1. Collective management of natural resources and agriculture

For as long as people and cultures have managed natural resources, collective action has produced systems of efficient and effective offtake, as well as offering potential for sustaining natural capital and valued flows of ecosystem services (FAO, 2016b; Folke et al., Reference Folke, Carpenter, Walker, Scheffer, Chapin and Rockström2010; Kelly, Reference Kelly1995; King, Reference King1911; Li Wenhua, Reference Li2001). A wide range of different types of more sustainable agriculture and land management have recently been developed and implemented, most centring on the notion that making more of existing land by sustainable intensification and collective action can result in greater and synergistic co-production of food and ecosystem services (Benton, Reference Benton, Pritchard, Ortiz and Shekar2015; FAO, 2016c; Foresight, 2011; Maréchal et al., Reference Maréchal, Baldock, Hart, Erjavec, Rac, Vanni and Mantino2018; Pretty et al., Reference Pretty, Benton, Bharucha, Dicks, Butler Flora, Hartley and Wratten2018). Yet at the same time, agriculture and land management have also contributed to biodiversity loss, nutrient loading of the biosphere, climate forcing, depletion of aquifers and surface water and pollution of air, soil and water (IPBES, Reference Díaz, Settele, Brondízio, Ngo, Guèze, Agard and Zayas2019; Rockström et al., Reference Rockström, Steffen, Noone, Persson, Chapin, Lambin and Foley2009, Reference Rockström, Williams, Daily, Noble, Matthews, Gordon and Smith2017).

Humans have a long history of developing regimes and rules in both hunter–gatherer–forager and agricultural communities to protect and preserve natural resources in a steady state (Berkes, Reference Berkes2020; Cummings et al., Reference Cummings, Jordan and Zvelebil2014; Denevan, Reference Denevan2001; Kelly, Reference Kelly1995). These diverse and location-specific rule systems form informal institutional frameworks within communities, legitimated by shared values. These social frameworks have regulated the use of private and common property throughout history, such as by defining access rights and appropriate behaviours (Ostrom, Reference Ostrom1990). Where these systems are robust, they can maintain productivity and diversity without the need for external legal enforcement: compliance derives from shared values and internal rules and obligations (Agarwal, Reference Agarwal2018; Bagadion & Korten, Reference Bagadion, Korten and Cernea1991; Gunderson & Holling, Reference Gunderson and Holling2002). In some agricultural systems, there is evidence that social structures have sustainably governed resource use over millennia, such as subak irrigation groups in Bali (Yekti et al., Reference Yekti, Schultz, Norken, Gany and Hayde2017) and irrigation tank groups in Tamil Nadu (Mosse, Reference Mosse1992). Elsewhere, the structure of farms in landscapes has been shown to shape wider social and political participation, such as in the classic study of small and large farmed communities in California (Goldschmidt, Reference Goldschmidt1946, Reference Goldschmidt1978; Lobao, Reference Lobao1990): social connectedness, trust and participation in community life were greater when farm size was smaller.

However, many of these inherited and legacy institutions have been undermined by choices made by the modern agricultural political economy: social institutions have been ignored, co-opted, undermined and deliberately broken (Cernea, Reference Cernea1991; Wade, Reference Wade1989). The emergence of neoliberal forms of economic development prioritized the competitive choices and actions of individuals rather than cooperation (Dorling, Reference Dorling2020; Uphoff, Reference Uphoff1992) and framed the approaches to technology adopted during the green revolution (Conway & Barbier, Reference Conway and Barbier1990). In some cases, state institutions were imposed on farmers as the price for obtaining modern varieties, fertilizers and pesticides, such as in Malaysia and the Philippines (Palmer, Reference Palmer1976); in others, local institutions lost power and withered, such as kokwet water systems in Kenya (Huxley, Reference Huxley1960), warabandi in Pakistan (Bandaragoda, Reference Bandaragoda1998) and common property resources in India (Jodha, Reference Jodha1990). The collapse of institutions allowed over-extraction through the unfettered actions of individuals (e.g., of groundwater in Gujarat; Shah, Reference Shah1990). Empty and paper institutions were also formed by states without local participation, such as for grazing in China (Ho, Reference Ho2016) or irrigation in Thailand (Ricks, Reference Ricks2015).

Further changes to the social structures of communities were fostered by the conditional policies of structural adjustment adopted by international finance institutions from the 1970s and 1980s resulting in the destruction of public institutions (Crisp & Kelly, Reference Crisp and Kelly1999; Forster et al., Reference Forster, Kentikelenis, Reinsberg, Stubbs and King2019), and by the adoption of the Training and Visit (T&V) system of agricultural extension (Benor et al., Reference Benor, Harrison and Baxter1984). The T&V system was built on a linear diffusion model (or transfer of technology), first implemented on the recommendation of the World Bank in 1967, and resulting in disbursement to 512 projects valued at US$3 billion over 1977–1992. Structural adjustment brought free-market policies to 135 countries between 1980 and 2014, causing severe impacts on inequality (Forster et al., Reference Forster, Kentikelenis, Reinsberg, Stubbs and King2019). At the same time, forestry management had also become centralized into state and private enterprises that took little account of existing cultural institutions and norms of co-management (FAO, 2016b). This era has been called the height of the “Capitalocene” (Haraway, Reference Haraway2015; Moore, Reference Moore2018).

