Monteduro, 2015;Gazzola, 2017;Greco, Nocco, 2022).In the legal system the most common qualification of "food" is in terms of nutrition as a consumable good; in other disciplines it is considered in terms of a common good aimed at the realization of basic needs for a plurality of individuals (Mattei, 2011;Lucarelli, 2013;Rodotà, 2013).Food attracts and involves different rooted knowledge and disciplines, ranging from agricultural production to anthropological aspects, from the history of gastronomic traditions to the regulation of the characteristics of its quality, from environment to consumption models and agri-food markets, encompassing the nutritional aspects and those related to personal well-being, including access to food, social and cultural values, conviviality, health, and waste reduction.What is certainly innovative is the proliferation of food policy (FP) initiatives in Italy.We refer, for example, to food strategies, food plans, food councils, food districts and food communities, just to name the more relevant ones in the current debate.Far from being systematically and rigorously defined, these concepts are often confused or even used as synonyms.Some formulations convey differently structured and formalized experiences depending on the contexts in which they are adopted or specific regulations and financing possibilities, as in the case of food districts and food communities in Italy, where food-related initiatives are spreading across the country, thanks to a relentless activity both from grassroots movements and local institutions (Dansero et al., 2020).Against this backdrop, defining the concepts, the perimeters of action, the normative background, and the role of actors can contribute to providing a common framework for the development of future initiatives and, subsequently, disambiguate some expressions.In this regard, we have dug into the utilization of the lemma "food"1 within three main fields: the scientific literature, institutional documents, and legislation, limiting our research to the expressions that revolve around the FP realm.
We followed a threefold approach: (1) a literature analysis to establish the state of academic research on food systems in its multidimensionality; (2) a review of the existing national legislation to detect the utilization of food policy-related lemmas; (3) a computational linguistic analysis applied on institutional documentation to explore how cities and territories are using concepts and definitions in the grey literature.The results show that the construction of narratives around the topic of food systems planning is under the spotlight, with particular emphasis on principles, background premises and governance aspects.Nevertheless, the risk of marginality for the agricultural sector in such discourses and narratives is highlighted.

Food policy narratives: between homologation and innovation
A key element influencing the relevance of any pathway to improve food systems (FSs) and make FPs effective is the narrative built upon them.In achieving sustainable FSs, it is increasingly relevant the way agricultural, and any other component of the supply chain, is represented in the productive, social, and environmental discourse.The impact of narratives on food systems is so powerful that Land and Heasman coined the expression "food wars" to refer to the emergence of conflicts over the global politics of food and health.These contrasts refer to emerging agendas and competing positions that seek to attract investments, public support and political legitimacy (Land, Heasman, 2015).Anderson and Rivera-Ferre talk in their work (2021) about the narrative of extractive vs. regenerative agriculture.The two systems have different practices and consequential spillovers throughout the FS, not only at the production stage, but also in markets, distribution systems, and even diets, affecting labour markets, agri-food research and policies.They focus on how different narratives can deeply affect choices and implications, and how such narratives help framing a specific issue in a specific problematization.According to the authors, some narratives have been more effective than others, such as, for example, the neo-liberal or laissez faire ones compared to the eco-friendly or "social well-being" ones.Mattioni et al. (2022) refer to the conceptual framework of sustainability transitions to mean "the long-term, multidimensional and fundamental transformation processes through which established socio-technical systems shift to more sustainable modes of production and consumption" (p.48).The focus is on the dominant narrative of the agrifood system, which has traditionally been dominated by large corporations as the industrial mode of production, although more recently other forces have been arising "from within" the dominant view and trying to change the narrative, bringing inside forces that can destabilize the dominant regime.These forces are mainly local governments and institutions, which are becoming the attractive poles of changes in the relationship among agricultural and rural areas, food production and consumption, supra-national and local policies governing the change processes.
What is happening today with new approaches of the transition "from within" in the production sector and local policies is very similar to what had happened before with the new rural development and agroecology.In the sphere of rural development, the changing narrative has been from an agricultural-based economy to the commoditization of the rural territory and landscape.The potential consumers of these new "goods" are both the local communities and outsiders, because of the power of the new framework built around rural areas and the narratives presented, which include the local actors and promote local development based on endogenous forces and human capital (Fløysand, Jakobsen, 2007).In Europe (Shacksmith et al., 2011), as well as in developing countries (Ashley, Maxwell, 2001), narratives became so powerful and effective in re-conceptualizing rural development that telling "successful stories" about experiences and case studies has been considered heuristically enough to prove the good quality of results and their reproducibility in different contexts.
Something very similar happened a bit later with new agricultural approaches such as agroecology (Barrios et al., 2020), in which it is possible to identify four main entry points upon which to build a structured process using visual narratives that help to identify socioecological transition trajectories: biodiversity, consumers, education and governance are identified.
A similar process is going on in the case of FPs.The diffusion of (di)lemmas turning around food policies as organized intervention systems is mostly due to a very well-constructed discourse and supportive narratives that establish a sort of inevitable abandonment of old paradigms and embracement of new theoretical frameworks; on the other hand, they encourage and support the diffusion of a new shared language, new practices and the acknowledgment of new actors that, in turn, gain momentum and create a new scale of shared values and priority in the economic, social and environmental spheres.

