1. Why reductionism makes sustainability unsustainable

Why are global efforts to achieve sustainability failing? From climate change to the ongoing sixth mass extinction, states and corporations publicly commit to often unambitious targets that are missed nevertheless (Howes et al., Reference Howes, Wortley, Potts, Dedekorkut-Howes, Serrao-Neumann, Davidson, Smith and Nunn2017). The political costs of such failures are mostly negligible. Meanwhile, internationally, parties with neoliberal growth ideologies that range from conservative to neo-fascist enjoy ballot success with programmes advertising their indifference to problems that researchers show to threaten all life on earth, e.g. Trump's support for coal (USA) and Bolsonaro's goal of accelerating exploitation of the Amazon (Brazil). In this paper, we argue that the reductionist anthropocentric focus on human needs at the core of the sustainability concept limits our ability to meet those needs. In short, by reducing sustainability to a concern for human needs without acknowledging their intrinsic, complex interdependence with more-than-human needs, humanity is unable to meet the needs of the present, while also risking that future generations of all species will be unable to meet their own needs. Only a broadened concept of sustainability that includes more-than-human well-being is likely to successfully support diverse life on earth, because species' needs are inherently and irreducibly interdependent.

The current sustainability concept focuses on a perceived generational conflict, namely between those currently living and those yet to be born, rather than situating this relationship in broader contexts of intrahuman and interspecies relationships. We will focus here on the underexplored latter, although we recognize that the former kind of relationship is far from caring for all human beings, since poverty, inequality, homelessness are left out of many narratives on sustainability (Gonella, Reference Gonella2019). Although the wording in the Brundtland Report does not explicitly limit the needs of the present to human needs, this emerges from the discourse around sustainable development which is often reported as in contrast to the notion of sustainability (Latouche, Reference Latouche2004; Springett, Reference Springett2013). The report named ‘the satisfaction of human needs and aspirations’ as ‘the major objective of development’ (Brundtland, Reference Brundtland1987, p. 41). Living beings beyond the human are listed alongside soils, waters, and atmosphere, reduced to ‘natural systems that support life on earth’ (ibid., p. 42), in essence embracing a form of Cartesian dualism. Not unique to the development context, Morton (Reference Morton2007) argues more romantic notions of ‘nature’ that are tied to environmentalism also have the effect of framing nature as something ‘out there’ and engagements with it as essentially in a consumerist aesthetic mode, rather than engagement that recognizes the complex ecological entanglement between human and non-human species. The current sustainability concept thus focuses on meeting human needs, recognizes the dependence of human needs on other organisms; yet, fails to conclude that the needs of non-humans must also be met. Such reductionism of complex, interdependent ecosystem and species entanglements to a monolithic ‘other’ can be seen as one form of anthropocentrism. This term has been widely used in the environmental literature, yet its usage has been problematized (Kidner, Reference Kidner2014). Kidner asserts that the tendency of mainstream societies to focus overwhelmingly on human needs above all others owes largely to the domination exerted by industrialism and the industrial symbolic system that emerged within the specific historical frame of Western modernity. Human beings or putative ‘human nature’ lack any intrinsic greedy or individualist natural feature. On the contrary, within the long-term view of human existence, the real human centric way of thinking has tended to value a healthy natural environment and recognize the relationality among people and non-human inhabitants of the natural world. The actual tendency towards viewing the natural world in terms of resources for exploitation indicates the specific emergence of a hegemonic industrialist and imperialist human-centric world view. The critique of this paper thus aims not at sustainability's concern with human well-being, but at the reductionist attempt to satisfy human needs and aspirations without considering the associated complex and interdependent multispecies entanglements. Understanding human needs without a multispecies context is likely impossible, as has long been recognized and argued by various First Nations and Indigenous scholars and philosophers (Graham, Reference Graham1999; Little Bear, Reference Little Bear2000; Todd, Reference Todd2016; Watts, Reference Watts2013; Whyte, Reference Whyte, Heise, Christensen and Niemann2017). These insights and knowledges have been ignored and even erased by mainstream Western scientific approaches. Yet, today these are increasingly recognized as vital, their methodologies and findings are slowly being replicated and verified (Deloria, Reference Deloria2018; Woelfle-Erskine, Reference Woelfle-Erskine2019). Decades of research across humanities and natural and social sciences now point to the inseparable multispecies entanglement of human needs (Albrecht et al., Reference Albrecht, Sartore, Connor, Higginbotham, Freeman, Kelly, Stain, Tonna and Pollard2007; Flies et al., Reference Flies, Skelly, Lovell, Breed, Phillips and Weinstein2018; Hanski et al., Reference Hanski, von Hertzen, Fyhrquist, Koskinen, Torppa, Laatikainen, Karisola, Auvinen, Paulin, Mäkelä, Vartiainen, Kosunen, Alenius and Haahtela2012; Keniger et al., Reference Keniger, Gaston, Irvine and Fuller2013; Rupprecht & Byrne, Reference Rupprecht and Byrne2014; Soga & Gaston, Reference Soga and Gaston2020; Swanson et al., Reference Swanson, Lien and Ween2018; Woelfle-Erskine, Reference Woelfle-Erskine2019). Beyond direct effects on human health, strong evidence confirming complex interactions and interdependencies also mandates applying the precautionary principle in assuming that all species are indirectly linked in some ways. Even those in extreme environments such as deep sea vents are still part of tremendously complex food webs and trophic cascades (Govenar, Reference Govenar2012). Human-induced global changes to the earth system, such as climate change, ocean acidification and microplastic pollution, further make unlikely the existence of species being wholly independent and unaffected (Capra & Luisi, Reference Capra and Luisi2016; Doney et al., Reference Doney, Fabry, Feely and Kleypas2009; Hale et al., Reference Hale, Seeley, Guardia, Mai and Zeng2020; Pecl et al., Reference Pecl, Araújo, Bell, Blanchard, Bonebrake, Chen, Clark, Colwell, Danielsen, Evengård, Falconi, Ferrier, Frusher, Garcia, Griffis, Hobday, Janion-Scheepers, Jarzyna, Jennings and Williams2017; Steffen, et al., Reference Steffen, Broadgate, Deutsch, Gaffney and Ludwig2015a). The ecosystemically-related zoonotic origin of the recent Covid-19 pandemic represents another hint in this direction (Bonilla-Aldana et al., Reference Bonilla-Aldana, Dhama and Rodriguez-Morales2020). By failing to account for both the direct and indirect interspecies relations that underpin human and more-than-human well-being, the reductionist anthropocentric sustainability concept thus allows conditions vital for its success to go unmet. This becomes clear through an examination of how decisions and actions surrounding sustainability are primarily made.