Concerns over the cost of ignoring local institutions and group approaches emerged, with project evaluations showing that the creation of farmer and rural institutions led both to sustained performance after project completion and to more efficient and fair use of natural resources (Cernea, Reference Cernea1987; Ostrom, Reference Ostrom1990; Uphoff, Reference Uphoff1992). New forms of participatory inquiry and systems of collective learning and action were field-tested, putting farmer knowledge and capacity to experiment at the centre of practices for improvement (Chambers, Reference Chambers1989; FAO, 2019; Pretty, Reference Pretty1995). By the mid-1990s, the linear diffusion model was increasingly seen as ineffective: non-adopters had been termed laggards, extension staff had become poorly motivated and research systems had been prevented from becoming learning systems (Antholt, Reference Antholt1994).

Since then, a wide range of new forms of social organization have been intentionally formed to support transformations in agricultural landscapes. These have sought to build political strength for land rights, to protect against resource extraction, to increase market strength and power (such as through formal cooperatives), to link farmers and consumers through food chains and to re-establish forms of co-management for natural resources (Berkes, Reference Berkes2020; Ostrom, Reference Ostrom1990). All of these structures are forms of social capital (Coleman, Reference Coleman1990), in which it is recognized that personal relations of trust, reciprocity and mutual obligation can result in actions and change that benefit larger numbers of people and farmers, particularly those ignored or disadvantaged by past forms of development. In some cases, these have been supported by novel public policies that reversed decades of state control by devolving decisions to local communities (Bawden, Reference Bawden, Wright, Camden-Pratt and Hill2011; FAO, 2016a; Rahman, Reference Rahman2019); in others, they have been organized to prevent the actions of the state (Veltmeyer, Reference Veltmeyer2019).

Here, we assess the emergence of social groups within particular geographical territories, with a focus on group-based learning and co-management for integrated pest management (IPM); forest, watershed and irrigation; and groups and platforms for microfinance, innovation and direct connections with consumers. We seek to address two key research questions: to what extent have efforts to form social groups for agriculture and land management within defined territories resulted in the formation of persistent collective groups? And do the worldwide numbers indicate improved possibilities for transitions towards sustainable agricultural development that will lead to improved outcomes for farmers and the environment?

2. The emergence of social groups in support of sustainability

A wide range of advances in agricultural and land sustainability have been made in the past two decades, with a range of terminologies and priorities. These include calls for a doubly green revolution (Conway, Reference Conway1997), for alternative agriculture (NRC, 1989), for an evergreen revolution (Swaminathan, Reference Swaminathan2000), for agroecological intensification (Garbach et al., Reference Garbach, Milder, DeClerck, Montenegro de Wit, Driscoll and Gemmill-Herren2017) and for agroecological movements (Giraldo & Rosset, Reference Giraldo and Rosset2018), for evergreen agriculture (Garrity et al., Reference Garrity, Akinnifesi, Ajayi, Weldesemayat, Mowo, Kalinganire, Larwanou and Bayala2010), for save and grow agriculture (FAO, 2011, 2016c) and for sustainable intensification (Godfray et al., Reference Godfray, Beddington, Crute, Haddad, Lawrence, Muir and Toulmin2010; Smith, Reference Smith2013). All of these have in common a desire to optimize the use of natural, social, human and financial capital while also being vigilant about the direct effects of agricultural and land management practices on these assets. Sustainable forms of management thus seek to use and enhance these capitals and reduce the costs of externalities on ecosystems and human health. Most of these also emphasize outcomes applying to any size of enterprise and not predetermining technologies, production type or particular design components (Weltin et al., Reference Weltin, Zasada, Piorr, Debolini, Geniaux, Perez and Schulp2018).

A recent global assessment of sustainable intensification (Pretty et al., Reference Pretty, Benton, Bharucha, Dicks, Butler Flora, Hartley and Wratten2018) indicated that systems of agricultural management undergoing fundamental redesign produce beneficial outcomes over sustained periods of time across differing ecological, economic, social and political landscapes. Redesign is, however, as much a social and institutional challenge as it is a technical one (Gliessman & Rosemeyer, Reference Gliessman and Rosemeyer2009), as there is a need to create and make productive use of human capital in the form of knowledge and capacity to adapt and innovate and of social capital to promote landscape-scale change, such as for positive contributions to biodiversity, water quantity and quality, pest management and climate change mitigation. As ecological, climatic and economic conditions change and as knowledge evolves, so must the capacity of farmers and communities improve to allow them to drive transitions through processes of collective social learning (Hill, Reference Hill1985).

Social capital has become a term used to describe the importance of social bonds, trust and reciprocity and collective action through institutions (Putnam, Reference Putnam1995). It was defined by Coleman as “the structure of relations between actors and among actors” that encourages productive activities (Coleman, Reference Coleman1990); by Bourdieu (Reference Bourdieu, Bourdieu and Richardson1986) as “a durable network of institutional relationships of mutual acquaintance and recognition … to membership of a group, which provides each of its members with the backing of collectively-owned capital”; and by Bhandari and Yasunobu (Reference Bhandari and Yasunobu2009) as a “multifunctional phenomenon comprising stocks of social norms, values, beliefs, trusts, obligations, relationships, friends, memberships, civic engagement, information flows and institutions that further cooperation and collective action for mutual benefits.”

