Corps de l’article

In 2002, at an international workshop that assembled Siberianist ethnographers from Russia, North America, and Eastern and Western Europe, some workshop participants expressed regret that Siberian ethnography is “treated as a kind of exotica within the discipline of anthropology, and not given credit for theoretical sophistication or looked to for comparative models” (Gray, Vakhtin, and Schweitzer 2003, 195, 206). No doubt, a number of aspiring Siberianist anthropologists, including those who did and did not attend that workshop, had the ambition to change that – to transport Siberia from the margins of the discipline to its theoretical center.

Closed to foreign researchers since the rise of Stalinism in the 1930s (with very few exceptions), Siberia was opened again in the 1990s, after the dissolution of the Soviet Union. Anthropologists from North America, Western Europe, and Japan, most of them young scholars, augmented the ranks of their Russian and East European colleagues, bringing with them their own favored theories and methods (Alymov and Sokolovskiy 2018; Funk 2018; Gray, Vakhtin, and Schweitzer 2003; Vakhtin 2006; Vitebsky and Alekseyev 2015). This occurred just as a number of new trends emerged in the field of anthropology, one of which was the turn to ontology by Eduardo Viveiros de Castro and Philippe Descola. Could their new approaches to perspectivism and animism, developed in their research on the hunters and gardeners of Amazonia, be applied to the hunters and herders of Siberia? This was the question posed by the scholars whose work we address in this review (see also Brightman, Grotti, and Ulturgasheva 2012).

In this article, we provide a critical assessment of contributions by anthropologists who have taken an ontological approach to research in Siberia.[1] This includes both ontologically-informed ethnographies of Indigenous people and comparative studies that draw on Siberian data in broadly conceived analyses of ontological difference. What is the anthropology of ontology, and how has it been applied in ethnographic research in Siberia and in theoretical syntheses based on that research? Which anthropologists have taken ontologically-oriented approaches to Siberian field sites and data? What have proponents of ontological approaches accomplished through their research regarding Siberian people? What difficulties arise upon reviewing texts written in this vein? These are our questions. We do not mean to imply that research on ontological themes or employing ontologically-informed methods is the only or even the most important recent development in the anthropology of Siberia. It is, however, a development that has attracted much attention and requires a critical review.

“The anthropology of ontology” is a label referring to varying trends with diverse roots and some commonalities (Descola 2014; Holbraad and Pedersen 2017; Kohn 2015; Latour 2009; Scott 2013). “Ontology” is about being or conceptions of being; and all proponents of ontological approaches in anthropology are concerned with different ways of being in the world or worlds that humans share with other intentional beings.[2] Ontologically-oriented anthropologists often disagree, however, about the location of “the key site of ontological difference” (Holbraad and Pedersen 2017, 177). Are ontological differences comparable to the social, cultural, political, and economic differences that anthropologists document routinely, though located “at a deeper level, where basic inferences are made about the kinds of beings the world is made of and how they relate to each other” (Descola 2014, 273)? Or does ontological difference subsist in “the activity of anthropology itself” (Holbraad and Pedersen 2017, 177)? In other words, is ontologically-oriented anthropological inquiry primarily descriptive and typological, or is it first and foremost “a methodological intervention” by means of which anthropologists might succeed in enabling “ethnographic material to reveal itself” to them, if they allow it “to dictate its own terms of engagement, … guiding or compelling … [them] to see things that … [they] had not expected, or imagined, to be there” (Holbraad and Pedersen 2017, 4–5)?

