1A century after Schumpeter ( 1983) and Weber ([1904-05] 2008) placed it at the heart of their analyses of capitalism, the entrepreneur occupies once more a central place in various currents of sociological research: in economic sociology, of course (see, in particular, Stark et al. 2009; Zalio 2009), but also in the sociology of social movements, sociology of public policy, sociology of science or the sociology of organizations. References have been made to “moral entrepreneurs” (Becker 1963), to “policy entrepreneurs” (Cobb and Elder 1972), or to “public policy entrepreneurs” (Kingdon 1984) to subsume actors who try to problematize and bring a (social) problem into the public arena (and especially to place it on the public agenda). Some neo-institutionalist authors meanwhile forged the concept of “institutional entrepreneurs” (Eisenstadt 1980; DiMaggio 1988; Oliver 1991) or are content simply with the word “entrepreneur” (Beckert 1999; Fligstein 2001; Greenwood and Suddaby 2006; Lawrence and Suddaby 2006; Fligstein and McAdam 2011) to describe those actors playing an active role in the transformation of institutional rules. Finally, the sociology of science, especially actor-network theory, has sought to describe the role of “entrepreneurs”/ “innovators” in the reconfiguration of the areas of scientific problems and networks (Latour 1988; Callon, 1986).
2Over and above the diversity of contexts and actors to which the concepts of entrepreneurship and entrepreneur have been applied, these sociological approaches share, more or less explicitly, a set of assumptions and conclusions. Firstly, entrepreneurs are designated in such perspectives as disturbers of the “social order.”  Whether they are introducing a new process, a new mode of exchange, a new organizational form or an innovative product in a market, are trying to promote a new challenge, a new cause or a new method on the public agenda, or striving to establish new institutional rules within an organization or field, entrepreneurs disturb the spaces in which they operate, break routines and subvert institutionalized power relations and stable hierarchies. Then, if they often try and sometimes fail to change the social order, it is because they “envision new institutions as a means of advancing interests they value highly yet that are suppressed by extant logics or rationales” (Greenwood and Suddaby 2006: 29). In fact, entrepreneurs are readily thought of as actors who are “interest-driven, aware and calculating ” (ibid. 29); see also Beckert (1999), Fligstein (2001), Fligstein and MacAdam (2011), they differ from other social actors in their desire to increase the realization of their interests in the transformation of the space in which they operate rather than by its reproduction.
3Another widely shared research result presents the entrepreneur as situated in a unique position in social space, which is located on the boundaries of several territories.  These approaches find a former French use of the term “entrepreneur” as “one who encroaches on” the domain of another, but also a more modern usage tends to present him as one who “entre-prend,” “inter-esse” in the words of Callon (1986), that is to say someone who has a hold on, or is caught between different worlds, objects, rationales… If entrepreneurs are able to modify the institutional logic that structures the social space, it is especially because this social space is divided into sub-areas as well as linked, by many boundaries or hinges (Abbott  2001, 2005), and organized according to varied cognitive, normative, political and economic rationales, even if conflicting ones. Entrepreneurs can be defined as “economic actors who, to serve their own interests, or those of their business, play (in the range of contexts in which they are observed) on the plurality of relational configurations, contexts, social framings and organizational arrangements of the activity, holes, boundaries to identify, create, or duplicate evaluations, generating innovations and price differences on which to build a sustainable profit” (Zalio 2009: 602). Although it focuses mainly on economic entrepreneurs, we believe that this “topological definition” can be generalized to all those actors who intend to exploit the heterogeneity and fragmentation of social spheres to build their business and feed their ambitions.
4Two types of entrepreneur emerge more clearly from this heterogeneous body of research: the entrepreneur as an “intermediary,” who ensures the flow of resources between worlds governed by values, norms and institutional logic of various kinds and the entrepreneur as a “translator,” capable of (re)combining resources in the different worlds through which he moves in an original way to produce new objects or new rationales. These two approaches, which were developed from some quite convincing case studies, have three main limitations, however. The first is that they fail to understand cases where the different universes/worlds/fields are not (yet) willing to accept radical change, though they are not hostile to the emergence of entities or elements that are deviant, dissonant or innovative. The second, which is an extension of the first, is about what these approaches implicitly or explicitly suppose is involved in situations where change has taken a sufficiently complete form that actors can unambiguously identify and characterize it as such. Finally, and most importantly, that by focusing their attention on how the entrepreneur acts on the areas he connects, these two approaches often overlook how the entrepreneur is reciprocally acted upon by the social worlds that he puts together. Building on the work of Padgett and Ansell (1993) on the one hand, and Star and Griesemer (1989) on the other, we develop a third type of entrepreneur, one we describe as the “boundary-entrepreneur.” This entrepreneur is characterized by his position on the boundaries of multiple worlds under tension, his ability to reproduce and reinforce many boundaries and his role as a “boundary object” open to the projections and manipulations of various actors involved in these worlds. Unlike the mechanism of translation (Callon 1986), the success of this entrepreneur is not based so much on his ability to involve other actors in his programme and to reconstruct a new world of which he would be the centre, as on his ability to play with its multiple identities and positions to configure his enterprise in accordance with the contours and rationales of the worlds that he connects.
5The study of the actors and the organization that developed the programme “Together Let’s Prevent Childhood Obesity” (“Ensemble, Prévenons l’Obésité des Enfants”–EPODE) allows us to provide an example of this figure of the boundary-entrepreneur.  This programme achieved a prominent place among among public health operators in just five years of existence. Launched in 2004 in ten “pilot cities,” it was, by 2010, implemented in some 226 municipalities. It has also obtained official recognition of considerable symbolic significance, when the French Health Minister decided to award the title National Plan for Health and Nutrition (“Plan National Nutrition Santé” PNNS) to all cities that joined EPODE, and agreed to serve in the Board of the association. This success  was not however a foregone conclusion. Indeed, the programme sought to build bridges between compartmentalized worlds often in conflict, and threatened two institutions: one that ensures that public health is a core competence of the State (Bergeron 2010; Bergeron and Nathanson 2012) and in which municipalities have little involvement in its roles  (other than those related to the implementation of basic health policies); the second which establishes that private actors, and a fortiori those considered to be part of the problem, cannot genuinely take part in public health campaigns. However, the programme aims not only to make municipalities into key players in the implementation of the national policy to fight against obesity, but it has also the distinction of being jointly funded by contributions from its member municipalities and by grants from partner companies who mainly come from the food industry and supermarket sectors. Add to this the fact that the steering group is provided by an association under the French 1901 law (non-profit organisation), but also by a private agency for marketing and health communication. The EPODE is therefore an “incongruous” sign (in the sense of the “not appropriate” of the neoinstitutionalists), of the commitment of private actors and local authorities in the field of public health policies.
6While this positioning at the boundaries of worlds in tension could have meant that the programme would be stillborn, we argue that in fact it has been the key to its success. The entrepreneur is in fact obliged to neutralize potential conflicts in developing what Padgett and Ansell (1993) call a “robust action:” working to expand and diversify partnerships, individual characteristics, organizational components and initiatives, he manages to make both its identity and action ambiguous. Like the Cosimo di’ Medici studied by these authors, the entrepreneur considered here is characterized by its multiple positioning and multivocality: thus each action or presentation by EPODE can be interpreted differently by different people. EPODE has been the object of very diverse projections,  as the various partners (national and local) have understood the programme (both literally and figuratively) in quite specific ways. Far from trying to “control their behaviour” (Latour 1987: 108), as would the entrepreneur-strategists of actor network theory, anxious to keep their innovation inviolate, the (individual and organizational) promoters of EPODE have instead left these various appropriations to grow. The relative plasticity of the innovation in question derives from the combination of all kinds of entities (norms, rationales, images, symbols, goods, services, activities, etc.) without any synthesis being carried out. Although this combination can be presented as a preliminary step towards synthesis (which however will not necessarily be realized), we intend to show how such a strategy proved favourable to the initiation and development of the programme.
7After explaining more fully the two models which dominate the literature on entrepreneurs and clarified the contours and features of the third model that we offer, we try to explain how it has been developed in the case of the EPODE programme. Thus we devote the second part to describing the multiple faces that can be seen in the people, the organizational components, and the constitutive initiatives of the EPODE programme. Then we make our perspective symmetrical by focusing on how this boundary-entrepreneur is “entrepris,” that is to say, managed and set up by its partners. Based on these analyses, we finally return to the concept of “boundary-entrepreneur” and show how it differs from the two classic models of the intermediary and translator.
An actor at the boundaries of different worlds: figures of the entrepreneur
8From a reading of the seminal work of Simmel on the stranger up to the most recent types of the “boundary spanner” (Crozier and Friedberg 1980) and economic entrepreneur, one can draw the lesson that the entrepreneur (of change) is an actor often on the boundaries of different worlds, whose entrepreneurial job is to “play” with these boundaries.
When the boundary spanner is an entrepreneur
9Reflections on the boundaries of social entities are at the origin of a range of conceptualizations of the entrepreneur. Whether these boundaries separate several countries (Péraldi 2001; Tarrius 2002), several cohesive networks (Padgett and Ansell 1993; Burt 1995), several professions (Abbott 1988; Jamous, Commaille and Pons-Vignon 1969) and ecologies (Abbott 2005), several organizations (Crozier and Friedberg, 1980), the different levels of the same organization or a (public) field of action (Sabatier 1988; Muller 1990; Nay and Smith 2002), one of the central ideas of these theories is that the definition and in turn the transformation of these social spaces are often produced by actors at the margins, where exchanges occur between these spaces and their environments, but also where ultimately their precarious and contingent existence is played out, as the makeshift product of the delimitation of an “inside” and an “outside” (Abbott  2001).