These aspects of social infrastructure act as resources for individuals to realize personal and community interests. As social capital lowers the costs of working together, it should facilitate cooperation. Individuals have the confidence and the means to invest in collective activities, knowing that others will do so too. They are also less likely to engage in unfettered private actions that result in resource degradation, though this is no guarantee that tragedies of the commons will not occur (Wade, Reference Wade1989). Social capital can also have a ‘dark side’, with exclusion and elite capture resulting in non-democratic outcomes for some (Putzel, Reference Putzel1997; Reddy & Reddy, Reference Reddy and Reddy2005; Verma et al., Reference Verma, Sonkar, Kumar and Roy2019). It may also be deployed deliberately to offset the existing structures of states and international institutions (Forssell & Lankowski, Reference Forssell and Lankoski2015). The literature emphasizes the importance of building relations of trust, reciprocity and exchange, agreeing common rules and sanctions and developing connectedness through groups (Pretty, Reference Pretty2003; Veltmeyer, Reference Veltmeyer2019).

Social capital is thus generally seen as a prerequisite for the sustainable management of resources and for the development of approaches and methods across all geographical territories (Agarwal, Reference Agarwal2018; Leisher et al., Reference Leisher, Temsah, Booker, Day, Samberg, Prosnitz and Sunderland2016; Waddington et al., Reference Waddington, Snilstveit, Hombrados, Vojtkova, Phillips, Davies and White2014). It does not, however, guarantee sustainable outcomes. It is common for fishing communities to want to believe that fish stocks are not being eroded, even though the evidence might indicate otherwise. Not all farmers know that the application of insecticides harms populations of beneficial natural enemies. In The Netherlands, farmers recently organized a backlash to demonstrate against environmental objectives and the addressing of climate change by farm policy (van der Ploeg, Reference van der Ploeg2020). It is also true that not all transformation towards sustainability requires the formation of local social capital: a simple intervention heuristic in Vietnam (“no-spray in first 40 days of rice cultivation”) contributed to farmers cutting pesticide use on rice (Escalada & Heong, Reference Escalada and Heong2004); and the aerial releases of parasitoid wasps (Anagyrus lopezi) in Africa to control cassava mealybug did not require active farmer involvement (Wyckhuys et al., Reference Wyckhuys, Wongtiem, Rauf, Thancharoen, Heimpel, Le and Palao2018).

Many forms of social capital have emerged in support of transitions towards greater sustainability and equity. These include transnational farmer movements, such as La Vía Campesina with 200 million families represented worldwide (Martínez-Torres & Rosset, Reference Martínez-Torres and Rosset2014), national land rights and anti-land-grab movements, such as Movimento dos Trabalhadores Rurais Sen Terra and the resettlement of 0.37 million families on 7.5 Mha over 10 years (Veltmeyer, Reference Veltmeyer2019), national rural unions (Welch & Sauer, Reference Welch and Sauer2015) and agroecology and social movements (Veltmeyer, Reference Veltmeyer2019). In some cases, these have led to active conflict and ‘peasant wars’ (Giraldo & Rosset, Reference Giraldo and Rosset2018; Levien et al., Reference Levien, Watts and Yan2018). At the same time, organization around food has advanced in the form of food sovereignty and justice movements (Edelman et al., Reference Edelman, Weis, Baviskar, Borras, Holt-Giménez, Kandiyoti and Wolford2014; McMichael, Reference McMichael2013) and alternative food networks (AFNs) and alternative food movements (AFMs), particularly from urban food production landscapes, and with many involving consumers as well as growers/farmers (Desmarais & Wittman, Reference Desmarais and Wittman2014; Forssell & Lankowski, Reference Forssell and Lankoski2015; Hoey & Sponseller, Reference Hoey and Sponseller2018; Plieninger et al., Reference Plieninger, Kohsaka, Bieling, Hashimoto, Kamiyama, Kizos and Saito2018; Saulters et al., Reference Saulters, Hendrickson and Chaddad2018; Si et al., Reference Si, Schumilas and Scott2015).

Our focus here is on a subset of this social capital, specifically social groups within defined geographical territories (Flora & Delaney, Reference Flora, Delaney, Sjöblom, Andersson, Marsden and Skerratt2012; Ostrom, Reference Ostrom1990). We use numbers of established groups as a proxy for social capital within communities, as each provides the context for innovation, negotiation and experimentation, bringing together individuals with different skills and knowledge. Such groups also require forms of engagement by professionals (researchers, extensionists, advisers) largely different from those dominant in the previous era of transfer of technology.

The concept of system redesign implies the establishment of new knowledge economies for agriculture and land (MacMillan & Benton, Reference MacMillan and Benton2014). It is clear that the technologies and practices increasingly exist to provide both positive food and ecosystem outcomes: new knowledge needs to be co-created and deployed in an interconnected fashion, with an emphasis on ecological and technological innovation (Willyard et al., Reference Willyard, Scudellari and Nordling2018). There have been many adaptations in terminology for these systems of co-learning: farmer field schools (FFSs), learning labs, science and technology backyard platforms (STBs), science field shops, junior life schools, innovation platforms, farmer-led councils, agro-ecosystem networks, farmer cluster networks, joint liability groups, landcare groups and epistemic communities. What is common to these social innovations has been an understanding that individual farmers, scientists, advisors and extensionists also undertake a transformative journey. Their worldviews are challenged and change, resulting in the formation of broader epistemic communities of common interest (Norgaard, Reference Norgaard2004) that utilize, synthesize and apply knowledge and skills from many sources. For sustainable outcomes, cognitive social capital in the form of beliefs and worldviews also changes.