Despite disagreeing about how to answer these questions, ontologically-oriented anthropologists share a common critical stance vis-à-vis the received anthropological wisdom, especially that of the cultural relativists. The ontologists’ common point of departure is their critique of the nature/culture distinction, which is thought to bear witness to a hidden contradiction within cultural relativism. In subscribing to the “Western” distinction between “nature” and “culture,” cultural relativists are said to assume that, while cultures are plural, the underlying world of nature is uniform and accessible to us through the methods of the natural sciences (Henare, Holbraad, and Wastell 2007: 9, 11; Descola 2014: 272–273). By privileging scientific knowledge of a uniform nature, anthropologists, like other Westerners, are said to introduce criteria for determining the veracity of various understandings of nature. Therefore, despite their profession of empathetic understanding for members of other cultures, relativists, supposedly, view understandings of nature that deviate from scientific knowledge – for example, the “representations” of the Indigenous people whose lifeways they are investigating – as false or true only in a metaphorical sense. Descola, Viveiros de Castro, and other proponents of the anthropology of ontology reject the nature/culture distinction, viewing it as an expression of a Western or naturalist bias. Rather than remaining trapped in a “naturalist” ontology (Descola 1996; 2013) with its “infernal dichotomies – unity/multiplicity, universalism/relativism, representation/reality” (Viveiros de Castro 2011, 129) – anthropologists, according to these authors, should abandon their weak form of epistemological analysis (showing how people view the world, more or less adequately) and turn to ontology (revealing the worlds they live in).

We turn now to those who have attempted, through research in Siberia, to achieve the goals set by Descola and Viveiros de Castro, among others. The authors whose works we review include only those Siberianist anthropologists who feature in their publications the term “ontology” as a major analytical concept or who cite the chief proponents of ontological approaches in anthropology. If we seem to ignore some prominent representatives of this research orientation, that is because they have had little discernable impact on Siberian studies. We also exclude a number of Siberianist anthropologists who share with the ontologists the same subject matter – spirits, human-animal relations, rituals, etcetera – without taking an ontological approach; and we leave out ontologists who do research in Russia but not in Siberia. Pedersen, a specialist of Mongolia, and Descola, an Amazonist, are included, because they have developed comparative schemes, regional or global in scope, in which Siberia figures prominently. Even given this strict delimitation of our subject matter, we have still had to make a selection – one that we think is representative of a larger body of work.

After commenting on contributions by selected authors, we cite some of the criticisms that have been made of the anthropology of ontology, reflecting briefly on how they might apply to ontologically-oriented research in Siberia. Then, we illustrate some of these critical points by examining how two of the authors under review – Philippe Descola and Rane Willerslev – have drawn on older ethnographies, particularly the works of Waldemar Bogoras (in Russian, Vladimir Bogoraz) on the Chukchi.

Examples of Ontological Approaches in the Anthropology of Siberia

Pedersen on Totemism and Animism in Siberia

Fairly recently, Morten Axel Pedersen argued that “ontological difference” should be sought in “the activity of anthropology itself,” not in “the differences anthropology has posited between cultures” (Holbraad and Pedersen 2017, 177); but, in an earlier publication, he does indeed seek ontological difference between cultures or people. In “Totemism, Animism and North Asian Indigenous Ontologies,” he examines various “theories or understandings about what exists” among people inhabiting the area stretching from “the taiga and tundra regions of … Northeastern Siberia and the Russian Far East” to “the steppe regions of the Mongolian Plateau” and “the forest-steppes of … the South Siberian Transbaikal region” (Pedersen 2001, 411–413). Specifically, he asks whether the typological distinctions suggested by Descola (1996) – “animism,” “totemism,” and “naturalism” – and Viveiros de Castro’s (1998) notion of “perspectivism” are useful in characterizing the ontologies of the Itelmens, Koryaks, Chukchi, Yukaghirs, Eveny, and Evenki in the north and the Halx, Darxad, Buriats, and Altaians in the south. In pursuing the answer to this question, Pedersen acknowledges that he is bracketing out the effects of “over 300 years of Russian, Chinese, and Manchurian presence” in an attempt to reconstruct “conceptualizations of human and nonhuman social life” that existed prior to those effects or continued to exist in spite of them (Pedersen 2001, 412).

