On the front lines in food policy: Assessing the role of neighbourhoods for food systems transformation in the Montreal food polity

This paper reports a multi-year design-based implementation research (DBIR) that examines practical issues, challenges, and innovations faced by the Montreal food polity in transforming food systems for alleviating food insecurity in vulnerable populations. Community organizations in three geographically distinct neighbourhoods were engaged in three distinct city-level collaborative engagement initiatives (coalition of neighbourhood roundtables; place-based philanthropy initiativeCIP; food system policy council-C-SAM). The latter city-level initiatives stemmed from different historical and institutional contexts and afforded different types and amounts of capabilities in support of community organizations. Our results underscore the rich diversity not only in how local communities organize themselves over time but also how they welcome or not scaling-up or capacity building initiatives like CIP and C-SAM. As part of the same complex and dynamic adaptive system observed at a given stage of its evolution, individual organizations and collaborative platforms observed in this research all had their respective historical trajectories and future aspirations in terms of composition, capabilities, goals, achievement and challenges. Contributions to food systems research concepts are three-fold: Isomorphism, Discursive Frame, and Decoupling between Norms and Action. Our research demonstrates that neighbourhoods, like nation-states, exhibit different pathways to adoption, adaptation, and decoupling action from norms when cities become part of an international regime. The outcome of cities signing on to new international agreements are similarly symbolic in nature.


Introduction
A surfeit of research on food systems demonstrates the central role that local organizations play in the formation of, innovation in, and responses to complex problems of engagement in food systems transformation (Handforth et al., 2013;Bazerghi et al., 2016;Enns et al., 2020). Yet, despite significant mobilization and action across a diverse set of contexts, the ability to scale up and scale out successful initiatives has been more limited. This appears to be linked to the weak ties between local community organizations and with commercial actors also operating in food systems at municipal, provincial, national and global levels (Levkoe, 2015). Governance and policy studies suggest that these disconnects and limitations are not unusual, as food constitutes one of the 'wicked problems' of multi-scale, ambiguous, and seemingly intractable policy change (Hammond and Dube, 2012).
At the same time, there is ample evidence to suggest that local initiatives can potentially and indeed significantly chart a path for transformation (Addy and Dube, 2018). For instance, neighbourhood food networks (NFNs) that reach out to other neighbourhoods with similar challenges provide important sources of mutual support, resource mobilization, and serve as building blocks for wider transformation (Blay-Palmer et al., 2016). The success of local food movements can also be measured in the growth of formal organizations, proliferation of planning and policy processes at various levels of government, and in the establishment of food policy councils (Blay-Palmer, 2009).
In this paper, we argue that these advances, while significant, have also themselves differentially impacted community outcomes by channeling food movements into new norms and policy structures. Thus, while the increasing structuration and formalization of the food policy sector has expanded the normative influence of community organizations and food initiatives, it has also reshaped neighbourhoods and their ability to translate new ideas and innovations into meaningful, long-term systemic change. This is impacted first by the normative frameworks and discourses of existing international agreements focused on food security. Thus, despite very real transformations in intent, priorities, and perspectives, these discourses shape cities' adoption of international policies on food systems.
Yet, as we also demonstrate, decoupling between commitments and implementation in policy occurs not only within national policy, but also at the municipal level with non-governmental organizations and local social movements that face barriers in framing their claims in globally-defined terms. In fact, regardless of a state's opportunities in implementation and openness to innovation, success or failure can be better explained by the local rather than the national context. This is, after all, where the connection between ideas and action hits the proverbial road. Policies may be adopted or resisted by community organizations who are both implicated and often responsible for their implementation.
To assess the role of neighbourhoods in this shifting policy landscape, we draw on research in sociological institutionalism, which demonstrates the role of both normative transformation and institutionalization in non-state and civil society contexts. We employ the discursive and urban turn in sociological institutionalism to address the formalization and institutionalization of local food initiatives and movements into municipal food polities which include, but are not limited to, the expansion and adoption of municipal food policy networks. In doing so, we highlight both the success of global, normative frames on the discourse of food policy (from, for example, the Milan Urban Food Pact) and an expanding set of expectations in municipal food policy councils while, at the same time, demonstrate the differential adoption and implementation of meaningful outcomes by location.
In Montreal, the field setting for this study, the global context and emerging cultural norms on sustainable food systems and food movements contributed to the structuration and formalization of municipal and civil society food policy beginning in 2017. While these changes reinforced and extended the legitimacy of neighbourhood coalitions to formulate and shape local food policy, they also redirected and channelled the priorities and relationships between organizations in many neighbourhoods. In some cases, neighbourhoods attempted to resist and mitigate these new structures while at the same time expressing commitment to new frames and norms in food policy.
In this paper, we assess the differing responses and outcomes of local food security tables in Montreal following these changes to assess the impact of the increasing formalization of food policy on differential outcomes in local communities. We begin by reviewing the literature on place-based food initiatives, neighbourhoods, and polity studies. We then provide an overview of the Montreal food policy and food movements' structure, focusing on three features of the municipal institutional framework; neighbourhood food security roundtables, the development and contested process of the Conseil système alimentaire montréalais (C-SAM), or Montreal food policy council (FPC), and the launch of the Montreal Collective Impact Project (CIP), a place-based strategic philanthropic initiative. Together, these constitute an overlapping yet increasingly structured and formalized set of relations that form a municipal food polity. Our methodological framework, drawn from design-based implementation research (DBIR), has helped us map these relations and neighbourhood responses. Through research co-creation with three food security tables, we present findings from field notes at community meetings over the course of one year and forty-four interviews with community organizations.
Our results show that, while local communities found opportunities within these new arrangements, it sometimes came at the cost of neighbourhood-driven innovation. Indeed, two neighbourhoods reorganized the governance of their tables to align with the expectations of funders and municipal policy structures as they became 'socialized' into norms set by the CIP. We conclude with directions for understanding how decoupling between new norms and implementation occurs in municipal polity contexts and suggests pathways for research in neighbourhood-driven food system innovation and transformation.
Literature review (Re)shaping outcomes: Food movements in the world polity In the world-polity literature, researchers ask a central, simple question in relation to international and intergovernmental governance and state authority: why is it that the state apparatuses of so many nation-states, with such disparate economies, histories, and polities, look so structurally similar? In over two decades of analysis and research, a partial answer consistently highlights the expanding role of intergovernmental agreements and the growth of international non-governmental organizations which, post-World War II, increasingly shaped and structured a stateless, global civil society organized around liberal, Western, universal values as norms of engagement (Meyer et al., 1997;Boli and Thomas, 1997). In this, new norms are proposed, adopted, reinforced, and institutionalized through movement campaigns on issues such as human rights, then are reflected in the establishment of international agreements and statelevel bureaucracies. In research on global norms and food, for example, scholars have documented the expansion of animal rights, global campaigns against controversial food consumption practices (Lien, 2004), and the assertion of cultural rights to food as a response to animal rights frames (Oh and Jackson, 2011). In this, one of the essential insights of polity studies demonstrates not only the role of culture and norms in shaping government institutions and behaviour, but also how institutionalization and the formation of governance structures channel non-government organizations' engagement with responses to, and action in, movements and policy.
Yet, at the same time, polity research also addresses many of the contradictory and, indeed, hollow victories U.N. treaties and agreements represent. While governments may sign onto new agreements to play the global civil society game and establish agencies to channel participation, many also exhibit varying levels of decoupling of state action and policy from their symbolic commitments to international agreements. Indeed, this 'hypocrisy paradox' is arguably a concomitant and constitutive feature of the expansion of new institutional norms (Fallon, Aunio, and Kim, 2018). Dubé Vol. 8 No. 2, August 2021   231 The importance of place and neighbourhoods in food systems transformation Neighbourhoods and local environments play an important role in reflecting and shaping food transformation (Charreire et al., 2010). This is particularly the case for large cities and urban planners who, in the tradition of Jane Jacobs (1961), have long concerned themselves with understanding what makes neighbourhoods and thus cities thrive. A substantial amount of work on food deserts, food swamps, and food environments addresses how mobility and availability of fresh food options factor into our everyday decisions about the food we eat (Mercille et al., 2013;Luan et al., 2015;Yang et al., 2019). These decisions are profoundly localized in urban environments, wherein the location of grocery stores and availability of public transportation in a neighbourhood can have a significant impact on the consumption of healthy food for the local population (Zenk et al., 2009). Many studies and thus policy intervention have been waged on the insight that inequality and racialized geographies in cities translate to fewer affordable, healthy options for residents of poor neighbourhoods (Raja et al., 2008;Walker et al., 2010). Walker et al. (2010), in their systematic review on food deserts, for example, find overwhelmingly that there are fewer grocery stores with fresh, affordable options in poorer communities. Policy implications and thus food systems transformation from this perspective has focused on using geo-spatial analysis to prioritize and site fresh food access points to improve options in disadvantaged communities (Powell et al., 2007). However, policy initiatives and responses based on this perspective have produced mixed and contradictory results at best. For example, Abeykoon et al. (2017), in their meta-analysis of grocery store interventions, found that while improved food access increased neighbourhood satisfaction, its impact on health outcomes were limited. Similarly, Alcott and colleagues (2019), in testing models of grocery store interventions, found that changes in the food environment reduce 'nutritional inequality' by only ten percent in studied communities. In Montreal, as part of policy interventions focused on improving affordable food access, several programs increased amounts of fresh food yet the impact on behaviour and outcomes was limited. In a substantive review of food environment approaches in Montreal, Robitaille and Paquette (2020) concluded that the active participation of community and commercial partners is a key lesson in establishing and siting access points as observed during informal exchange among community organizations, wherein new initiatives launched by external actors quickly were jettisoned because they did not reflect the culture, community mobility, and social dynamics of the neighbourhood.
A more diverse literature on food movements and mobilization emphasizes the relationship between citizens, organizations, and actors and their efforts to transform their communities (Holt-Gimenez, 2011;Levkoe, 2015;Wekerle, 2004;Wittman, 2011).
In this, neighbourhoods face complex realities in seeking to both implement and expand innovative ideas with governance models and arrangements often being an impediment to innovation and change. A community organization with a bright idea that proposes and/or participates in alternative food initiatives (AFIs) at the local level can thus face not only the challenge of testing and implementing a new initiative, but of also navigating an existing structure and set of relationships under the auspices of good 'governance' that mitigate their potential impact.

