Knowing bears: An ethnographic study of knowledge and agency in human–bear cohabitation

Introduction

This article explores a case of human–wildlife cohabitation in the Rodopi mountains of Bulgaria, wherein people and brown bears (Ursus arctos) have adapted to living together in relative harmony. The glaring deficiencies of conventional strategies for biodiversity conservation that seek to strictly separate people from wildlife via creation and enforcement of protected areas have been widely documented for quite some time (see e.g. Brockington et al., 2008; Büscher and Fletcher, 2020). Salient issues include PAs’ frequent inability to effectively preserve endangered species (Wuerthner et al., 2015) and negative social consequences in terms of displacement and marginalization of local people (Brockington et al., 2008). Recognition of such issues has provoked growing calls for new conservation strategies that instead promote coexistence, or cohabitation, of humans and wildlife in shared or at least overlapping spaces (e.g. Frank et al., 2019; Pooley et al., 2017). Such spaces would of course necessitate “mutual adjustment by both humans and wild animals as they learn to live together” (Boonman-Berson et al., 2016: 192). Yet how such mutual adjustment for cohabitation can be encouraged and governed remains contested. This is especially true with respect to large carnivores, which are central to many conservation efforts due to their status as keystone species helping to regulate overarching ecosystems but which are often particularly challenging to cohabitate with due to the threat they are seen to pose to human life and livelihoods (Dickman et al., 2013; Pooley et al., 2017).

Within such discussions, questions of the relative roles and responsibilities of humans and nonhumans in shaping interactions remain particularly contentious. While conservation has conventionally been understood as something humans do on behalf of other species, a growing body of “more-than-human” research challenges this perspective as “anthropocentric,” arguing that nonhumans should be considered “co-constitutive actors” of the spaces they occupy (e.g. Ampumuza and Driessen, 2020; Boonman-Berson et al., 2016; de Silva and Srinivasan, 2019). Based on this understanding, some go so far as to assert that a “multispecies ethics” demands that nonhumans be actively included in decision-making concerning such spaces’ governance (Celermajer et al., 2020; Cochrane, 2018). Cochrane (2018), for example, advocates that human trustees be appointed to represent the interests of nonhumans in policy negotiations.

We engage with this discussion via a case study of human–bear cohabitation in Bulgaria’s Rodopi Mountains. Within the global spectrum of human–wildlife interactions (cf. Frank et al., 2019), this case is rather unusual in that there are currently no significant conflicts between people and the brown bears inhabiting the village surroundings, and which we therefore characterize as a landscape of tolerance (see Toncheva et al., in press). Consequently, this case can be considered a relatively rare instance of peaceful cohabitation, both within Bulgaria and the wider Balkan region (where human–bear conflict is elsewhere more prevalent; see Toncheva and Fletcher, under review) as well as with respect to human–wildlife interaction more broadly (Frank et al., 2019). The factors responsible for this situation are thus illuminating for discussions concerning how to facilitate coexistence as a conservation strategy more broadly.

A previous article addressed the interaction between local knowledge, spatial relations and ecotourism delivery in facilitating cohabitation in the area (Toncheva et al., in press). This article builds on the previous by focusing on the micro-dynamics of human–bear interactions to explore in more fine-grained detail how humans and bears come to know and respond to one another in learning to peacefully cohabitate.

Our analysis adopts a political ecology perspective employing ethnographic fieldwork to emphasize multispecies entanglement in highlighting the respective roles played by humans and nonhumans in co-producing specific forms of interaction (see Boonman-Berson et al., 2016; de Silva and Srinivasan, 2019; Margulies and Bersaglio, 2018; Margulies and Karanth, 2018; Srinivasan and Kasturirangan, 2016). Grounded in this approach, our analysis demonstrates that while both humans and bears appear to mold their behavior in relation to their sensing of the other’s behavior, it also shows that knowledge of bears’ behavior is ultimately always interpreted by humans in conservation planning. Moreover, different groups of stakeholders hold different knowledge of bears that influence their respective attitudes and behavior toward the animals. This discrepancy is particularly significant with respect to knowledge held by local residents and scientific researchers, reflecting a common contestation between “indigenous” and “scientific” knowledge in informing natural resource management strategies more broadly (Agrawal, 1995). In exploring how these various issues influence human–bear relation in the field site, our study thus raises important questions concerning how to incorporate bears (and other nonhumans) within conservation decision-making, and whose knowledge should be privileged in the process.

We begin by situating our case study within overarching discussions concerning appropriate conservation strategies to facilitate cohabitation. We then explain our methodology emphasizing multispecies entanglement. Following this, we outline the results of our study, explaining: (1) how humans gain knowledge of bears; (2) how this knowledge differs among different groups of stakeholders; (3) how knowledge is translated into different forms of engagement; and (4) how knowledge is employed to also interpret bears’ perceptions of and reactions to people. We conclude by highlighting the implications of our analysis for understanding and enacting cohabitation moving forward.

Conservation governance for cohabitation

Within current discussions of coexistence and cohabitation, a key question concerns the relative roles played by humans and other species in developing cohabitation arrangements. While historically conservation has been understood predominantly as something people do on behalf of wildlife, increasingly this “anthropocentric” view is challenged as both unethical and inaccurate in its obfuscation of the role nonhumans play in shaping the spaces and processes (both ecological and social) in which they participate (Ampumuza and Driessen, 2020; Barua, 2016; Boonman-Berson et al., 2016; Collard and Dempsey, 2013; da Silva and Srinivasan, 2019; Margulies and Karanth, 2018). Pooley et al. (2017: 517) thus assert,

Examples of tolerance toward predators arise not only from the beliefs and behaviors of humans but also from animal behavior. This suggests a different kind of focus on animal behavior (i.e., how individual animals and societies of animals have adapted to the human societies they interact with).