3. Assessment methods

For this assessment of territory-based social groups, we have analysed agriculture and natural resource systems worldwide, drawing upon both published literature and the knowledge and networks of the co-authors. We searched online research platforms for published literature in Agricola (USDA National Library; https://agricola.nal.usda.gov), Agris (UN FAO; http://agris.fao.org/agris-search/index.do), CAB Abstracts (https://www.cabi.org/products-and-services/publishing-product/online-resources), Google Scholar (https://scholar.google.com) and Google (https://www.google.co.uk), Scopus (https://www.scopus.com/home.uri) and Web of Science (https://clarivate.com/webofsciencegroup/solutions/web-of-science) for published records over the past 10 years, and we drew on the collective knowledge of the assessment team and their personal contacts regarding further unpublished material from government and non-government initiatives. Projects and programmes in all countries were eligible. We selected terms for searches drawn from our knowledge of programmes in the field: social + capital; sustainable + agriculture; sustainable + intensification; joint/participatory + forest + management; agroforestry + groups; integrated + pest + management; farmer + field + schools; watershed + management; conservation + agriculture; irrigation + management; water + user + groups; pasture/grazing + management; microcredit/microfinance + groups; innovation + platforms; participatory + methods/approaches; farmer + organisations/institutions. A number of international analyses were drawn upon (e.g., of farmer fields schools: FAO, 2019; van den Berg et al., Reference van den Berg, Phillips, Dicke and Fredrix2020b; of community forestry: FAO, 2016b).

We organized the findings into eight functional categories of redesign, each with different types of enabling social intervention (Table 1). We report on data gathered from 122 initiatives in 55 countries across 6 continental regions, and we have applied no lower limit to the number of groups reported per initiative. Of the 55 countries, 13 were in industrialized countries and 42 in less-developed countries.Footnote ⁱ

Table 1. Eight categories of social group interventions for sustainable agriculture and land management.

Note: Previous assessments of social capital used 5 categories (mapping here onto (1)–(4) and (6)) (Pretty & Ward, Reference Pretty and Ward2001). A global assessment of sustainable intensification used seven categories (mapping here onto (1)–(5), (7) and (8)) (Pretty et al., Reference Pretty, Benton, Bharucha, Dicks, Butler Flora, Hartley and Wratten2018).

We gathered data on numbers of social groups, numbers of farmer members and numbers of hectares under interventions for co-management. We have excluded data on groups where there is evidence of misreporting or the phenomenon of paper or empty groups – those reported to meet political targets but that do not exist on the ground (Ho, Reference Ho2016; Ravindranath & Sudha, Reference Ravindranath and Sudha2004; Ricks, Reference Ricks2016; Ricks & Arif, Reference Ricks and Arif2012). An unintended outcome of positive policy support for group formation has led to some inaccurate reporting to meet targets in some locations, such as for participatory irrigation management and forest co-management. We have also not made assumptions about intended adoption: for example, an EU Directive (2019) now requires all farms to use IPM, but preparations for implementation have not yet led to significant uptake of agricultural practices that significantly benefit ecosystem services (Buckwell et al., Reference Buckwell, Nordang Uhre, Williams, Polakova, Blum, Schiefer and Haber2014). We have not included reporting of area of land under co-management where group data do not exist, such as for large tracts of forest now under community-based forestry (FAO, 2016b).

As indicated above, we also did not include here analysis of non-territorial social capital in the form of cooperatives, farmer organizations, federations and movements. Many rural and agricultural cooperatives (focusing on milk processing, input supply, collective marketing and sales) are not geographically based. In the USA, there are 2047 rural cooperatives (though down from 10,040 in 1950) (USDA, 2018), in Brazil there are 1620 (with 1 million members) (Dias & Teles, Reference Dias and Teles2018), in China there are 2.2 million (MARA, 2019) and in the UK there are 420 rural cooperatives (6% of UK farm market share by value; market shares by farm cooperatives are higher in The Netherlands (68%), France (55%) and Spain (45%)) (Cooperatives UK, 2018). In a number of countries, dispersed clusters of farmers collaborate on common research interests, but they may not result in natural capital improvements within whole territories. In India, there are >3000 farmer producer organizations with 0.3 million members, but these tend not to be geographically based (SOIL, 2017; Verma et al., Reference Verma, Sonkar, Kumar and Roy2019). Federations, food networks and movements are also effective at transforming agriculture and food systems across and within countries, working at both national and international levels, such as in Canada (Desmarais & Wittman, Reference Desmarais and Wittman2014; Levkoe, Reference Levkoe2014), in the European Union (EU) and in Japan (Plieninger et al., Reference Plieninger, Kohsaka, Bieling, Hashimoto, Kamiyama, Kizos and Saito2018). La Vía Campesina works across 81 countries (182 member groups; La Vía Campesina, 2020) and the International Partnership for the Satoyama Initiative works across 71 countries (258 member groups; Kozar et al., Reference Kozar, Galang, Alip, Sedhain, Subramanian and Saito2019). These forms of social capital are not included in this assessment, but they comprise a rich field of study for further work on transitions towards sustainability in agriculture and land management.