The Study
This study draws on multiple methods and levels of engagement with food security tables and municipal policy in Montreal from 2017-2019. The authors are guided by the good food principles articulated by Food: Locally Embedded, Globally Engaged (FLEdGE); in particular, producing community-driven research that both connects people and feeds cross-sector and community collaboration in the production of food policy. 1 These principles oriented both our work with local communities and food tables in Montreal as well as our methodological framework for community-driven engagement. In the development of a broad portrait of the food systems landscape in Montreal, we also draw on the two years of research-community partnership with food security tables led by the Dawson Food Justice and Sustainability (FJS) Hub. FJS has hosted cross-neighbourhood exchanges, mini-grant support for local food security tables on research and data needs, and events co-organized with the C-SAM in the development of its strategic plan.
To develop a clear and deep understanding of roundtable responses in three neighbourhoods, we adapt the model of design-based implementation research (DBIR) in educational research on innovation and apply its central principles and insights to food systems and movements research (Fishman et al. 2013;Penuel et al., 2011). DBIR is grounded in ongoing collaboration between researchers and practitioners with the goal of understanding the practical issues, challenges, and innovations that arise in the process of implementation. Central to this is the insight that new research does not easily nor does it wholly translate into action without innovations and adaptations to theory on the part of practitioners who are key practitioners in translation, adaptation, and implementation.
While there is not a single DBIR methodology or method, four key principles guide the approach: 2 They are: Dubé Vol. 8 No. 2, August 2021

233
• a focus on persistent problems of practice from multiple stakeholders' perspectives.
• a commitment to iterative, collaborative design.
• refining theory and knowledge through practice and implementation.
• capacity building to sustain change.
In this project, we employ the principles of DBIR in the educational environment to guide partnerships and research in the community and neighbourhood environment. Our central goal is to develop capacity within local food security tables as well as co-design a research approach that supports their collective efforts to implement meaningful change. To do so, the foundation of this partnership is predicated on first supporting local communities and organizations in the practice of research. Community organizations and representatives from the three neighbourhoods were members of the research team along with the project leads and research assistants. The team met on average once per week to develop the DBIR-adapted approach, discuss ongoing questions and issues as they emerged in the research, and co-author reports and presentations. Within the neighbourhoods, community research team members were embedded within the food security table, coordinating organizations to bridge the everyday experiences and challenges of community mobilization and cross-neighbourhood and research insights. First, the research team met to design the study and develop tools for data collection. We then presented our methodological framework to the food security tables for feedback and to discuss consent for the data collection methods. All data collection was carried out by community-based research team members who critically reflected on their experiences to develop an organizational autoethnography of the food security tables (Doloriet and Sambrook, 2012). To identify persistent problems and perspectives within the neighbourhood, communityresearch team members took field notes at food security roundtable meetings and other neighbourhood meetings on food as appropriate. After data collection, team members drew on the organizational autoethnographies and field notes to develop an interview protocol tailored to the neighbourhood context and the emerging perspectives on problem orientation. The research team then completed semi-structured interviews with food coalition members. Responses to interviews were coded in relation to the major perspectives and themes that emerged from semistructured interviews. Finally, the research team presented preliminary findings to the food security table for feedback and to feed decisions on collaboration, planning, and policy at the local level. Overall, field notes were collected at nineteen neighbourhood meetings and fortyfour interviews were completed at the three sites of study. The vast majority of these interviews were completed with representatives of community organizations that participate in collective planning and information-sharing focused on emergency food aid. However, in some cases table participation also included traditional religious organizations with food bank programs as well as a new cohort of organizations focused on alternative or social business models such as coops. Representatives from these initiatives were also interviewed to understand the changing nature and definition of food security within the neighborhoods. Dubé Vol. 8 No. 2, August 2021 234 Finally, several additional representatives who were key to food security policy and initiatives but not employed at or responsible for service delivery in each neighborhood were included in this study. These included elected officials, administrative agents, and individuals focused on collaborative planning or coordination. Overall, of the forty-four individuals interviewed, thirtyone represented local NGOs, seven were affiliated with local government as counselors or bureaucratic agents, three were from local religious organizations, two represented cooperative or social business initiatives, and one interview was completed each with a foundation and school representative. These are summarized in Table 1 below. Interview questions were asked of each respondent based on field notes and, in particular, the persistent issues and questions discussed at table meetings in the year before interviews took place. All interviews for the three sites took place between June 2018 and September 2019. The vast majority of interviews lasted one hour and took place at a time and place of the respondent's choosing. In some cases, however, the interview was far longer, lasting up to two hours in length. In the case of the West Island, some interviews occurred by phone when the respondent could not meet in person. This was an accommodation to address difficulties in transportation for some respondents in the large geographic territory of the West Island.