Local food policies urban-rural linkages
To understand the role food plays in our societies, it is necessary to consider the production and supply side, while enlarging the fields of action and investigation to the demand and places of consumption.The British urban planner Caroline Steel (2015) states that to understand cities correctly, we must observe and analyse them through the lens of food.The hypothesis underlying the present study is that food is one of the privileged areas for analysing, understanding and guiding the development paths of the territories in their complexity.Nevertheless, the lack of common definitions and clear logical frameworks and, subsequently, of coordination of mutual understanding between the actors involved, has led to ineffective outcomes with respect to the objectives (Fattibene et al., 2023).This is exacerbated by a growing disconnection between food production and consumption that affects many spheres: economic, geographical, symbolic, cultural and political (Bricas et al., 2017).In this context, stronger and continuous connections between rural areas and cities, as well as a systemic planning of (peri)urban agriculture, can mitigate or reverse these trends through various economic and policy levers (Vaarst et al., 2017).Several studies have shown that cities and local authorities are a key component of the sustainability agenda (Steel, 2008;Fattibene et al., 2019;Moragues-Faus, 2021).Urban markets can strengthen the connections between city and countryside, giving rise to innovations, both in terms of commercial opportunities and from an ecological and territorial protection standpoint; eco-systemic and social services offered by farms and especially by peri-urban agriculture can also contribute to enhance the interlinkages between rural and urban areas, at the same time reducing the dichotomy on which most of the past development models and paradigms were built.Rural areas are also increasingly involved in local development paths through the lens of food: this is an approach already used in consolidated rural policies, such as the LEADER, which has enhanced short food supply chains and re-localization of FSs as an opportunity to promote the territory (Kneafsey et al., 2013).These themes are also reflected in the long-term vision for rural areas, which envisages, among the four areas of intervention identified in the Action Plan, the diversification of economic activities and the improvement of the added value of food-related activities (agrotourism, agri-food, agriculture).The Common Agricultural Policy, for the 2023-2027 programming, has adopted a broader spectrum of objectives, explicitly mentioning nutrition, diets and food (Specific Objective n. 92 ) (EU, 2021), consistently and in compliance with the Farm to Fork strategy.
Experiences aimed at defining FS planning strategies are, therefore, widespread in both urban and rural areas, although the necessary integration of different territorial systems is increasingly recognized.The manifesto of the Italian Network of Local Food Policies states that the "local" nature of FPs encompasses not only cities but the relations and the urban-rural continuum in a territorialist vision of urban policies."Local" refers to both a geographical context and a space of action (Dansero et al., 2020) and is intended as a context with blurred boundaries, as an intermediate level, to be constructed and legitimated by a variety of actors (individuals, groups, local communities) and multi-scale policies (from the municipal, to the national, European and global levels).Such complexity reflects the liveliness of the Italian context (Fattibene et al., 2023) when it comes to analysing the number and maturity of initiatives.However, many experiences carried out at the local level demonstrate the abundance of ideas, resources, projects, and actors involved, often capable of generating considerable ferment and excellent opportunities to activate synergies, but they also risk being ineffective if a common framework is not defined.