This ecocentric position reflects an overarching trend within social scientific scholarship to espouse “more-than-human,” “multi-species” or “posthumanist” perspectives that challenge “human exceptionalism” to instead understand humans as intimately entangled with other entities all of whom contribute to shaping the “assemblages” they comprise (e.g. Barua, 2016; Braidotti, 2013; Collard, 2014; Hurn, 2012; Whatmore, 2006). A central figure in such discussions, Bruno Latour, thus asserts that such a more-than-human perspective demands “a completely new cosmopolitics” corresponding to the “new relations between humans and non-humans” this perspective implies (2011: 80). Within this new cosmopolitics, all entities would be considered “actants” who participate equally in constructing a given “actor-network.” A central concern here is hence to “democratize” or “distribute” agency beyond the human subjects to which this concept has conventionally referred to instead attribute it, in Latour’s term, to “any thing that does modify a state of affairs by making a difference” (2005: 71). From this perspective a growing variety of nonhumans, including animals (De Waal, 2016; Haraway, 2007), plants (Gagliano, 2018) and even various entities traditionally considered inanimate (e.g. rivers, CO₂) (Bennett, 2010) have been deemed agential actors (or at least “actants”). Some go further still to assert that ostensibly supernatural “earth beings” can also exert agency in human affairs (De la Cadena, 2010, 2015).

Imported into conservation debates, this perspective enjoins us to understand nonhumans as actors who, together with humans, “co-produce” a shared environment and in this way influence conservation management strategies including promotion of cohabitation (Ampumuza and Driessen, 2020; Boonman-Berson et al., 2016; da Silva and Srinivasan, 2019; Hinchcliffe, 2007; Margulies and Karanth, 2018). It implies that humans and nonhumans can and should share a common space in which nonhumans are considered not “other” beings but rather fellow inhabitants of a shared world (Boonman-Berson et al., 2016; Hinchcliffe, 2007). For some advocates of this approach, its importance is also in terms of moving conservation policy and practice beyond its traditional grounding in an “anthropocentric” perspective that privileges human perspectives and needs above those of other species (e.g. Washington et al., 2018). Such anthropocentrism is commonly contrasted with “biocentric” and “ecocentric” perspectives, respectively, the former seeking to give equal importance to all forms of (animate) life and the latter going beyond even this to also include ostensibly inanimate elements of the nonhuman environment within the fold. As Washington et al. explain, “An ecocentric worldview is one in which human and nonhuman organisms, species, ecosystems, and ecosystem processes are all understood to have moral value” (2018: 368). This is seen by advocates to constitute the correct basis for conservation politics.

For some proponents of cohabitation, consequently, the approach goes much further to demand the prospect of granting agency and decision-making authority to nonhumans in conservation governance processes. This position is in line with broader proposals asserting that processes of decision-making and conflict resolution more generally should be expanded to include nonhuman entities. In this way, longstanding calls to foreground social justice concerns in conservation (Martin et al., 2015) are countered by insistence on the need to also considered “justice for nonhuman nature” (Washington et al., 2018: 370). Variants of this position come under different labels, including “ecojustice” (Washington et al., 2018), “multispecies justice” (Celermajer et al., 2020), “ecodemocracy” (Cullinan, 2003) and “Earth jurisprudence” (Burdon, 2011). As one group of advocates explains, “our proposal embraces the entitlement of all life to have a say in the globe-girdling exploits of current human adults” (Treves et al., 2019: 140). Such proposals build on Latour’s (2004) popular argument that the “politics of nature” should be understood as enacting a “parliament of things,” that is, deliberation among all elements of a given assemblage, both human and nonhuman.

These proposals also coincide with a growing campaign to include “rights of nature” within existing political processes. Common examples include Ecuador’s inclusion of rights of nature—Pachamama—generally within a revamped 2008 constitution (Rival, 2010) and New Zealand’s designation of a particular river—Whanganui—as a legal person (Chapron et al., 2019). Boyd (2017), indeed, asserts that growing adoption of rights of nature approach constitutes a widespread “legal revolution that could save the world.”

Whether approached through the lens of justice or rights, ¹ a central question within such discussions concerns how exactly to include nonhumans within prescribed political processes. Most advocate appointing human “trustees” to represent nonhuman perspectives within political fora. Cochrane (2018), for instance, proposes a “sentientist cosmopolitan democracy” envisioned as “a global political system made up of overlapping local, national, regional and global communities comprised of human and non-human members.” He insists that this system should be “participative, deliberative and representative” and “include dedicated representatives of non-human animals whose job should be to translate the interests of animals into deliberations over what is in the public good for their communities.” ² As Treves et al. explain of their own proposal for “just preservation,” appointment of representatives would depend on “their ability to understand ecology, ethics, and the manifold, complex, dynamic interactions of humans and non-humans” (2019: 138).

