Slow Food as one in many a semiotic network approach to the geographical development of a social movement

Slow Food is a global, grassroots movement aimed at enhancing and sustaining local food cultures and traditions worldwide. Since its establishment in the 1980s as a local protest movement in Italy, Slow Food evolved into a global movement composed through countless local ‘grassroots’ activities intersecting with more ‘top-down’ umbrella orchestration and framing. This paper explores the many faces of this one-in-many movement by focusing on the circulation and variation of Slow Food ideas and practices across the world. The core argument is that these ideas and practices effectively capture and steer the manifold affective moments emerging from local and network activities. The affective–effective conversions are traced here by applying a semiotic-network approach to a large corpus of Slow Food websites. Thus adopting a novel approach and methodology of tracing the geographical development of a social movement, the paper reveals both grassroots and global patterns of change and diffusion, and zooms in on specific nodes, connections and practices that play a key role in the movement’s development. In doing so, we develop a research strategy that is better able to make sense of complex and non-linear processes of geographical diffusion-innovation.


Introduction
In 1986, a demonstration was organised by Arcigola, a newly established subdivision of Arci, the cultural and recreational federation of the Italian Left (Andrews, 2008).
The demonstration was a response to the intended opening of a McDonald's near the Spanish Steps in the heart of Rome. This would be the second branch of this fast food multinational to open in Italy with more to follow. The demonstration was a wider political statement against fast food and the argument of Arcigola was soon developed into the 'Slow Food Manifesto' by the Italian poet and writer Folco Portinari. Aptly, the Manifesto was first published in the Italian newspaper Il Manifesto in November 1987 and, according to Andrews (2008), it was this publication that exported Slow Food's ideas beyond Italy and set in the beginnings of the movement.
In December 1989, Slow Food was officially launched as an international movement in Paris. Nowadays the movement has over 1500 local chapters (the so-called convivia) in more than 150 countries worldwide. It owns a publishing house, maintains two magazines and operates a service company. Moreover, it has co-created not-for-profit entities such as the Slow Food foundation for Biodiversity, a university for gastronomic studies, a wine bank, the Terra Madre Foundation and others. Terra Madre is a network of Slow Food communities ranging from farmers and fishers to academics with an annual conference focusing on knowledge exchange and skills building. It also organises the so-called Presidia, devoted to the preservation and defence of rare foods, as well as events such as the Salone del Gusto, Slow Cheese, Slow Fish and many others (Simonetti, 2012: 169). The Salone del Gusto, a biannual marketplace for producers and consumers of good, fair and clean food, welcomed over 227,000 visitors during its most recent edition in September 2018 (SlowFood.com 2018), shows that Slow Food is now 'a true multinational entity, capable of raising funds on a large scale, of concluding cooperation agreements with governments and large corporations, and of mobilizing politicians and prominent personalities with divergent political opinions' (Simonetti, 2012: 169).
So, within a period of 25 years, Slow Food not only developed a global presence but also has taken on manifold faces, stemming from the many ways Slow Food practices have been enacted in local to global settings. At the grassroots level, such practices entail cooking clubs, political campaigning, book reading and debating events, education projects, markets and community service, among other. At the international level, Terra Madre, Salone del Gusto and other international conventions mix food markets, cooking and eating activities with seminars and strategic discussions. The faces emerging from such activities range from a 'high cuisine', epicurean face advocating the 'slow' enjoyment of high-quality food, notably of regional origin, and the protection of rare herbs and spices, to radical agendas attacking the global food industry. For example, Slow Food has intervened in discussions on the reform of the Common Agricultural Policy of the EU, but also invests heavily in educating children about food and cultivating food gardens in schools, villages and on the outskirts of cities in 25 African countries. From a pragmatic point of view, the cooccurrences of such diverse practices and faces hardly pose a problem. What they do prompt, however, is a continuing discussion on what face(s) the movement seeks to display, also as part of ambitions to engage more in direct political action. The internal diversity of Slow Food and the many faces it seeks to display sometimes cause tensions, in particular when several of the movement's 'faces' are simultaneously acting upon a specific Slow Food setting or practice. For instance, the epicurean or 'convivial' (that is, social/communal) face that many Slow Food Chapters in North Western Europe actively broadcast sometimes seems to disturb the more coordinated attempts of the movement's international Headquarters to engage in strong political activism and campaigning, such as campaigning against north-south inequalities or GMO agriculture (Siniscalchi, 2014).
In this paper, we aim to better understand the geographical diffusion and infusion of the Slow Food Movement and the ways in which the multiple 'faces' of Slow Food shape this one-in-many movement. We explore this by examining the circulation and variation of a variety of ideas and practices across the world and look into how these ideas and practices are enrolled in local settings. In so doing, we aim to make sense of how the complex processes of geographical diffusion-innovation help the Slow Food Movement to develop and reach-out globally.
Theory: Social movements as 'one in many' In the field of social movement research, the idea of a social movement as a collective consisting of 'many faces in one' is not new (see, among others, Tilly, 2001, quoting Barnes 1984 or the work of Cumbers and Routledge (2013) on global justice movements). In fact, many social movement studies have focused on how global forms of collective action result from interlinking grassroots practices, ideas and identities into a larger, often internationally oriented, collective (Bandy and Smith, 2005;Castells, 2012;Hendrikx, 2014;Juris, 2012). They have explored how internal differences are negated by specific forms of organisation, leadership and strategies of framing (Benford, 2013;Haug, 2013;Keck and Sikkink, 1998), and how this has helped to shape collective movement identities (Polletta and Jasper, 2001). Social movement studies have thus provided significant insights into transnational social movement development, most notably from a point of view that highlights organisation, coordination and stability (Chesters and Welsh, 2006;Davies, 2012). Yet, a key challenge for this literature remains how to understand and explore social movements as a form of collective action that is deeply and essentially rooted in diversity.