Underrepresentation of actors and stakeholders in decisions and actions around sustainability issues can be identified as a leading cause of failure. Through negotiation, parties with an interest (e.g. corporations pursuing business opportunities, or residents improving their quality of life) aim to see their interests furthered or protected in the outcome. In negotiations, participation of stakeholders is seen as vital in achieving successful compromises (Hadorn et al., Reference Hadorn, Hoffmann-Riem, Biber-Klemm, Grossenbacher-Mansuy, Joye, Pohl, Wiesmann and Zemp2008). In democratic elections, the circle of those who may participate has been successively widened in terms of age, social standing, and gender (Przeworski, Reference Przeworski2009). In the context of sustainability, multi-stakeholder dialogues have been explored as new modes of governance that aim to counter democratic and implementation deficits (Bäckstrand, Reference Bäckstrand2006). We are not claiming that all of this is necessarily sincere and effective, yet an increasing participation of human actors in decision-making is at least present in debates. In political economy, the issue of the labour of living beings beyond humans, including both value derived from being alive as well as reframing beyond-human contributions as more than use value, is also receiving increasing attention by scholars (Barua, Reference Barua2017; Kallis & Swyngedouw, Reference Kallis and Swyngedouw2018). However, needs and interests of living beings beyond humans remain insufficiently represented, if at all. For example, in a scientific context, a tree only enters negotiations indirectly, as a local source of timber, an abstract tool for carbon storage and sequestration, or (less frequently) for cultural ecosystem services such as its perceived aesthetic, cultural, health-giving, or religious properties (Daniel et al., Reference Daniel, Muhar, Arnberger, Aznar, Boyd, Chan, Costanza, Elmqvist, Flint, Gobster, Gret-Regamey, Lave, Muhar, Penker, Ribe, Schauppenlehner, Sikor, Soloviy, Spierenburg and von der Dunk2012). Its recognized value is therefore wholly dependent on how useful humans perceive it to be (Davidson, Reference Davidson2013). Absent from the negotiations are the tree itself as well as the multitude of species interacting with and depending on it, from fellow plants and root-associated microbes to lichen, insects, birds, and others. Consequently, this lack of representation in the negotiation leads to outcomes skewed against the trees' interest (and that of its natural stakeholders) (Donoso, Reference Donoso2017), which in turn results in an overall unsuccessful compromise for everyone. The way actors are excluded from negotiations through their labelling as resources in the reductionist sustainability concept thus leads to a failure to represent and protect the complexity and interdependence inherent in natural systems. This is most visible in conceptual representations of the sustainability concept.