Layers of federations can be important in ensuring local priorities reach upwards to influence policy and practice. In Andhra Pradesh, for example, the 830,000 women's self-help groups (SHGs) are organized into village-level federations (of 15–30 SHGs) each, and these are organized into distinct federations of 40–60 village organizations (Bharucha et al., Reference Bharucha, Mitjans and Pretty2020; Kumar, Reference Kumar2017). In Japan, 1000 teikei purchasing groups are linked to organic and natural farming and have organized into federations, with some leaders coming to be elected as members of parliament (iPES-Food, 2016; Kondoh, Reference Kondoh2015).

This assessment of social group formation also does not imply that numbers of farms and hectares are fixed: on the ground, there will be a flux in numbers resulting from both adoption and dis-adoption. This may arise from farmer choice and agency, but equally from the actions of vested interests, input companies, the consolidation of small farms into larger operations, changes in agricultural policy or shifts in market demand and discrepancies between on-paper claims and what farmers have implemented. We have excluded data on groups formed during the assessment period but since abandoned by changes in development assistance funding and/or national priorities. In the 1990s, for example, 4500 catchment groups were formed in Kenya by the Ministry of Agriculture; the catchment approach ended in 2000 and evolved into a National Agriculture and Livestock Extension Programme with common interest groups, but since then the focus has changed away from these groups (Pretty et al., Reference Pretty, Toulmin and Williams2011). In Southeast Asia, some recorded participatory irrigation groups later also become ineffective (Ricks, Reference Ricks2015, Reference Ricks2016).

4. Outcomes: group numbers

We organized the findings into eight categories of agricultural and land management intervention that are contributing to the emergence of new knowledge economies (Table 1). Across the eight categories and 122 distinct initiatives, we recorded 8.54 million intentionally formed social groups worldwide. These comprise groups collectively managing 299 Mha of agricultural and non-agricultural land. This represents growth in these types of groups from 0.005 million at the end of the 1980s (primarily in participatory irrigation management) to 0.48 million in 2001 (Pretty, Reference Pretty2003; Pretty & Ward, Reference Pretty and Ward2001), and now to 8.54 million by 2020 (exponential fit: R = 0.982). Figure 1 shows the marginal increase between 2000 and 2020 in the groups in each of the eight categories.

Fig. 1. Increase in numbers of social groups in eight categories of sustainable agriculture and land management (2000–2020).

Most social groups surveyed contain 20–25 members (range 15–30), with occasional numbers progressing towards 100 (e.g., for microfinance groups). Group sizes have remained constant over time. Small groups provide for more effective communicative interactions and permit the agreement of common goals and practices, the sharing of planning and evaluation and the agreement with norms and rules that work for all. As groups progress to larger sizes, they are generally more effective if divided and then federated. Small groups of approximately 25 members are generally able to survive with the presence of small numbers of free-riders (those that do not actively contribute to collective outcomes but benefit from these outcomes) (Dannenberg et al., Reference Dannenberg, Löschel, Paolacci, Reif and Tavoni2015). Using the mean membership of 20–30 people per group, this assessment suggests that there are 170–255 million members of social capital groups providing both private and public benefits. Though not all of these are farmers (e.g., non-landowners with rights to jointly managed forests), a midpoint (assuming membership of 25 per group) represents 39% of all 570 million farms worldwide (Lowder et al., Reference Lowder, Skoet and Raney2016). The distributions of groups and areas are shown in Table 2 and Figures 2 and 3. The majority of groups have been formed in less-developed countries (98.2%), as is the majority of the area (93.6%). The distribution of groups across six continental regions is shown in Figure 2.

Fig. 2. Social groups formed in categories of agricultural and landscape redesign (122 initiatives, 55 countries): disc area = Mha.

Fig. 3. Social groups formed across six continental regions (122 initiatives, 55 countries): disc area = Mha.

Table 2. Social groups and land area across eight categories of agriculture and land management (2020).

5. Outcomes: documented impacts

Across all of the literature, there is considerable evidence of improvements within agroecosystems, landscapes and farm household economies as a result of the formation of intentional social capital within geographical areas (Agarwal, Reference Agarwal2018; Cernea, Reference Cernea1991; Leisher et al., Reference Leisher, Temsah, Booker, Day, Samberg, Prosnitz and Sunderland2016; Ostrom, Reference Ostrom1990; Uphoff, Reference Uphoff1992; Waddington et al., Reference Waddington, Snilstveit, Hombrados, Vojtkova, Phillips, Davies and White2014). Comparisons have been made between projects/programmes working with groups compared with individuals, with prior degraded natural systems (e.g., forest or eroded landscapes), with prior agroecosystems harmed by compounds used in agriculture (e.g., harmful pesticides) and with agroecosystems with legacies of low productivity (e.g., that have not seen productivity improvements in recent decades).