Three problems, in particular, stand out. First, notwithstanding recent developments in the field -notably the move towards network, discursive and geographical approaches (Diani, 2013;Juris, 2004;Laclau and Mouffe, 1985;Miller 2013) -mainstream social movement theory is still largely dominated by linear and categorical ways of thinking (Chesters and Welsh, 2006;Della Porta and Diani, 2006;Lockie, 2004). We feel that social movement theory still approaches movements, their contexts and the actors involved in their development, too much as stable and autonomous entities that develop in predictable and sequential ways. Even in their more recent work, authors such as McAdam and Tarrow (Tarrow, 2005, Tarrow andMcAdam 2005;particularly McAdam et al., 2001) tend to study movements on the basis of a more general set of concepts and categories. Their goal is to find common processes and mechanisms, finding similarities and universalities across an enormous landscape of radically different social movements worldwide (for a critique see Davies andFeatherstone, 2013 or Flacks, 2003). Although this work offers important insights into how movements may function on a more abstract level, we still feel that such an approach reduces social movements too much to a set of generic processes and 'underlying mechanisms' (such as, brokerage, certification, social appropriation, identity-shift -see, for instance, McAdam et al., 2001). Aiming to explain, in general terms, the emergence and behaviour of any social movement (or as McAdam labels it, social body of political contention), this subset of the literature sidesteps the many micro-practices, differences, multifarious actors, shifting relations and wider contexts through which every movement is also shaped. To put it differently, with such an approach it is extremely hard to portray movements as truly dynamic relational constructs that, in their core, always differ from political arena to arena, from time to time, from place to place, from practice to practice, and from actor to actor (e.g. Della Porta, 2013;Diani, 2013;Fligstein and McAdams, 2011). A more recent shift (also involving the authors discussed above) towards a more synthetic approach to movements, and in particular towards social-network/relational driven approaches, has not entirely done away with these limitations. In her much cited work, Saunders (2007), for instance, limits her relational study of the environmentalist movement by defining, at the onset, its composite elements as stable categories (institutionalised and non-institutionalised organisations, pressure groups), thus relying on sharp definitions of what is a part of the movement and what is not. The analytical focus is on 'the patterns of networking' emerging between different types of actor-organisations (radical, reformist, conservationist, . . . ) and how these impact the movement itself. This sketches a rather taxonomical portrait of the environmentalist movement, and withholds us from a more inclusive, dynamic view on the many actors, the fuzzy relations, shady and twitching borders, and the intricate interactions that also and continuously shape it. Similarly, in their characteristic social network analysis of the 'Free Al Jazeera staff' movement, Isa and Himelboim (2018) concentrate on binary network-relations (i.e. a relation is 'on or off'), and focus only on specific 'categories of actors' (core-actors, elite, non-elite, social-mediators, etc.) to determine how information and influence flows through particular routes and in so doing shape the #FreeAJStaff movement. Although this allows them to extricate important insights on key movement strategies in a digital age, it also blurs from view the many micropractices, differences and non-linear engagements that also help the movement to play its important political role.
Second, as far as social movement theory has paid attention to the transnational diffusion of social movements, and the circulation of the constitutive ideas, practices and identities that shape them, they have primarily done so from a 'one-directional' and sequential point of view. Ideas and practices are seen as shaping movements in a linear progression: ideas are sent out from a 'central' sender (an inventor, a movement entrepreneur) towards a specific adopter by means of specific channels of communication (the internet, a charismatic leader, a social movement organiser) (Chabot and Duyvendak, 2002;Hendrikx, 2014). In their much-cited work on the sit-in movements of the 1960s, Andrews and Biggs (2006), for instance, examine the diffusion of the innovative 'sit in' practice as a relatively stable form of practice, which then spreads out through coordinated action (movement organisations) and stable networks of established social/media interactions. This produces the image of an unaltered and strategic transfer from sender to receiver location. Similarly, in their renown work on 'The Dynamics of Contention', McAdam et al. (2001) portray the diffusion process within social movements as a relatively mechanistic process operating through a limited set of stable transfer channels, going through easy emission, transmission and adoption, enabling messages and scripts to travel without friction (see also Tilly, 2005). Yet, in practice, such travels are far more complex: they are gradual, often a distinction between senders and adopters cannot be made (see Featherstone, 2008), the direction of diffusion is sometimes turned around, and there are many unexpected developments that may speed up, slow down or even break chains of diffusion (Featherstone, 2008;Hendrikx, 2014). More fundamentally, it is impossible, both substantively and analytically, to separate innovation from diffusion, since new ideas and practices (i.e. innovations) always emerge from creative processes of association where circulating heterogeneous elements coming from elsewhere (in time and space) are combined and assembled in new inventive ways (Gherardi and Nicolini, 2000;Nicolini, 2010;Steinberg, 2008). Critically, items become and perform through their 'reach', which Allen (2016) conceives in terms of a dynamic, multifaceted topology. Reach, in this sense is the way in which any idea or practice is able to 'act at a distance', across space and time (see also Latour, 2005). This is not necessarily directed in a hierarchical spatial sense (from the top/centre to the bottom), but operates more often, in a much more fluid, dynamic (yet 'intensive') and relational way (Allen andCochrane, 2010: 1074). It works through affective social relationships, through powers of seduction and affection, through (re-)connection and dis-connection, and through their re-assembling in new surroundings. In his own words: 'Topological reach is thus a malleable feast in respect of its spacing and timing' (Allen, 2016: 51). To understand the development of ideas and identities across space, hence, one needs a more flexible and more pragmatic approach that focuses on the actual, empirical trajectories of 'circulation' through which ideas and practices travel-and-vary, thus 'becoming' a widespread social movement.