MATERIALS AND METHODS
To capture different dimensions of the food policy narratives in Italy, a content analysis of the literature and documents set using a combination of deductive and inductive coding techniques (e.g., Elo, Kyngas, 2008) has been conducted.The deductive analysis was guided by a matrix comprising lemmas -both in the English and Italian languages -associated with the topic of "food" mainly used in public debate, policy and legislation that could be shared by the various actors involved -scientists, practitioners, and policy makers (Table 1).The hypothesis underlying this methodology is that conducting an analysis on three types of sources which use different languages, would help to capture the complexity of the narratives relating to food policies in Italy.
Therefore, although regularly used, other lemmas have been excluded from the research, such as "food security", "food system", "food environments" or again "food strategy", because they could have a broader and more general meaning or could be assimilated with others already included in the list, such as "food plan" and "food policy".
Thus, with the aim of identifying similarities and differences in their use, as well as the frequency of the expressions, a desk analysis was conducted on papers and articles from the scientific literature, in the existing Italian legislation on the subject, and on documents issued by local governments and development agencies.
Given the nature of the analysed source materials, it has been necessary to use different methods of content analysis to capture the meaning of the selected locutions within each type of documents, as shown in Figure 1.In the case of the scientific literature and legislative set, the analysis was guided by a second matrix (Table 3) including lemmas extracted by the Milan Urban Food Policy Pact (2015).Alongside deductive coding, inductive coding was used to capture additional themes during the analysis process, such as Urban Regeneration or New Technologies and Innovation.In the case of the institutional documents, a computational linguistics software (IRaMuTeQ) was used to investigate the correlation between the terminologies used, visualizing groups of homogeneity and thematic relationship.

Literature analysis
An initial literature review was conducted through the SCOPUS database, using three selection criteria: i) territorial dimension (Italy); ii) temporal dimension (after 2015); iii) thematic dimension (lemmas)3 .Each selected lemma listed in Table 2 was searched in English and in Italian.It was also chosen to set a time range that could provide the study with material of recent publication and dissemination, setting 2015 as the starting date (ii).Expo2015 was considered a milestone for the diffusion of food-related issues in Italy: the Milan Urban Food Policy Pact remains one of the main legacies of the exposition.
The research results were then validated, eliminating contributions not considered relevant for the purpose of the analysis.In addition, articles included in more than one list were assigned to the most appropriate one, according to a careful evaluation by the research team.
The same search was also applied to the Web of Science (WoS) platform, which produced fewer results than Scopus and no additional material was added to the previous list.

Normative analysis
At the national level, the only forms of territorial governance related to food are Food Districts and Food and Biodiversity Communities of agricultural and food interest, which have been regulated by national and regional laws over the years as rural development policies have evolved.
The reconstruction of the national regulatory framework was based on "Normattiva" portal and regional regulatory databases and regional institutional websites.For "Normattiva", the locutions in Table 2 were used to identify those regulated by specific laws, i.e., "distretto del cibo" (food district) (4 hits) and "comunità del cibo" (food community) (1 hit).
As far as Food Districts are concerned, there is an official register managed by the Ministry of Agriculture (MASAF) and implemented by Regions and Autonomous Provinces that contains 185 districts.In addition to the national law (l.205/2017), only the implement-ing regulations issued at the regional level (n=15) were analysed, leaving out laws issued at the regional level before the national law (n=16).In addition, "Community Charters" formalising Food Communities (n=13) and national implementing legislation (l194/2015) were also analysed.

Institutional documentation analysis
For the institutional literature, the analysis has been limited to the lemmas FP, "food council" and "food plan", as from a preliminary screen, they are the main domains in which local administrations have produced documentation, as shown by the most recent territorial analysis studies of Italian FPs (Dansero et al., 2020;Dansero, 2022) and the experiences gathered by the Italian Network on Local Food Policy.The analysis was run using IRaMuTeQ (R INTERFACE for multidimensional analysis of texts and questionnaires) computational linguistics software.It is based on R software and python language and provides users with statistical analysis on text corpus and tables composed by individuals/words.The Similarity Analysis has been utilized, a graphical output that uses graph theory concepts to represent the co-occurrences of words in a corpus.The corpus has been constructed by extracting from each consulted documentation the parts where a definition or indications on relationships among policy instruments have been provided.Subsequently, the segments have been reorganized according to the relevant variables, i.e. the city/territory and the three terminologies considered.Given that the language of the consulted documents is Italian, the corpus has been constructed accordingly, in order to not distort the results, as the software is language responsive.Nevertheless, the translation of the displayed terminologies has been provided in Appendix 1.