CFS/RCÉA
Interviews were then coded for emerging themes and from key informants to guide the production of the presentation for each site. Finally, comparative and cross-cutting themes were identified, coded, and included in the report for each site to address commonalities in issues across the three territories.
Below, we discuss the formation and structuration of a food 'polity' in Montreal beginning in 2006 and continuing to the present. This included three overlapping, place-based perspectives on social change along with institutionalization of engagement. We then turn to the impacts of these shifts on our three sites of study: Notre-Dame-de-Grace, Verdun and the West Island in Montreal.  Dubé Vol. 8 No. 2, August 2021   235 The international context and Montreal: The Milan pact To date, over 210 cities globally have adopted the Milan Urban Food Policy Pact (2015), an international agreement opened in 2015 with UN support that institutionalizes cities as signatories enabling them to take action on new norms in food politics. Launched in 2014 by the city of Milan, the pact was specifically proposed to address the increasingly important role that cities play in representation and policy. A central aspect of the pact was to empower cities and mayors to take collective global action by adopting principles for healthy, sustainable food systems (Dubbeling, et al. 2015). With a proposed thirty-seven recommended actions in six categories, signatories commit to coordinate in policy and action to develop food systems that are "inclusive, resilient, safe and diverse, that provide healthy and affordable food to all people in a human rights-based framework, that minimise waste and conserve biodiversity while adapting to and mitigating impacts of climate change" (Milan Pact, 2015). Cities also commit to documenting progress through indicators defined for the six categories: governance, sustainable diets and nutrition, social and economic equity, food production, food supply and distribution, and food waste. The initial agreement in 2015 was adopted by over 100 cities worldwide. At present, it includes over 200 cities globally, including Montreal. Through this, signatory cities have adopted the discourse and norms of the pact, participate in global meetings that legitimize their roles as central actors in the global commons, and adopt policies to meet the six goals. Notably, the goals are universal in their definition and in the use of a common set of indicators to measure progress. They even exhibit isomorphism in food policy, establishing similar governance structures and relational arrangements in wildly different venues and contexts. This is demonstrated, for example, in the proliferation of food policy councils, both across North America and internationally. However, it is not clear whether and how this expansion has impacted existing local movements, organizations and long-standing community food networks within cities.

Montreal food context: Neighbourhoods, municipal policy councils, and systems impact
In the formation of civil society institutions and food systems policy in Montreal, three interrelated movements-each committed to mobilization and social transformationdemonstrate distinct histories and trajectories in food systems change. Two of theseneighbourhood roundtables and the Collective Impact Project (CIP)-place food on a menu of related social problems, ostensibly to be tackled together. The third-the C-SAM-explicitly formed in response to both 'bottom-up' and 'top-down' pressures from food movements to organize a coherent, systemic response at the municipal level in food policy and mobilize over 200 community actors in doing so. Dubé Vol. 8 No. 2, August 2021 236

CFS/RCÉA
The locations, histories, and orientations are all stories of increasing formalization and structuration in the Montreal food system that, in turn, can be characterized as a municipal food polity (Meyer, 1997). The primary mobilizing structures that give voice to and have become a central actor in localized food policy options are part of longstanding tables of concertation (coordination) at the neighbourhood level in Montreal. Local tables initially emerged out of neighbourhood mobilization in communities hardest hit in the economic and social crises of 1980s Montreal. Over time, neighbourhood roundtables became the primary means for front-line service organizations and community groups to share information, coordinate activities, and advocate on behalf of residents in relation to social policy. In identifying the most significant social challenges, neighbourhoods established roundtables on employment, youth, seniors, housing, and food security. Roundtables also self-identified the 'place' of their responses and defined neighbourhood members via historical, social, and geographical relationships that resonated with local residents and citizens.
These loose, unstructured, and community-led mobilization efforts sought to sustain their impact by institutionalizing the neighbourhood roundtable model of cooperation as well as through the continuous multiplication of the roundtable model across neighbourhoods.
In 1996, local roundtables sought to affirm and institutionalize this work by forming the coalition montréalaise des tables de quartier (CMTQ), a Montreal-wide coalition of neighbourhood tables. In 2006, the Montreal Initiative to Support Local Social Development established a framework for annual funding for roundtables and further formalized and institutionalized the CMTQ and roundtable model as the primary vehicle of community engagement and mobilization to address poverty and social exclusion.
Within the municipal polity, neighbourhood roundtables represent local interorganizational cooperation among religious organizations, charitable groups, non-profit organizations, local health and welfare agencies, schools, and local housing authorities. Dubé Vol. 8 No. 2, August 2021   237 They are also the primary vehicle for community-led planning to neighbourhood challenges and crises by organizing regular meetings, setting up neighbourhood-specific priorities, and publishing local resource 'bulletins'. Local tables may be self-organized and 'independent' and/or coordinated by an established social development organization within each neighbourhood. Each table is a member of the CMTQ. The CMTQ meets at least once per year to discuss cross-neighbourhood responses, provide feedback on local policies, negotiate collective policy positions on issues when necessary, and, most importantly, respond to new applications for recognition to form a neighbourhood This history and structure for community action is significant in addressing the outcomes and responses to the CIP and C-SAM we discuss below. Importantly, however, it is of note here that community cooperation focused on food has been and continues to be framed in the discourse of food security. This was a product of the challenges of the 1980s crisis as well as the dominant normative model of charity and emergency food aid. As we discuss below, this model has been challenged and transformed as a result of alternative food movements and other critiques of food security paradigms. Dubé Vol. 8 No. 2, August 2021 238 Figure 2: The CMTQ Territories, Coordination Tables, and Food Tables Place-based philanthropy: The collective impact project