The Brundtland concept of sustainability can be divided into three parts, which comprise the three elements of mainstream visualizations (Figure 1): society (needs of the present), environment (ability of future generations to meet their own needs), and economy (the method through which needs are met). As Wu (Reference Wu2013) notes, these visualizations demonstrate how reductionist-anthropocentric sustainability-based varieties are concerned with how the three dimensions interact (triple bottom line), whether one can be substituted by the other (weak sustainability) or not (strong sustainability). Recently, such research has been further developed by breaking the environment and society aspects down into more detailed categories, a prominent example being Raworth's (Reference Raworth2017) doughnut economics model (Figure 2). The progression of these visual models allows several observations. First, the role of the economy is relegated from a separate entity in the triple bottom line model to a subset of society in strong sustainability, and reconceptualized as an interface in the doughnut model. Second, the depiction of the environment transitions towards an entity encompassing rather than existing alongside humanity. These changes arguably bring the models closer to portraying the interdependence of species, but several issues remain.

Fig. 1. Visualizations of the sustainability concept (adapted from Wu, Reference Wu2013).

Fig. 2. Doughnut economics model representing sustainability as a space for humanity that does not overshoot ecological ceilings while providing a social foundation (Raworth, Reference Raworth2017).

The foremost issue of both classic sustainability visualizations (Figure 1) and newer ones (Figure 2) is their reductionist view of the environment, and with it all life beyond humans, as a resource for exploitation and tool to achieve human ends. Strong sustainability oversimplifies its representation of the environment as it collapses all species beyond the human into one residual, binary ‘non-human’ category, simultaneously conflating them with non-biological elements such as the atmosphere, geosphere, and hydrosphere. In contrast, the doughnut model only represents the environment through the different ways it is affected by humans, borrowing from the planetary boundaries model (Steffen, et al., Reference Steffen, Richardson, Rockström, Cornell, Fetzer, Bennett, Biggs, Carpenter, de Vries, de Wit, Folke, Gerten, Heinke, Mace, Persson, Ramanathan, Reyers and Sörlin2015b). Non-humans are only referred to indirectly through biodiversity loss and food. Newer definitions of landscape sustainability reviewed by Wu (Reference Wu2013) remain resource- and human-focused. Another issue of all the visual sustainability models examined here is their tendency to hide the complexity present in a system. Just as there are many human societies with radically different social and economic organizations as well as different relationships to living beings beyond humans, living beings have been observed to vary significantly in their behaviour within species (Escobar, Reference Escobar1998, Reference Escobar2018; Kothari et al., Reference Kothari, Salleh, Escobar, Demaria and Acosta2019). Finally, the interdependence and agency of living beings remain absent in all visualizations of sustainability concepts.

2. Towards a multispecies concept of sustainability

The reductionist anthropocentric sustainability concept has dominated scientific and political discourse, but in the original meaning of the concept, many other approaches to sustaining human life exist and may be more likely to yield success. For example, the reduction of non-human beings to resources contrasts with a view of non-humans as actors with agency that cannot simply be managed as resources, but must be negotiated and compromised with as beings in their own rights, with their own needs and interests (Chapron et al., Reference Chapron, Epstein and López-Bao2019; Davies & Riach, Reference Davies and Riach2018; Johnson & Larsen, Reference Johnson and Larsen2017; Rose et al., Reference Rose, James and Watson2003). Such views can be found in societies, and belief systems all over the world are implemented in diverse forms through laws, customs, teachings, and traditions, and are often intricately linked with what has been labelled traditional ecological knowledge in the academic discourse. Although a detailed review of alternative approaches to sustainability are beyond the scope of this paper, previous studies suggest that many of these approaches have a vastly better historical track record of success in achieving sustainability (Berkes et al., Reference Berkes, Folke, Gadgil, Perrings, Mäler, Folke, Holling and Jansson1994, Reference Berkes, Colding and Folke2000; Escobar, Reference Escobar2011; Fraser et al., Reference Fraser, Frausin and Jarvis2015; Rose et al., Reference Rose, James and Watson2003). Moreover, recent conceptual advances in integrating the interdependence inherent in multispecies interactions have been made in fields primarily associated with humanities and social sciences, from social theory to anthropology, geography, and philosophy (Braidotti, Reference Braidotti2013; Houston et al., Reference Houston, Hillier, MacCallum, Steele and Byrne2017; Kirksey & Helmreich, Reference Kirksey and Helmreich2010; Locke & Muenster, Reference Locke, Muenster and Jackson2015; Morton, Reference Morton2007; Ogden et al., Reference Ogden, Hall and Tanita2013; Puig de la Bellacasa, Reference Puig de la Bellacasa2017; Tsing, Reference Tsing2015; van Dooren et al., Reference van Dooren, Kirksey and Münster2016; Wolch et al., Reference Wolch, West and Gaines1995). These advances are heavily indebted to the rich knowledge found across diverse, often Indigenous human cultures resulting from a slowly but steadily increasing visibility and representation of Indigenous knowledge in academic discourse (Graham, Reference Graham1999; Larsen & Johnson, Reference Larsen and Johnson2016; Little Bear, Reference Little Bear2000; Todd, Reference Todd2016; Watts, Reference Watts2013; Whyte, Reference Whyte, Heise, Christensen and Niemann2017). In this paper, we focus on one such conceptual advance, known as multispecies or more-than-human thinking. In a definition of multispecies ethnography (but not limited to ethnography), Locke and Muenster use the term for