For both Descola and Viveiros de Castro, “animism” involves viewing nonhumans such as animals and spirits as intentional social beings, even if they differ from humans corporeally. However, each of these authors has his own way of opening up the category of “animism” for further analysis – Descola (1996; 2005; 2013), by contrasting it with other types of ontology in a comparative approach, and Viveiros de Castro (1998, 482), by developing the concept of “perspectivism,” which he describes as a “somatic complement” to “animism.” Paraphrasing his senior Brazilian colleague, Pedersen (2001, 421) gives the following rendition of “perspectivism:” humans see animals and perhaps also spirits as “others,” but these “others” (presumably) see themselves as humans and humans as “others.” Often, animals see humans as predators, while spirits see them as prey.

In contrast to “animism,” “totemism” is an ontological orientation that makes use of “empirically observable discontinuities between natural species to organise, conceptually, a segmentary order delimiting social units,” each of which incorporates specified humans and nonhumans (Descola 1996, 87). Both “animism” and “totemism” contrast starkly with the ontology that Descola (1996; 2013) calls “naturalistic” and Viveiros de Castro (1998) and Pedersen (2001, 414) call “Western” – an ontology based on a radical distinction between the intersubjective, distinctively human world of “culture” and the objective world of “nature,” the latter of which is accessible to human subjectivity via the methods of natural science (in the understanding of Descola, Pedersen, and Viveiros de Castro, among others).

With reference to his own fieldwork, to classic ethnographies by Jochelson (1908, 1910–26) and Bogoras (1904–09), and to more recent syntheses by Roberte Hamayon (1990) and Caroline Humphrey (1996), Pedersen concludes that “the ontologies of Northern North Asia … are predominantly animistic in nature,” while “the ontologies of Southern North Asia … are predominantly totemistic” (Pedersen 2001, 411). As in pre- or non-ontological contributions to the anthropology of religion, these contrasting orientations correspond to differences in social organization as well. Animism prevails among “small egalitarian bands of reindeer-breeding hunters,” who “organize the world horizontally (through notions of charismatic leadership, egalitarian ethos, bilateral descent, direct exchange, an orally based shamanistic religion, etc.);” while totemism is found in the “deeply hierarchical tribal empires of the Mongolian grasslands,” which “organize” the world “vertically (through notions of inherited leadership, a hierarchical ethos, patrilineal descent, indirect exchange, a script-based Buddhist religion, etc.)” (Pedersen 2001, 411, 419–420).

On the basis of data provided by Jochelson and Bogoras, Pedersen finds “perspectivism” (which he sees as a particular form of animism) in northern North Asia, for example, among the Chukchi and Koryaks; but, he notes, it is largely absent, or has undergone structural transformation, among the herding people in southern North Asia. In the north, as in Amazonia, one finds “extra-human perspectivism (humans becoming nonhumans and vice versa),” while, in the south, one finds only “inter-human perspectivism (humans becoming other humans)” (Pedersen 2001, 421). Descola (2005, 2013) takes up and develops the notion of structural transformation underlying the spatial distribution of ontological differences.

Descola on Northern Siberia as a Transitional Zone

Taking theoretical reformulations based on his own research in Amazonia as his point of departure, Descola has developed a comparative framework, global in scope, which he presents in Par-delà nature et culture (2005) – in English translation, Beyond Nature and Culture (2013). Within this framework, Northern Siberia plays a key role as a transitional zone between North America and the southern regions of Siberia, bordering on Mongolia.

Before arriving in Siberia, Descola unveils his comparative schema, based on distinctions among “four major types of ontology” (he has added one since 1996), which he describes, generally, as “systems of the properties of existing beings,” serving as “a point of reference for contrasting forms of cosmologies, models of social links, and theories of identity and alterity” (Descola 2013, 121). He distinguishes these four types from one another by asking whether or not humans see nonhumans as having similar or dissimilar “interiorities” and “physicalities” – whereby “interiority” refers to the supposed presence or absence of the soul, consciousness, subjectivity, or intentionality, while “physicality” refers to bodily form. The four types are: animism (similar interiorities, dissimilar physicalities), totemism (similar interiorities, similar physicalities), naturalism (dissimilar interiorities, similar physicalities), and analogism (dissimilar interiorities, dissimilar physicalities).