Yet how exactly this expertise should be attained and assessed remain unclear. The discussion thus invokes another longstanding debate concerning the very different ways that various groups of humans investigate and gain knowledge of a more-than-human world (Forsyth, 2004; Goldman et al., 2011). Within this debate, a common divide distinguishes indigenous from scientific knowledge (see Agrawal, 1995). While this dichotomy has been widely challenged (see e.g. Agrawal, 1995; Berkes 2008), it remains prevalent in myriad environmental disputes, wherein local forms of knowledge are often subordinated to technocratic understandings of human–nonhuman relations grounded in scientific knowledge production (e.g. Hunn et al., 2003; Nadasdy, 2005). The relative authority of different modes of knowledge production is thus considered another important dimension of justice—called recognition or epistemic justice—within environmental policymaking (Martin et al., 2016; Massarella et al., 2020).

Such disputes are prevalent within discussions of human–wildlife conflict (HWC) and coexistence in particular. HWC is commonly approached technocratically as a simple “management issue” addressed via two main strategies: (1) modification of animals’ behavior (often by radical measures, such as killing); and/or (2) prevention of activities that overlap in space (e.g. by fences, zoning, relocation, etc.) (see Hodgson et al., 2020; Treves and Karanth, 2003). Yet growing evidence of “cultural (nonscientific) explanations” for the presence or absence of HWC in different contexts signals the importance of also attending to “the roles of culture and values in human–wildlife coexistence,” and in particular to different “cultural ways of thinking about animals” (Pooley et al., 2017: 513, 514, 515; see also Dickman et al., 2013). Numerous researchers thus emphasize the importance of integrating local systems of knowledge concerning how to manage natural resources and human–nature relations more generally—what in the literature is commonly termed traditional or local ecological knowledge (TEK or LEK)—to support effective biodiversity conservation including cohabitation (e.g. Berkes, 2008; Berkes et al., 2000; Hunn et al., 2003; Nadasdy, 2005). ³

If nonhumans are included as actors within conservation policymaking, consequently, the question of whose knowledge should be considered the basis for authoritative understanding—and therefore “expert” representation—of nonhuman perspectives and interests becomes pertinent. Indeed, various researchers point out that the very framing of attribution of agency to nonhumans as an innovative perspective within more-than-human scholarship is only possible via denial of the fact that numerous nonwestern people have regarded nonhumans in similar ways for millennia (see e.g. Hornborg, 2015; Sundberg, 2014; Todd, 2015; Watts, 2013). Yet whether nonhumans can indeed be understood sufficiently by humans for them to claim to accurately represent nonhuman interests via whichever mode of knowledge production remains unclear. Soper, for instance, asserts that “what we map back onto ‘nature’ … is necessarily to some degree a projection of our own self-understanding, and of the meanings attributed to our specific ways of doing things” (1995: 170). Hence she finds “something dubiously anthropocentric about [ecocentrists’] readiness to assume that human needs, desires and capacities give us a direct access to knowledge of their ‘analogies’ in the worlds and life-styles of other species” (1995: 171).

In what follows, we explore these contentious issues in relation to our case study of human–bear interaction near Yagodina village in the Rodopi Mountains. Based on this analysis, we make four central arguments in relation to the above discussions. First, that whatever the degree of agency attributed to nonhumans in human–wildlife interactions, ultimately understanding of nonhumans’ behavior, and policymaking following from this understanding, relies fundamentally on the perspectives and agency of human actors. A corollary to this is that, second, despite the aspirations of many posthumanist researchers to “flatten” disparities between humans and nonhumans, unequal power relations must still be taken into account in human–wildlife relations. Third, that any appointment of human trustees to represent nonhuman interests in policymaking should be therefore understood not as nonhumans speaking through people but rather as people speaking for nonhumans, with all the tricky issues of bias and interpretation this implies. Fourth, that in incorporating knowledge concerning nonhumans’ perspectives and behavior into decision-making, discrepancies between “local” and “expert” knowledge must also be mediated and power differentials between different forms of knowledge acknowledged. Taken together, these four propositions cohere to suggest that ultimately efforts to incorporate nonhumans into conservation policy-making must inevitably reinscribe some of the anthropocentrism proponents of this approach aim explicitly to avoid.

In the next section, we begin to develop the analysis informing this set of conclusions by explaining the ethnographic methods employed to explore the “multispecies entanglement” in our study.

Researching multispecies entanglement

The study is grounded in four months of ethnographic field research conducted by the first author between June and September 2018 in the village of Yagodina. To investigate the humans’ point of view, qualitative social scientific research encompassed 30 semi-structured and semi-directive interviews with informants selected via snowball and purposive sampling to include different groups of relevant stakeholders (among permanent residents of the village encompassing diverse age groups): hunters, ecotourism guides, employees in tourism, pensioners and children, among others. Some interviews were tape-recorded, while others, following informants’ preferences, were documented in field diaries. Interview data were complemented by administration of 17 detailed questionnaires assessing LEK concerning brown bear’s ecological and cultural salience to another set of respondents. These questionnaires were not intended to pursue representative sampling of the total population for statistical analysis but merely to complement in-depth interviews with comparative qualitative material collected from a broader range of local residents. In addition, participant observation was conducted throughout the research period, including the accompanying of two bear watching trips (part of an ecotourism excursion) for two full weeks (see Toncheva et al., in press). In the following analysis, we draw on this material to compare and contrast various local residents’ views of and reactions to bears’ behavior with results of scientific research and planning in order to highlight similarities and differences in how bears are understood and negotiated.