10We can locate the origin of these reflections in Simmel’s “The Stranger.” The latter describes the stranger as someone whose “position in this group is determined, essentially, by the fact that he has not belonged to it from the beginning, that he imports qualities into it, which do not and cannot stem from the group itself.” ( 1950: 402, 2004: 53). Although the origin of this concept refers to space and territory, Simmel does not reduce the figure of the stranger in a geographical sense, for him the stranger is involved in a “a special proportion and reciprocal tension” (ibid. 59, 408) between the “nearness and distance” inherent in any social relationship. This figure of the stranger was taken by Park (1928) and Hughes (1949) as the “marginal man” straddling two disparate social worlds. For Park (1928), he embodies the progress of civilization, because he is the carrier of a meeting of cultures between that of his origin and that of his arrival, he is the person through whom change happens because he is not a prisoner of his ties like the “local man.” Hughes (1949) highlights the status challenges for this “marginal man” whose universe of reference differs profoundly from that of his universe of attachment.
11Several arguments have been put forward to explain the importance of these actors in the transformation of social worlds. One explanation relates to their structural position: because they are on the boundaries of several social worlds, they have the capacity to mobilize resources and networks on both sides of the boundary. This is why, according to Simmel, “the stranger everywhere appears as the trader, or the trader as stranger. As long as the economy is essentially self-sufficient, or products are exchanged within a spatially narrow group, it needs no middle-man: a trader is only required for products that originate outside the group” ( 1950: 402, 2004: 54). One can again understand the term “trader” in the wider sense of “intermediary” because he can be thought of as one who has access to resources and networks beyond the actors in a particular social group. He is an obvious intermediary for economic exchanges but also for negotiations between social groups in conflict, “soft commerce” being the domain of merchants as much as of diplomats.
12A second argument concerns the ethos of these actors. Simmel and Park saw them as representatives of cosmopolitanism and the “urban mind,” a mind that is not rooted in a single country or a single cultural logic, but that is built instead in the meeting of cultures. Devoting their attention to other objects in a different theoretical perspective, Jamous, Commaille and Pons-Vignon (1969) and later Crozier and Friedberg (1980), and many others after them, highlight the lesser embedding of those that they call “boundary spanners” in the organizational and institutional rationales specific to a social space. They are more likely to reflect on (in the sense of “mirror” and “analyze”) the different worlds in which they also act. Such reflexive capacity helps their perception of the limits and constraints of the rules and makes it easier to believe that they can be changed. Reflexivity is thus, in many works, conceived as an essential competence to facilitate change (Giddens 1984; Sewell 1992; Beckert 1999; Fligstein and McAdam 2011), although its genesis and unequal social distribution are not fully explained (see, however, Mutch 2007). The boundary character of social position helps us to precisely specify the reasons for the selective distribution of this competence between actors in a field: lesser involvement is conducive to maintaining a critical distance, an “objectivity” according to Simmel and Park, which allows the actors to better understand the contradictions and flaws inherent in the institutional logic and thus to consider what would otherwise be inaccessible or unthinkable alternatives: “(r)evolution becomes possible once institutions, however fragile or robust, are no longer perceived as inevitable” (Clemens and Cook 1999: 449). 
13A third argument refers to the status(es) of these actors. Hughes (1949) highlights the status contradictions that characterize these hybrid individuals and tends to present them as incompatible. Their integration into a (sub) social space must often be accompanied by the negation of one of these features under penalty of being excluded or stigmatized. In fact, these actors must frequently deal with having a dominant status in the social space from which they come and a dominated status in the social space they seek to enter.  These contradictions and the dissatisfaction that result from them could be analyzed as a powerful force for social change, whether it occurs through individual accommodations or by the creation of social movements.
14As Greenwood and Suddaby have emphasised, “(l)ow embeddedness combined with high ‘interest dissatisfaction’ explains why actors might be motivated to consider change, but it does not explain the circumstances that precipitate them to do so” (2006: 29). To go further, we must look at the work carried out by these boundary spanners. How and under what conditions, have they managed to play on their position to reverse the relationship of domination to which they are subject and to turn into a true entrepreneur? In other words: what is it about the boundary that makes it not just a constraint for these players but also a matchless entrepreneurial resource?
Entrepreneurial work at the boundaries: just an intermediary or a real translator?
15To describe this entrepreneurial work, the distinction made by Latour (1994, 2006) between intermediary and mediator is helpful: “An intermediary, in my vocabulary, is what transports meaning or force without transformation: defining its inputs is enough to define its outputs. … Mediators transform, translate, distort, and modify the meaning or the elements they are supposed to carry” (Latour 2005: 39, 2006: 58).  The entrepreneur may therefore be defined 1) as a “runner,” that is to say, an intermediary who simply passes objects, ideas, practices, etc., from one side of the boundary to another, without changing the shape or nature of what circulates and the spaces between which these objects circulate, or 2) as a “translator” or “combiner,” whose action leads to a joint processing (more or less radical and non-homogeneous) not just of the entities in circulation but also boundaries and territories in which these entities circulate.  These two types of entrepreneurial work differ in their content, form and consequences. 
16Inspired by Simmel, various researchers have attempted to analyze the mechanisms that have enabled immigrants to become trans-boundary entrepreneurs (Portes 1995; Péraldi 2001; Tarrius 2002). According to them, immigrants and migrants are required to use their ability to move between economically and socially heterogeneous worlds to trade goods and generate profit by playing on valuation differences within these worlds. “Goods,” do not only mean physical assets that give rise to a monetary value, but also the cognitive, moral and political resources that make up institutional logics. Thus, the mediators analyzed by Muller (1984) and Jobert and Muller (1987) also use a transfer (mediation) of the logics from a global set of references to a sectorial set of references.  In fact, for these entrepreneurs, the boundary is less a constraint to be abolished than a resource that can be exploited. These actors are thus tempted to develop strategies to become “obligatory passage points” (Callon 1986) between the different social worlds which they interconnect. 
17However, it is common for entrepreneurs to do more than ensure the neutral circulation of goods on both sides of a border. While they are able to make themselves into “obligatory passage points” this circulation will not be free-moving and the boundary is relatively secure. The transfer of items between two social spaces can then be accompanied by a translation process redefining their characteristics within the cultural and institutional context of the social space of arrival. The entrepreneur works on the direction and power of the items he carries from one world to another. He becomes a mediator whose work has an impact on the nature of items in transit but also, even more importantly, on the spaces of circulation. This figure of the entrepreneur as an innovator (or of the innovator as entrepreneur) has been used in many fields of the social sciences. It is found first and foremost in the work of Schumpeter ( 1983). He describes the entrepreneur as a non-conformist, the opposite of a manager, but gives little detail that would help to theorise the basis of his action (Zalio 2009). In order to go beyond the heroic figure that he sets up, we must look closely at entrepreneurial work. However, the sociology of science through the mechanism of translation (Callon 1986), the sociology of organizations through the mechanism of “bricolage,” combining new elements with older elements of the field, which facilitates their acceptance (Rao, Morrill and Zald 2000; Dacin, Goodstein and Scott 2002; Rao, Monin and Durand 2003; Castel and Friedberg 2010), political science through the notions of transcoding (Lascoumes 1996), of mediator (Muller 1984, 1990; Jobert and Muller 1987), of “intermediary actor” (Nay and Smith 2002) and “policy-broker” (Sabatier 1988) and economic sociology with “recombinant” actors (Stark 1996; Stark et al. 2009; Padgett and Powell 2012) have each in turn proposed heuristic frameworks for thinking about this work of mediation. Although they are located in different theoretical frameworks, these authors agree that the strength of entrepreneurs lies in their ability to articulate heterogeneous worlds governed by different or even openly confrontational rationales. According to them, the entrepreneur is the person who is able to upset the boundaries between social worlds by redefining the institutional logics and rationales governing these worlds.
18The implicit assumption of these approaches, which is particularly clear in Callon (1986), is that they present the entrepreneur as an interested actor who manages to use his environment to serve his “cause.” However, as pointed out by Star and Griesemer (1989), the limitation of this model is to consider that the entrepreneur is the only “true actor” confronted with an environment consisting of passive agents that he attempts to enlist and to align on his agenda. All the research that we have referred to has generally paid little attention to how these entrepreneurs were perceived, used and attracted by other actors and how this contributed to the success of their business. By adopting an ecological approach that breaks with the heroic model of the entrepreneur, to locate him in a “struggle for opportunities for action” and relying on the thoughts of Star and Griesemer (1989) and Padgett and Ansell (1993), we will now try to sketch another model of the entrepreneur who seems more likely to correspond to the case described by the EPODE programme.
A “boundary-enterpreneur?” Ambiguities, projections and robust action
19The concept of “boundary object” developed by Star and Griesemer (1989) had an audience that far exceeded the scope of the sociology of science (Trompette and Vinck 2009). It helps to understand how actors driven by heterogeneous and conflicting viewpoints can be coordinated in a collective action while maintaining their own identities and viewpoints.  In the case studied by these authors, the coordinated production of a collection of animals in a natural history research museum was carried out not through coercion, the imposition of representations or fragmented actions, but achieved through the production of objects strong enough not to break during their circulation and malleable enough to be invested in and accepted by all the actors despite their different viewpoints and practices. Drawing on this analysis, and taking the example of EPODE, we argue that entrepreneurs may also manipulate their identity and their “presentation of self” (in Goffman’s sense 1956) as a “boundary object” in order to link heterogeneous social worlds, without attempting to change the rationale controlling each of these worlds, or the identities of the actors who inhabit them. 