CFS/RCÉA
In January 2016, eight foundations in Montreal launched the Collective Impact Project (CIP), a collaborative philanthropic initiative led by Centraide of Greater Montreal (Centraide) that established a governance framework for coordinating financial support for neighbourhoods and communities to "intervene directly" and "catalyze changes in a complex environment" (Pole and Fontaine, 2017). The initiative quickly added non-financial partners, including the city of Montreal, the Direction régionale de la santé publique de Montréal (DRSP; Montreal Regional Public Health Department), and the Coalition montréalaise des tables de quartier (CTMQ) (Montreal Neighbourhood Tables Coalition), in order to provide strategic advice and direction to the effort. As a coordinated initiative that essentially pools the funds of several foundations, the CIP represents the largest government, philanthropic, and non-governmental partnership in Montreal aimed at transforming local communities to reduce poverty and achieve social development. In this, food security and food systems transformation are central priorities and thus avenues of funding and support. Dubé Vol. 8 No. 2, August 2021 239

CFS/RCÉA
While new to the local organizations in Montreal and, as discussed below, presenting a significant change and challenge to their coordination and work in local food security, the CIP partnership is part of a far broader movement across Europe and North America taking place over the past five years towards strategic philanthropy focused on comprehensive community change (Phillips and Scaife, 2017). Central to this shift is a reorganization of relationships between funders and community organizations in community development towards place-based and coordinated interventions. While there is no systematic review and accounting of the number of place-based strategic philanthropic initiatives at present, Phillips (2019) estimates that there are over 2,000 community foundations dedicated to place-based philanthropic giving in over fifty countries.
Place-based philanthropy proposes three major re-orientations to social development: via whole-systems approach to community change, an emphasis on community-and thus capacitybuilding and an emphasis on bringing in new partners for long-term, sustained revitalization (Gamble, 2010;Cabaj, 2011). Collective impact frameworks provide a 'road-map' to achieve this by setting conditions to guide funding and measure success. In Montreal, this framework consists of five such conditions: a common agenda, mutually reinforcing activities, continuous communication among the partners, a shared measurement system, and a support structure to coordinate work (Pole and Fontain, 2017). In practice, this translated into seventeen opportunities for funding for the thirty neighbourhood roundtables. In the spring of 2016, foundation partners committed $23 million over five years to support the CIP and seventeen neighbourhood roundtables were selected for a first round of funding. Importantly, the CIP partners and lead agency, Centraide, have adopted a model of strategic philanthropy usually practiced by corporations in their orientation and structure through the CIP, wherein their collective funding commitment is conceived and structured as an 'accelerator' of change for local communities.
Below, we address the impact of the support and the lack thereof by the CIP on roundtables. Here, we note that the 2016 launch marked an abrupt shift for many neighbourhood roundtables who needed to meet the five conditions of the framework to receive funding. Importantly, these conditions involve demonstrating a commitment to and success in a structural and normative framework for collaboration between neighbourhood organizations.

SAM)
Following an informal dialogue on municipal food policy lasting over ten years, the city of Montreal officially announced in October 2018 the creation of the Conseil-système alimentaire montréalais (C-SAM), or Montreal Food Policy Council. Dubé Vol. 8 No. 2, August 2021 240 Beginning in 2011, informal dialogues were formalized as a municipal process through the office of public consultation (OCPM, 2012). These public consultations and the slow coalescing of the C-SAM as the governing structure brought together over 200 organizations in Montreal to provide insight into and contribute to the goals and strategic plan to guide the C-SAM's work shaping the roles, orientation, and local engagement with the network.

CFS/RCÉA
Officially, the C-SAM is the lead organization for food policy, in which governmental agencies and administrative bodies such as the Department of Public Health participate as de jure members of the governing council, but for which the C-SAM plays a supra-institutional role in guiding policy that is inclusive of, but not restricted to, the municipal administration. 3 Thus, while municipal administration(s) at the borough level are responsible for enacting policy and establishing plans for other policy measures, food policy is distinct in its independence, both at the neighbourhood level within local communities as well as at the 'supra' municipal level in their participation in the C-SAM.
In terms of governance models, the C-SAM represents a 'multi-stakeholder' model that includes regional and local governments, municipal agencies, foundations, and community organizations. While the C-SAM's main role has been to set priorities for policy through consultation on and approval of a strategic plan every two years, it has expanded to include other roles and partnerships in the face of crises and specific challenges. Meanwhile, as part of the health-promoting initiative called Montréal, Métropole en Santé, the C-SAM has the approval of the city of Montreal. It does not have an official 'mandate' to determine or implement food policy for Montreal. As such, unlike other municipal policy directives, the city and its boroughs are not accountable for implementing the strategic plan or achieving specific goals in relation to it. This highlights the two essential functions for the council: (a) in building consensus across a broad cross-section of 200 organizations to participate in its strategic planning process and (b) serving as a normative and consensual framework for action on food policy. This is important when considering that the city of Montreal signed the Milan Urban Food Pact-an agreement that includes specific indicators of progress-and that the city of Montreal has played specific role(s) in guiding and legitimating the role of the C-SAM as the food policy organization. That said, the C-SAM is a representative organization for the Montreal region that includes both these boroughs and several independent municipalities and was designed to give equal voice to government and non-governmental organizations. In this role, one notable impact is that while the C-SAM sets a strategic framework as a council, it does not mandate policy for the city. The city administration and boroughs, in fact, can choose not to follow the strategic plan and/or simplify fail to implement its goals. Dubé Vol. 8 No. 2, August 2021

241
One of the clear outcomes of the C-SAM's normative leadership on this, based on the Milan Pact, has been to redefine food security as a discursive and organizational framework for policy to first include urban agriculture initiatives and goals then redefine the overall goals of municipal commitments to align with the six categories of the pact. Key to this has been a transformation through the C-SAM process to a food systems lens and institutional agenda. This orientation, while legitimized and amplified by Milan, represents a significant departure from the food security and place-based philanthropy frames forwarded by the CMTQ and CIP organizations in Montreal.
As we discuss below, in the formation of a polity, this translates to the C-SAM as the primary guardian of Montreal's accountability to the Milan Pact all the way to engaging neighbourhood organizations to achieving particular goals. In this, the C-SAM concerns itself with governance, participation, and, most importantly, with serving as a conduit between global and local norms. This has not been without controversy and, in particular, potential decoupling between norms and action.

The Montreal food 'polity'
Beginning in 2006 with the formal, ongoing commitment to channel funding for neighbourhood coordination through the CMTQ, increasing and overlapping structuration of civil society organizations coalesced in the three main partners discussed above. In the process, they have reinforced each other's legitimacy and autonomy as vehicles for engagement in the Montreal food system. Thus, the CMTQ has become the main channel of participation and access from neighbourhoods to municipal-level policy. As an organization, it is a partner in the CIP and in the governance structure of the C-SAM. Correspondingly, the members of the C-SAM, including especially philanthropic and municipal partners, are also the primary drivers of the neighbourhood 'accelerator' model. This reflects, far more than a formal governance structure, a food 'polity' in Montreal, where, in the absence of formal governmental authority, these actors exercise normative, cultural, and 'soft power' along with institutionalized mechanisms of engagement. They thus (1) provide the primary discursive lens through which food systems transformation is framed and (2) 'channel' the participation of local organizations invested in food systems transformation into the formal, overlapping structures. While they do provide opportunities for local communities to access funding and support as well as have a formal seat at the table in municipal food policy, it remains to be seen whether these structures facilitate and accelerate broader transformation. Key factors in that broader transformation are the neighbourhood and table responses to the formalization of the C-SAM and their roles within it. This may have differential impacts as channeling and structuring participation meet neighbourhoods' responses to this formalization. Dubé Vol. 8 No. 2, August 2021 242 Neighbourhoods: Participation, channeling, and innovation

CFS/RCÉA
To gain a fuller understanding of the ways in which these arrangements have impacted communities as well as how communities have responded, we turn to the experiences and perspectives of three food tables in the Montreal polity: NDG, the West Island, and Verdun.
Here, we discuss first the precipitating, or background formal and informal participation, of local communities in collective action associated with their local food system.