We argue that multispecies thinking applies beyond the realms of social science and the humanities. It provides a powerful frame for developing an improved sustainability concept built on the interdependence of life, not the supremacy of a single species – a task we take up in this paper. We develop a multispecies concept of sustainability in three steps. First, we discuss the normative aspects of multispecies sustainability, arguing its necessity to achieve normative goals whether they are limited to human flourishing or explicitly include non-human flourishing. We then propose six principles which underpin a new definition and models that visualize the interdependence of species. Second, we develop four different future scenarios as a thought-experiment to explore why futures cannot be derived from an axis based on resource sustainability alone, but differ radically based on another axis: ‘reductionist anthropocentric to multispecies’. Third, we explore how multispecies sustainability can be applied to real-world problems using research on human–Japanese honeybee co-shaping of the landscape and insights from the Healthy Urban Microbiome Initiative. Finally, we conclude with directions for further research.

As sustainability is at its core an ethical concept with normative assumptions, we begin by examining whether multispecies sustainability requires its own distinct ethical standpoint. Because a comprehensive discussion of the ethical complexity of more-than-human ethics lies beyond the scope of this paper (Cohen, Reference Cohen2012; Droz, Reference Droz2020; Puig de la Bellacasa, Reference Puig de la Bellacasa2017), we only compare two simplified normative premises: (1) that human well-being should be sustained, a view often taking a utilitarian perspective in focusing on individuals, and (2) that both human as well as more-than-human well-being should be sustained, a view often rejecting simple utilitarianism in favour of a distributed, network approach to well-being. In the latter case, the need for a multispecies-inclusive concept of sustainability follows directly from the normative premise, and such relational ethics have long informed First Nations and Indigenous ethical frameworks (Graham, Reference Graham1999; Little Bear, Reference Little Bear2000; Watts, Reference Watts2013; Woelfle-Erskine, Reference Woelfle-Erskine2019). However, given the variety of moral statuses assigned to more-than-humans across human individuals and groups, disagreement on their precise nature might threaten commitment to strive for multispecies sustainability. Moreover, arguments in mainstream, capitalist political discourses often stress the need to prioritize human well-being and economic growth (Hickel, Reference Hickel2018). We thus believe making the strongest case possible for multispecies sustainability benefits from an argument that does not rely solely on valuing more-than-human well-being as a normative premise. Instead, following our critique of the reductionist understanding of sustainability above, we argue that traditional and Indigenous knowledges as well as scientific evidence support the hypotheses of complex, fundamental interdependence between species, and of human flourishing as fundamentally dependent on other species. To achieve sustained human flourishing as a normative goal will thus require abandoning a reductionist approach in favour of a multispecies approach to sustainability, regardless of the moral status assigned to more-than-humans based on individual or group beliefs. This may explain why, as outlined above, over time many cultures and societies have developed mechanisms of engaging with more-than-human actors. Here we thus propose multispecies sustainability not as an argument for changing values and beliefs held to include more-than-humans (although such an argument certainly merits consideration), but as a broadening of the ethical concept of sustainability required to account for the fundamental interdependence of species' wellbeing, and ultimately achieve sustainability goals.