Literature analysis
The research, conducted in the period September-October 2022, resulted in 25 eligible papers out of the 180 originally selected.Table 2 shows the search results categorized by lemmas.The overall number of articles selected is more numerous than those analysed, because some of them were found in more than one search by locution, therefore assigned to only one according to the title/keywords chosen by the Authors.
The first stage of the literature analysis consisted of the construction of a matrix composed of thematic areas considered relevant for their relationships with the identified lemmas by the authors of the 25 articles considered.Some of these themes were subsequently merged for similarity, resulting in a total of 18 thematic dimensions.The 18 dimensions were then aggregated into 3 thematic clusters: Food Safety and Security, Territorial and Local Dimension, Food Supply Chain (FSC) Management (Table 3).
FP and "food networks" cover all clusters in a significant way, whereas "food communities" and "food movements" are less present in the three clusters.Food safety and security is the most covered thematic cluster in the 25 papers (70 times); followed by FSC Management and Territorial and Local Dimension (45 and 40 times).Consequently, issues dealing with the Food Safety and Security cluster are prominently considered within four out of six lemmas, while the cluster turns out to be poorly relevant for "food communities" and "food movements".
The relevance of the relationship becomes less intense for the other two clusters, turning out to be strong only for "food policy" and to a lesser extent for "food networks"; as for the other 4 lemmas it is rather weak.Territorial and Local Dimension is scarcely found in "food councils" and "food movements", and it is rel-atively low in "food communities" and "food plans".Similarly, FSC Management issues have -quite surprisingly -little presence in "food plan" and "food councils", while they are relevant to "food policy" and "food networks".The definition of Urban Food Policy (UFP) is only addressed by one article related to "food policy", while the concept of "prosumerism" emerges exclusively for "food networks" related papers.
Food policy and food networks show a rather even distribution within the three clusters, whereas a marked polarisation on some clusters emerges for the other lemmas.Food plans and food councils in the first cluster; food communities in the second cluster; food movements in the third.However, these results could be due to the limited number of papers selected by the literature review.For this reason, a more in-depth analysis has been conducted considering only the two lemmas food policy and food networks.In the food policies-related literature, various papers present these initiatives as best practices or case studies developed in specific local contexts.The common trend is to explore their characteristics or compare the various experiences, rather than analysing in depth the socioeconomic and agri-food contextual background from which they stemmed.
Looking at the first thematic cluster, three themes dominate: quality, social benefits (safe and healthy food), and food loss and waste.Regarding the first one, some authors consider quality as one of the qualifying elements of food policy local experiences (Minotti et al., 2022;Calori et al., 2017;Mazzocchi, Marino, 2019); others use the concept referring to the composition of the menu in Public Procurement Policies (Mazzocchi, Marino, 2019).However, food quality can be interpreted in different ways, as mentioned by Andreola et al. (2021), by producers, standard or critical consumers, experts or other actors.Saviolidis et al. (2020), for example, report that quality is a possible solution identified by a stake- holder as a direct regulation policy activity to improve food security and enforce high quality standards.Forno and Maurano (2016) show how the new economic circuits are pushed and supported by growing groups of consumers who consider quality and wholesomeness of food as central.The theme of social benefits is often linked to the topic of redistribution and solidarity (Minotti et al., 2022;Andreola et al., 2021;Arcuri et al., 2022;Borrelli et al., 2017;Giambartolomei et al., 2021;Alberio, Moralli, 2021); while Guarascio (2022) emphasizes the role played by solidarity purchasing groups.Finally, food loss and waste is addressed as a crucial issue in relation to urban FPs by the majority of papers (Calori et al., 2017;Forno, Maurano, 2016).In particular, Fassio and Minotti (2019) emphasize the topic within a circular economy sphere, while Giordano et al. (2022) explore the role of alternative networks in reducing the phenomenon.Several papers jointly explore the three selected themes of the first cluster: quality, social benefit (safe and healthy food) and food loss and waste (Minotti et al., 2022;Andreola et al., 2021;Borrelli et al., 2017;Mazzocchi, Marino, 2019;Spadaro, Pettenati, 2022;Alberio, Moralli, 2021).
In the second thematic cluster, attention is mainly focused on the connection with the territory.In food policy related papers, some authors focus their attention on analysing specific territorial food governance policies and processes, investigating territorial relations and their weight (Minotti, Cimini et al., 2022;Mazzocchi, Marino, 2020, Andreola et al., 2021;Arcuri et al., 2022;Fassio, Minotti, 2019;Calori et al., 2017, Giambartolomei et al., 2021;Spadaro, Pettenati, 2022) but also on contextual factors driving the success of local productions (Vaquero-Piñeiro, 2021).In Saviolidis et al. (2020) territoriality is proposed by stakeholders involved in their research as a way to promote a more locally focused approach for rural areas' needs and support the transition towards a more sustainable production.For Rossi et al. (2021) promotion of food and territories connections or knowledge of territorial production and consumption systems links are crucial for the creation of a new food culture.
In the Food network related literature, Sacchi et al. (2022) outline that this kind of experience allowed a deeper knowledge of local territory and the creation of a network of local actors.Other authors use the territorial approach to examine local innovation of social groups engaged in organic production and consumption (Alberio, Moralli, 2021), or territorial implications and participation behind alternative food networks for local development (Guarascio, 2022;Forno, Maurano, 2016).Finally, in the third cluster dimension, both locutions are marked by a focus on sustainable agricultural practices and short food supply chain types, including local markets.For instance, some authors include the first theme in the analysis and discussions on food policies processes, objectives and key points (Mazzocchi, Marino, 2020; (Guarascio, 2022;Sacchi et al., 2022).Short chains and local markets are also included in the debate on food policies as key areas (Minotti et al., 2022) or, in relation to food networks, as a way to reorganize the production and consumption of fairer products (Guarascio, 2022).These themes analysed above are also addressed to different extents by a group of papers recorded for the following locutions considered: Food Plans, Food Communities, Food Movements4 and Food Councils (Vittuari et al., 2017;Cretella, 2019;Renna et al., 2018;Orria, Luise, 2017;Berti, Rossi, 2022).