NDG, the West Island, and Verdun
All three sites of study vary in the size of their territories, populations, budgets, socioeconomic status (SES) and other attributes. We summarize some of the basic elements of these indicators in Table 2 using 2016 census data. As is clear, the West Island encompasses a far larger territory with a population that is comparatively wealthier than Montreal in general and the other neighborhoods included in the study. Additionally, they vary in their organization of and participation in local food security tables as well as cooperation with one another within the table territory. While they have all also participated in the consultation process organized by the C-SAM in 2019, they differ on their participation in the CIP; both Verdun and the West Island received CIP funding while NDG did not. They were all established at wildly different times in Montreal's history and thus their own neighbourhood's history and organization: NDG was founded in 1998, the West Island in 2015, and Verdun in 2018. They also represent varying degrees of formalization and integration into the CMTQ structures and relationships. We turn to each of these below for a brief overview of their histories and participation in food governance. Dubé Vol. 8 No. 2, August 2021 243

CFS/RCÉA
The NDG Food Security Coalition was founded in 1998 by the CLSC NDG 4 Montréal-Ouest to regroup organizations working in food security in the neighbourhood and facilitate the sharing of resources, information, and best practices. In 2014, the Coalition, with support from a Department of Public Health measure, produced a study, which identified three key sectors in NDG lacking access to fresh produce. Following a public consultation, the Coalition, with the NDG Food Depot as the fiduciary and implementer, launched a project to realize mini mobile markets in the Walkley and St-Raymond neighbourhoods. Major changes in the neighbourhood and orientation of the table occurred with the growth of the NDG Food Depot (now the Depot), which merged with Action Commauniterre and the Boîte à Lunch project to establish a stronger food security organization with an expanded mandate comprising food assistance and education, urban agriculture, and community kitchens. One consequence of this was a reduction in membership in the table; by 2016, the Food Depot was one of the only organizations in NDG with a food security mandate. The Coalition chose to review its own mandate in light of the partners' capacity to participate at the Coalition and to act on food security. The Coalition mandated consultant Jean-Frederic Lemay to produce a study including recommendations on the future functioning of the Coalition. After performing a SWOT 5 analysis and other studies, the working committee recommended that with the limited resources available, the Coalition hold meetings two times per year and reducee the action planning cycle from 3-5 years to one year, with only two actions chosen as priorities.
Due to its own internal restructuring, the CIUSSS withdrew its community organizer as co-coordinator of the Coalition in 2018. As NDG Community Council became sole coordinator of the Coalition in the spring of 2018, they recommended to the Coalition that it not accept two of the recommendations of the working committee for lack of resources and capacity to carry out the recommendations. The Coalition developed an action plan at the beginning of 2019.
At the same time, member organizations were far more active independently. The Depot, for example, launched the Boite à Lunch program in 2016 only to expand the model to other neighbourhoods in successive years. In this case, however, the funding, the program model, and the expansion of the program to other neighbourhoods was launched, managed, and operated exclusively by the Depot and not with partners at the coalition level.
Unlike NDG, the In 2015, it published a report of the population of the South of the West Island (Ziuleva, 2015), which included a diagnosis of the needs of vulnerable populations followed by their Territorial Social Development Action Plan addressing the issues discovered in their report and diagnosis: health and social services, poverty and social exclusion, food security, housing, and transportation. The Food Security Committee was created after that. In 2018, the TQSOI published a report on poverty and food insecurity in the West Island. The awareness campaign Make the Invisible Visible orchestrated by the Food Security Committee that included members of the TQSOI, the Community Resource Centre (CRC), and various other community organizations, ran from June to December 2018. That campaign included a short documentary called Hidden Hunger launched in November 2018. Both the campaign and the documentary were successful, and in January 2019, a public forum on food security was held to consider the next steps to continue working on food security in the West Island. The following fall, the TQSOI, the CRC, and the Bread Basket (a community organization focused on food security) came together and formed a new governance model to increase collaboration and productiveness of the multiple committees involved with food security. In 2020, a Forum on Food Security was organized by the new Food Security Committee of the TQS to present food projects, initiatives, and services to the members. In 2015-2016, the Table de Quartier du Nord-Ouest de l'île de Montréal (TQNOI) joined forces with the TQSOI and Concertation Ouest de l'Île (CODI) to apply to the Collective Impact Project from Centraide. This reflects a broader perspective on the island of Montreal that has historically and conventionally defined the "West Island" to include the areas in both the north and south of the West island. The CIP gave a boost to the collaboration in the West Island to tackle three previously identified issues: housing, transportation, and food security. The latter was chosen as the starting point for community initiatives bringing together the two tables. In 2017, the collaboration with the TQNOI for the PIC came to an end and, in 2019, due to internal conflicts and other organizational issues, the TQNOI was disbanded as a CMTQ organization. The collective process at the TQSOI focused a great deal on examining the different options available for the table and the concertation in the wake of this development. 6 Verdun is both a locally and historically-defined borough located on the island of Montreal and is a recipient of CIP funding. Verdun sans faim (VSF), the food security table, was established in 2018 following the dissolution and disbanding of a previous food table. It is neither a non-profit or non-governmental organization, but rather a loose, informal coalition that is independent of the poverty and community development organization-the CDSV (Concertation en développement social de Verdun), which is the primary contact organization in Verdun for CMTQ membership. 6 Dubé Vol. 8 No. 2, August 2021 245

CFS/RCÉA
In terms of its structure and history, the Verdun table is distinct from both NDG and the West island in several significant ways: (1) local tables were initially established as independent entities that self-organized and, previous to the CIP funding, had previously registered as an independent NGO representing the local food organizations in order to carry out projects in the neighbourhood. Following the completion of a major food bank project, the food security itself disbanded and subsequently formed a new, informal network to regroup, reorganize and focus on sharing ideas at the neighbourhood level; (2) while several organizations merged to become an important community food organization focused entirely on food programs and policies, Verdun does not have one, large overriding organization focused exclusively on food security as its mission. As a result, organizations at the table all faced resource limitations and issues associated with participation when, at the same time, their missions were not primarily focused on food security; and (3) because the Verdun food security table was originally formed as an independent group and reformed as an informal network, they lacked the capacity to apply for funding as well as the ability to legitimize the network in carrying out simple functions such as calling a meeting. At the same time, the local organization that provided support in other territories could not help the Verdun table with basic administration because of its historical relationship to the issue tables. In short, as the tables had always been independent and self-organized, local organizations had previously insisted that the CDSV adopt by-laws limiting their ability to provide coordination assistance to the tables.
Over the course of this study, these tables agreed upon and exhibited wildly different missions, models of cooperation, and varying degrees of engagement from members in the formation and implementation of local food policy. Both the West Island and Verdun received funding from the CIP during this period. NDG, however, did not. That said, all were present for and participated in the collective planning and consultation process launched in 2018 by the C-SAM to set municipal priorities for a Montreal-wide action plan in food policy. And all, as we discuss below, were impacted by the discursive and normative mandate in the Montreal polity focused on place-based transformation. We turn to the impacts of the tables' participation in and work in Montreal below. As each table worked in the Montreal CMTQ context, each was also impacted by the informal networks across neighbourhoods as well as the values of CIP in shaping their local action plans, priorities, and disagreements for policy