Third, in line with its predominantly categorical and linear perspective, the literature on social movements has generally adopted a binary language, and an accompanying analytical strategy, of micro-macro opposition and of local-global hierarchy to explain social movement developments (Cumbers et al., 2008;Lockie, 2004). Closely associated with this, social movement studies have tended to focus on either the micro-or the macro side of movements. On the one hand, there are studies that focus on specific (micro) cases of socialmovements (and social movement organisations), without systematically exploring the ways they are also shaped by wider interactional networks and processes that take place outside of the specific case and movements under study. By proposing 'case studies' as the ultimate form of social movement analysis, Snow and Trom (2002), for instance, suggest that focusing on particular 'movement instances' (p. 149) or 'a stretch of time in a movement's career' (p. 146) should help to build a reflective 'case of' the movement proper. On the other hand, an increasing range of studies focuses on the global and transnational manifestations of social movements, concentrating how particular actors and organisational networks are able to mobilise support at a global scale (Bandy and Smith, 2005;Castells, 2012;Keck and Sikkink, 1998). There has been less interest in the autonomous and diverse role played by a larger variety of (connected) grassroots practices in shaping heterogeneous collectives (Lockie, 2004;Melucci, 1996). As Lockie (2004: 47) notes, unless greater clarity can be provided over what it means to attribute agency to a collective subject the concept of 'social movement' may easily be reduced either to a reified category of structural analysis or a voluntaristic outcome of individual motivation Indeed, the literature seems to gloss over what Diani has called 'the fundamental characteristic of social movement experience', namely that of 'being part in a conflict which is at the same time embedded in specific, local orientations, interest, identities, but at the same time exceeds their boundaries, while maintaining the freedom and individuality of specific actors' (Diani, 2000: 8). What is lacking in social movement analysis, hence, is a focus on the intermediate level of movement processes, a focus on the interactions that bridge the gap between grassroots and collective dimensions of social movement development (Chesters and Welsh, 2006). It is at this intermediate level that different voices from with a movement make their 'presence', in ways that, due to technological (digital communication), but also other (political-institutional) reasons, have become more subtle, varied and incisive (Allen, 2016).
In this paper, we aim to respond to these three challenges. We aim to do so by outlining an approach to social movements that focus more strongly on the meso-level dynamics within social movements. We argue that the Slow Food Movement should be seen as an expression of social and global complexity, united and constituted by its internal diversity. Drawing on relational thinking, we develop a 'semiotic network approach' that helps us to highlight the emergent nature of social movement becoming, and that focuses attention to the dynamic and performative interactions between a multiplicity of components and actors (ranging from the grassroots to the collective) composing, changing and extending the Slow Food Movement across the world (cf. Davies, 2012). In doing so, this paper traces the particular time-space evolution of the movement at the level of activities and debates.
Our data consist of a comprehensive corpus of all Slow Food websites, complemented with ethnographic and interview data. 1 The analysis reveals both national and global patterns of repetition and difference, and zooms in on specific nodes and connections that play a key role in the movement's development.

Semiotic practices and networks
Our analysis of the movement's evolution focuses on mapping and tracing the development and proliferation of what we see as 'semiotic practices'. In brief, a 'semiotic practice' refers to a code ('way of thinking and doing') in an expressed form, through which the practice can travel and diffuse. As Latour (1996) explains, semiotics is the basis of the mobility of actornetworks; semiotics provides order and capacity of path-building, without reliance on the role of meaning or external intervention. Put differently, semiotic practices provide entities with 'access and travel codes', shaping them as actor-networks 'whose elements define and shape one another' (Law, 2007: 146). In the Slow Food Movement, the 'convivial gathering' is perhaps an insightful example of a semiotic practice that has a presence throughout the movement. The word convivium refers to the importance of community and communal practices within the movement; places where producers and consumers meet, discuss, and . . . eat. For Slow Food, convivial gatherings are 'the basic building blocks of the movement' that bring 'the Slow Food philosophy to life through the events and activities [that are] organized in local communities' (about section, SlowFood.com 2018). From simple shared meals and tastings, to visits to local producers and farms, conferences and discussions, film screenings/festivals and more, almost all chapters of Slow Food worldwide engage in 'conviviality'. As a semiotic practice, one can see it being used and shaped throughout the various geographical and social presences of the movement across the world, where each Slow Food community uses the idea and practice in their own (yet shared) ways. In the empirical section of the paper, we will focus more strongly on the travels of the semiotic practice of the 'school garden' as it travels through the Slow Food Movement.