To propose a preliminary definition of multispecies sustainability, we build on study by Davies and Riach (Reference Davies and Riach2018) who, to our knowledge, first proposed broadening sustainability into a multispecies context. Their analysis of bee–human relations in industrial beekeeping concluded with the question what bee-focused, multispecies sustainability would entail. Here, we attempt to generalize the findings of this analysis in the context of other studies towards a broader multispecies sustainability, that nevertheless affirms the complexity and flexibility inherent in multispecies relations. Multispecies sustainability might thus be grounded in these six principles:

1. (1) Needs of one species cannot be met independently, but rather require needs of other species to be met. The state of two or more species' interdependent needs being met can be called multispecies well-being.

2. (2) Multispecies well-being emerges from and depends upon a set of complex relations shaped by the agency and transformative potential of all members involved, even if this agency and potential is expressed in different ways. From this derives the need for multispecies stakeholders; especially stakeholders representing the needs of other species in human multi-stakeholder spaces.

3. (3) Multispecies well-being exposes the logical fallacy of maximizing human well-being at the expense of others in a trade-off calculation. Well-being is relation-based, not resource based, thus not a zero-sum game. Meeting the needs of the present should enhance, not compromise, the ability of future generations to meet their needs with the goal of increasing multispecies well-being over time.

4. (4) Multispecies well-being is too complex to be controlled top-down. Following Ashby's law of requisite variety (Ashby, Reference Ashby1956; Beer, Reference Beer1979; Pickering, Reference Pickering2010), a management system requires equal or higher variety than the system it seeks to control. Diverse, changing, interdependent, and inseparable needs of multiple species possess very high variety. Anthropocentric management systems would thus require equal or higher variety to regulate in detail the complexity involved in achieving multispecies wellbeing. Abstractions can help to attenuate variety, but will not be effective without the participation of autonomous local multispecies actors. Effective management therefore requires the enrolment of these actors to realize viable co-existence.

5. (5) Diverse, interdependent, changing, and inseparable needs can only be met through adapting, self-regulating systems. This calls for systems based on representations of and experimentations around continuously renegotiating complex, entangled multispecies interests. Such systems rely on and respect multispecies agency, and aim to provide to all species the operational autonomy necessary to meet their needs (Droz, Reference Droz2019). Many best practices have been developed by Indigenous peoples and are part of traditional ecological knowledge systems.

6. (6) Different species have very diverse anticipatory features and capacities – from the cellular level to the level of multiple communities of interacting species – that anticipate future conditions (Poli, Reference Poli2010). Human anticipatory capacities, based on self-reflexivity and empowered through language and technology, are unique in some ways, but also limited in other ways when compared to the anticipatory capacities of other species. Moreover, human anticipatory capacities have mostly been used to dominate and exploit other species. Multispecies well-being therefore seeks to draw on, translate between, and combine the different ways of anticipating futures that exist among different species, while seeking to employ the specific anticipatory capacities of the human species for better futures for the entire earth system.

Based on these principles, we propose the following preliminary definition:

How might this definition of multispecies sustainability be represented visually? One way might be to focus on the interdependency at the core of the concept (Figure 3). Here, visual elements (biosphere, microbial societies, plant societies, etc.) are dependent on those containing them, and affected by those they contain. The model thus may be read as human societies depending on but also affecting plant and animal societies, as well as fungi, microbes and other life forms. Each element contains in itself a high diversity and complexity, as indicated by the use of the plural form for ‘societies’ formed by different kinds of living beings. In contrast to the classic models reviewed above, this model explicitly identifies formal economies as one aspect of human economic activities overshadowing informal practices that may be more sustainable (Jehlička et al., Reference Jehlička, Kostelecký and Smith2013; Smith & Jehlička, Reference Smith and Jehlička2013). Yet visually, the model remains centred on humans. Another approach might be to take inspiration from recent updates to the biological tree of life (Hug et al., Reference Hug, Baker, Anantharaman, Brown, Probst, Castelle, Butterfield, Hernsdorf, Amano, Ise, Suzuki, Dudek, Relman, Finstad, Amundson, Thomas and Banfield2016) and centre the earth system and elements in it to emphasize the diverse, entangled agency shaping them (Figure 4). Similar to the definition proposed above, both visual models should be regarded as preliminary and are offered as first steps and an invitation to the sustainability community to improve them. In the next step, we apply this model to examine how multispecies sustainability affects the generation of future scenarios.

Fig. 3. Visual model of multispecies sustainability focused on interdependence. Elements depend on those containing them, and are affected by those they contain.

Fig. 4. Visual model of multispecies sustainability emphasizing shared agency in shaping the earth system.