Normative analysis
The analysis of the normative has identified two regulated headings in Italy, "Food District" and "Food Community".The lemma "district" was introduced by Law 317/1991 on "Interventions for the Innovation and Development of Small Enterprises", which dictated a formal definition of "Industrial Districts" and provided for them a detailed discipline for identifying areas and participants.The lemma "food district" was introduced by Law 205/2017, which rewrote art.13 of Legislative Decree 228/2001 "Orientation and modernisation of the agricultural sector" and it defines four types of districts: a) rural districts; b) quality agri-food districts; c) local production systems (LPS) characterised by a high concentration of small and medium-sized enterprises (SMES); d) LPS with an interregional character, e) LPS in urban and peri-urban areas; f) LPS characterised by the integration of agricultural and other activities; g) LPS with sustainable production; h) organic districts.
The lemma "Food Community" (FC) is codified by Law 194/2015, art.13.FCs are defined as local spheres resulting from agreements between farmers; solidarity purchasing groups; schools and universities; research centres; associations for the protection of biodiversity quality; school canteens; hospitals; SMES; public bodies; catering and commercial businesses.Agreements may have as their object (art.13,paragraph 3): a) study, recovery and transmission of knowledge on genetic resources of local food interest; b) implementation of forms of short supply chain; c) study and dissemination of organic farming practices or those with a low environmental impact and aimed at saving water, reducing carbon dioxide emissions, improving soil fertility and reducing the use of packaging; d) study, recovery and transmission of traditional knowledge relating to agricultural crops, natural seed selection to cope with climate change and proper nutrition; e) creation of educational, social, urban and collective gardens.
The aims of food districts and FC are reported in Table 4.
In the analysis of the regulations, 14 of the 18 thematic areas already present in the literature analysis were used, excluding those that were not pertinent, to which a further 2 themes were added, considering the relevance for their relations with the identified lemmas.For similarity, 16 thematic dimensions were identified and grouped into the 3 thematic clusters: Food Safety and Security, Territorial and Local Dimension, Food Supply Chain (FSC) Management (Table 5).With regard to the lemma "district", there is a quite even distribution in the three clusters even if the third one is less relevant than the others, while as far as communities are concerned, the second thematic cluster Territorial and Local Dimension has more weight than the other two, while the third cluster, relating to FSC Management, is marginal.
In the first cluster, all regional food district regulations focus on quality and, with the exceptions of Lazio and Tuscany regions, also on food loss and food waste  Food Districts (art. 13, comma 1, D. lgs. 18 228/2001) Food Communities (art. 13, Law 194/2015) Food Districts are established to promote territorial development, cohesion, and social inclusion, foster the integration of activities characterised by territorial proximity, ensure food safety, reduce the environmental impact of production, reduce food waste and safeguard the territory and rural landscape through agricultural and agri-food activities.
Food communities aim to: 1) raise public awareness, implying the protection and enhancement of agricultural and food biodiversity.2) support agricultural and food productions obtained from risk, both managed by breeders and farmers registered in the National Network as "Custodians" (art. 4 of Law 194/2015) and not registered in it.3) promote behaviour to protect biodiversity of agricultural and food interest.
and safety.The community charters identify quality; education; redistribution and solidarity economy as key items.The cluster Territorial and Local Dimension is the most relevant one for the two locutions under analysis, which all focus on territory and landscape & cultural heritage.In the case of FCs, Protection of biodiversity and Cooperation and collective forms are also themes present in all the Community Charters.In the third cluster, districts concentrate above all on sustainable agriculture, short food supply chains and local markets and partially (Calabria, Campania, Piedmont, Sicily, Umbria and Veneto) on distribution and logistics as well as (Calabria, Campania, Lombardy, Tuscany, Umbria and Veneto) on new technologies and innovation; while FCs partially focus on this cluster where FSCs and local markets have a greater weight.