20The article by Padgett and Ansell (1993) on the background to the coming to power of the Medici in Florence in the fifteenth century provides another set of arguments that help to clarify the concept of “boundary-entrepreneur.” Using the descriptions by contemporaries of the behaviour of Cosimo de Medici, Padgett and Ansell define him as a character who was “sphinxlike,” a master in the art of obscure and cryptic speaking, and whose words could always be interpreted in multiple ways. The opponents as well as the allies of the Medici projected their strongest hopes and deepest fears on Cosimo, leaving him the opportunity to confirm or deny their choice depending on the circumstances. According to the authors, this behaviour was largely connected to the position occupied by the Medici in the Florentine networks of elites, on the boundaries of (at least) two social worlds that rejected each other and who might at any time start a war whose outcome would be difficult to predict. In this context, multivocality was a way to keep options open as long as possible to ultimately opt in the end for whichever camp was going to win: when war became inevitable, the Medici were able to use their networks among the “new men” and other patrician families in disgrace in order to wrest power from the old families and establish their government.
21This case illustrates how an entrepreneur can play with the ambiguity surrounding his business or his identity to arrange a plurality of options. It also shows that the activity of a entrepreneur cannot be reduced to its economic dimension: the Medici certainly made alliances with other families which were definitely related to economic and financial exchanges, but also to more political exchanges (through marriages and access to the status of citizen of honour), so the “robustness” of their action was as much about the multivocality as the multiple positioning of the Medici. Finally, this case??”like the work of Star and Griesemer??”tends to reverse thinking about the entrepreneur by being interested not only in how he uses other actors, but also on the “projections” of which he is the object.
22We will now try to show the heuristic value of these two theoretical insights in order to explain the success of the EPODE programme.
Portrait of the entrepreneur as Janus: multi-positioning and multivocality at the heart of entrepreneurial strategy
23In this section, we adopt the point of view of those who developed the EPODE programme. Following the example of Stark (Stark et al. 2009), we define the entrepreneur as a collective actor, composed of two central individuals of the programme, Dr. Jean-Marc Horet and Sylvie Distin,  as well as different organizational bodies (a marketing agency, a non-profit law 1901 Association Fleurbaix-Laventie-Ville-Santé [FLVS], a scientific committee and a mayors’ Committee which all together form the national coordination of the programme).  This entrepreneur shares two points in common with Cosimo di’ Medici as described by Padgett and Ansell (1993). The first point concerns the position that these entrepreneurs occupy, at the junction of many isolated and sometimes conflicting worlds. For Padgett and Ansell, multivocality is inseparable from multiple positioning: this is because the Medici are embedded in multiple, loosely connected networks of economic, political and marital relationships, with the result that their discourse may appear ambiguous to the ears of their partners. The promoters of the EPODE programme have also played this multiple-positioning by trying to build bridges between areas of science (at European, national and local levels), public policy and private enterprise. It is indeed difficult for the partners and opponents of EPODE to assign it specific and stable identities, interests and rationales for action.  But while this ambiguity and the unpredictability to which it gives rise (Crozier and Friedberg 1980) are the result of the plurality of meanings that can be given to the rare and mysterious words of Cosimo di’ Medici, they are born here from the very large number of identities that the promoters of EPODE can adopt.
24We will show how the ambiguity characterizing the identity of individuals and of the organization that runs the EPODE programme are the result of a strategy of multiplication of positions and viewpoints within a world where there is little or no communication. We then sketch the model of a entrepreneur akin to Janus, the two-faced God who guards the gates of Heaven and Earth.
Presentation of the programme “Ensemble Prévenons l’Obésité des Enfants”–EPODE (“Together, Let’s Prevent Childhood Obesity”)
In addition to the annual collection of statistics of the weight and height of children in the city, EPODE claims to be leading a large number of local prevention activities. These activities are of great variety, more or less specific, more or less structural, as evidenced by the work identified in the four cities where a monographic study was conducted: organizing the “taste week,” making fruits and vegetables available in school canteens where children can use self-service facilities, nutritional education work in schools, development of baby gym classes in nurseries, organization of cookery workshops in neighborhoods and schools, development of solidarity gardens, organizing sporting events, organizing of accompanied trips so students can walk to school (termed the “walking bus” [pédibus]), reduction or elimination of morning snacks in schools and nurseries, reorganisation of school courses to promote physical activity, once a week “healthy tea-time snack” [goûter] served to schoolchildren, closing a street to increase the walking distance to go to school, etc.
From an organizational point of view, the programme is coordinated nationally by an association called Fleurbaix-Laventie-Ville-Santé (FLVS), with a marketing and communication agency specializing in health (that we refer to as “the Agency” in the rest of this paper) and a Committee of Experts. This coordination is positioned at the interface between the local level??”the city being considered as the relevant territorial level to lead activities designed for children and their families??”and national and public institutions (see Appendix 2, Figure 1). The FLVS association has a contract with the Agency, that commits it to train project managers in the cities, to propose ideas for activities (in the form of campaigns and information packs), to centralize the annual collection of data on the height and weight of children and to produce teaching aids in the form of fact sheets on nutrition.
The implementation of the programme in a city follows the signing of a charter of engagement between the city and the FLVS Association and the payment of an annual fee equivalent to 6,000 euros (for a community made up of municipalities, the contribution ranges from 6,000 to 9,000 euros). Cities agree, firstly, to recruit under their own budget, a “head of EPODE project” or to appoint one of their own employees to devote a variable portion of their working time (usually half-time), and, secondly, to establish a “steering committee” (consisting of at least one doctor, a dietician, a school doctor and a school nurse). In return, the city enjoys the benefits of the Agency, but must find an additional budget to implement local prevention activities.
The programme benefits from the contributions of institutional partners to participate in the financing or simply provide it with official support. The PNNS and the various ministries (Ministry of National Education, Ministry of Health and Sports, Ministry of Food, Agriculture and Fisheries, delegate to the City and the Urban Renewal Department), but also professional associations (French Association of Ambulatory Pediatrics, French Society of Pediatrics, National Academy of Medicine) have been associated with the programme for the implementation of policies and the involvement of local actors (especially teachers and health professionals). Although these institutions have agreed to lend their support to the programme and collaborate in certain activities, they were not willing to finance the activities of the national coordination except in specific cases. Part of the programme is financed by contributions from its member municipalities, but a significant portion of funding received by the national coordination comes from corporate donations and private foundations (about one million euros per year in total), with the majority coming from the food industry: Nestlé France Foundation??”founding member??”and, from 2006, the Carrefour International Foundation, Orangina Schweppes, Ferrero France, the corporate Foundation ISICA (retirement and provident society for the food industry), and Solvay pharmaceuticals. The mayors of the EPODE cities are grouped in an association called the Club of EPODE Mayors, and connected to the national coordination.
The programme initially took the form of an experiment in ten pilot cities before being quickly (even before the end of the experimental phase) extended to any voluntary city (226 member cities in 2010). It has also been developed in a dozen foreign countries and been funded by the European Commission to develop European guidelines for good practice in local policies on obesity prevention.
EPODE was renamed “Let’s live healthily” (Vivons en forme) in 2012 and the association has commissioned a new advertising and marketing agency, founded and directed by Sylvie Distin, (who we present later), after she left the agency. Dr. Horet, also described later, sits on the Board of Directors of the Association. Our paper does not address this change, since our survey was conducted prior to these events. A consultation of the website, however, shows that the organization and the key players seem to be a continuation of what we describe.
Two individuals with many faces
25Two people have played a central role in the design, promotion and setting up of EPODE: Jean-Marc Horet and Sylvie Distin.  Examination of their respective careers helps to understand how the programme has supported the production of the multiple identities that these players have used to adapt their presentation of self to their partners and in trying to legitimize their project and expertise in the field of prevention of obesity.
26Jean-Marc Horet is an endocrinologist-nutritionist. He served as a practitioner in private practice and in the Department of Endocrinology of the University Hospital of Lille from 1991. He often describes this experience as a practicing physician as a foundational one, as he had to confront the limits of medicine alone in its capacity to modify the behaviour of obese children and adults. After visiting a Californian city that had established a community cholesterol screening and prevention programme, he decided to launch a similar programme in the Nord-Pas-de-Calais Region. With the Head of Endocrinology from the University Hospital, he set up a study in 1992 in two villages of the department (Fleurbaix and Laventie). It aimed to measure the impact of a nutrition education approach among schoolchildren on the occurrence of cardiovascular disease in these children and their parents. For this research, called the Fleurbaix-Laventie Ville-Santé study (FLVS), partnerships were established with researchers from INSERM (French national medical research institute). The position occupied by Dr. Horet in the programme had multiple dimensions: he could be seen as a research entrepreneur, involved with his team to maintain relationships with local actors to facilitate the completion of the study, he participated in the design of the study and most importantly he provided the communication of its results to scientific conferences and to the private partners who funded the study. This image of a scientist or at least a sponsor of a scientific study received the seal of consecration in 2009, when an article of which he was a co-author was published in a second-rank international journal of nutrition, which shows a significant decrease in childhood obesity in the two villages involved.
27The work of Dr. Horet was not limited to science. The implementation of the study required skilled political work to effect an alliance with the local elected officials and inform and involve doctors, teachers and the public. He saw this fifteen years of involvement as an implementation of the study which was at least as important as reducing the inhabitants’ obesity, for it shows that citizens can participate in a long-term public health program, once activities can be undertaken that are closer to their daily lives, and with the support of local mayors.
28Dr. Horet also founded the FLVS association, headed by the Chief of Endocrinology in the university hospital service (to ensure the scientific monitoring of the study), and a private company that he created (to provide the financial and human resources needed to conduct the study). Originally intended to receive funding from private partners, the company has gradually developed a nutrition consulting business with pharmaceutical companies. Besides easy access to the “Address book” of the sector, the position of “Managing Director” had another advantage: it created another dimension of the presentation of self that Dr. Horet could make of himself, a facet of his identity likely to attract future private partners to EPODE by reassuring them about his ability to understand the specific problems of the private sector and run a successful business.