3. Multispecies sustainability-derived future scenarios: a thought experiment

How might a different concept of sustainability affect future pathways? In the context of sustainability, scenarios are frequently used to identify desirable and undesirable potential outcomes (Merrie et al., Reference Merrie, Keys, Metian and Österblom2018; Vervoort et al., Reference Vervoort, Bendor, Kelliher, Strik and Helfgott2015; Wiek et al., Reference Wiek, Binder and Scholz2006). In a thought experiment, we here consider two axes to structure four different possible future scenarios. The horizontal axis represents sustainability, from a condition in which the means required to meet needs are shrinking (left) to one where the means are growing (right). Given the inherent uncertainty about future needs, it seems wise to consider simply maintaining the means as a bare minimum condition for sustainability. Many examples show living beings, including but not limited to humans, succeeding in managing their surroundings in a way that enhances their ability to meet their needs over time (Jones et al., Reference Jones, Lawton and Shachak1997). The vertical axis represents the contrast between a reductionist-anthropocentric approach (bottom) in which only human needs count (and more-than-humans are seen as resources that can be perfectly controlled and managed, rather than beings with agency whose needs require representation), and a multispecies-oriented (top) approach as outlined by the principles above. The scenarios we arrive at show that both axes radically affect the outcomes (Figure 5).

Fig. 5. Future scenarios along two axes: unsustainable/sustainable and reductionist-anthropocentric/multispecies-oriented.

4. Applied multispecies sustainability I: a policy framework for co-shaping landscape with bees

Scenario-based thought experiments show that many possible futures exist. To increase the likelihood of realizing a future similar to the ‘shared futures of multispecies well-being’ scenario, the multispecies sustainability concept can underpin policies aiming to improve human–nature relations. In particular, taking multispecies agency and well-being seriously requires new policy tools that integrate these factors, as such tools may influence whether wildlife–human interactions turn into sites of conflict or co-flourishing (Houston et al., Reference Houston, Hillier, MacCallum, Steele and Byrne2017; Rupprecht, Reference Rupprecht2017). We here present two examples where multispecies sustainability is applied in real-world efforts to improve multispecies well-being.

Beekeeping has developed over thousands of years and can be a symbiotic, multispecies relationship, in which bees enjoy increased colony survival and honey provides humans with nutrition and medicinal uses. Its capitalist industrialization is one example where reductionist anthropocentrism and the absence of a multispecies perspective is now endangering this relationship due to a multitude of factors (Davies & Riach, Reference Davies and Riach2018). This need not be the case. In many traditional and alternative practices, humans and non-human stakeholders interact in complex ways and co-shape both ecological and policy environments in which the interactions take place. Based on 2 years of transdisciplinary research into Japanese honey bee keeping in Japan, some of the authors drew on Raworth's (Reference Raworth2017) doughnut economy concept (Figure 2) to develop a multispecies stakeholder policy framework to define a space where humans and bees can coexist (Figure 6). Preliminary analysis of beekeeper surveys showed these practitioners to often consider bees as partners in co-shaping the landscape rather than livestock, and through this practice taking on the role of stewards mediating between bees, humans, flowers, pests, and predators. With the goal of creating a holistic policy tool based on these insights to inform how multispecies stakeholder interactions might be organized, four groups of stakeholders were identified: bees, beekeepers, residents/consumers, and multi-level governance actors. Although the doughnut economy model is based on a social foundation and ecological ceiling, our Japan-based research showed both ceiling and foundation to be socio-ecological in nature. Limiting ceiling and foundation factors were thus identified for all four stakeholders. The process highlighted targets for policy intervention as well as interdependencies in the system that can enable or prevent positive outcomes. For example, urban beekeepers rely on the support of the public for beekeeping as a foundational factor that can be fostered through educational policies. A lack of environmental awareness of local residents paired with a high perceived risk from bees could, on the other hand, lead to restrictive overregulation.

Fig. 6. Conceptual draft of a multispecies stakeholder policy framework to support identifying and creating holistic policies for human–wildlife coexistence.