The overwhelming evidence from the field and reported in the published literature is that collective management of resources can lead to redesign and also result in net increases in system productivity. There have been few counterfactual examples, such as where groups could have been formed to increase resource exploitation or extraction (e.g., water or forest capture). In Malawi, for example, where village management committees were imposed without taking account of existing institutional arrangements for resource management, this resulted in the clearance of trees, heightened conflict within communities and the destruction of existing institutions (Kamoto et al., Reference Kamoto, Clarkson, Dorward and Shepherd2013). There are also other institutions, corporations and groups of individuals engaged in resource depletion to serve private and generally short-term interests.

It is also clear that any social capital established in the form of groups can lead to suboptimal outcomes for certain population subgroups. By definition, groups comprise members, and those outside of the groups may be excluded from the benefits of membership. This phenomenon of ‘the dark side of social capital’ (Coleman, Reference Coleman1990; Putzel, Reference Putzel1997) has seen both elite capture (the already wealthy or more powerful individuals using groups to strengthen personal benefit at the expense of others), exclusion (group membership restricted to only some members of a population or location) and negative selection (where individuals are actively excluded). Nonetheless, the majority of the literature points to the benefits of social capital to (1) individuals and groups/communities, (2) agricultural systems and (3) wider landscapes and ecosystem services.

1. (1) To individuals and groups/communities: evidence of changes to personal capabilities and growth, to worldviews and locally generated resource availability:

   • • Emergence of new leaders of groups, especially by women (Agarwal, Reference Agarwal2018), and changes in the relationships between women and men (Westerman et al., Reference Westerman, Ashby and Pretty2005).

   • • Positive role of women leaders in group effectiveness and conflict resolution over common resources (Coleman & Mwangi, Reference Coleman and Mwangi2013).

   • • Changes in the worldviews of farmers (Campbell et al., Reference Campbell, Alexandra and Curtis2017; van den Berg et al., Reference van den Berg, Phillips, Dicke and Fredrix2020b) and of scientists and extensionists working with farmers in novel innovation platforms (Zhang et al., Reference Zhang, Cao, Li, Zhang, Wang, Liu and Mi2016).

   • • Increases in the savings and repayment rates of members of microfinance groups (BRAC, 2019; Rahman, Reference Rahman2019).

2. (2) To agricultural system net farm productivity: evidence of increased system outputs and reduced input needs:

   • • Increases in crop productivity, such as by FFSs on all crops (Chhay et al., Reference Chhay, Seng, Tanaka, Yamauchi, Cedicol, Kawakita and Chiba2017; FAO, 2019), and in grazing and pasture productivity (NRC, 2010).

   • • Increases in tree and agroforestry cover on farms (Bunch, Reference Bunch2018; Garrity et al., Reference Garrity, Akinnifesi, Ajayi, Weldesemayat, Mowo, Kalinganire, Larwanou and Bayala2010; Reij et al., Reference Reij and Smaling2008).

   • • Reductions in the use of pesticides in IPM (Pretty & Bharucha, Reference Pretty and Bharucha2015; Yang et al., Reference Yang, Zhao and Shen2014).

   • • Adoption of organic and zero-budget systems (Bharucha et al., Reference Bharucha, Mitjans and Pretty2020; Reganold & Wachter, Reference Reganold and Wachter2016).

3. (3) To natural capital and key ecosystem services: evidence of increased productivity and reductions in the use of harmful or potentially harmful compounds and releases:

   • • Increases in irrigation water availability and efficiency of use (Ricks, Reference Ricks2016; Zhou et al., Reference Zhou, Deng, Wu, Li and Song2017).

   • • Improvements in forest productivity of wood, forage and secondary products (FAO, 2016b; Ravindranath & Sudha, Reference Ravindranath and Sudha2004).

   • • Increases in carbon sequestration in soils by conservation agriculture (CA) (FAO, 2011; Lal, Reference Lal and Dent2014).

   • • Reductions in surface water flows and soil erosion (Reij et al., Reference Reij and Smaling2008).

6. Key findings for the eight functional categories

7. Securing sustainability

We have shown that over the past two decades a variety of novel social infrastructure has created platforms for collective transitions towards greater sustainability of agriculture and land management amongst rural communities across the world. These have led to greater flows of knowledge and technologies and built trust amongst individuals and agencies. The cumulative increase in numbers of social groups from 0.5 million to 8.5 million over two decades implies that there have been transformations in capacity and personal benefit combined with improved environmental outcomes for agricultural landscapes.

The marked difference in implementation and uptake of social capital between industrialized and less-developed countries is striking. In industrialized countries, farmers have also tended to be self-organized into value chain-based groups of common interest rather than into groups within specific geographical territories, though this could change with growing interest in policy support for landscape-scale change to deliver public goods, such as in the UK's 25 Year Environmental Plan (Defra, 2019). The latter needs facilitation and support, a particular challenge where investments in extension are small or where public extension systems no longer exist. Nonetheless, where geographically based groups are formed, both productivity and natural capital outcomes can be substantial (Maréchal et al., Reference Maréchal, Baldock, Hart, Erjavec, Rac, Vanni and Mantino2018).