29While Dr. Horet most often played the dual role of first herald and then “recruiter” of cities for EPODE programme, it was a second person, Sylvie Distin, who was in charge of the daily management of the programme and relations with local communities, contributing also to promotion of the image of the initiative with these actors. Like the career of Dr. Horet, the occupational trajectory of S. Distin is relatively unusual and allows her to present herself in different ways. After a postgraduate degree in management of cultural institutions, she first worked at the Ministry of Culture, then as director of a dance company. She then joined a business selling tableware where she was involved in advertising its franchise network.  After this period, she created her own marketing and communication business in the field of infant nutrition, and this agency has organized numerous events for the French Committee for Health Education and for major food groups about the education of children to taste and nutrition. These events led her to work with public and private organizations as well as with scientists, allowing her to become familiar with the representations of these different actors. Following the launch of EPODE, she has concentrated on the publication of articles about social marketing, which aims both to give external coherence and scientific credit to the method and to defend her own participation in its development.
30This involvement in situations where they often occupied marginal or even “dominated” positions,  made it possible for both of these entrepreneurs not only to build a large network of varying relationships, but also a large repertoire of presentations of self on which they could play to maximize the opportunities for strategic alliances. When involved in the national media and political arenas, J.-M. Horet could thus highlight his status as a specialist medical practitioner on issues of obesity  or as a scientist who had conducted research on the question, to highlight the public health rationale that drives the EPODE programme. Similarly, when they appear at scientific conferences on public health and social marketing, J.-M. Horet and S. Distin present themselves as researchers unveiling the results of an unprecedented “life-size experiment.” But they do not hesitate to declare themselves to be public policy entrepreneurs to local authorities or as the spokespeople of the communities in national, political or media arenas. Last but not least, they fully capitalized on their images as entrepreneurs, executives of an advertising agency, when it came to enlist large companies in the programme.
A multi-headed organisation
31This multiplication of identities and of positions does not only involve these two individuals, it extends into the formal organization of the programme which includes entities with different legal statuses and objectives.
32At the heart of the programme is a large marketing and health communication agency. This firm, which counts among its clients large agribusinesses, and that designed the PNNS logo, has recruited Jean-Marc Horet and Sylvie Distin as executives. Its employees conceived and developed its content and ensured the implementation of various tasks for the municipalities presented above. However, the Agency is officially only putting into action a programme whose management is provided by the FLVS association. In theory, it is the association that determines the general guidelines of the programme and which is expected to contract with mayors (represented on its Board of Directors) and private partners.  The national coordination of the programme therefore has two highly complementary faces: that of a not-for-profit law 1901 association, and that of a private for-profit company.
33In addition to this institutional tandem, EPODE is “represented” by the EPODE Mayors Club, a political showcase, and by the Committee of Experts, a scientific showcase. The EPODE Mayors Club was established in 2006 on the initiative of the mayor of one of the ten pilot cities and Sylvie Distin, with the aim of this club being, as a 1901 law association, to allow exchanges between cities in the programme, to facilitate the extension of the programme to new cities and to ensure its promotion to national political institutions, through the organization of conferences in the French Senate and National Assembly. For its part the Committee of Experts was an extension of the scientific committee which had provided monitoring of the methodology followed in the former epidemiological study. With the implementation of EPODE, its role is now to ensure that the documents and actions that are developed are in line with the recommendations of PNNS and advances in nutritional research, but it also seeks to legitimize the programme on a scientific level, being composed of researchers recognized and involved in public programmes, first and foremost the PNNS. 
34This multiple positioning of the organization deepens as it doubles up with that of Sylvie Distin and Jean-Marc Horet. While the developers of the programme are both executives of the marketing agency, they are also involved in various ways in other parts of the programme: both as members of the Committee of Experts; Sylvie Distin as Secretary of the EPODE Mayors Club; Dr. Horet as the main spokesman of the association as well as of the Committee of Experts.
35Again, this increase in organizational components that are authorised to speak and act on behalf of EPODE helps to blur the identity of the programme and those who are part of it. As we have shown elsewhere, the presence of a dual structure at the head of the programme functions as a “moral filter” that “symbolically erases, in a … more or less effective manner, the commercial nature of the programme [and allows] the exchange of financial and symbolic resources” between the various participating poles (Bergeron, Castel and Nouguez 2011: 224). By formally separating the member municipalities and private funders, the association reassures the former as to the “public and disinterested” nature of the operation, while allowing (legally and symbolically) the latter to pose as patrons of a public scheme. For its part, the participation of the Agency in the implementation of the programme is used by the developers of the programme as a guarantee of a seriousness and professionalism, and even more so that it can rely on the recommendations of a Committee of Experts. Finally, the Club of Mayors plays a central role in promoting the programme to mayors and to representatives of foreign countries tempted to join EPODE, but also in defending it against the sort of attacks for which the programme can be a target;  the programme can thus be presented as the addition of public and local initiatives for which EPODE was basically the catalyst and facilitator, and the private partners were the co-financiers.
36This strategy of multiplication of identities is also evident in the way the programme’s promoters depict the partnerships with public and private actors. On the website as well as the brochures describing the programme, there is always a banner on which all of the partners’ logos are lined up, without any real distinction of status between them. The logos of ministries, Directorate general for Health and Consumers of the European Commission, the mayors’ associations (Association of Mayors of France, EPODE Mayors Club), associations of doctors and private companies, without specifying the nature or the terms of their participation in the programme. The juxtaposition of these logos generates an image of equity between the partners (Bergeron, Nouguez and Castel 2011) helping to neutralize tensions and disputes that could potentially divide them (the programme unites these different actors around the prevention of childhood obesity) and helping to add, by transitivity, the effects of their authority to legitimize the programme and to make the identity of the programmme ambiguous, or at the very least uncertain.
A programme with multiple pillars
37When the promoters of EPODE were asked in 2008 to specify their methodology to establish European guidelines, as part of the European EPODE network, they highlighted “four fundamental pillars.” 
38The first pillar defines EPODE as the application of social marketing principles to public health. To fight against obesity, particularly childhood obesity, the programme intends to act on the environment of children and families. The programme presentations are often accompanied by a diagram in which the family unit is literally surrounded??”even besieged…??”by actors who may have an influence on behaviour (“schools, health professionals, businesses, child-welfare professionals, voluntary association network, school meals, traders and distributors, local producers, media, other local actors, etc.”) (see Appendix 2, bottom of Figure 1). For the promoters of the programme, if all these actors were to emit the same public health message and thus saturate the environment in which families live, the consumption habits of these families will be influenced by them.
39The second pillar focuses on “political mobilization.” The programme is defined as “network engineering” for coordinating all of the aforementioned local actors. The success of the operation would be based on the political will of the mayor and the ability of the local project manager to overcome the divisions between the various municipal services and local stakeholders to make the prevention of childhood obesity a cross-over project and the source of social cohesion, with the help of the tools of national coordination. From this point of view, the programme’s ability to open up the local political and social space around a common cause is just as important as the reduction of childhood obesity.
40The third pillar is scientific evaluation. Each year a measurement of the weight and size of schoolchildren is organized in all schools of member cities, the aim of these measurements is to provide as exhaustive as possible a statement of the prevalence of obesity in the cities and to measure the results of the initiatives underway. Although members of the National Coordinating group clearly distinguish the EPODE programme they describe as “evaluated intervention” from the scientific study previously conducted by Dr. Horet, this “scientific” aspect is of no less crucial importance in their eyes as well as those of the mayors in the cities: it is a key source of information about and legitimacy of the intervention.
41The fourth pillar is the public-private partnership. Considering that the public and private actors can contribute to the fight against childhood obesity and that a private company can support the design and the outcome of a public health programme, the promoters of EPODE present themselves as entrepreneurs of the market principle in the field of public health. As we shall see, this latter register of action has undoubtedly been the aspect most challenged by the traditional actors of public institutions and non-profit associations. But these different issues did not prevent development of the programme. This rhetoric was used as an argument to convince and attract other actors in the field, such as mayors, who are more used to having to deal with private interests in municipal politics (Le Galès 1995) and have thus found the financial resources that would not be provided by the state or the European Commission, which promotes this type of approach in its White Paper on the prevention of overweight and obesity (A Strategy for Europe on Nutrition, Overweight and Obesity related health issues 2007).
42Although these pillars are most often presented as complementary, we would point out that the weight given by its promoters to each have varied according to the interlocutors they faced. Thus, transformed into modules, these pillars can be selected and grouped into “packages” based on promoters perceptions of the expectations of the partners. This is obviously the case of public-private partnership that, although it is fully endorsed by the promoters of the programme has not always been implemented in foreign countries:
Question: “Do the Australians keep to the same philosophy, the same pillars as EPODE?
Answer: “Exactly. There is just the public-private partnership that is not retained but is transparent enough for now, because for the political launch of the project, the minister did not want to be bothered with the ongoing controversy about this partnership with the opposition. It is a much more sensitive issue over there than here at home, so the minister said: ‘I really agree with this public-private partnership, because I think that for the sustainability of things it is what we need to do, I think that in the medium term the Australian programme will evolve into a public-private foundation, and there it will not be a problem. But now we really want to kick things off so I do not want to bother with that’.”.
44These strategies for the modularization of pillars also apply to the scientific evaluation of the programme: when they had to deal with complaints by the project managers, teachers and nurses of some cities who felt that the annual procedure of weighing and measuring children is too expensive in time and money, the promoters of the proposed programme suggested only taking these measurements every two years. The mayors of the cities involved have finally decided to keep doing them annually, because they believed the data produced by the programme was sufficiently useful to them. Nonetheless in considering the possibility of re-negotiating this part of the work, members of the national coordination appeared to recognize that scientific assessment was ultimately not an essential pillar of the programme.
45This modularization has been accompanied by a significant modulation in the ways of referring to the programme and what it undertakes in terms of the social spaces that concern its promoters. Thus, in a pamphlet of 2006, EPODE presents itself as “a methodology for mobilizing local actors to prevent childhood obesity;” during events organized by French Institute for Health Education and Promotion (INPES), it turns into “action research programme in health education, whilst in the Social Marketing World Congress, it was presented finally as ‘an approach,’ to social marketing and networking to permanently prevent childhood obesity.”