Institutional documentation review
The analysis of the institutional documentation has been focused on 12 cities/towns and 1 grouping of Municipalities (Madonie).The documentation taken into consideration concerns three lemmas, as they represent the domains in which local institutions have rooms for planning: "FP", "food council" and "food plan".As shown in in Table 6, some territories have only one tool, such as Aosta and Bergamo which have adopted a Food Council.Other territories have instead produced institutional documentation on several areas, thus covering more completely the areas of governance of local food policies.
The Similarity Analysis result, performed with IRa-MuTeQ, offers a descriptive analysis of how the themes present in the corpus have been distributed (Figure 2).To enable the reading of results by an international audience, Appendix 1 contains translations of the lemmas contained in the graphical representation in Figure 2.
In terms of levels of governance, "food policy" is the most frequent lemma and represents an "umbrella" issue around which the implementation tools revolve.In fact, the locutions "food plan" and "FP" belong to the same area of homogeneity, which is strongly connected with the purple leaf -the area with greater connection homogeneity -in which the lemma "action" is the most represented.This highlights the strong operational nature that characterizes these two governance tools, especially in terms of activating institutional tools."Food council" is instead positioned in another area of homogeneity, in which words that refer to the representativeness of cities and spaces for collective participation co-occur.The lemmas "political" and "food", despite recurring consistently in the text, belong to another area of homogeneity, even more distant than that characterized by the word "action", in which there are cooccurrences with arguments concerning integrated and coordinated territorial planning.Finally, in an even more peripheral area of homogeneity, the terms "local", "food" and "system" recur quite frequently and are strongly connected.Less frequent is the word "sustain- able", loosely linked to "nutrition".Indeed, the thickness of the branches represents how much the words are connected to each other.Finally, it should be noted that the words most closely linked to the production system in the strict sense, i.e. "agro" and "fish", are very distant and weakly connected.
The clusters show in a synthetic way how the documents are focused on governance and actors to be involved in the FPs issue (yellow, red and violet), while food and its production are far away, highlighting the relative marginality of this important aspect in the current discourse around the food policies.