Also including the bears within this multispecies research design proved more complicated, however. Acknowledgement of the importance of approaching different entities comprising a “multispecies-network” (Latour, 2004) on equal terms has provoked discussion regarding the most appropriate methodology to account for both human and non-human actors within such a network. As Pooley et al. (2017: 516–517) explain, “Animals have a physical existence independent of humans, and for the purposes of management, it is essential to understand their physiology, behavior, and ecological relations.” While researchers largely agree that investigation in these terms requires methods exceeding the conventional social science research toolkit (Barua, 2016; Lorimer, 2015), demanding cooperation between the natural and social sciences (see Barua and Sinha, 2019; Hodgetts and Lorimer, 2015), there remains disagreement concerning how exactly such investigation should be executed. For most, the ideal of multispecies research is to employ appropriate methods to rigorously investigate each species (and their interaction) simultaneously (Barua and Sinha, 2019; Hodgetts and Lorimer, 2015). In practice, however, there exists a widespread “bias towards ethnographic methodologies involving participant observation of humans in their interactions with nonhumans,” with understanding of nonhumans’ behavior in many studies remaining “dependent upon the goodwill, expertise and field sites of scientists” (Hodgetts and Lorimer, 2015: 287; see also e.g. Margulies, 2019). For multispecies research involving ethnography, in particular, therefore, the “[t]he potential for ethnographers to write fantastic animals as they want them to be, rather than what they might (intangibly) be, is problematic” (Madden, 2014: 289).

Such issues are frequently complicated by the nature of the nonhumans under study. Most relatively balanced multispecies studies to date have been performed in relation to domestic animals such as cows (Grandin and Johnson, 2006) and dogs (Haraway, 2007), or those in captivity (e.g. in zoos or rehabilitation centers) where they can be easily observed (e.g. Parreñas, 2018). In-depth engagement with nonhumans of this sort is more difficult to replicate in the case of free ranging animals such as large carnivores, like bears, which are reticent, roam widely, and hence difficult to observe directly. In such cases, a multispecies approach requires more indirect methods, such as use of visual methodologies (see Margulies, 2019), technologies for remote monitoring and study of molecular markers (Hodgetts and Lorimer, 2020)—or, as is common, reliance on natural scientific data collected and published by other researchers (Boonman-Berson et al., 2016). All of these strategies pose additional questions regarding the extent to which humans and nonhumans are equally represented within the resulting research.

In our study, limitations imposed by the difficultly of observing brown bears directly and lack of published data concerning the animals’ ecology and behavior in this area meant that including the animals as actors within the multi-species network under investigation required reliance on multiple interviews ⁴ with an ecologist who has conducted research on bears there for many years. In addition to directly observing the bears both in the wild and from a bear watching hide, this ecologist has utilized a variety of other non-invasive techniques including natural sign surveys and camera trapping. The former involves recording and analyzing the natural signs (tracks, scats, tree rubs, digging, etc.) “deposited over the course of an animal’s normal daily activities” (Heinemeyer et al., 2008: 45). This generates general information about the presence, distribution, and relative abundance of bears, as well as specific data concerning their diet and seasonal foraging behavior. Camera trapping, defined as “the use of remotely triggered cameras that automatically take images and/or videos of animals or other subjects passing in front of them” (Rovero and Zimmerman, 2016: 1), provided additional insights, particularly into bear behaviors such as activity patterns and breeding. Aggregated over time, such data help to assess whether and how anthropogenic influences impact and alter bear behavior, as well as to identify the main zones of human–bear cohabitation in the region, that is, the specific localities where villagers and bears share a common space and hence are most likely to come into direct contact.

In addition to providing insight concerning bears’ behavior, discussions with this ecologist afford comparison between knowledge of bears grounded in scientific investigation and the LEK of villagers lacking scientific training. Yet while this distinction is useful for heuristic purposes, as Agrawal (1995) among others cautions it should not be considered a strict dichotomy as in practice neither local nor scientific knowledge is homogenous while different forms of knowledge also overlap and hybridize in complex ways. As the ecologist himself pointed out:

It must be remembered that I have spent a vast amount of time myself in the mountains and core bear habitats during the last 30 years, and actually therefore, although a ‘foreigner’ have my own LEK. So it might be argued I am presenting LEK as well!

Knowing bears

Often considered a keystone species (Van Valkenburgh and Wayne, 2010, the brown bear is also considered particularly sensitive to human influence (Woodroffe and Ginsberg, 1998). This increases its vulnerability due to such factors as habitat loss and degradation, depletion of prey, persecution, hunting and exploitation (Karanth and Chellam, 2009; Temple and Terry, 2007). Bears’ conservation is therefore “a significant challenge in landscapes which are as densely populated and heavily modified as those found in much of Europe” (Temple and Terry, 2007).

The village of Yagodina is surrounded by forest, which is excellent habitat for brown bears whose numbers have increased significantly in recent years, according to local residents whose encounters with bears have become much more frequent during this time. This increase has been facilitated by the agricultural decline experienced during the last 30 years (under postsocialism), during which huge amounts of previously cultivated land have been abandoned (Toncheva et al., in press). Further reasons for this increase include the fact that since 2002 brown bears have been granted protected status under both Bulgarian and EU law ⁵ and the extremely high endemic biodiversity in the surrounding forest. The great variety of species provides an abundance of food for bears, such as wild berries (strawberries, raspberries, cornelian cherries, bilberries, etc.), various roots and grasses, ants, and so forth. This abundance contributes to the lack of conflict and successful cohabitation while also causing various interactions since local livelihoods also comprise forest activities to gather wild foods.