The four types of ontology, which correspond to distinct “theories of identity,” are supplemented by “models of social links” or “modalities of relation.” These include “exchange,” “predation,” and “gift-giving” as well as “production,” “protection,” and “transmission” – categories to which Descola refers, first, in specifying actual ontological variation from one people or region to another and, second, in analyzing the transformation from one type to another (Descola 2013, 309–335).

In mapping the spatial distribution of ontological variables, Descola (2013, xix) remains largely in “an ethnographic present,” eschewing historical or evolutionary implications – even in the chapter entitled, “Histories of Structures,” in which he traces the distribution of variables over an area stretching from North America, across the Bering Strait, to Pedersen’s two halves of North Asia. His goal is to “envisage a spatial continuum of comparable societies as if they were transformations of one another, … [l]eaving aside questions of genesis and antecedent causality” (Descola 2013, 366). Northern Siberia figures as a transitional zone between two regions in which aboriginal ontologies correspond more closely to one of Descola’s pure types. Animism is characteristic of Native North America; but, as one moves across the Bering Strait, animistic ontologies display indices of gradual change, which, by the time one arrives at the borders of Mongolia, result in a transformation to analogism.

In the following, we summarise very briefly Descola’s analysis of spatially distributed ontological variation by contrasting different master spirits and their relations with animals and humans. A being known as Caribou-Man among the Montagnais (or Innu) of northeastern North America provides an example of “gift-giving animism.” Caribou-Man has a protective relation with wild caribou, but he is willing to offer some of his protégés to hunters, “expecting no compensation, provided the ethics of hunting remain respected” (Descola 2013, 367).

From the hunting people of North America to the Exirit-Bulagat, horsemen and cattle herders of Buryatia, Descola traces a transition in subsistence, social organization, and ontology. Among the Exirit-Bulagat, “the egalitarian relations between collectives of human and nonhuman persons” that are characteristic of North America “have been supplanted by vertical relations of differentiation that Pedersen,” citing Descola 1996, “calls ‘totemic’” but that “turn out to be much closer to what I call the analogist system” (Descola 2013, 371–372). In both its social and cosmological aspects, the world of the Exirit-Bulagat is “divided into separate hierarchical and complementary units, each of which is attached to particular sites and composed of a mixture of entities of various kinds” (Descola 2013, 374).[3]

The dominant spiritual presence among the Exirit-Bulagat is Buxa Nojon or Lord Bull, an original progenitor who sired the ancestors. This divinely human lineage is indicative of a further transformation in relations of “protection,” which now entail superordination and subordination. Just as livestock are subordinated to and protected by humans, so humans must subordinate themselves to their ancestors and to the gods from whom they descend, making sacrifices to them in order to gain their protection for themselves and their herds (Descola 2013, 376–377).

For Descola, the Chukchi occupy an intermediate position between the animists of North America and the analogists of Buryatia, both geographically and ontologically. Indicative of their intermediacy is their double orientation toward two types of spirit master simultaneously. The Chukchi combine the hunting of wild reindeer with the herding of semi-domesticated reindeer.[4] Each category of reindeer has its own spirit master. The master of the wild reindeer is Pičvu’čin[5], who belongs to the class of spirits known as ke’le [kêly] in the singular and ke’let [kêl’êt] in the plural. Although he resembles the North American Caribou-Man, Pičvu’čin also differs from him markedly, insofar as wild reindeer killed by Chukchi hunters are not gifts but objects of exchange, which must be compensated with various offerings (Descola 2013, 369). In contrast, domesticated reindeer are under the protection of “benevolent powers” known as va’irgit [vagyrgyt].[6] Descola (2013, 370) describes the va’irgit as “impersonal and localized manifestations of cosmic vitality with which no kind of interaction is possible.” Chukchi, he says, do not exchange with the va’irgit, as they do with Pičvu’čin; rather they sacrifice domesticated reindeer to them. Here, as among the Exirit-Bulagat, sacrifice is an index of analogism, though, among the Chukchi, this ontological regime is only “embryonic” (Descola 2013, 371).