Successful projects already resemble the proposed framework, and demonstrate that such bee–human interactions are only the tip of an iceberg of complex multispecies interactions involving the entire span of taxonomic life: diverse pollinators, vegetation as sources of nectar and pollen, diverse species benefiting from reductions in pesticide use and regenerating soil biomes. ‘For the Love of Bees’ (For The Love Of Bees, 2019), led by vision-holder Sarah Smuts-Kennedy, is a multifaceted project in Tāmaki Makaurau, Aotearoa (Auckland, New Zealand) which invites residents to imagine their city as the safest city in the world for bees. As an artwork and platform for collaboration, the project provides an example of what a cohesive (yet welcoming and open to chance encounters) programme of activity for the creation of a shared future of multispecies wellbeing could look like. FTLOB initiated and brought together diverse projects: the OMG Organic Market Garden (a pilot urban farm on a precarious inner-city site); a network of small-scale compost hubs throughout the city; a free bee school; the Urban Farmers Alliance (a peer-to-peer mentoring platform); advocacy for behaviour change in weed control; multiple pilot regenerative parks in schools and art centres throughout the city; and a public programme of events. FTLOB combines these into one cohesive action through the lens of bees, describing their approach as follows: ‘We take a bee's eye view on nature's operating system … to reveal innate interconnectedness and the power of small actions … leaving you inspired and capable to co-create safe spaces for bees and all life’ (For The Love of Bees, 2019). The project co-creates human–wildlife coexistence space (Figure 6) by progressively folding in collaborators and interested parties while generating and spinning-off new initiatives. In the process, different actors (e.g. schools, art centres, and restaurants) move from being passive-supportive members of the public to becoming new types of active stewards. The project's positioning as an artwork and social sculpture (Biddle, Reference Biddle2014) has thereby allowed it to bridge the four quadrants of the framework, bringing together human and non-human stakeholders. As these two examples show, the seemingly two-species human–bee interaction is thus an entry point for people and their institutions to explore direct and indirect ecological connections in the web of life, initializing a multispecies process of building knowledge and capacity for the more-than-human planning practices Houston and colleagues call for (Reference Houston, Hillier, MacCallum, Steele and Byrne2017).

5. Applied multispecies sustainability II: the Healthy Urban Microbiome Initiative

The second example of applied multispecies sustainability showcases the Healthy Urban Microbiome Initiative. There is now widespread awareness of the crucial role that microbial communities play in shaping the health of humans and other species. At its narrowest realization is the impact of the gut microbiome on everything from how we process food (Ridaura et al., Reference Ridaura, Faith, Rey, Cheng, Duncan, Kau, Griffin, Lombard, Henrissat, Bain, Muehlbauer, Ilkayeva, Semenkovich, Funai, Hayashi, Lyle, Martini, Ursell, Clemente and Gordon2013) and pharmaceuticals (Vétizou et al., Reference Vétizou, Pitt, Daillère, Lepage, Waldschmitt, Flament, Rusakiewicz, Routy, Roberti, Duong, Poirier-Colame, Roux, Becharef, Formenti, Golden, Cording, Eberl, Schlitzer, Ginhoux and Zitvogel2015) to our physical (Honda & Littman, Reference Honda and Littman2012) and mental (Valles-Colomer et al., Reference Valles-Colomer, Falony, Darzi, Tigchelaar, Wang, Tito, Schiweck, Kurilshikov, Joossens, Wijmenga, Claes, Oudenhove, Zhernakova, Vieira-Silva and Raes2019) health. However, there is a growing awareness of the ways human microbiomes are connected to the microbiomes of the surrounding environment (soil, air, plants, animals, etc.), which are connected to the surrounding biosphere, atmosphere, geosphere, and hydrosphere (Figures 3 and 4). Foundational to all these concepts is the awareness that biodiversity stabilizes ecosystems and allows them to maintain function in the face of threats and shifting environmental conditions (Tilman, Reference Tilman1996). Although an in-depth discussion of the role of diversity in driving ecological stability is beyond the scope of our paper, evidence for pervasive indirect links between species and the application of the precautionary principle in assuming interdependence rather than independence suggest treating no individual species as expendable without ecological consequences. Importantly, biodiversity of the human microbiome is beneficial to human health; people exposed to greater microbial diversity are less likely to have allergies (Hanski et al., Reference Hanski, von Hertzen, Fyhrquist, Koskinen, Torppa, Laatikainen, Karisola, Auvinen, Paulin, Mäkelä, Vartiainen, Kosunen, Alenius and Haahtela2012; Ruokolainen et al., Reference Ruokolainen, von Hertzen, Fyhrquist, Laatikainen, Lehtomäki, Auvinen, Karvonen, Hyvärinen, Tillmann, Niemelä, Knip, Haahtela, Pekkanen and Hanski2015) and asthma (Ege et al., Reference Ege, Mayer, Normand, Genuneit, Cookson, Braun-Fahrländer, Heederik, Piarroux and von Mutius2011). By emphasizing the importance of (microbial) biodiversity for human health, the Healthy Urban Microbiome Initiative (HUMI; http://www.HUMIcity.org) provides an opportunity to showcase the co-benefits between human health, conservation, and multispecies well-being and provides a template for applying this knowledge in ways that benefit health.