As argued by Allen (2016) and others (Anderson, 2016;Pile, 2010), such assessing and travelling is not only a matter of circulation and translations, but also of persuasion and precognitive, radical open connectivities. One way to conceive and operationalise this is by distinguishing between 'affective' and 'effective' moments of semiotic practices (cf. Deleuze and Guattari, 1986). In their affective moment, practices originate from intimate encounter, which is open, creative and resourceful (Pile, 2010). The effective moment, on the other hand, involves expressive processes of 'signification' (Griggs and Howarth, 2014;Laclau and Mouffe, 1985), 'inscription' (Latour and Woolgar, 1986) and 'coding' (Delanda, 2006), that help ideas and practices to travel through the network. Effective moments arise when intimate practices turn semiotic, by 'coding' concrete activities in many of its local instances and by 'circulating' these codes through the ties and nodes of the movement's network. This results in 'trajectories' of proliferating activities, thus shaping the movement (Massey, 2005). Such trajectories often start with inspiring ideas, prompting explorations on how to obtain resources and support and how to enrol them, leading to new organisational strategies, routines for planning, for communication, financing, etc. In the Slow Food Movement, we can recognise such affective-effective articulations and 'conversions' in all the nodes of the movement, from the most central ones (Slow Food's headquarters in the Italian city of Bra) to the most outer ones (small 'peripheral' convivia in the Caribbean or Southeast Asia).
Besides the articulation of the intimate-potential (affective) and expressed-mobile (effective) moments, trajectories of proliferating semiotic practices manifest another important characteristic, namely a continual shifting between more concrete and abstract moments. This refers to what in assemblage terms is called '(de)territorialisation' (Deleuze and Guattari, 1986) and to what, in ANT terms underpins the creative processes of 'translation' (Latour, 2005). In more concrete faces, semiotic practices diffuse through concrete, physical encounter, with a stronger role of the affective, and through concrete scripts and devices (inducing effect). More abstract faces, on the other hand, entail generic narratives and values ('health', 'sustainability') and conceptual-synthetical frameworks and codes ('participation', 'internationalisation'). While these moments and faces should be considered as mutually constitutive, always operating in tandem (Allen, 2016), they present useful analytical entry points. This can be traced, in particular, by zooming in onto the processes of encounter, circulation and alignment occurring in the activities at the nodal points in the wider networks, such as convivia and (inter) national associations and meetings.
Using the concept of semiotic practices, we explore Slow Food as an emerging new social movement aspiring to be deeply rooted in affective settings, manifesting strong flexibility and diversity, as well as work as an effective 'machine' (Siniscalchi, 2014) yielding coherent and comprehensive results to gain societal and political recognition and clout. More specifically, Slow Food faces a crucial challenge of how to square two major tendencies. On the one hand, it seeks to develop and promote a kind of 'responsible hedonism' (Slow Food International, 2007), with a strong emphasis of the enjoyment of eating and tasting in a way that takes on key concerns about health, sustainability and (bio)diversity in a primarily affective way. On the other, the movement seeks to develop a stronger collective political voice, targeting major societal problems and issues such as obesity, environmental degradation and land-grabbing through concrete means of expression. In dealing with these tendencies, Slow Food is markedly Janus-faced, being simultaneously affective and effective.

Affective movement
The affective movement crucially stems from the double dynamics of (1) the generative, creative processes in which grassroots groups assemble together different circulating elements into concrete, intimate activities and, (2) the ways in which these give rise to specific 'structures of feeling' (Anderson, 2016) propelling and continuing local practices. Grassroots groups are heterogeneous assemblies fusing resources (money, meeting spaces, food products and practices, websites, etc.), individuals (members, leaders, inspirators, etc.), and ideas and practices (potluck dinners, political campaigns, community projects, convivia guidelines, movement ideologies). This fusion brings to presence certain key emotions and affects through which members become attached to the movements' set of values and ambitions. Together, this shapes what Pile (2010: 5) calls an 'affectual geography'. 'Elitist' settings in North America and Western Europe giving rise to highbrow eating clubs contrast to 'radical' settings in Latin America and Africa where Slow Food gives rise to concerns about land grabbing and the domination of Western (and increasingly Chinese) corporations of food production and retail.
Besides contextual diversity, geography also brings the movement together by providing the opportunity for networking and gathering. Geography supports different kinds of nodes and networks -physical and affective -which play key roles in selecting and proliferating certain semiotic practices. This presents a bridge to the other side of the movement, namely 'as one'.

Effective movement
The movement acquires coherence, and thus becomes 'one' in a twofold way. Firstly, the movement's coherence emerges through multiple networked exchanges and circulations involving the proliferation of mobile 'semiotic practices' discussed above. In other words, it emerges through the day-to-day concretisation of emerging and circulating ideas, concepts and practices in different places across the world. Mobility comes with variation. Through their connectivity and adaptability, some semiotic practices (e.g. school gardens, Terra Madre day) are able to circulate more easily through Slow Food's networks of exchange. Other semiotic practices (e.g. Slow Fish) have more difficulties to travel either because the movement' sites are less receptive to their codes, or less able to adapt their codes locally. The result of these circulations and articulations is a specific (geographical) patterning of Slow Food concepts and practices that endow the movement with a particular diversity and coherence worldwide. This patterning is never final and exclusive, since there are always other semiotic practices that may start to circulate, providing alternative Slow Food identities and meanings. Nevertheless, it temporarily results in the prevalence of certain ideas and practices, and the relative marginalisation of others, thus providing the network as a whole with certain key symbols and narratives.