West Island: Participation and collaboration
The West Island presents particular challenges to thinking about the role of neighbourhoods in food systems transformation; as a large territory encompassing a predominantly suburban population, it is difficult to make the case that one food security roundtable can somehow coordinate across ten boroughs and municipalities. Dubé Vol. 8 No. 2, August 2021 246 Yet the TQSOI now represents wildly different communities and needs in food security coordination and planning. That said, it is also clear that, prior to the participation of this 'neighbourhood' in the CIP process and support by that funding to define and deliver a food security agenda, very little collaboration took place between organizations across the terrain, which resulted in a less engaged and active table on the issue of food security broadly. This was reflected in the perspectives of one organization as they became involved in the CIP (PIC) process. When asked about the role of the Related to this, one of the most puzzling 'disagreements' that emerged from our interviews in this community arose in relation to the question of collaboration. While the issue of collaboration was a source of great frustration for some organizations, it appeared to be barely a concern for others.

CFS/RCÉA
As one coordinator of an organization with close proximity to other table members related: "…So, some of them have participated and been engaged quite a bit, but some of them never really joined our collective effort. So, we tried to get the information from different ways by doing surveys by doing questionnaires or having short conversations here and there, but it's never been an ongoing process with some of them. So of course, it limits the amount of solutions and impacts that we can actually have on the system as a whole." [Interview,Respondent 20,West Island] But when other organizations that were less connected or close to the main table were asked about collaboration, they responded that they did, in fact, collaborate with other groups. This was despite the fact that they were criticized by the organizations above for not being collaborative enough. More importantly, this basic division between organizations repeatedly arose as a point of frustration. If viewed from the position of membership in the CMTQ roundtable, this is a contradictory and strange disjuncture. Dubé Vol. 8 No. 2, August 2021 247 However, as we coded the interviews and organized them in relation to the CIP, a pattern emerged: those that were more 'central' to the table and had been heavily engaged in the CIP process emerged as the critics, while many more organizations similarly oriented yet not involved were criticized. Thus, while their orientation or actions in the food system may not have changed, their definition of and orientation towards one another as transformative actors did. They also perceived that the CIP would bring recalcitrant or 'lagging' organizations around. As one volunteer food bank coordinator responded when asked about the role of the CIP But yeah, there's definitely some organizations that really are focused on what they're doing and don't want to be part of research, don't want to, you know, they'll network with who they need to network with, and that's it. But I do think that's changing and especially with like the PIC project, and the research that's being done. There's definitely more like there's been a lot of awareness [...] And even the organizations that don't really want to work with someone are like, "oh, okay, there's, there's something bigger going on out there" and "oh, they're all working together. Like, I guess I should be part of that too." [Interview,Respondent 22,West Island] Interestingly, one of the unintended consequences of this has been to reinforce the CMTQ structure on the West Island and the territory while also revealing that there is wide variation of connection to identification to the table. As a partner to the CIP in which 'neighbourhoods' are defined by CMTQ boundaries rather than administrative or municipal districts, the prospects for deeper collaboration between organizations in transformative projects are more limited. 8

Verdun: Negotiation and adaptation
Like the West Island, Verdun is a CIP-funded table, with support for collective planning, decision-making, and priority-setting by community organizations. Similar to the West Island, Verdun Sans Faim-the food security table for Verdun-established food policy and food insecurity as priority issues for the neighborhood and thus their CIP plan as part of this process.
Unlike the West Island, although VSF was formally established in 2018, Verdun had a long history of coordination and cooperation among community organizations in the neighbourhood to address social and economic issues. Dubé Vol. 8 No. 2, August 2021 248

CFS/RCÉA
As discussed above, a previous, formally incorporated table (as an NGO) existed before VSF, but was disbanded as the table members began to question its rationale at the end of the period of administering a grant. In the aftermath of this formalization and role as administrator of the grant, members of the table reflected on how the project changed the rationale, activities, and coordination among local organizations and decided that they needed to stop carrying out projects. Quickly, however, they also immediately began to meet through VSF following this decision. When asked about this, multiple members of the table discussed the need and desire to share ideas, information, and thus coordinate in an informal manner. As one member reported: We decided that we were too focused on the project when we became responsible for funding, grants, and administrating the grant. We thought this would be a good direction, but realized we were wrong. At the same time, we thought we still wanted to network and share information with each other. We just didn't want to be focused too much meeting time on talking about coordinating projects and wanted to spend more time networking" [Personal correspondence, 2018].
Notably, therefore, when the CIP was announced in 2017, Verdun organizations had just recently mobilized to relaunch a food security table in the neighborhood. As organizations reflected on their experiences with the previous table and the decision to create a new table, members were particularly sensitive to debates about whether and how they would manage an application to the CIP that required a shared mandate, coordinated activities, and a place-based, Verdun-wide project proposal.
A complicating factor in relation to this was the historical role of the table and the primary fiduciary organization in Verdun as part of the CMTQ: the Concertation en développement social de Verdun (CDSV). This is a key difference from other neighborhoods and tables in this study: unlike the West Island and NDG, Verdun's tables in food security, housing, and other issues had and continue to be independently organized by local organizations. Each table could thus, in their view, more easily mobilize without the priorities of the CDSV or CMTQ spilling into their work. When the CDSV was founded in 2000, like other "concertation" organizations and tables, it was formed to represent the collective issues of the neighbourhood from the 'ground up', or as an organization that would amplify the plans and activities of independent tables. Each issue table thus sends one representative to coordinating meetings hosted by the CDSV to identify each table's priorities, activities, and concerns. However, written into the CDSV's by-laws is also the prohibition of the CDSV playing any role in coordinating the issues tables. Thus, at the moment that VSF was in a fragile state, the cooperation between organizations was both a long-standing historical fact and thus expectation in the neighbourhood. Certain that they needed VSF and a table, however, members quickly disagreed about what the table's mandate should be and how it should be organized.
At the same time, all members expressed interest in and attended sessions hosted by the CSAM to set priorities for the city-wide plan. Dubé Vol. 8 No. 2, August 2021 249 While they were sometimes critical of this process, member organizations supported the goals of municipal planning and embraced the need for local organizations to be involved in the process.