46As with the status of Jean-Marc Horet and Sylvie Distin, or the different organizational components, the pillars of the programme appear to be so many faces that can alternatively or additionally be mobilized according to the publics and partners involved. But to better understand the success of this enterprise, we need to move our gaze from the entrepreneur to its environment. If the EPODE programme was capable of being inserted at the boundaries of heterogeneous and conflicting social worlds, it was also and above all because of the capacity of the programme’s partners to “project” their own interests and rationales of action on the programme and its promoteurs.
The entrepreneur “entrepris:” multiple projections and selective appropriations by the programme partners
47We would now like, as Star and Griesemer (1989) invite us to do, to make our approach symmetrical and look at those who are supposed to be “entrepris” (managed and set-up) by EPODE. As we recall in Appendix 1, we were surprised during the interviews we conducted with local and national, public and private partners, to see how difficult it was for these partners to identify the actors who are “hiding” behind the acronym EPODE, their interests and rationales of action. We were also surprised to find that, in the absence of a precise and comprehensive definition of the programme, these actors were asked to project their own rationales on the programme, choosing from the many facets of the programme that they perceived which correspond best to their interests and/or values. Thus, unlike the actors used by the translator (Callon 1986), the accession of partners to the EPODE programme lies not in its ability to align their interests with his, but, on the contrary, in its ability to lend itself to all projections and serve the interests of its partners.
A nationalized public health programme? National representation faced with EPODE
48From the birth of the EPODE programme, its promoters have sought to forge partnerships or obtain the patronage of multiple departments (Health, Education, Youth and Sport, City) and influential parliamentarians, whether on health issues (Gerard Bapt) or financing (Pierre Méhaignerie). But that recruitment was possible because, at the same time, the prevention of obesity had acquired a prominent place on the public agenda (Bossy 2010). Faced with the lack of financial and human resources but also the political impossibility of conducting a vigorous regulatory policy due to opposition from the food industry to measures that would be too restrictive (Bergeron, Castel and Saguy 2011; Bergeron and Jouzel 2011), parliamentarians as well as corporate actors were looking for initiatives that would enforce a “real” commitment in the fight against obesity. Thus, in 2004, the EPODE programme was discussed and presented to parliament and ministries as an interesting initiative and one worthy of being encouraged and extended.
49However, two dimensions made ownership of the programme by the government or parliament problematic. On the one hand, the public-private partnership, which is a pillar of the EPODE programme, has been the subject of controversy, with some public programme partners holding that agribusiness should not be involved in the resolution of a problem of which they were considered to be the main cause. For example, the right-wing UMP deputy Valérie Boyer refers in a report to “[t]he distrust of INPES vis-à-vis the programme ‘Together, Let’s Prevent Childhood Obesity’ (EPODE) … INPES takes the view that it cannot envisage conducting a prevention campaign in partnership with the food industry” (Boyer 2008: 142). On the other hand, although the Ministries of Health, Sports and the City (and later of Agriculture and Food) were partners in the programme, they were not its leaders. In several ways, EPODE could be seen as a competitor to PNNS and its own programme “PNNS Cities” had the same rationale but seemed harder to develop. In order to “publicise” and “to establish state control over” the EPODE programme, it was necessary to efface the strong participation of private companies and municipalities in the creation, financing and implementation of the programme. The multivocality and ambiguity that characterize EPODE business and entrepreneurs thus gave support to national actors to take selective ownership of the programme by projecting their own representations of its origin and its operations.
50We can trace these selective appropriations through the presentations that were made to the French National Assembly and the Senate by elected representatives and ministers at the launch of the programme in 2004:
Mrs. Poletti (UMP deputy of the Ardennes, question N° 35512 addressed to the Minister of Health, published in the Journal Officiel [JO] of 09/03/2004): “… The health education programme EPODE (Preventing Childhood Obesity), supported by the Ministry of Health and led by a group of doctors will soon be implemented in ten cities with 600,000 inhabitants. So she asks him if it is possible to consider extending this system to the whole of France.”
Response from the Department of Health (published in the JO of 23/05/2005): “… In many regions, experiments are being conducted on the strategy for the management of childhood obesity, bringing together around the child and his family a network of the multiple professional skills of actors in the city and the hospital. The EPODE (Preventing Childhood Obesity programme), being conducted in ten cities, is a good example. Among these cities, six have recently signed up with the Minister of Health and Chairman of the Association of Mayors of France to the charter ‘Active PNNS City’.”
“In recent years, the public authorities have recognized the dangers that are concealed within the rising phenomena of overweight and obesity. But public responses (National Plan for nutrition health, the EPODE initiative, new attention focused on nomadic nutrition in the school setting, current campaign for physical exercise) were too widely dispersed, and very weak in relation to the threat. It is therefore necessary to introduce a national plan for the prevention of and fight against obesity, setting objectives, and informing people widely and especially regularly, generalizing successful experiments such as the EPODE initiative to the entire country, involving local communities, food manufacturers, local health actors and Education, and ensuring that there is regular evaluation of results.”.
52If we are to rely on these statements, the EPODE programme has been presented successively as a “health education programme,” an “experiment” or an “initiative” whose origin and implementation came from a “group of doctors,” the “Ministry of Health” or the public authorities in general, which belong to a large body of “public responses” such as the Charter “Active PNNS City” and that should be extended throughout the country. Thus, projections by parliamentarians and government do not dispel the ambiguity characterizing the entrepreneur behind EPODE but instead use this ambiguity to appropriate the eventual success of the programme, by suggesting that they are or will soon be its main driver. Similarly, local authorities and food manufacturers see their role in the functioning of the programme minimized, since they are not presented as actors but rather as targets to be “recruited” in the context of the programme, which is described as an experiment initiated by the central government rather than as a local initiative in the Minister’s response.
53By deepening the ambiguity about the identity of the entrepreneurs behind EPODE the Ministry of Health left itself the possibility of minimizing its own involvement in the event of an official questioning of the programme, or on the other hand to go further in terms of appropriation by “nationalizing” the programme if it were to prove successful. In 2009, after the programme had published results showing a decrease in the prevalence of childhood obesity over five years in nine pilot cities of the programme, Roselyne Bachelot, then Minister of Health and Sports, decided to strengthen this appropriation, establishing equivalence between the titles “PNNS Cities” and “EPODE Cities” and bringing in a representative from the Ministry of Health to the Board of Directors of the FLVS Association, and representatives of PNNS to the scientific Committee and the new Committee on public-private partnership at the head of the programme.  This decision can be interpreted both as a clarification of the role of the state in the programme and as a reinforcement of the ambiguity typical of the identity of the organization, since the association would from that point be a mix of actors from non-profit associations, representatives of local government, private enterprise partners and representatives of the state.
54Thus, far from reducing the ambiguity that typifies the programme’s identity, the projections of governmental and parliamentary actors have helped to reshape the contours of this identity in a way that is more favourable to them.
Private initative for the common good? Business and European Commission faced with EPODE
55The entry of firms (mainly agribusiness) into the EPODE programme can also be explained by the setting of issues about obesity as part of the public agenda in the early 2000s and by the strong pressure on these companies considered, rightly or wrongly, to be the main cause of it.
“Obesity is a problem that we saw beginning to appear at the start of 2002-03. Previously, we were not, as a business, conscious of this public health problem. We had to take on board these concerns in a fairly rough way, since we had been blamed for them very directly, particularly in the media. It was at the same time that the public also became aware of the rise in obesity in children. They were looking for direct responsibility. This led us, at first, to question our level of responsibility, to understand that this was a very complex issue, and that, rather than finding out who is the culprit, since there certainly is not just one, we should rather all pull together and work together at different levels and find solutions.”.
57The programme performs several functions for the food manufacturers and the health insurance companies involved in financing EPODE. The agribusiness partners of EPODE see it as an initiative promoting other solutions to obesity than a simple stigmatisation of their products alone and as an alternative to the tax or regulatory constraints with which they could be targetted. By having a preferential way of pointing to the role of individuals and local communities in reducing the obesity epidemic, is de facto a way to divert attention away from their own contribution to the genesis of the problem. From this point of view, it is not a trivial point to note that they are also satisfied by the growing public focus on physical activity as a decisive solution to the problems of obesity. As for food education, EPODE has the additional advantage, based on the tools of social marketing, to be promoting the idea of pleasure as a condition for behaviour change rather than insisting on the stigmatization of particular foods.
“We are not against nutritional information. We put it on for those who request it. But I do not think this is the solution to all problems. Rather we need to give some fair information on a healthy lifestyle??”, and a healthy lifestyle is a wide area: it is what we eat, but also how we behave in our everyday lives, and it is also physical activity. This is where we come to with the EPODE philosophy, which from the beginning advocated these values and now has evidence to prove that it is effective.” (Interview with the Head of Public Relations of a food firm partner).
“I think that educational value and individual responsibility can be the sources of solutions to public health problems. I do not believe that regulatory or legal constraints, whatever their form??”the prohibition of sale, increased taxation, prohibition of advertising??”could have an effect on long-term effective management of the acute problems for certain categories of people. … When I first discovered EPODE I saw that the programme contained all of the factors on which we could work. First, an educational programme that addresses the problem early in the relationship between the consumer and the product, since it operates at the level of children. This is a programme that does not isolate the child in the management of consumption of the product. Which educates him on complementary behaviours to the consumption of the product, whether in eating habits or physical behaviour. That promotes the consumption of healthy foods such as fruit and vegetables. And that is a perfect fit with the profile of my products, since most of my products contain fruit.”.