Our exploration of human–bear relations begins with documentation of stories told by the elderly population suggesting that there have always been bears around the village; people between 70 and 90 years old remember encounters that occurred during their parents’ lifespan. Still, assertions that bears are more numerous now suggest that their presence was less obvious under socialism, due partly to the fact that they did not enjoy protected status and were allowed to be hunted and killed. According to local hunters, who are most familiar with the bears around Yagodina, the latter now number about 10–13 just in areas around the tiny village, meaning that bears could be encountered at any time in any part of the surrounding landscape. As many people believe so many bears were never present before, this forces both humans and bears to adapt to a new situation in which both species must learn to live together.

As the bears are widely considered fellow inhabitants of the shared space by local residents, a perspective evidenced by the multiple narratives related below, the animals are an important and often discussed topic. The village’s relatively small size (less than 500 people) means that everybody knows one another and that the village square functions as a center for exchange of news. While not everyone has seen a bear themselves, bears are therefore part of local life. Most encounters are occasional meetings by chance, occurring in different locations outside the village boundaries, predominantly to local hunters or people involved in other forest activities. The hunters are of particular interest for the present study, as their activities (stalking other non-protected species) bring them into close contact with the bears. Therefore, they not only hold the most knowledge of bears but also are de facto “managers” of human–bear relations in the area as well as of the ecotourism developed around bears (see Toncheva et al., in press).

The most famous direct encounter between a local man and a bear happened around 40 years ago and seemed to be known to everyone, being told so many times that it has already gained some mythological elements. It was narrated in different versions, so here we present the version containing the most common elements:

One old man was walking in the forest, Old Sakalche we called him, in the direction of Koru Dere. There was a big hole there and he wanted to have a look and check what is inside. There he saw a bear with two small cubs. When he looked, the bear came out, and caught him by the waist. He had a poyas ⁶ and under it - a tobacco box. The bear grabbed him by that box and dragged him around. There was a rock wall there and the bear threw him down. He started shouting. The bear then descended and buried him under some branches. On the other side there were two people who started shouting: ‘Hey, leave the man’. The bear took her cubs and left in direction Greece. He was then saved.

People are aware of the bears’ presence by “reading” various signs left in the surrounding landscape:

Once we went with my husband to collect mushrooms and stopped by a fountain for lunch. My husband said: Look at these stones, the bear has been here!

Apart from ecological factors, another very important contributor to peaceful cohabitation is the people’s attitude, marked by tolerance or, as some informants claimed, “care” for the bears. For this reason, we define this as a landscape of tolerance (Toncheva et al., in press). The majority of informants believe that people and bears can, indeed, peacefully coexist while some go even further:

We can live together, without any problem.

Local knowledge of bears

Our research shows that very little information has been preserved from previous generations, partly due to the fact that the number of bears and hence contact with them during the socialist period was much lower. Consequently, very few respondents refer to what they know about bears as inherited from older generations. When analyzing responses to questions regarding knowledge of bears, we observe substantial agreement among respondents on general knowledge and uncertainty (i.e. giving multiple answers) when referring to more specific questions related to the bears’ ecology.

To begin with, there is no disagreement that bears are permanently present around the village, and while not everyone has seen a bear themselves, all have seen signs of bears such as scats, prints, broken branches on trees, ransacked ant nests, or flipped stones. However, the number of bears estimated by different respondents varies widely from 2 to 20.

In terms of their character, the bears are described as either shy or brave. But when describing bears’ size, there is a large variation, ranging from 100 to 600 kg and from 0.80 to 1.50 m (when standing on four legs). When compared to the average size of bears in the Rodopi as measured by ecologists (weight of 100–250 kg and shoulder height of approximately 0.75–1.1 m), one can observe a tendency for people to overestimate the size of bears they have seen, or to assume that any bear encountered will be larger and thus more frightening than in reality. People more accurately describe bears’ color, describing them as brown, grey or black, as ecologists indeed report that coat color can look different under different light conditions or during different seasons. A few villagers claim they can distinguish male from female bears, explaining that males seem bigger, and that the faces of females are more rounded and beautiful.

Concerning bear movement, most villagers are aware that bears normally move slowly through the forest and more quickly only in case of danger (when they are scared or sense the presence of people). However, there is no consensus about the period (day or night) during which bears are most active. Furthermore, estimates of the daily distance travelled by bears in their search for food range between 5 and 50 km, whereas ecologists claim bears tend to cover much smaller distances, typically less than 5 km per day.

While many respondents are aware that bears rub and mark trees, the significance of this form of behavior is not well understood. It is most often interpreted as evidence of bears marking out and controlling exclusive territories, believed to be between 2 and 100 km², with the bear who can mark highest on the tree being the territory’s owner. Yet camera trap evidence suggests that rather than these trees being marked as “no trespassing” signs, they function more as “bulletin boards” for exchange of chemical messages.

The habitats people associate with bears include rocky places, forests, and meadows, in particular areas with raspberries and bilberries. While everyone agrees that bears hibernate, there is no agreement about the hibernation period. Some estimate it as beginning from the first snow, typically in December, which corresponds with the ecologist’s findings, but others place it as early as September. Dates of the bears’ emergence from hibernation are given as between February and May. No knowledge exists concerning specific details such as bears’ breeding season, whether males and females pair for life, or when young are born, apart from a few hunters who correctly mentioned that the young bears are born while the bears are still in the den. Even then, however, there is discrepancy concerning whether this occurs during winter or spring. The number of cubs was correctly estimated as between 1 and 3, as was the fact that cubs are brought up by the mother and remain with her for up to 2–3 years. However, there was no precise knowledge concerning bears’ lifespan, with estimates ranging between 10 and 40 years.