The power of narratives vs. real food system transformation
Discursive strategies are fundamental to the development of "attractive visions of alternative futures" (Turnheim, Geels, 2012), thereby building the cultural legitimacy of the new system (Mattioni et al., 2022).The main findings of our study show that the construction of narratives around the topic of FS planning is experiencing a particular momentum.In fact, the construction of complex governance structures around food-related policies is accompanied by wide-ranging policy documents, in which the prevailing narrative focuses particularly on the principles, background premises, and the frames in which structures and policy tools should take place.In general, FP is a recurring expression, a much evoked and attractive theme for both the scientific community and policymakers.This attractiveness can be explained by its multi-disciplinary and multi-sectoral nature, capable of evoking symbolic, cultural, and political elements that have a high collective appeal.This is particularly relevant in Italy, where despite the growing contribution of solid political and social analysis, the theme of food is subject to a trivialization, which suffers also from a stereotyped use of customs, traditions and narratives (Girardelli, 2004).
However, local FP interventions and actions are identified more for their best-practice character rather than as systemic measures responding to well-defined and detailed policy conditions.The Milan Urban Food Policy Pact (MUFPP) is paradigmatic in this regard.Indeed, in addition to creating an administrative and political culture around local food policies, and particularly around monitoring systems, what is particularly stimulating is the exchange of practices and learning between signatory cities.Indeed, through Milan Pact Awards, the MUFPP encourages action, facilitating the emergence of best practices, making them evident to the community with a function of inspiring the action of other signatory cities.Still, the results, particularly evident in the analysis of scientific papers, show that most food plans import good practices and apply solutions deemed effective a priori, instead of acting structurally on FSs.As already happened in the case of narratives around rural development (Fløysand, Jakobsen, 2007) and, more recently, agroecology (Barrios et al., 2020), also in the case of FP the programmatic documents are flattened on a "positive prejudice" bias with respect to the goals and needs of more sustainable and democratic local food systems.However, the translation of these principles into actions, resources, roles and trade-offs is  ).On the one hand, the plurality of actors potentially involved implies a great capacity for structuring complex and branched participatory systems; on the other, one of the main problems of food systems is the relationship between the multidisciplinary of food and the policy tools available, still calibrated on "silo" visions, i.e. sectorized and fragmented (James, Friel, 2015).
The textual analysis carried out on the institutional documentation reveals the absence of lemmas related to the analysis and understanding of FS.This aspect calls for a re-composition and rethinking of the sciencepolicy interfaces based on credibility, legitimacy, and diversity of knowledge (Turnhout et al., 2021).In this regard, the recent initiatives of Food Atlases should be mentioned, which in addition to mapping local FSs, from the urban (Dansero et al., 2018) to the metropolitan scale (Marino et al., 2022), act as tools to support policies, often promoted by researchers and scholars and endorsed by local administrations.

Agriculture in local food systems: the elephant in the room
Mapping works and studies such as Food Atlases have the merit of providing updated information on local FSs starting from the agricultural sector up to the postconsumption phases, passing through transformation, distribution, logistics and markets.This is particularly relevant since our analysis highlights the risk of marginalization of the agricultural sector within FP discourses, which instead represents the essential component underlying the functioning of local FSs based on systemic and circular approaches.Indeed, it can be stated that the ecological transition of FSs mainly passes through the redefinition of sustainable agricultural models and the reconnection between agriculture, food, and the environment (Lamine, Dawson, 2018).When speaking of "food", the downstream stages of the supply chain (see Table 3) are emphasized, with agriculture being neglected.This may be due to several reasons: on the one hand, the difficulty in identifying local planning tools capable of harmonizing with agricultural policies at any level which have direct or indirect impacts on agriculture (e.g.land use policies, direct payments, local bans on pesticides use, fiscal incentives for agriculture, etc.); on the other, a progressive interest towards alternative forms of food consumption and governance is guiding the interests of researchers and representatives of civil society.This latter aspect, despite having the advantage of systematizing and identifying the success factors of good practices, risks focusing on niche phenomena which have structural limits in growing in scaling-up, such as some forms of direct sales, Solidarity Purchasing Groups, Community-Supported Agriculture, Collective Farmer Shops, and so on.These trends risk overlooking and neglecting those structural components of FSs channelled through conventional and largescale distribution and retailing systems.In our opinion, they must be considered, analysed and questioned for a real ecological transition.The risk is that, in the absence of shared knowledge between science and policymaking, the strongest and most decisive components of FSs are left solely in the hands of the market.This, in turn, may entail a disempowerment of policy instruments and a lesser democratization of local food systems, but also an operative risk, given the growing corporate concentration and power in the global FS (Clapp, 2021).
This problem is also found in the regulatory analysis, as far as food districts are modelled according to public-administrative systems (where the active subject is the Region or any other administration) rather than private-business ones and, for this reason, they often appear inadequate for the needs of the territories where they operate.Food districts refer to a plurality of regulation and intervention tools, sometimes non-homogeneous, characterized by the overlap between rules aimed at regulating and supporting the phenomenon and rules mainly affecting other disciplinary areas.Moreover, only in a few cases the procedure follows bottom-up processes that are consistent with the model of locally-based and self-governed district organization that should be promoted.Of the terms examined, the one that contains the most references to the agricultural sector is the Food Communities (FCs).In Italy, FCs are regulated by a specific law with the aim of promoting sustainable models of food production and distribution while respecting the environment and biodiversity.They work to spread good agriculture and good food practices, to defend and protect local traditions and culture and to enhance the area, adopting their own mission.However, FCs, although regulated by specific rules, can have very different characteristics, given the diversity of possible agreements that can be made, deriving from the different relationships that can be established between the various subjects involved.