Secondly, the movement seeks to present itself as 'one' by framing its activities and practices through sets of nodal codes, i.e. signifiers and narratives that express coherence, identity and visions in a more abstract-synthetical way. A core example is the movement's universal slogan of 'good, clean and fair'. More broadly, framing occurs through the positioning of Slow Food in terms of wider societal discourses and signifiers such as 'sustainability', 'fair trade' and 'biodiversity'. This positioning puts a movement irrevocably in the political domain, internally as well as externally. Internally, the choice to align the movement more explicitly with any signifier or narrative will undoubtedly be supported by some and opposed by others, prompted by both affective and effective articulations. We can see this in the way local 'intimate' encounters, as well as circulating semiotic practices are discursively linked to specific narratives. An example is how a practice like 'Slow Food school gardens' is framed in terms of either health, education, or local development (Hendrikx et al., 2017). Externally, the movement positions itself in relation to prevailing narratives on food, health and sustainability. In the case of Slow Food this is an alternative of a global world of interconnected food production and consumption, pitted against the global domination of industrial food chains and supermarkets. As Slow Food president Petrini (2012: preface) argues: [ . . . ] I do not believe it is utopian to imagine a system like this in the future, made up of many local economies with a network established among them, thus not at all closed off and not necessarily self-sufficient, but completely open to exchange and innovation. If more people learn to become more familiar with food culture they will come to a greater understanding of the profound, irreplaceable value that food has, through the myriad of unseen interconnections between man and the land, and then the change will come almost spontaneously, because this is a common-sense solution.
These two developments, the circulation and development of semiotic practices and the articulation of the movement's defining narratives or discourses, happen in tandem throughout the movement. Its geography is a like a pulsing body, expanding into new territories while also holding on to 'established' core positions and narratives. In this expansion, the circulation of semiotic practices, articulation of key codes and the anchoring in broader narratives are crucial for Slow Food's reach as a powerful political voice. In fact, to make the international political aim of changing the food system into 'good, clean and fair' valid and powerful, it is paramount that the movement's more abstract faces connect to local concrete practices and actions. This is why the effective circulation and translation of semiotic practices between different contexts is so instrumental in the consolidation and versatility of Slow Food worldwide.
To comprehend the movement's continual conversions in some more detail, the remainder of the paper will take two opposite viewpoints: zooming out to map sites and practices, and zooming in onto one iconic Slow Food practice, namely School Gardens.

Zooming out: The web presence of Slow Food
To get a comprehensive overview of its semiotic practices within a surveyable amount of time, our project makes extensive use of the web presence of the movement. To do so, we have manually collected, stored and analysed data from over 400 Slow Food websites, focusing on ambitions, events and practices (2014)(2015)(2016). The map shown in Figure 1 combines the movement's information on the location of all convivia around the world with our own analysis of the web presence of the movement. The spheres on the map are linked to particular countries, and the size of each sphere is determined by the total number of convivia in that particular country. The different shadings denote which part of the total number of convivia has (or has not) a presence on the Internet. Convivia with a web presence are represented in the lighter shading; countries with a larger share of convivia with a web presence, like the United States or Japan, are represented by a predominantly light sphere. Countries with few convivia with a web presence, like Germany and Brazil, mainly have a darker shade. The map also shows other related activities, projects and events which are clustered in five groups: Cittaslow, Earth Markets, Terra Madre, the Youth Food Movement and various other related websites.
If we explore how the websites relate to each other, we can observe the geography with clear centres (USA and Italy, and individual websites like Slow Food international) and peripheries ( Figure 2). This map shows the hyperlinks between 800 sites and is made using dedicated web crawling software. This map has some clear limitations, since, for instance, the translation of offline practices into online websites is not one to one (although, see Park and Thelwall, 2003). Yet, it does provide us with a good approximation of the movement's development and its functioning in effective terms (for a more detailed explanation of this analysis, see Hendrikx et al., 2017).
From the 800 Slow Food websites active across the world, we have selected three national clusters for further analysis. These are Slow Food in the USA (220 websites), Slow Food in the UK (22 websites) and Slow Food in the Netherlands (11 websites). The three national clusters have been selected for a variety of reasons. First, to examine the proliferation and circulation of specific semiotic practices, our concern was to avoid issues of translation, thus focusing on English speaking countries and on Dutch speaking convivia, as Dutch is our mother-tongue. Direct translation is a risky practice in discourse and content analysis, as it might erase the experiential, socio-cultural meaning and interpretation with which a word, concept or idea has become endowed in its 'original' language (Sch€ affner, 2004;Van Dijk, 2002). Second, each of the three samples represents a different 'cultural' part of the movement, whereas Slow Food in the USA manifests itself as being more activist and community oriented (Figure 3), Slow Food in the UK focuses more strongly on biodiversity and 'forgotten' foods, and Slow Food in the Netherlands is associated more strongly with epicurean values. 2 Third, convivia in the USA, in the UK and in The Netherlands are organised via a national Slow Food organisation rather than directly via the central organisation in Italy. This gives the respective national movements in these countries relatively more leeway and autonomy to diversify and open up its identity and practices.
The websites of Slow Food groups in each of the countries have been analysed on the basis of over 40 indicators. Among other things, these indicators cover general and organisational issues, programmes, activities and campaigns, and issues concerning the mission and ambition of the convivium (Table 1). With this spectrum, these items reveal spatial patterns of affective-effective conversions, through which rooted aspirations and intimacies are turned into circulating semiotic practices conveying more abstract ambitions and directions.