The majority of people believe that bears are predominantly vegetarian, feeding on fruits, fungi, roots, grass, honey, ants and, only exceptionally, on livestock. Bears are believed to like sour food, hence their preference for ants, and for this reason there is a widespread belief that before feeding on meat, a bear will first leave or bury it, so that the meat becomes sour. In general, it is believed that bears do not kill deliberately, only “by chance,” the exception being so-called “problem” bears which are seen to predate sheep, cattle, horses, wild boar, deer, and chamois.

Hunters’ special knowledge

Among local residents, hunters are the group that holds most knowledge regarding bears, therefore we explore their knowledge in a devoted section. This special position is due to the fact that hunters are the group experiencing most frequent encounters with bears due to their practice of encircling the perimeter of a specific hunting zone which could be any space within the surrounding forests. As one hunter related:

Most likely you meet a bear during hunting, because we surround a large area and there is no way that the bear doesn’t come out.

The hunters’ LEK is the result of practical experience, or received from older hunters as well as, to an extent, from literature, media, and television (many claimed to have seen bear programs on National Geographic). Most of their knowledge corresponds to the advice given by conservation agencies concerning how people should react when interacting with bears. For instance, what hunters advise as result of own experience, is compatible with information about appropriate behavior during human–bear encounters presented on noticeboards in the region (erected as part of a project on damage reduction for other areas—see Златанова и др., 2013). These state that behavior during an encounter should not be aggressive to ensure that bears do not feel threatened and therefore attack; that one should not react by running away, looking the bear in the eyes, or making loud noises, but instead by keeping calm and slowly retreating.

Hunters’ practical knowledge is also evidenced by their observation of details regarding bear behavior such as bears’ proficient climbing ability:

They climb very well. We used to place barrels at 2-3 meters height with a pipe and barbed wire around. They used to climb on that and destroy the barrels.

Interpreting the bears’ point of view

Despite the fact that, as our key informant ecologist admits, one can never know what exactly the bears think or know about humans because one “cannot think exactly from a bear’s perspective,” research concerning bears can still provide us with some ideas regarding their perspective and attitudes vis-à-vis humans (while also acknowledging that this is always ultimately filtered through people’s own subjective interpretations of the bears’ point of view). Bears would certainly, he claims, build up a knowledge about “when tourists use trails, jeeps drive around, people use the forests, where they go” and then adjust their behavior in order to avoid conflict. The ecologist claims that bears most likely view humans as another type of bear inhabiting their environment, and thus interact with people much in the same manner as they interact with another bear. Ecologists therefore assess bears’ knowledge primarily via the latter’s reactions and behavior.

As with humans vis-à-vis bears, avoidance is seen to be the primary reaction of bears to human presence. Being generally under threat from people, bears are described by biologists as living in a “landscape of fear” (Laundre et al., 2001). The ecologist thus explained that “bears don’t want to meet people or to have anything to do with people at all. They regard humans the same way that they see another bear … and try to avoid conflicts.” If a bear meets a person, he maintained, she wants to avoid conflict and keep the person who is a potential threat at a safe distance. He described bears’ reactions to human encounter as typically following three stages:

Most bears would run away.

According to ecologists, this sequence of bear behavior is simply a form of visual communication aimed at avoiding direct conflict, and not a sign of aggression and imminent attack as most people misinterpret it. It is this misunderstanding, and the failure of humans to understand the “language of bears,” that in ecologists’ view can lead to unnecessary danger and conflict.

While, as previously noted, Yagodina residents’ attitude towards bears is primarily one of tolerance, the ecologist claims that this is reciprocal. The bears around the village, he claims, are, due to the predominantly tolerant attitude towards them and the fact they are not hunted, “used to people appearing and don’t see them as a threat.” For this reason, when encounters between bears and humans occur, the bears do not feel directly and immediately threatened, and therefore respond with flight rather than a decision to fight, as they know that they can safely move off into cover and have no need to engage in potentially risky direct confrontation or attack.

For the bears, the main signs of human presence are likely human activities occurring in the forest during which people can be seen, heard, and smelled by the bears, such as in the case of the bear hide which is managed by people (see Toncheva et al., in press). Ecological data suggest that bears

know very well, they can sense that people put this down [the food used to attract bears to the hide], that there are people connected with it … they are used to the occasional people going there, people picking up mushrooms, etc.

From the bears’ perspective, the ecologist thus asserts:

Successful cohabitation is if people just get along, doing their things in the forest and let bears do their things. If you meet a bear, just ignore it. And this is what a bear wants, to be ignored … and if you do that it will do the same.

According to the ecologist, an important factor for the bear is its prior experience with people and hence whether it perceives them as dangerous. The bears’ protected status in the region ensures that they are not likely to be killed or hurt, contributing to their perception of people as simply a disturbance rather than a direct threat. This means that bears will likely feel that they have the time to move away from humans in the same way they would retreat from encounters with other bears. By contrast, as seen among bear populations in other parts of the country (and around the world) where bears are routinely harassed, scared off, and shot with rubber bullets, such so-called aversive conditioning can actually cause more conflict between humans and bears. Scaring the bears—and in particular the use of pain as an aversive management tool—may increase the critical distance within which bears feel imminently threatened. This creates a much higher chance that a bear will attack during an encounter. But if, by contrast, bears are used to cohabiting with non-threatening people who are simply living and working in the region, they will come to see humans as benign, enhancing trust and respect and helping to create a landscape of tolerance (see Toncheva et al., in press).