59The programme is also seen as a way of asserting the social responsibility of the company vis-à-vis its customers and its employees. For example, the health insurance mutual isolates and affirms the “front-line players in public health” dimension of EPODE to present this partnership in the eyes of its members as an asset and ex ante investment in prevention and to demonstrate that it is not just concerned with the passive role of repayment of ex post liabilities for medical care. Its directors add, in a new EPODE presentation, that this is an experimental laboratory for their organization in public-private partnerships, an increasingly valued and encouraged form of action in which they expect to see significant development in the years to come. Some public bodies such as the European Commission specifically put forward the public-private partnership and, more generally, the ability of the “private-sector” to contribute effectively to the production of public goods when it comes to valuing EPODE’s initiative in their relevant arenas: European public health authorities carry out the task of adapting the identity of EPODE to the neoliberal economic grammar that dominates the space of European discussions on health issues (Guigner 2008; Bergeron 2010).
60Yet, while the public-private partnership is displayed as a pillar of the EPODE program, its promoters have tried to reduce the area of contact between partners for what they call “ethical” reasons by managing their modes of participation in the adventure. The charter of the public-private partnership that must be signed by any company that wishes to fund the programme sets out a series of prohibitions (“clear rules of engagement”) on “programme content,” “implementation” and “communication about the programme.” Private companies undertake not to interfere in the design of EPODE, but also??”and this is significant??”not to organize activities directly with local communities. The largest number of prohibitions concern communication and publicity: the partner is committed to “submit any proposed communication referring to the EPODE Committee partners, as well as to the EPODE Mayors Club, whatever the mode of communication, except those intended for internal use only by the partner; … not to link the EPODE programme to any product brand promotion; in cases where it would set up corporate communication and publicity, this will only mention the approach; include the partner logo on all communication materials that it would be required to print” (from the Charter of the contract with partners). Although this charter might have appeared to be a tactless form of censorship in the eyes of private partners, given the scale of funding involved, it was on the contrary accepted and respected as one of the strengths of the programme.
Question: “Is the strict separation between private partnership and the policies advocated by EPODE something that suits you?
Answer: Yes. This is essential. EPODE will only be effective and valued because of this committee [public-private partnership created in 2009] which completely separates things and forbids us to take part in educational programmes, in what is done within EPODE. Even though it does not prevent us also having a specific role as a social actor within EPODE. … Apart from the people in the EPODE network, the project managers, there are very few people who know that our company is a partner.
Question: Do you use this in your advertising and publicity?
Answer: It is used only in corporate advertising and publicity. In EPODE’s own communication we are only listed under the heading of Club partners. Even on the EPODE site, you have to click again to go to Club partners to see that behind it there is our business. It must be the 4th click or something like that. It is not at all visible. And we didn’t try to increase our visibility.”.
62For this food company, the programme is valued not so much for its ability to convey a positive image of the company as by its ability to change the understanding of the determinants of obesity and therefore the target of public action. Involvement in the programme is presented both as an affirmation of the social responsibility of the company and as a way to redirect this responsibility onto consumers and governments. In fact, adherence to the principles supported by the programme by private partners is so great that the projection takes the form of an erasure of its private dimension. The Charter of public-private partnership, by clearly separating the two types of partners, thus appears as a means to protect public health from the influence of private interests, but also and mainly as a way for private companies to redefine in the same process the framing of public health problems, the targets and methods for public action on the matter.
The power of the local? Municipalities faced with EPODE
63Selective appropriations, or even misunderstandings regarding the identity of the programme and those who are involved with it are even more pronounced in the case of municipalities that are programme partners. The multiple positioning and multivocality that characterizes the promoters of the EPODE programme are fully reflected in the multiplicity of positions and interests of the mayor members. Thus, the member cities of EPODE differ in their geographical positions,  their demographics,  and the political affiliation of their mayor.  This variety is even more about the reasons why these mayors have joined the programme and how they have deployed it in their own city. Thus, behind the “sacred union” around the “necessary” fight against obesity, can be seen strong and often widely varying differences in political understandings of the problem and solutions that should be applied to it.
64Some mayors see EPODE as a way to introduce and advertise a municipal public health policy, over and above the single case of obesity and overweight. It is a resource that can be used to “legitimately” call for conducting a local public health policy that develops without, or even against, the programmes developed by the central government:
“[EPODE] is a demonstration of the political gap today about the choices to be made between a national policy dictated by the powers that be, whether right or left, and the reality at the groundroots, which controls the application or the organization of the policies that are dear to it. Ultimately, I think it will be a conflict between the fact that to make savings in national management, we pool, we concentrate, we remove, while the French need closeness, a varied life, to recognize themselves, to live in the area they have chosen, and not in this concentration camp world, for which there is no justification??”I am not criticizing, but it is a fundamental societal choice. EPODE is a typical example of this. … Finally, for programmes like this??”and I’m sure this is true in other areas of everyday life??”the right person to gear this down and the best inductor is the mayor in his city.”.
66For this rationale, the EPODE programme is designed as a paradigmatic example of a new form of local public policy, the fight against obesity becoming a means more readily than an end. Elected officials see the proposed methodology as an instrument for networking and coordination of the various actors in the city and to improve social cohesion, in the very general sense of the term.
“[EPODE] is one tool among many, and proposes an interesting partnership with the Ministry of Education, where it is not always easy to be heard. EPODE will help us measure pupils’ weight and size at schools in two poor neighbourhoods, every two years from kindergarten (age 3) to CM2 (age 10-11). We will also support this programme to strengthen the involvement of associations and sports clubs and consolidate partnerships, for example with the Chamber of Agriculture to build channels promoting a healthy and balanced diet.”.
68In contrast, a number of politicians have made EPODE a springboard to build or strengthen their parliamentary or ministerial career. For these mayor-deputies, or senator-deputies, the EPODE programme is a resource used to legitimize their participation in parliamentary committees on health issues. Ten mayors of EPODE cities are members of such committees. This is the case of Brigitte Bout, mayor of one of two cities that participated in the original FLVS study, who has since become a Senator and relied on this experience to claim the contested status of expert in the Senate on public health policy. The EPODE programme is thus often presented as a variation of the right to experiment local initiative and as a means to enlighten and enrich the policies of the state.
“I am also convinced that decentralization and experimentation are more effective than centralized systems. At first, I saw that the Ministry of Health was not very keen on the idea. This is always the problem??”what we do not control, we do not like! At the Ministry of Health and to the Director, I said it was a very good programme that appeals to responsibility. And when Valerie Boyer, not knowing the system, was a bit harsh, perhaps because the National Health and Nutrition Programme seemed more attractive on paper, I said, ‘you’re wrong, we must enhance the EPODE experience, which is an experiment in which you engage the responsibility of elected officials, families, personal responsibility.”.
70Another way to understand the diversity of projections/appropriations of which EPODE may be the object is to observe the location of the programme within the municipal system. From one municipality to another, the management of EPODE can be located in the department in charge of academic and extracurricular life (including school meals), the departments in charge of social action (including communal Social Action Centres), the municipal health services in charge of youth and sports (including leisure centres), city policy, departments of urban development and the environment, or even directly within the office of mayor. As for the project managers, they can be a trained dietician, nurse, social worker, educator or a planner… Which is not to say that this location isolates and permanently fixes how EPODE is seen locally, but one can interpret the variability of these institutional locations as a further sign of the cognitive plasticity of the project: it can indeed be “framed”??”as defined by studies of framing problems (Snow et al. 1986; Benford and Snow 2000)??”as a social project, health, nutrition education and/or sport, entertainment for children and teens, city planning or for building social cohesion, etc.
71For many of the local actors involved in it, the identity of the programme is not defined as much by the actors who conceived and organised its national coordination as by the actors who are responsible for it locally, foremost among them the mayor and the project manager. In fact, members of the National Coordination are hardly ever involved in the locality. Their role and identity as well as the organization of this coordination are only known, most often, to the municipal teams (even though they are never silent or hidden), the project managers and local officials. Even when considering the national coordination of the programme, many local actors we interviewed tend to define it as an offshoot of government, whether it be PNNS or the Ministry of Health, or as a project that is hard to define. 
“I thought it was mostly actors involved in health at the medical school level, etc. … The Ministry of Health, a public initiative, I did not say because it seemed obvious to me.”
Question: “How do you see this business?
Answer: This is just a project, with experts behind it: doctors, nurses, dietitians, more people in contact with children and families. For me it is not a business, it is a shared project. Led by whom? We have a doctor who manages almost everything, but there are several people to help him.”
“I do not know who is behind EPODE I think it is a combination of several municipalities, which are included in EPODE.”
These actors have often been surprised to discover that this initiative had a private origin, when we asked them.
“Well, we did not feel like that because there was already the town hall, it is the town hall and their people who come and are involved, dieticians, and Social Security, so it’s true, I’ve never asked the question, it never seemed abnormal [laughter] frankly, we are usually all a little reluctant, but then I did not really feel… I did not even know that it was private [peals of laughter].”
76It is true that the ostentatious signs of the fact that the programme belongs in the public domain are legion (whether in the form of logos displayed on flyers and posters in many public institutions or the number of public-sector actors directly involved in the implementation of the programme). If they have long since learned to deal with private interests in municipal policies (Le Galès 1995) and welcome the additional financial resources to be found in these partnerships, local politicians are reluctant to expand on the private origin of the programme in order to avoid the risk of the initiative suffering discredit amongst local actors.
“Yes, many people do not know that there is private financing, yes I know. But it is not us who will tell them. You know, in France private financing is frowned upon. But without private financing, it would be the death of EPODE.”
78The success of EPODE can finally be explained as much by the ability of its promoters to show many faces as by the projections and selective appropriations to which it has been subjected.