The database includes basic geographic information of each of the groups studied. This allows us to plot the different convivia on a map, and attach to these sites, by means of GIS programming, data about their different ambitions, identities and activities. More specifically, we have used this data to highlight the movement's concrete ambitions and practices in terms of political engagement (politically activist, community oriented, campaigning for biodiversity and health), the type of events organised (markets, food workshops, films, eating and cooking activities) and an engagement with forms of food production (farming, gardening, school gardening, agriculture development and education, food traditions, food sovereignty). The themes local convivia engage with are not exhaustive, for reasons of space we have chosen here for three of the categories most ubiquitous within our corpus. Below, we have visualised one issue per theme for each of the three countries. While our data are dynamic, in the context of this paper we can only show static maps.
The maps zoom in onto specific national and regional clusters of Slow Food groups across the world, providing a means to assess their main characteristics and diversity at bird's eye view. From such a perspective, we can observe that Slow Food groups in the USA and in the UK convey more political activism than those in the Netherlands. Slow Food in the USA is focused strongly on community service in comparison to those in the UK and in the Netherlands. In terms of concrete events organised, we can see that almost all of the convivia in the Netherlands are engaged with organising markets, which takes place to a lesser extent in the other two countries. Almost all of the convivia in the corpus organise eating and cooking events for their local convivia on a regular basis. Gardening and school gardens are themes that are more dominant in the Anglo-Saxon world, yet these practices are slowly diffusing throughout the Slow Food Movement. To provide more detail, below we zoom in onto the trajectory, circulation and development of the diffusion of the School Gardens semiotic practice. This serves as an example of the emergence of movement difference and coherence through the relational semiotics of the Slow Food Movement.

Zooming in: School gardens
Since the early days of the movement, food and taste education is an important aspect of what Slow Food does. In June 1992, Slow Food launched a programme, called 'Taste Week', which turned Italian classrooms into workshops and kitchens to involve youngsters in supervised tasting and practical food experiences (Morgan and Sonnino, 2008;Petrini, 2001). This initiative culminated in May 1997 with a conference in Rome entitled 'Dire fare gustare. Discorsi, progretti, esperienze intorno all'educazione sensoriale' ['saying, doing and enjoying. Discourses, projects and experiences of sensory education'] which eventually led to the publication of the Rossano Nistri's book Dire fare gustare in 1998 (Petrini, 2001). Both the conference and the book are nowadays seen as the beginning of Slow Food's taste education projects.
Meanwhile, in Berkeley, California, the Edible Schoolyard (ESY) was founded. In 1995, Alice Waters, organic food activist, author and owner of the renowned Berkeley restaurant 'Chez Panisse', and Neil Smith, then principal of the local Martin Luther King Jr. Middle School, joint forces to transform an unused plot of land in back of the school into a garden project that would involve students, teachers and community. The space was transformed into an organic garden and a kitchen classroom was developed to provide an intimate environment for students, producing vibrant local affectual geographies. At ESY Berkeley, students participate in all aspects of growing, harvesting, and preparing nutritious, seasonal produce during the academic day and in after-school classes. The King Middle School garden serves as a model for other Edible Schoolyard affiliate programmes that are being established around the USA. There are affiliate programmes located in New Orleans, Los Angeles, San Francisco, Greensboro and Brooklyn. The wider network of the edible education movement encompasses 2118 locations predominantly located in the USA. School Gardens are considered as vital activities to confront children (and their parents) with the idea of healthy, locally produced, original food.
In 2001, only one year after its foundation, Slow Food USA embraced the idea of food and taste education and led the first national project promoting School Gardens. An Educational Committee was set up to launch the project nationwide, and, in just two years, 30 school gardens sprang up around the USA. Entitled 'Garden-to- Table', the project has since expanded to include after-school cooking programmes and farm tours. Soon after, the initiative became popular in Europe as well. At the International Slow Food Congress in 2003 it was decided that every convivium in the world should work to set up School Garden projects, and later in 2006 at the Slow Food Italy National Congress in San Remo it was resolved to create 100 gardens in Italy. In doing so, the concrete School Gardens activity became the subject of coding, inscription and counting, able to extend it affectual geography to different players in the network, allowing the international movement to set targets and produce statistics.
While School Gardens thus reached out, they did so with considerable variation. Our inventory revealed that Slow Food is involved in nearly 100 of about 435 school gardens existing in Italy. Together with the students, teachers, parents, grandparents, and local producers are involved in the project, creating the learning communities for transmission to the younger generation of knowledge related to food culture and environmental protection. The educational themes of the school gardens are similar; yet they are adapted to the culture and territory of each country. Slow Food views all schools with gardens as part of the same large network of learning communities. In the USA, there is a national programme focused on the processing of products in the kitchen and on the supply of products for the school canteen. In Cuba, not far from Havana, a network of urban gardens was established with support of Slow Food International (Slow Food International, 2010). The project involves a school and more than 30 families. In September 2009, a network of 12 schools in 8 European countries (Belgium, Bulgaria, France, Ireland, Poland, the United Kingdom-Northern Ireland, Romania, Spain) was established. In the UK, the Slow Food school-garden idea did not gain much foothold. Although several convivia started to organise children's taste and education events, and a general children's Taste Adventure project was launched by Slow Food UK in cooperation with a local Slow Food group in Oxfordshire, only a few UK convivia are engaged with (school) gardens directly. One London convivium has set up a Slow Food global schools 'twinning' project (2013), to match schools with others in the UK as well as further afield 'so that the young people can share their stories on differing traditional foods, growing techniques and water availability through creative writing, pictures, photos and Skype calls' (Slow Food UK, 2013). Also, the convivium of Slow Food Oxon (Oxfordshire) expresses an ambition to help set up various school gardens in the region. In the Netherlands, of the currently (2020) existing 16 convivia, only a few have engaged with educational projects for children, and only one of them involves a gardening project. A communal garden, which also works with students and children to generate and share knowledge about local urban gardening is the only active Slow Food gardening project noticeable in the Netherlands (Slow Food Limburg, 2018).