The landscape of tolerance cultivated in this area, in sum, appears to function namely because it is reciprocal, enacted, and experienced similarly by both species. When asked whether local attitudes toward bears were predominantly positive, the ecologist thus responded:

Very much yes, and I think this is contributing to the lack of conflicts here. Because people are showing tolerance to the bears, bears are showing tolerance to the people.

Learning to live together

The preceding discussion demonstrates the significant extent to which humans and bears within the research site have established particular cohabitation strategies as a result of their interactions and adaptation to living together. While basic protection of the species by the state seems to function as a successful precondition for conservation here, lack of concrete management strategies imposed from outside has given people and bears the freedom to establish their own mechanisms of mutual adaptation to which both are understood to actively contribute (see Toncheva et al., in press for more explanation of these mechanisms). Their bottom-up nature has allowed these mechanisms to be shaped by specifics of the local context (Toncheva et al., in press). Both people and bears read one another’s signs as proof for the presence of the other, and based on this reading appear to react in a similar manner, predominantly by attempting to avoid encounters. While not only professional ecologists but also village residents spend significant time and energy studying bear behavior, ecological research suggest that bears also gain knowledge about human behavior, which they apply when adjusting their own behavior to facilitate cohabitation. Similar to local hunters (as well as possessors of TEK more generally), therefore, bears’ knowledge of humans can be considered experiential knowledge that is “validated by testing in practical circumstances for its effectiveness” (Usher, 2000: 188).

Direct and indirect (i.e. reading signs signaling the other’s previous presence) encounters thus constitute an ongoing adaptation process for both people and bears by creating particular practices and meanings: for people concerning how to behave when meeting a bear; and for the bears, how better to avoid encounters with people, as well as awareness that benign people are not a threat and thus flight rather than fight is the most suitable reaction when an encounter occurs. Within the cohabitation space, both actors can be understood to “read” each other’s signs, such that indirect encounters result in bears reacting to the sound and scent of human presence, while people assess bears’ presence via prints, flipped stones, caves, and so forth. Both sets of actors incorporate these signs into their practices of avoidance, increased mutual awareness, and predictability. The latter, resulting from consistent human behavior, seems to be essential for the bears, and hence a major factor for successful cohabitation. The ecologist thus asserts that “as long as people are predictable, as they go in the forest and do their own business and don’t disturb the bears people would be predictable, bears would sense them, know there is someone … and continue eating.” Humans, meanwhile, translate their knowledge into active measures and strategies for cohabitation, such as provision of food for the bears and the modification of feeding barrels to keep bears from destroying them.

Substantiating previous research (Huber and Pedersen, 1997), the study demonstrates that knowledge of bears among the population of this remote mountain region both differs from and overlaps with current scientific or expert knowledge. LEK in this context can therefore be considered hybrid knowledge (Nygren, 1999) including elements of both scientific and folkloric origin, being obtained simultaneously from outside sources, inculturation in local life and through practical experience. Our research demonstrates that, far from comprehensive or coherent, LEK remains fragmented in relation to bears’ ecology, ranging from possession of facts corresponding with current scientific knowledge to overestimation of various dynamics (e.g. bear size) to simple vagueness and uncertainty.

Among local stakeholders, however, we have identified one group that seems to have achieved unique “proficiency,” namely the hunters whose LEK originates primarily from direct experience and “living with” the bears. This practical knowledge is the most functional in addressing human–bear relations as it come into use not only when applying adaptation measures, but also via transmission to other segments of the local population (i.e. by informing and advising people on various bear-related issues). More extensive experience and knowledge, we have shown, results in an understanding of the bears more consistent with scientific knowledge, particularly in terms of bears’ behavior and reactions to humans, which, the ecologist asserts, is a key factor for successful cohabitation.

By the same token, it became clear that lack of practical experience resulting in more fragmented and superficial knowledge—or knowledge received only from other sources (books, media, TV, etc.)—heightens fear. Despite such fragmentation, bears occupy an important place in local culture, demonstrating intimate and vital relations between people and the surrounding environment. The depth and utility of local practical knowledge, in particular, supports assertions of the importance of LEK for conservation management and especially for addressing human–wildlife conflicts (Berkes et al., 2000; Pooley et al., 2017). Yet even the fragmented parts of the more superficial LEK exhibit an attitude of respect for bears as a reflection of relationships between humans and wildlife established over centuries.

In the case of hunters’ specialized knowledge, we are able to see most clearly the importance of practical knowledge for cohabitation. The research has shown how cohabitation practices both create and originate from particular experiences and associated meanings. Support for this idea is illustrated by the findings demonstrating how less knowledge of bears, as a result of fewer interactions between particular stakeholders and the bears, is much more likely to result in negative attitudes and emotions. By contrast, the group possessing the deepest knowledge of bears—the hunters—is not only bear “experts” but also de facto “managers” of local human–bear relations. That this is a locally developed system of knowledge that both derives from and informs an historical process of “living together” demonstrates the importance of attention to diverse cultural attitudes, values and practices for encouraging human–wildlife coexistence (Pooley et al., 2017; Turnhout et al., 2013).

In developing this analysis, our study also demonstrates that including animals as an integral component of the multispecies network is important for revealing how both knowledge and mechanisms for cohabitation are co-produced by both humans and bears. Members of both species establish particular practices and behavior on the basis of experience from encounters, by accumulating particular knowledge concerning one another’s behavior and habits. Adopting a “more-than-human” perspective that grants bears an adequate role in the network of relationships has enabled the study to acknowledge the perspectives and roles of both sets of participants in the process.