This shift towards sustainable redesign in agriculture and land management has been successful where individual worldviews have changed, emerging from the processes of co-production embedded in groups. Many programmes have built on the principles of adult learning, social ecology, liberation education and epistemic change. Social capital can provide a supportive context for transformations, both in practices (behaviours and choices) and personally (the inner journey) (Bawden, Reference Bawden, Wright, Camden-Pratt and Hill2011; Norgaard, Reference Norgaard2004). Through experiences in the world, each person comes to see and know it from a particular epistemic position that reflects a set of assumptions about reality. These assumptions and worldviews shape the way each person chooses to act and behave. Such epistemic change is called for in turbulent times, as there is an inseparable interconnection between cognition and action (Fear et al., Reference Fear, Rosean, Bawden and Foster-Fishman2006). It has previously been argued that social groups, movements and campaigns comprise an ‘immune system’ for the planet (Hawken, Reference Hawken2007), in that they offer platforms for collective action and larger-scale action towards greater sustainability and equity.

Social media and mobile platforms for information will play complementary roles in information access and exchanges, as well as in helping to keep people connected (FAO, 2019). The term ‘sustainable’ suggests an incorporation of the need for improvement (e.g., to well-being, food production and natural capital), and thus it requires the need to change the way individuals think about and come to know about the world (Norgaard, Reference Norgaard2004). To date, epistemic communities and networks of social capital have been established in many locations and could build distributed expertise and trust over time (de Bruijn & Gerrits, Reference de Bruijn and Gerrits2018; Granjou & Arpin, Reference Granjou and Arpin2015), particularly where there is a greater number and diversity of engaged actors (Grêt-Regamany et al., Reference Grêt-Regamany, Huber and Huber2019; Hazard et al., Reference Hazard, Steyaert, Martin, Couix, Navas, Duru, Lauvie and Labatut2018). Social capital in a variety of forms could help to open up science to innovation, particularly where problems are complex and solutions unknown and where the values of all actors are salient (Richardson et al., Reference Richardson, Durose and Perry2018).

Nonetheless, there will be constraints and countervailing pressures. Land tenure and secure rights are preconditions to local people making long-term investments in natural and social assets. When Burkina Faso and Niger granted rights to individuals to use their own trees as they wished (mid-1980s), this resulted in an increase in tree cover as there were now incentives for the long-term rather immediate resource extraction (Godfray et al., Reference Godfray, Beddington, Crute, Haddad, Lawrence, Muir and Toulmin2010; Sendzimir et al., Reference Sendzimir, Reij and Magnuszewski2011).

Though state and international organizations have contributed to the advance in numbers of social groups, the evidence for positive actions from the private sector is weak, with many not matching up to their own statements on corporate social responsibility (Elder & Dauvergne, Reference Elder and Dauvergne2015). There is evidence (from the field) that pesticide companies have promoted and run FFSs precisely to sell more product, and thus not to develop forms of IPM that reduce negative impacts on the environment and human health. It has been argued that big retail power will not be interested in social groups in rural areas unless they serve their own purposes, and thus that the ‘dance of the supply chain’ (Freidberg, Reference Freidberg2020) cannot produce progressive outcomes. At the same time, apparent inflows of foreign direct investment into poorer countries may continue to lead to outflows of capital in profits and returns on investments (Veltmeyer, Reference Veltmeyer2019).

We have found that social groups have emerged from both government and non-government contexts. Some have required critical changes to policy or regulation, often being more effective at the state or district level. Changes to water rights allowed for the emergence and spread of participatory irrigation; changes to forest and tree use rights were essential platforms for joint and participatory forest management programmes to be established and devolve decision-making to local people; changes to lending assumptions allowed banks to lend to NGOs and social groups, which then provided security rather than individuals. Further policy changes and support will be important to help these projects spread. Extension systems, for example, will need to adopt more cooperative models rather than seeking only to work with compliant individuals (Clark et al., Reference Clark, Bean, Raja, Loveridge, Freedgood and Hodgson2017).

It will be important to be mindful of the past failures of state organizations that have undermined, ignored or suffocated local resource-based institutions (Levien et al., Reference Levien, Watts and Yan2018; Jodha, Reference Jodha1990; Palmer, Reference Palmer1976) or have created paper or empty institutions (Ho, Reference Ho2016). Many members and activists in social and agroecological movements would also argue that it is the structures of the world economy (and its capitalism) that prevent effective transformations towards sustainability and equity (Giraldo & Rosset, Reference Giraldo and Rosset2018; Moore, Reference Moore2018). Nonetheless, as social groups federate into higher-level structures, they too are able to act to deliver greater agricultural and natural capital benefits, as well as returns to farmers (Kondoh, Reference Kondoh2015). Though we were unable to gather data on the gender mix between and within social groups, it is clear that mixed groups of women and men are more effective (in terms of farm and/or forest productivity) than single-gender groups, and groups of women are more effective than groups just consisting of men (Agarwal, Reference Agarwal2018; Leisher et al., Reference Leisher, Temsah, Booker, Day, Samberg, Prosnitz and Sunderland2016; Westerman et al., Reference Westerman, Ashby and Pretty2005). Programmes seeking to form social groups will thus need to be aware of how to ensure full and proper participation by women.