In 2010, the Slow Food School Garden movement helped to establish an ambitious Slow Food Programme, namely the 'Thousand Gardens for Africa'. After thorough preparation, the programme was launched at the international Terra Madre convention in 2010. The programme envisaged the building of school gardens across Africa, as well as community gardens within villages and elsewhere. The ambitions of the programme went beyond youth education. It also aimed at reviving and protecting local agricultural traditions and practices, countering the colonisation of food production and retail by international corporations and technological standards. In the words of Roger Bello, Cameroonian delegate at Terra Madre 2010: For years the Westerners have explained to us that everything we were doing was wrong. They said that our techniques were inefficient, that our products had to be replaced by more productive crops. Now we hear the opposite. At Terra Madre they told us our story is important, that we have to recover the wisdom of our ancestors, be proud of our roots, cultivate our own grains and the fruits selected by our fathers. These are new words for us. (Bello, quoted in: Slow Food International, 2011: 4).
Alongside the global ambition, the programme entailed the development of a handbook (script) and country seminars, the building of learning networks, a public sponsoring scheme, supervision by the Slow Food Foundation of Biodiversity, and extensive publicity via various websites of Slow Food and other organisations, media coverage and dedicated events at international Slow Food meetings. So far, over 600 gardens have been established. From Senegal to Ivory Coast, school gardens help to feed children with local and healthy food, while also enhancing traditional knowledge. In Uganda, for example, a project of Slow Food Mukuno involves 17 schools and offers over 620 students the opportunity to learn and grow local products, which are then cooked in the school cafeteria. Likewise, village gardens have popped up across the continent (Slow Food International, 2014).
'Thousand Gardens of Africa' has received a further boost at the 2012 Salone del Gusto/ Terra Madre convention in Turin. The results and future plans have been presented and discussed in detail. The programme is looking for institutional sponsorship from corporate as well as public organisations. National and local Slow Food organisations are summoned to support programme and to help the number of gardens to grow beyond 1000. The practice thus presents a major impetus, as well as vehicle for the 'global' ('North South') ambitions of the Slow Food Movement.
For Petrini and Slow Food Headquarters in Bra, the Thousand Gardens in Africa programme formed an ideal way to give more substance to the political side of Slow Food as a Social Movement. It fitted in Slow Food International's key drive to move away from the 'softer' epicurean side of the movement towards a more impactful activist role. This is reflected in Petrini's remarks about the Salone del Gusto. In a key note speech, Petrini states that Slow Food will start to prioritise the Terra Madre conference, which is about politics, campaigns and contents, over the Salone del Gusto, which is more about tasting, the products itself and conviviality (field notes Salone 2014). Since 2016, this is even reflected in the renaming of the biannual event changing it from Salone del Gusto to 'Terra Madre Salone del Gusto'.
The focus on coordinated international campaigning, in another example, the Thousand Gardens development project, did, however, not result in praise from all (grassroots) parts of the movement. As one national organiser says: Slow Food as a movement seems to be divided in, say, the epicurean side (the Foodies) and the Terra Madre side ( . . . the political side that is about 'all those who are hungry' unite!) . . . In our Chapters, most just want to enjoy themselves and were interested high quality food; they were interested in good quality butter . . . if you want to have a really good quality porc-sausage you need to go to this and that address . . . We did not want to go into development aid via the Thousand Gardens project; many of us said: 'we already give a lot of our money to development projects via different channels, Slow Food should not become a development organisation. (field notes Salone, 2014) Such claims, when made at the international board meetings of Slow Food, however, were not appreciated. It undermined the international political strategy of Slow Food: We discussed it extensively in [our country]. So, the next year I conveyed our standpoint to the international board meeting: This is not appealing to us in [our country], this is the believe of the majority of our national members . . . We shall therefore not support the Thousand Gardens programme from [our country] . . . So, Carlo said, that this was not possible. We could contribute less to the programme, but nothing that was not to be allowed. (field notes Salone, 2014) This testifies to how, despite the proliferation of school-garden semiotic practice in all its concrete and abstract faces, there were clear boundaries to its affectual geography.

Conclusion
Social movements present networks of change, operating on the basis of a delicate mix of diversity and coherence, driven by continual conversions between intimate encounters and presences (affective) and semiotic practices and networks (effective). Diversity stems, in particular, from the manifold ways in which a movement prompts concrete activities at the grassroots, even when resulting from the local adoption of concrete scripts and devices and more abstract codes and narratives. Yet, because of the reach thus gained, diversity also supports coherence, through the establishment and stabilisation of movement-wide semiotic practices, encompassing standard scripts and coding as well as 'global' stories, slogans and symbols. In embracing this coherence-through/in-diversity, the 'one in many'), we have gone beyond the categorical thinking still prevailing in social movement literature.