Conclusion

What, finally, does our analysis suggest concerning appropriate strategies of conservation governance for facilitating cohabitation from a multispecies perspective? Firstly, our research demonstrates that bears’ knowledge and behavior, and how this is taken into consideration by people in developing cohabitation strategies, is ultimately interpreted through the lens of human perception and its inevitable anthropocentric bias. Bears’ perceptions and reactions are understood through human concepts—fear, danger, avoidance, trust, signaling, etc.—that may or may not have much to do with how bears understand their own experiences and actions, dynamics that are essentially unknowable. While it may indeed be possible to overcome this interpretive bias to some degree by presenting nonhumans’ behavior on its own terms free from expert translation (e.g. via direct presentation of data from camera traps and other forms of remote sensing; see Kirksey et al., 2018), to include understanding of this behavior within subsequent policymaking necessitates further human mediation. The inevitability of such mediation, we believe, should be more widely acknowledged and addressed by those advocating ecocentric conservation approaches.

While some commentators frame contestation concerning the relative emphasis placed on social and ecological justice within conservation policy as about the rights of people versus rights of nature, therefore one could equally claim that the dispute is not really about nature so much as rather about which human values and interests to privilege in policy. Thus Soper calls on ecocentrists to acknowledge

the element of self-interest that underlies their preservationist attitudes … Even ecocentric arguments tacitly rely in their defence of the ‘intrinsic’ value of nature on the appeal to human aesthetic and moral sensibilities … when they are calling on us to value nature ‘for its own sake’, they are also implicitly suggesting that the correct environmental ethic for human beings to adopt is one that values nature for its independence and without concern for its utility. (1995: 207)

Secondly, our study demonstrates that while bears and humans may both work to understand and cohabit peacefully with one another, this relationship remains far from equal, as humans retain (via their superior numbers and technology) far greater power to influence the nature of the interaction in the manner they choose, from simply killing bears (as occurred most frequently in the past) to refraining from such killing (as they do today). While bears remain physically more powerful than humans, bears’ power to shape cohabitation strategies beyond individual encounters with specific (unarmed) humans remains far more limited. Our analysis thus suggests that rather than endeavoring to level or “democratize” all interactions entirely, attention to power relationships and inequalities remains an essential focus of discussions concerning cohabitation as well as multispecies entanglement more generally.

If this is the case, then, thirdly, proposals to appoint human trustees to represent nonhuman interests in conservation decision-making should be understood as these trustees speaking for nonhumans in the same way one would do e.g. for a dependent child (i.e. trying to understand and act in terms of what the other would want while ultimately retaining decision-making authority) rather than merely channeling nonhumans’ views, as some advocates of this approach seem to suggest (e.g. Cochrane, 2018; Treves et al., 2019). And this in turn raises the question of how human representatives for nonhuman interests might be selected and appointed within such a process. It is clear, for instance, that bears could never be organized to appoint their own representatives, democratically or otherwise. How, then, might such appointments be made, and on the basis of which knowledge or expertise?

It is here that, fourthly, the relationship between different forms of knowledge becomes important. Our analysis has shown that in the case in question, there are substantial differences between the knowledge possessed by local villagers and scientific researchers concerning bear’s behavior and appropriate forms of engagement following from this knowledge, as well as significant differences in levels and forms of knowledge among various groups of villagers. Whose knowledge should be privileged in conservation policymaking, and particularly in relation to who speaks for and on behalf of bears within such policymaking? Should scientific knowledge be considered the gold standard and LEK validated to the extent it coincides with that, as is common in conservation planning elsewhere (e.g. Hunn et al., 2003)? Or does LEK have an equal right to assert its intrinsic validity for policymaking as a form of epistemic justice whether or not it coincides with scientific data? Our study suggests that this key question demands more explicit discussion and debate within proposals to represent nonhumans in decision-making.

In sum, consequently, our analysis suggests that anthropocentrism, far from an evil to be transcended, is in fact unavoidable in humans’ engagement with nonhumans. Rather than seeking to evade or denounce this perspective, consequently, it may be both more honest and more productive to instead fully embrace it and acknowledge that it is our own human interests and perspective in terms of which we ultimately act first and foremost even if we also take (our interpretation of) nonhumans’ own views into account in the process. It does not follow, however, that anthropocentrism necessarily means “seeing all of nature as just a resource for human use,” as some claim (Washington et al., 2018: 368). On the contrary, anthropocentrism can be enacted in different ways, from asserting this sort of absolute and instrumental domination to working to help other lifeforms live other sorts of lives in co-production with latter. This is certainly not to claim that humans actually do wholly dominate all other forms of life in the ways that are often assumed (by certain commentators on the advent of the Anthropocene, for instance; e.g. Lynas, 2011). From another perspective, after all, a wide variety of other organisms and entities, from microbes to mosquitos, can be seen to engage in their own distinct processes with little human control, or even via the use of humans for their own ends, as the ongoing COVID-19 pandemic among other examples clearly shows.

Within any approach, however, the responsibility of human subjects to choose how they will enact their anthropocentrism remains inescapable. Denying this entails disavowing this responsibility and hence one’s duty to think seriously about the ethical and political implications of how this responsibility should be operationalized. Rather than claiming to speak for nonhumans—or to claim that they are speaking through us—therefore, perhaps we humans must instead simply decide among ourselves how to treat nonhumans, as respectfully and responsibly as we can. We invite other researchers to respond to the questions and provocations raised in this analysis as we continue to collectively think through both the ethics and practicalities of multispecies cohabitation.

Highlights

Outlines the factors responsible for peaceful human–bear cohabitation in this case.