CFS/RCÉA
In this context, it should be noted that, unlike the West Island, Verdun's table benefited from one, important attribute in its negotiations: while the West Island encompassed over ten boroughs and municipalities and a relatively large territory, Verdun' table and administrative borough boundaries were[?] exactly the same. Both Verdun residents and organizations have strong identification with the neighbourhood and identify with each other primarily by this affiliation and thus identity. Verdun is thus highly integrated territorially and historically but was not institutionalized as a food security table in 2018 when it applied for a funding predicated on a whole-system, place-based approach to planning and policy. The result, as all interviewees remarked, was initially disastrous: VSF almost disbanded in the process and there was a considerable amount of in-fighting between organizations over priorities, perspectives, and competition for resources.
In the first year, after several rounds of negotiation among members and between VSF and the CIP, work on food security was eventually funded through the CDSV which, in turn, created an entirely new process and organization to plan and manage CIP projects. After setting priorities in food, housing, and education, the CDSV launched the système alimentaire Verdunois (SAV) as the primary coordinating body for food security in this process. Notably, whether and how to participate in this process was a key source of tension and negotiation for VSF. Some saw this as an opportunity, while others worried about duplication of services and lack of coordination among local actors.
As one local counselor with a key role in the table remarked, Alternatively, other members of the VSF worried about duplication of services as well as being sidelined given the increasing role of the SAV and thus the CDSV. Dubé Vol. 8 No. 2, August 2021 250 This reflected, additionally, a more general concern that duplication of services is an issue in Verdun impeded both collaboration and cooperation on food system issues. As one respondent representing a long-standing local organisation reflected, In the lead up to the application submission, members of VSF were evenly divided about whether and how to respond to the CIP process. This initially had the effect of creating more tension, conflict, and suspicion among members of VSF as they each jockeyed to be the lead organization for the funding call and thus garner most of the financial resources offered by the CIP. 9 Eventually, to resolve this conflict, members approached the CDSV as the fiduciary administrator, a decision which, in turn, led to the SAV as a parallel process focused on coordinating activities for the CIP. All other planning, coordination, and collaborative work is now under the auspices of VSF, while management of the grant is coordinated by partners in the SAV. The upshot of this decision and the successful CIP application is that there are now two coordinating tables in Verdun focused on food security. While the focus of their work is different, both VSF and the SAV are populated by the same organizations as members. Both also have a mission to coordinate and collaborate with one another in order to address food security and the local food system. In short, while there is perhaps no duplication of agendas or activities for Verdun, there is a duplication of organizational relationships, of mission and values, and coordination between members within the neighbourhood.

NDG food mobilization, innovation, and transformation
NDG, among the three neighbourhoods under study, is the only non-CIP funded initiative. It is, however, highly integrated into the CMTQ model; the food security coalition had sustained cooperation for over 20 years and NDG served on the executive committee of the CMTQ. At the same time, members of the coalition were acutely aware of the CIP in other neighbourhoods. As one community coordinator for a large organization in the neighbourhood related: Dubé Vol. 8 No. 2, August 2021 251 "..Collective Impact so like this is like a buzzword right now, you know that like everyone's talking about how to create collective impact and Montreal started the PIC, the Project Impact Collectif, which doesn't cover our neighbourhood but I've seen it in action in Sant Michel and I'm over there. And I think it's a really interesting initiative. And something that I think we could learn from and, and get inspired from I think, as I said before, these really like systemic, complicated, complicated adaptive problems require a lot of creativity and a lot of collaboration and thoughtfulness to address and I think the more actual like money support, and, and then political or like, you know, like whatever, just like local support for collaboration and the frameworks and the structures that would support collective impact…" [Interview,Respondent 3,NDG] At the same time, despite the fact that the  Dubé Vol. 8 No. 2, August 2021 252 Yet another, more active organization, when presented with this perspective was strongly and negatively critical of a perceived lack of commitment on the part of this table member. In their view, attendance and participation in cooperative structures was one of the organization's responsibilities to the community which were, by definition, ends in themselves. The Food Security Coalition's problems were not new: they had been struggling since at least 2015 to revitalize what had, when the table was founded, been an energetic and engaged set of community relationships and excitement about cooperative community change. Members were at a loss about what happened, particularly as table meetings were successively reduced in duration and frequency to reduce what could be perceived burden of the time commitment.
At the same time, NDG has been a neighbourhood teeming with innovative experiments in food systems mobilization and action. Transition NDG joined the table and both reflected and reinforced the expansion of the table from food security cooperation to include a more diverse array of policy and possibility orientations. Nothing exemplifies this more than the transformation and growth of the NDG Food Depot from a food security organization to a Community Food Centre recognized across Canada. The merger with Boite à Lunch also benefited and facilitated the cooking program's capacity to think about and develop, test, and expand their success in NDG to four other neighbourhoods in Montreal. At the same time, the Depot has been one of the most active and committed members of the neighbourhood roundtable. This raises the question: what fosters innovation and food systems transformation? What extends and deepens the impact of new initiatives? We presented these perspectives and questions to the table and its members in October 2019 with further, specific questions to foster dialogue on the points of disagreement that emerged during the course of study. As a result of that dialogue, member organizations revised their meeting schedule to talk more often, formed a group to talk and share information informally between meetings, and recommitted to the mission.

Discussion
As each of these tables and community-wide responses represent, each of their histories, trajectories, norms, and coordination focused on self-defined neighbourhood territories in Montreal. Each incorporated representative table defines 'neighbourhood' differently. Thus, the West Island territory encompasses several boroughs and municipalities, while NDG encompasses one, historically defined community within one borough. Dubé Vol. 8 No. 2, August 2021 253

Isomorphism, discursive frames and norms
Neighbourhood roundtables were innovative responses to a challenging social context at a particular historical moment in Montreal where the most successful and thus longest standing neighbourhood roundtables were established in densely populated downtown and largely disadvantaged neighbourhoods. As a result, their territories were and are highly urban and correspondingly small. With the formalization of the roundtable model through the CMTQ and its subsequent extension into other milieus, both the sociodemographic and geographic contexts now vary from highly urban and poor to suburban and advantaged. Yet while roundtables have prioritized different social issues by neighbourhood, the formal structures and expectations for roundtable cooperation has been largely reproduced regardless of context. One of the themes that became clear across interviews and in neighbourhoods in Montreal was the excitement generated by the CIP 'accelerator'. Collective impact is a new, exciting buzzword and agenda that shapes the context and culture in which food organizations act. Neighbourhoods spent a great deal of time and resources aligning themselves with the perspective(s) and expectations of the CIP model. In this way, organizations and neighbourhood coalitions, similar to states and NGOs articulated in world-polity models, perceived the benefit of participating in the new orientation and governance as well as the new norms associated with place-based strategic philanthropy . They were required, in relation to this, to establish and propose a collective project that would ostensibly marshal the cooperative and shared goals of neighbourhoods in food systems transformation. A central feature of this was to, first and foremost, collaborate in the definition of a local food project that would also entail a coordinated response among local organizations in working together to address it. Funding was contingent on a collective project that organizations agreed upon and agreed to work on together on an ongoing basis.
Neighbourhoods with well-established governance structures and collaborations could both advocate collectively for their goals and determine the structure of evaluation and stages of support. It should be noted, in this regard, that these were neighbourhoods that were tightly coupled with the overarching goals and new norms represented by the CIP process. In the case of neighbourhoods with new or nascent neighbourhood approaches, the CIP process provided opportunities for established organizations to coordinate and collaborate with one another. In this, the structure and organization of TQSOI were central to spearheading this effort and bringing in organizations for planning purposes. However, while it did lead to a first collaborative effort with specific goals achieved, wide disparities persisted in organizations' definitions of and thus perceptions of collaboration in relation to food systems transformation. Organizations involved in the CIP process shared the same definition of collaboration to include horizontal planning and relationships, communication, and shared resource planning. Dubé Vol. 8 No. 2, August 2021 254