Returning to the boundary-entrepreneur: an obligatory passage point… without translation
79To conclude this point, we would like to return in the light of this demonstration, to the model of the boundary-entrepreneur that we have developed and what distinguishes it from the “translator” conceptualized by Latour (1984, 2006) and Callon (1986). Like the entrepreneur-strategist in Actor-Network Theory (ANT), the EPODE entrepreneur is indeed an obligatory passage point between the different actors. While they proclaimed the need to break down barriers between different levels of public action, and the areas of public policy and private enterprise, the promoters of the programme were in fact seeking to minimize the contact surfaces between partners by the management (of the terms) of their participation in the adventure: the Charter of public-private partnership aims to minimize the interaction between the two types of actors, although their logos rub shoulders on the official documents of the programme. The entrepreneur is also interposed between the local authorities and national political institutions. Regarding the implementation of the programme, the national coordination is the main contact between elected officials and project managers, and brochures provided and actions implemented by EPODE substitute in practice for those developed by the PNNS. This separation work also relates to the “political representation” of cities: although some EPODE mayors can make their voices heard at the national level directly through their office of parliamantary deputy or senator, it is actually the EPODE Club of Mayors that provides this representation function. The organization of conferences by the club in the Senate or National Assembly focuses and directs the political representation of the programme more surely than the not necessarily coordinated response of the 220 Mayors of the cities involved. Finally, the national coordination is set at the intersection of inter-city relations, since it provides and controls the movement of the merits and lessons that can be learned from various local initiatives: the national coordination is the sole organizer and controller of the meetings between local actors, especially during the biannual training for project managers and the annual conferences of EPODE cities. Thus, the rhetoric of breaking down barriers is actually accompanied by the formation of a network arranged in a star-shape, with national coordination at its centre: it asserts itself ultimately as the single point of junction of the different actors and almost exclusive control of the flow of information and resources between the reticular segments thus connected (see Appendix 2). Like the entrepreneurs described by Callon (1986) and Stark et al. (2009), EPODE is located at the border of worlds governed by conflicting values.
80However, the big difference with the entrepreneur-strategist of Actor-Network Theory is that our boundary-entrepreneur is hardly concerned that the recruited actors, “if they do take up [the statement], … transform it beyond recognition” (Latour 1987: 108). Whilst EPODE has managed to become an obligatory point of passage between different social sub-areas and different levels of action (municipal, national, European), it has not however worked to translate: our boundary-entrepreneur (the individuals who support it, the organization in which they live and the programme they promote) has not claimed to be a “translator,” whose activities would have led to a more or less radical transformation of boundaries and territories between which it circulates many entities. It has not tried to overturn the interconnected worlds and break the boundaries that isolate them, to reinvent and reorganize the whole thus deconstructed into a coherent, ordered whole according to a new scenario, and of which it would become the centre. On the contrary, by stacking up, without inventory and classification, identities (individual and organizational) and actions, so that everyone projects and derives from them what suits them, and by keeping separate the spaces that it connects, EPODE has fundamentally undermined the integrity of boundaries, the spaces they produce and institutional rules that organize them. It also can not be considered a mere “runner” in that as an obligatory passage point between worlds that are potentially conflicting, it permanently transforms some entities that it selectively circulates to adapt them to the institutions that organize the spaces thus interconnected, as much as it lets actors pick and choose the features that suit them.  The identity and nature of the boundary-entrepreneur, as well as those of the entities in circulation are, in the case studied here, constantly being re-worked, by EPODE and by its partners. 
81* * *
82As noted by Lamont and Molnár (2002), the success of the concept of boundary in social science, especially in studies of science and professions, can be explained by the fact that the relationship is a key dimension of social life: the concept has been extensively used to show how and to what extent scientists and professionals are able to demarcate themselves or to keep out other actors (defined respectively as “non”-scientists or “non”-professionals)  or to conquer or defend their jurisdictions in the sense used by Abbott (1988). The notion of boundary was less often used to study situations where, on the contrary, actors from different social worlds were able to cooperate, exchange or coordinate. In addition to the now classic concept of boundary object (Star and Griesemer 1989), more recent studies have revisited the model or figure of the economic entrepreneur (Stark 1996; Stark et al. 2009; Zalio 2009) using this notion of border. It seemed that this literature could be reread with profit, and in a somewhat unusual way for analysing entrepreneurs of public policy or, more generally, the so-called “change entrepreneurs” or “institutional entrepreneurs.”
83One could probably identify some actions and achievements showing that the promoters and organization of the programme have some traits of the “runner” or “re-combiner” entrepreneur more traditionally identified in the literature, it nevertheless seemed to us that the model of the “boundary-entrepreneur” allowed us to better describe and analyze what makes EPODE and what it does, and to understand how it came to exist, grow and spread in fields that could be much more resistant, and more aggressive to it. We also eschewed the term “blurring of boundaries,” often used and rarely specified (to the point that it is in itself blurred) when it comes to analyze the transformations and reconstructions of institutional rationales, particularly the reconstructions of state action. In the case of EPODE the concept of boundary-entrepreneur made it possible to better specify what playing with/around the borders means: multivocality and multiple-positioning which generate ambiguity, promote the expression of various projections favourable to the development of the programme. If synthesis can be a prospect for any innovation process, EPODE appears more readily as a motley stack of multiple theories, partners, initiatives and organizational components. Far from being a hindrance to its spread, this collection of all sorts of entities was more of a resource because it allowed its promoters to arrange a set of opportunities for alliances and to thrive in an environment segmented by many boundaries and divided-up by major conflicts. This multivocality also allowed programme partners to project their expectations on it, to choose their commitments and hence through modularity, to have hardly any ties to the components/entities of the programme that could jeopardize their investment if they had been non-negotiable at the outset. EPODE therefore enlists and coordinates without such an operation of association significantly transforming the identities and interests of the entities thus “entrepris.” Paradoxically, the strength of the programme has relied less on self-affirmation of an approach and a radically innovative identity as on its heteronomous openness to various confusing and sometimes contradictory movements of a composite environment that had to be married together rather than transformed. This actor has hardly disturbed the spaces in which it has intervened, or broken routines and subverted stable power relationships and established hierarchies. EPODE (that is to say the actors who are involved in it and the unique organization in which they live) is not least a boundary-entrepreneur, it is located at the border, in that it is itself a boundary object and it claims to be a border guard involved in the reproduction of many separations.
84We would like to conclude this paper by risking some hypotheses about the possible contribution of this study to the analysis of innovations. Outside of the stacking, the singular property of the innovation analyzed here, most of the authors (from Lascoumes to Stark) insist that innovation is the product of a process of (re) combination of old and new elements. We are tempted to think that innovation is perceived at the outset by those who are supposed to receive it, as in this essentially composite state: they receive (or can receive) its existing elements, recognize (or can recognize) a family resemblance which brings it closer to other entities and distinguish (or can distinguish) what is novel in the combination, they may also experience the discomfort, embarrassment or misunderstanding that can produce such a combination of new and old entities. A learning period is needed for the genesis and learning of dispositions to understand this composition of heterogeneous and undefined elements as a relatively coherent whole and to taste all of its flavours. Strictly speaking, this combination only becomes an innovation when it is recognized as such by those to whom it is addressed, that is to say, when those to whom it is exposed acquire cognitive patterns that make them understand it as such; it only becomes innovation a posteriori, when the synthesis is carried out subjectively, that is to say when it is no longer possible to distinguish the old elements from the new elements. We risk, under these conditions, the assumption that there is a period between creation and institutionalization,  during which the identity of the entity that will later become an innovation (if this is the case) is fundamentally ambiguous:  an ambiguity on which the innovators play to help it be gradually accepted and on which those who receive it rely, either to tame it or instead to try to change it.
Appendix 1. Study methodology
85In many ways, the results presented in this article reflect the serendipity theorized by Merton (Merton and Barber 2004). The origin of this research was an initial survey in 2009 by students of the Masters in “Sociology of organized action” at Sciences Po Paris, under the guidance of three supervisors (Hugo Bertillot, Fabien Foureault and Elsa Gisquet) and two scientific officers (Henri Bergeron and Patrick Castel) of the Centre de Sociologie des Organisations??”Centre for the Sociology of Organizations. This study, conducted at the request of officials of the EPODE programme and funded by the programme, aimed to analyze the characteristics of its implementation in four pilot cities. In each city, elected and administrative municipalities, doctors and nurses, civil society actors, actors of Education and representatives of the external services of the State, such as the Regional Directorates of Health and Social Affairs (Directions Régionales des Affaires Sanitaires et sociales –DRASS) or regional public health groups (Groupes Régionaux de Santé –GRSP) were interviewed. At the national level, we met with officials from the French Health Ministry and INPES. The company executives responsible for the programme were also interviewed, as well as some scientists with whom they work, and industry representatives who help to finance the programme.
86During these interviews, we were impressed to see the wide variety of definitions and characteristics given to the programme. The actors interviewed seemed to have their own idea of the nature of the programme and were often unable to give us the complete picture. For example, we seem to learn from local and international actors that the programme was carried out by a private company and was funded by food companies. If a fragmented view is understandable on the part of programme partners, it was surprising that the promoters of the programme themselves were persuaded to change their story depending on their interlocutors. The programme seemed like a kaleidoscope of which no actor appeared to have an overview and for which it was difficult to outline and identify the heart.
87In order to clarify what we then saw as gaps in our survey method, we initiated a second study conducted in the framework of the programme “Governing the behaviour of consumers. The case of sustainable consumption and the fight against obesity,” funded by the French National Research Agency and directed by Sophie Dubuisson-Quellier. The purpose of this second study, for which scientific responsibility was shared by the three authors of the article, was to trace the genesis and organization of the national programme, including questioning its inclusion in the French policy of obesity prevention. We conducted further interviews with the promoters of the programme (at the agency and the association), we consulted the archives of the association and we obtained various documents from the programme in order to better understand its genesis, but also changes in time and space, of the “presentations of self” developed by its promoters. We have attempted to view these presentations using the perceptions of them by the partners in the programme, whether the private companies that finance them, participating mayors, deputies, senators and representatives of the Ministries of Health and Agriculture which have been asked to decide on a possible “branding” programme. We conducted interviews with several representatives of these organizations and explored the online archive of the National Assembly and the Senate to see how these politicians had presented the programme. We also took part in the fifth EPODE cities conference held in May 2010, which allowed us to understand the views developed by these actors about the programme at a particular point in time. This second study has allowed us to support our hypothesis that EPODE has fed largely on the ambiguity surrounding the identity of its promoters and the nature of its actions.