As humans have created the built environments that now house over 50% of the population, we have altered the habitat for non-human life as well. Unsurprisingly, cities house animal communities that are less biodiverse than those in wilderness areas (McKinney, Reference McKinney2006) which can be reflected in the microbial communities in the air (Després et al., Reference Després, Nowoisky, Klose, Conrad, Andreae and Pöschl2007; Flies et al., Reference Flies, Clarke, Brook and Jones2020) and soil (Liddicoat et al., Reference Liddicoat, Weinstein, Bissett, Gellie, Mills, Waycott and Breed2019). As is predicted by the ‘biodiversity hypothesis’ (von Hertzen et al., Reference von Hertzen, Hanski and Haahtela2011), people living urban and westernized lifestyles are likely to have less diverse microbiomes (Hanski et al., Reference Hanski, von Hertzen, Fyhrquist, Koskinen, Torppa, Laatikainen, Karisola, Auvinen, Paulin, Mäkelä, Vartiainen, Kosunen, Alenius and Haahtela2012; Vangay et al., Reference Vangay, Johnson, Ward, Al-Ghalith, Shields-Cutler, Hillmann, Lucas, Beura, Thompson, Till, Batres, Paw, Pergament, Saenyakul, Xiong, Kim, Kim, Masopust, Martens and Knights2018) and also experience higher rates of many allergic, auto-immune, and inflammatory diseases (Flies et al., Reference Flies, Mavoa, Zosky, Mantzioris, Williams, Eri, Brook and Buettel2019). Indeed, in a randomized controlled mouse trial, mice exposed to trace levels of more biodiverse soils showed lower rates of anxiety-like behaviour (a mental health issue more common in cities), and this effect was modulated by soil bacteria that colonized the gut of the mice (Liddicoat et al., Reference Liddicoat, Sydnor, Cando-Dumancela, Dresken, Liu, Gellie, Mills, Young, Weyrich, Hutchinson, Weinstein and Breed2020).

Under HUMI, biodiverse urban green spaces can be used to (1) restore human health, (2) improve urban biodiversity, and (3) facilitate human–nature interactions critical to support a multispecies way of thinking (Flies et al., Reference Flies, Skelly, Negi, Prabhakaran, Liu, Liu, Goldizen, Lease and Weinstein2017, Reference Flies, Skelly, Lovell, Breed, Phillips and Weinstein2018; Liddicoat et al., Reference Liddicoat, Weinstein, Bissett, Gellie, Mills, Waycott and Breed2019; Mills et al., Reference Mills, Weinstein, Gellie, Weyrich, Lowe and Breed2017; Robinson & Breed, Reference Robinson and Breed2019). Although the focus of HUMI is on humans and microbes, itself an incredibly diverse category of life including archaea, bacteria, fungi, viruses, micro-eukaryotes, HUMI argues that biodiverse microbial communities depend on biodiverse plant and non-human animal communities, thus epitomizing multispecies sustainability. It is due to these interconnections that HUMI efforts focus on the restoration of ecosystems as a public health intervention (Mills et al., Reference Mills, Weinstein, Gellie, Weyrich, Lowe and Breed2017; Robinson & Breed, Reference Robinson and Breed2019, Reference Robinson and Breed2020); re-planting a diverse array of species can return a functional and biodiverse environmental microbiome (Baruch et al., Reference Baruch, Liddicoat, Laws, Kiri Marker, Morelli, Yan, Young and Breed2020; Mills et al., Reference Mills, Bissett, Gellie, Lowe, Selway, Thomas, Weinstein, Weyrich and Breed2020). These environmental microbiomes are formed in soil, the air, on plants, and on other surfaces, from which they can transfer to people. HUMI emphasizes that this is beneficial for human health (Grönroos et al., Reference Grönroos, Parajuli, Laitinen, Roslund, Vari, Hyöty, Puhakka and Sinkkonen2019). HUMI thus exemplifies an endeavour that is anthropocentric without being reductionist, instead applying a multispecies sustainability lens to strive for more-than-human mutual flourishing. Importantly, to create biodiverse urban green spaces, HUMI also underlines the necessity of cross-sectoral collaboration. Successful projects will be led by communities, supported by local government, and guided by public health and research professionals. This cross-sectoral collaboration forces different views and aims to be discussed and the potential for a multi-species well-being emphasis to emerge.