CFS/RCÉA
Other organizations, while they shared the same goals and overall view of food security along with a commitment to transformation, defined collaboration in varying and slightly different ways that were altogether less 'demanding' of time and resource investment, such as sharing information and organizational differentiation. While these are both valid models of cooperation, organizations involved in the CIP planning process perceived other organizations as 'not' collaborating and grew frustrated with their perceived lack of commitment. Yet the CIP model offers a directed, normative model of 'collaboration' that includes intense commitment for members. While this is part of the model, questions remain about how transparent this is to table members, who often voiced, in addition to the experiences with other organizations in the West Island, frustration at their understanding, and the resource demands, of such an intensive model. In the figure below, we identify key features and conditions related to whether neighbourhoods were consistent or exhibited tight 'coupling' between norms and implementation. In short, each neighbourhood varied in its internal organization as it became involved in the CMTQ, CIP, and C-SAM. Those that exhibited high internal coordination, such as NDG and Verdun, were far more able to negotiate their involvement in the CIP process and adopt or adapt new norms. In this, they exhibited more relative capacity to implement the goals brought into Montreal by the Milan Urban Food Policy Pact. That said, they did not always follow through in these. Verdun, for example, had low integration into the overall normative and discursive structure of these norms in Montreal even though it was funded as part of the CIP. The relatively low external integration drove more conflict between organizations and almost split the table apart at a particularly sensitive time in its history. Alternatively, the West Island, both established later and far less internally structured (both historically and currently) faced the challenge of integrating the north of the West Island and navigating the CIP funding. As a result, much of their agenda was driven by the norms and agenda of the CIP because the table was far less internally coordinated. Dubé Vol. 8 No. 2, August 2021 255

Figure 3: Isomorphism and coupling between norms and policy: Integration and internal coordination
Decoupling between norms and action As noted above, while food systems and place-based whole systems change has become a dominant normative framework for local organizations, food security roundtables exhibited far greater variation in substantive commitment to and participation of members by neighbourhood. This is despite the fact that many neighbourhoods increasingly prioritize food security as an important issue to address through collective action and continue to overwhelmingly support organizing through the roundtable model on food security issues. Some neighbourhoods exhibit strong engagement in and coordination through roundtables on food security issues while others persistently experience conflict, disengagement, and frustration in cooperation on food security issues. In the neighbourhoods we studied, there was significant evidence of the decoupling of policy formation and commitments to the CMTQ model from the realization of the goals of roundtables in practice. Dubé Vol. 8 No. 2, pp. 227-262 August 2021 256 While there was decoupling between normative commitments and discursive frames locally, however, this does not mean that they lack innovative ideas, the capacity or willingness to transform local food systems, or the ability to bring ideas to fruition, however. In fact, neighbourhoods are sites of innovation, experimentation, and thus the energy of food policy transformation. But they may lack the ability or capacity to document, scale up, and/or pitch their successes and innovations, particularly when these do not match the prevailing municipal and cultural food norms. Preliminary reflections and evaluations in the community as part of the CIP process affirmed that this was at least one persistent problem-the inability to realize the contributions of neighbourhoods and community organizations and scale up innovative ideas. In a 2017 evaluation of the CIP, researchers thus noted that the major limitation of the study was the underrepresentation of and thus failure to capture the perspectives of neighbourhoods in the analysis.

CFS/RCÉA
As within the global context, this suggests that the relative success of changes in the representative and normative commitments in municipal context depends greatly not just on local capacity, but also on resistance. In the C-SAM, local organizations continuously criticized the process as lacking transparency in public consultation. The CMTQ amplified the voices of neighbourhoods in both the CIP and C-SAM process, such that they were invited to actively engage and have a seat in the governance of the CIP and successfully changed the governance of the C-SAM to include more local voices and organizations. Yet these have also been limited: while the C-SAM have taken steps to address this in the new strategic plan with several consultation and engagement events, it is yet unclear how much the SAM can capture and harness the view 'from the ground'. Dubé Vol. 8 No. 2, August 2021   257 Conclusion: A convergence-by-design approach to account for path dependencies and building sustainable and resilient communities

CFS/RCÉA
Our results underscore the rich diversity in how local communities organize themselves over time as well as in how they welcome or not scaling up or capacity building initiatives like CIP and C-SAM. As part of the same complex and dynamic adaptive system, individual organizations and collaborative platforms observed in this research all had their respective historical trajectories and future aspirations in terms of composition, capabilities, goals, achievement and challenges (Addy et al., 2014;Addy and Dube, 2016;Dube et al., 2012). Accounting for such path dependencies (Struben et al., 2014) is critically important to understand and build upon this complex multiscale, multisector, and multijurisdiction dynamic.
Our study not only argues against a one-size-fits-all approach but calls for convergence-bydesign, organic approach to science and policy for bringing all actors around a common goal of supporting vulnerable communities. In Canada, there are large differences in food insecurity across urban areas within individual provinces (Tarasuk and Mitchell, 2020), mounting evidence such as ours that diversity at the community level also exists in local systems, impacting access to affordable and healthy food (Lake and Townshend, 2006;Rodriquez et al., 2016). Dubé Vol. 8 No. 2, August 2021 258 Going one step further than current implementation of research, convergence-by-design recognizes actors within and across disciplines and sectors need true interdisciplinarity, what the Canada Foundation for Innovation (2019) calls "convergence"-"the deep integration of disciplines, knowledge, theories, methods, data and communities" and the "deepening collaboration between researchers and research organizations in academia, the private sector and government and non-governmental organizations" to tackle complex problems Dubé et al., 2014a;Dubé et al., 2014b;Dubé, Lencucha and Drager, 2019;Dubé et al., 2020;Dubé, Pingali and Webb, 2012;Hammond and Dubé, 2012). Convergence thinking and practice demands person-in-systems thinking to identify the range of factors which are likely to facilitate the design, administration, and adaptation to the actors and contexts of each community. This next generation approach may be important in order to fully account for and respect such bottom-up energy while supporting and embedding these into the whole-of-society efforts to address the many grand challenges tied to food systems. Moving in this direction, a transformative innovation policy paradigm (Diercks et al., 2019) is progressively emerging to better account for the fact that such complex challenges concern all functional sectors of society and the economy, and that they take place not only at national but also at local, state, as well as global levels.