Appendix 2. Organisation of the EPODE programme
However, unlike the “heroic” model of the entrepreneur developed by Schumpeter ( 1983), sociologists do not reduce the entrepreneur to an individual overcoming adversity alone, but instead try to show that the entrepreneur is embedded in a group, when it is not itself a collective actor. It is this second approach to the entrepreneur that we develop here.
Such an assumption does not necessarily imply adherence to the “hard version” of the theory of the rational actor. As we will see later, we can assume an interested and strategic actor, but whose rationality is limited, and whose interests and strategies are largely defined by the context in which it operates.
We adopt a broad definition of territory as a social space separated by a material or symbolic border (for the distinction between these two types of borders, see Lamont and Molnár 2002) which prevents or restricts exchanges between individuals located on either side of the border.
The study, conducted using qualitative material (see Appendix 1) collected during two phases of investigation??”one in 2008-09 and another in 2010??”focuses on the genesis and development of the programme between 1992 and 2010. Since 2010, the programme has changed its name and has undergone several developments still underway and not fully completed, that we do not discuss in this article.
Success measured by the diffusion of the programme in France, but also by the recognition and adoption of the “EPODE model” in many other countries and institutions (Australia, Belgium, Spain, Greece, United States [with a visit by Michele Obama], European Commission, etc.). This diffusion does not mean, however, legitimation, as the two processes are not necessarily moving together (Colyvas and Jonsson 2011): EPODE must continue, in many circumstances, to justify its existence in an environment that is largely hostile to it or neutralize the normative tensions it creates (Bergeron, Castel and Nouguez 2011).
From a legal point of view, the only powers that are granted to them are those involving certain public and animal measures, administrative and technical control of hygiene, safety and environmental protection and, optionally, social assistance.
By “projection,” we mean the process by which the actors involved in EPODE or its partners are required to assign one (or more) identity(ies), one (or more) interest(s), one (or more) skill(s) and one (or more) model(s) of action to the programme’s promoters to meet their expectations, so as to model their own action on these projections. In this sense, the notion of projection that we hold is much closer to the notion of typification developed by Schutz ( 1987).
The recent book by Alter (2012) on “atypical bosses” (disabled, homosexuals, women, foreigners or French of foreign origin, self-taught, etc.) is directly along these lines. In particular, it stresses how the bosses were able to cultivate their position “between two worlds” to develop and enhance social skills (objectivity, critical distance, curiosity, sense of adventure, etc.) which the “normal” would not possess.
These “boundary spanners” are often given the “dirty work” in the moral division of labour (Hughes 1962). For the analysis of such a process in the case of cannabis dealing, see Nouguez (2003).
The terms “intermediary” and “mediator” have often been used as equivalents by political scientists (Hassenteufel 2008). The analytical distinction between the two terms made by Latour seems to highlight the profound differences between these two types of activity.
The distinction between the two types is, of course, sometimes tenuous in practice. For example, it is difficult to determine if the antique dealers discussed by Sciardet (2003) are more relevant to the first or second category: in fact, if the new value of the property is mainly due to moving into a new context (which would rather put them into the category of “runner” entrepreneurs), the work of development of the object by its exposure to others can be considered, to some extent, as a work of recombination.
The runner is very close to the “trader” that Austrian economists Mises ( 1974) and Kirzner (1997) describe as an actor able to use discontinuities in the flow of information to generate profits. In contrast, the figure of the translator would be more akin to the entrepreneur-innovator of Schumpeter ( 1983), who generates profit through his ability to transform the world. For an overview of analyses of the various authors, see Zalio (2009).
Again, the distinction is difficult to establish: many analysts see the work of mediators as close to that of “translators,” while others are more likely to see them as “facilitators” who circulate more than they transform or innovate. In any case a general conclusion is rather questionable and it deserves to be empirically determined.
For empirical examples, see especially Padgett and Ansell (1993), Burt (1995) and Massardier (1996).
Jasanoff (1987) has argued for a perspective which, on some specific aspects, is similar to “boundary objects.” It showed that the concepts of “science policy,” “risk assessment” and “peer review” are vague enough to allow for coordination between actors whose interests are not necessarily convergent, while having sufficient social strength to circulate in different worlds. Carpenter and Moore (2007) have the same approach to the concepts of efficiency and safety of medications continuously used by various actors of the FDA. See also Epstein (2007) and Jabko ( 2009) for similar arguments.
Star and Griesemer (1989) find it difficult to extend the concept of boundary object to people, because of the limited repertoire of roles they are likely to take on and the risk of the explosion of their personality presupposed by circulation within too many and too diverse a set of normative worlds. We advocate instead the idea that entrepreneurs, like boundary objects can develop the ability to stand on the border of several worlds and maintain multiple identities, or at minimum of presentations of self, which can marry the rationales of different worlds. This multivocality is probably uncomfortable, it requires consummate role-play, but this does not mean it is unsustainable, especially if it can rely on an organizational support.
Real names have been anonymised.
As Padgett and Ansell (1993) show very well, Cosimo di’ Medici is only the tip of the collective enterprise that was the Medici family and its network of business alliances. The sociologist who wants to escape from the heroic and individualistic reading of the entrepreneur must not only return him to the organizational and social context which provides him with social support (Zalio 2009), but also go further, and think of the entrepreneur conceptually as a collective actor (Stark et al. 2009).
Similarly, we had to abandon identifying and prioritizing the main drivers of the action of those two key individuals (economic interest, desire to influence public policy, axiological beliefs, etc.). Entrepreneurs do indeed adopt the same multivocality vis-à-vis sociologists as they do vis-à-vis other people they meet. Therefore, rather than trying, in a haphazard way, to isolate the “good” reasons among those which had been raised at various times by the promoters of the programme, we found an additional element in favour of our thesis, namely that the strength of our entrepreneurs is precisely their ability to resist any attempt to reduce their motivations to a single rationale for action.
Other actors have played a role in the implementation and promotion of the programme, but J.-M. Horet and S. Distin were the main spokespersons of the programme as much vis-à-vis local communities participating in the programme as vis-à-vis external partners, public and private. In return, the programme has played an important role in their own careers.
For her, EPODE is in fact the application of the franchise system to public policy.
This is the case particularly for Dr. Horet who referred in an interview to the contempt to which he has been subjected by INSERM scientists and University Hospital physicians, because of the “impurity” of the research he was attempting to set up and because he did not belong to the “club.”
Dr. Horet has also appeared on television shows such as Les Maternelles (The Kindergartens) and Allô Docteur (Hello Doctor) where he was described as a “doctor specialising in obesity.” Although he was not invited to speak about the programme but to respond, as a doctor, to questions about obesity (children as well as adults), these programmes were listed on the EPODE website under the title: “Talking about EPODE.”
Until 2005, the partnership contract agreements were signed jointly by the Agency, the association and the private partner and municipality. Since 2005, as the association had changed its the status to one of “association of general interest,” donations by private partners are eligible for tax deduction and from then on contracts are now signed only by the private partner and association, to which the Agency then bills its services.
The committee played an active role in the design of the first EPODE campaigns but its meetings have become less frequent over time.
This was particularly the case when the Boyer report was published (see below).
Although the programme’s promoters did not make their action one of its pillars, it is obvious that they were also involved in setting the public agenda (local if not national) on the question of obesity.
This decision of the Minister of Health is not only based on the right “results” of the programme, but also in a context of strong competition between the Ministry of Health and Ministry of Agriculture (renamed “Ministry of Food”) around control over the policy towards the battle against obesity.
While the vast majority of member cities are located in the Nord-Pas-de-Calais and Languedoc-Roussillon regions, the ten pilot cities belonged to ten different regions and there were in 2010 EPODE towns in fourteen regions.
The ten pilot cities in 2004 had populations of between 8,000 and 90,000 inhabitants. This was in part a wish of the programme’s promoters who intended to conduct the experiment in a variety of contexts.
While seven of the ten pilot cities of the programme were led by right-wing UMP Mayors in 2004, there were twenty Mayors who are members of the UMP five years later, five from various right-wing parties and one member of the centrist MODEM against fourteen left-wing Socialists, five Communists, an Ecologist and two from other left-wing parties (the other Mayors of EPODE cities being independents).
We asked a journalist making a documentary for France 2, who came to interview us about EPODE in mid-April 2012, this question: “Why are you doing a report on EPODE? What is your angle?” And the reporter’s answer was significant: “But what is EPODE? It is to understand what EPODE is… And what it does…”.
It is not really the operator of a simple transfer of PNNS recommendations to municipalities or private actors, for example, nor is it the neutral carrier of a market rationale to public actors. It more obviously creates a platform 1) filtering and adapting entities that are suitable to be circulated, and 2) facilitating participation by selective projections and appropriations.
As was strongly suggested by the journal’s reviewers, the empirical case analyzed here describes a particular figure or model of symmetrical work used by the programme’s partners: EPODE is designed such that it allows projections and selective appropriations, although we might however consider whether there might be driving forces that fundamentally alter the course of the programme or change the nature of the policies pursued. From this point of view, the empirical case is an excuse for these theoretical developments which should be seen as presenting a weakly “embarked upon” model of the entrepreneur.
See in particular the work of Gieryn (1983) and Jasanoff (1987).
This does not of course mean that the creation necessarily leads to institutionalization (Strang and Soule 1998); without also assuming that dissemination equals institutionalization (Colyvas and Jonsson 2011).
It goes without saying that the degree of ambiguity related to the composite nature of the entities, and conversely the degree of integration, varies depending on the case: we willingly accept that there is no equivalent for EPODE and for Pasteur’s theory (Latour 1988), for example.