Our second contribution to the literature here is the emphasis on complex diffusion in which extend ANT's emphasis on circulation and articulation to the important role of affectual geographies. Semiotic practices evolve in the meso-level network of the connected sites of the movement, supported by nodal events (global conventions), infrastructures and players such as seminal authors (like Petrini or Waters) To analyse these dynamics, we focus on the way affective encounters and activities become coded, and how emergent semiotic practices start circulating through the movement (such as cooking-eating-tasting events, book reading and debate sessions, and school gardens, among many others). Our research is focused on tracing semiotic-practices through the movement, basically examining the network as a whole, mapping the circulation at some level of detail, and zooming in onto more specific 'nodal points' that play a key role in (re)framing and (re)scripting practices and in (re)narrating the movement's scope and ambitions. Such nodal points include national and international organisations, meetings and forms of communication.
In our perspective, semiotic practices move through and constitute the meso-level of the movement, the network connecting its sites and nodal points. Our conceptualisation and operationalisation of this meso-level presents the third contribution to the debate on social movements. We have illustrated this movement and constitution using the case of the school gardens practice. Through following the development of this semiotic practice, from its pre-Slow Food birth in the USA, its consolidation by Slow Food USA, its adoption by other local and national Slow Food organisations in Italy and elsewhere, and its conversion in an ambitious development project, we map and trace the development of the activity as well as its wider meaning. This presents a trajectory of continual conversions. Local material encounters, of advocates launching the idea of a school garden, and of students, teachers and community members actually 'making their hand dirty', fuel emotions and sentiments on how food can be produced, distributed and consumed in a 'good, fair and clean' manner, resulting in vibrant local affectual geographies. Wider communication and translation brought this to presence in national and international Slow Food sites, thus fuelling the conversion into a semiotic practice, extending and transforming the affectual geography. In doing so, School Gardens became part of more organisational and political processes, intersecting and sometimes also competing with other practices and ambitions. What remains to be seen is how the global network will provide a conduit for mustering further support for School (and Community) Gardens elsewhere, such as propagated through 'Thousand Gardens of Africa'. This clearly does not present a one-size-fits-all trajectory. The internal diversity of the movement has made some groups more responsive to the idea of political campaigning that others. The political activism reaching out through the movement from Italy, for instance, did not chime well with the more epicurean interest of many of the Western-European groups.
This exemplary trajectory suggests that semiotic practices could be seen as codes that are subject to evolutionary processes, converting from 'niches' to more mainstream practices that may, in turn, be confronted with resistance and (out)competition. How a practice evolves, spreads, mutates and dies out can be understood only by tracing its development from site to site, from node to node, and by understanding its intimate and practical significance for local communities as well as the network as a whole. Co-evolving with these practices, through processes of abstraction, are the wider narratives of a movement. Importantly, these narratives remain closely articulated with local affects and activities. No central narrative, accordingly, can serve to fully control the development and proliferation of semiotic practices (Allen, 2016). Which practices are locally favoured is, to a large extent, resulting from local conditions and preferences. Certain stimuli, of a discursive as well as material nature, can obviously prompt certain initiatives, but the extent to which they really land and become successful is contingent upon how circulating scripts and frames are locally received and negotiated. In turn, where local activities meet (repetitive) success, and where this success is coded and becomes part of larger narratives, they may impact on the way the movement performs as a coherent whole.
In elaborating a semiotic network approach, drawing on a relational geographical perspective and using novel Internet search techniques, we have tried to overcome the classical divide in social network theory between structure-and agency-oriented approaches. Semiotic practices -as entities of circulation -provide a middle-range concept that helps to understand how forms of agency (activity building as well as nodal practices) help to build structure (material and discursive) as well as the reverse, against the background of affective settings and developments. We thus gain specific insights into the making and evolution of a movement as a diverse and connected 'grassroots' activity network, and the extent to which this is guided by actions taken at a movement's central nodes (organisational hubs and global meetings). The focus on encounter, coding and signification, moreover, sheds light on the twists and turns affecting a movement's political orientation and clout. This paper has shown how, as a core activity and symbol, 'School Gardens' have been promoted and broadcast by Slow Food International, meeting resistance at the grassroots level. It will be interesting to see how future developments in local-global communication and interaction will affect the practical, symbolic and political significance of such semiotic practice. Another interesting question is to which extent a semiotic network approach may shed new light on geographical processes of innovation, diffusion and transition also beyond the domain of social movements.

Highlights
• Studies the simultaneous diversity and coherence of the Slow Food Movement, a global, grassroots movement aimed at enhancing and sustaining local food cultures and traditions worldwide. • It responds to prevailing ways of categorical thinking about social movements by focusing more strongly on the circulation and variation of 'semiotic practices' (ideas, practices, ways of doing) of Slow Food across the world. • It shows a distinctive geographical patterning of the Slow Food Movement, which is infused with both grassroots and global patterns of change and diffusion, and zooms in and out on specific nodes, connections and practices that play a key role in the movement's development. • We show, for example, how, as a core Slow Food activity and symbol, 'School Gardens' have been promoted and broadcast by Slow Food International, meeting resistance, changes and alterations at the grassroots level.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the Dutch National Research Council NWO for providing the grant that made this research possible (under the label of the Complexity Program) and Dr Stefan Dormans for his participation in part of the fieldwork on which the paper draws.
Notes 1. Ethnographic data were drawn from our active participation in the movement both locally and internationally, in the period 2012-2016. In this period, we also held several interviews with key national, local and international Slow Food organisers (approx. 25 interviews), at various Slow Food events and private appointments, and gathered a wide range of documents, brochures and pamphlets used within the movement. 2. This selection has been made on the basis of the (strong) contrast that appears between these three different shapes of the movement. The existence of such different 'cultures' within the movement has been confirmed both by our own qualitative engagement within the international Slow Food Movement and through several of the interviews that we had with international slow food members.