FINAL CONSIDERATIONS
Our research confirmed the tendency of institutions to develop simplified, self-consistent versions of reality, as demonstrated by Rayner (2012).Furthermore, a standardization of the responses by institutions and communities to problems related to food has been detected, as already noted by Lazzarini and Mareggi (2021) when analysing Italian local food plans and strategies.The risk is that the programmatic documents end up being drawn up more on ideological positions than on scientific evidence.Such a risk would also widen the distance between scientific research and political intervention.Such prejudiced considerations risk emptying, banalizing or stereotyping the technical language, which should be more specific through policy instruments.We hypothesize that, in the face of this gap, individual practices and initiatives, welltreated and analysed in the scientific literature, could have Food (di)lemmas: disentangling the Italian Local Food Policy narratives taken place even in the absence of those high-level planning instruments identified by the analysis of institutional documentation.However, from this point of view, a counterfactual evaluation would be necessary to identify the different shapes that local FSs could have taken even in the absence of institutional documentation.This is linked to an intrinsic limit of our study.Indeed, the survey was conducted in the presence of some kind of formal documentation, while we know that many FP-like experiences develop spontaneously and, sometimes, without the aid of official documentation or institutional endorsement.
Furthermore, the research was limited to investigating the prevailing narratives within food policies, identifying the relationships between the lemmas.However, it would be appropriate to further develop the study, delving into the individual tools and comparing them with the real application in the territories.Indeed, we assume that, although still lacking regulatory tools and levers, narratives around FP have stimulated a "politicization" of FSs and a broader awareness and political and social culture with respect to local FSs and the connections between elements and stages of supply chains that have traditionally been treated and considered as silos.That being so, the ambition of this paper has been to stimulate building a common language and a shared vocabulary of lemmas around narratives and conceptual discourses on food policy.We deem it necessary to construct a genuine, shared and truly multidisciplinary approach to the broad topic of food.This paves the way also to future developments on analysis on FPs, especially regarding the specificities compared to experiences in other European states, the ability to scale-up food policies to a national regulatory level, the possible displacement and inconsistency effects between FPs and agricultural policies, and the representativeness of the agricultural sector and farmers within these processes.
not always a straightforward and shared path, especially when it comes to coordinating different political visions and approaches expressed by local actors.This bias has threatened the capacity to rely on a proper solid literature presenting results and impacts of the construction and systematization in real contexts for long time.Only recently, a robust body of literature has been established assessing and evaluating the impacts of years of policies whose effects were considered positive "per se".However, food systems face very complex problems linked to their governance and government(McKeon, 2021

Figure 2 .
Figure 2. Analysis of similarities in the institutional documentation.

Table 1 .
Locutions targeted by the desk analysis in English and Italian.

Table 2 .
Databases research results and selection by locution.

Table 3 .
Relevance of themes and thematic clusters by locutions.

Table 4 .
Aims of Food district and Food Community.

Table 5 .
Relevance of themes and thematic clusters within the lemmas (%).

Table 6 .
Consulted institutional documentation per city/town/territory.