A separate but important evidence base points towards the health and well-being effects of the greater trust and reciprocity that inhere when social capital is high. Social capital is known to have positive effects on well-being (Holt-Lunstad et al., Reference Holt-Lunstad, Robles and Sbarra2017) and on life satisfaction and longevity (Graton & Scott, Reference Graton and Scott2016; Layard, Reference Layard2020). Though not part of the recorded benefits of these social groups, it can reasonably be assumed that members will be receiving personal benefits over and above the functional improvements to farm productivity and income. The socially connected live longer and are happier (Holt-Lunstad et al., Reference Holt-Lunstad, Robles and Sbarra2017), and countries with higher levels of trust in other people are happier (WHR, Reference Helliwell, Layard and Sachs2019, Reference Helliwell, Layard, Sachs and De Neve2020). Volunteers who contribute to the well-being of others and to the quality of lived environments tend to have healthier lifestyles, lower incidence rates of mental ill health and to live longer (Anderson et al., Reference Anderson, Damianakis, Kröger, Wagner, Dawson, Binns and Cook2014; Borgonovi, Reference Borgonovi2008; Layard, Reference Layard2020). On the other hand, net well-being across populations is reduced by growth in inequity (Wilkinson & Pickett, Reference Wilkinson and Pickett2009, 2018), breakdown of social structures and support (Piketty & Saez, 2014) and lack of access to natural and green spaces (Mitchell & Popham, Reference Mitchell and Popham2008; Mitchell et al., Reference Mitchell, Richardson, Shortt and Pearce2015).

This platform of 8.5 million social groups distributed across 55 countries, but comprising 3% of the world's population, could comprise an opportunity to consider greater challenges, such as advances towards meeting the Sustainable Development Goals and addressing climate change. If different worlds are to be brought forth as a function of a quest to transform the way we live and consume, we will need to modify the epistemes that have come to dominate modern consumption cultures (Bawden, Reference Bawden, Wright, Camden-Pratt and Hill2011). Some social capital is already influencing global systems, resulting, for example, in mitigations of climate change, biodiversity loss and air pollution, as well as increases in net food production. Platforms of groups, for example, could engage in the co-production of new patterns of material consumption and ways of living within global boundaries and limits (Dorling, Reference Dorling2020; Jackson, Reference Jackson2009; Pretty, Reference Pretty2013). Some argue that large-scale advances in sustainability and equity are impossible if capitalism and class are ignored (Levien et al., Reference Levien, Watts and Yan2018).

It is clear that considerable changes will be required worldwide to limit the advance of the climate crisis, both in individual choices and behaviours and in the policies developed by all countries. An era of ‘degrowth’ may be needed (Gerber, Reference Gerber2020), and certainly of green restructuring of economies directly to reduce material consumption and substitute it with sustainable or green alternatives (Ivanova et al., Reference Ivanova, Barrett, Wiedenhofer, Macura, Callaghan and Creutzig2020). There may be, in short, possibilities of the good life within planetary boundaries (Dorling, Reference Dorling2020; Layard, Reference Layard2020; O'Neill et al., Reference O'Neill, Fanning, Lamb and Steinberger2018). We have not analysed the political philosophies or aims of these social groups. Clearly, individuals will have many reasons for organizing and taking collective action, and given the context for these changes, it is likely that many individuals will continue to support sustainability and equity outcomes. But there is no guarantee that such values will remain unchanged.

Can these groups survive and flourish? Threats to these groups will come from external and internal sources. External sources could include major social and economic disruption (e.g., following the COIVD-19 pandemic), climate-driven forced abandonment of farms and territory, policy changes in support of land grabs and large commercial monoculture operations (e.g., for oil palm) and state support for only empty or non-credible groups. Internal disruptors could include stresses arising from benefit capture by individuals, gender imbalances in benefits and farm abandonment in favour of employment in urban areas. Nonetheless, many advantages have been found in the sharing economies of connected food systems where goods and services are pooled (Miralles et al., Reference Miralles, Dentoni and Pascucci2017), such as a more even distribution of power, increased collaborative consumption, higher trust and more efficient use of resources. Agricultural transformations will be critical in the coming years both for contributing to reducing climate forcing and for mitigating negative effects. Some have called for adventurous food futures (Carolan, Reference Carolan2016). It would appear that social groups and movements have already created opportunities for individual and collective transformations.

8. Concluding comments

This assessment has shown growth in the numbers of groups engaged in platforms of innovative and sustainable management within geographical territories of engagement over the past two decades. These groups deliver individual and public benefits, improve well-being and natural capital and provide platforms for wider progress towards sustainability. These groups provide the basis for further progressive change towards sustainable policies and behaviours, with opportunities to help mitigate the advance of some global environmental challenges. We further note this social infrastructure has already changed worldviews and capacities to redesign towards sustainability and increased net productivity of agricultural and land systems.

Attention will need to be paid to ensuring that access to groups is equitable and that there is further research on the causative links between all forms of social capital and the emergence of more sustainable practices. In a number of contexts, social groups exist only on paper to meet policy objectives, and any increase of this phenomenon will undermine the wider goals of seeking further social capital formation. At the same time, some wider political and economic structures will make the formation of social groups harder to sustain. Nonetheless, the redesign of all agricultural and land management remains a critical global challenge, and though the growth in the number of groups has been substantial, in many cases supported by novel policies and regulations within countries, more support is needed to ensure best practice is spread to aid the transitions towards more sustainable and equitable forms of farmed and managed landscapes worldwide.