Descola’s analysis of this regionally distributed set of structural variants invites a number of critical remarks, which we reserve for the final section of this review (see also Ingold 2016; Kapferer 2014).

Willerslev Among the Yukaghirs

In a review of a work by Pedersen (2011), James Laidlaw (2012) notes oscillation in Pedersen’s understanding of the task that he has set for himself: Is he revealing the different worlds in which his field interlocutors live their lives, or is he describing their distinctive conceptions of the world? Rane Willerslev stands out as an author who displays this same ambivalence but tries to make it central to his analysis.

During his fieldwork among the Yukaghirs of the Upper Kolyma region from 1999 to 2000, Willerslev joined a small group of men hunting in the forests surrounding the village of Nelemnoye on the Yasachnaya River. In his subsequent publications, he illuminates hunting, Yukaghir-style, by “merging … two kinds of theories: the proposition of ‘Amerindian perspectivism’ (Viveiros de Castro 1998) and notions of ‘mimesis’ (Taussig 1993)” (Willerslev 2004a, 629).

Yukaghir hunters imitate reindeer and elk (that is, moose) in order to lure them within shooting distance (Willerslev 2007, 1). The term “mimesis” is, then, a felicitous choice, since one of its meanings is “imitative behavior of one species by another.” But Willerslev (2007, 9) prefers to follow Taussig in defining “mimesis” as “the meeting place of two modes of being-in-the-world – ‘engagement’ and ‘reflexivity’… ‘sameness’ and ‘difference,’ ‘self’ and ‘other,’ ‘me’ and ‘not me.’” This play of opposites is said to characterize the “world of the Yukaghirs,” who “must constantly steer a difficult course between … transcending difference and maintaining identity” (Willerslev 2007, 12).

In describing the hunter’s mimesis of his prey, Willerslev alternates between formulations corresponding to the two poles of the ambivalence that Laidlaw (2012) identifies. On one hand, Willerslev makes strong claims regarding the ability of Yukaghirs to become members of other species: “they can and do transform themselves into various others, both human and nonhuman, but must avoid total participation and confusion” (Willerslev 2007, 12). On the other hand, he sometimes suggests not that Yukaghirs “take on the body of a being from another species” but that they “regard it as possible” (Willerslev 2007, 89).

Aside from a few brief references to purported cases of metamorphosis, the examples that Willerslev provides take the form of cautionary tales in support of the statement that “taking on the body of another species can … only be done for short periods of time” because “it does pose risks” (Willerslev 2007, 89). One such tale is based on Willerslev’s own experience. When he and a partner were in the forest for two months, trapping sables, they “became increasingly obsessed with accumulating furs” (Willerslev 2007, 91).

Then one evening … Ivan said, ‘Can’t you feel it?’ ‘Feel what?’ I asked. ‘How we are turning into greedy predators, just like wolves. We have this need to kill more and more. Even if we had two hundred sables we wouldn’t feel satisfied, would we? Just like the devil, you see.’ He paused for a while. Then he added, ‘I suggest we calm down … and stop hunting for a week or so’

Willerslev 2007, 91

This intriguing story seems to be linked to the ethics of hunting and trapping, since, for Yukaghirs, “predators are ‘children of the devil’” and “a human person who [like a predator] kills recklessly and ignores the rituals intended to secure the proper circulation of game is called ‘a son of the devil’” (Willerslev 2007, 92). One might assume, then, that stories of this type serve to discourage overhunting by threatening those who overhunt with frightening consequences. But Willerslev’s ontological stance makes him hesitant to interpret his partner’s statements strictly in metaphorical terms:

[A]nthropologists, [who are] by temperament and training inclined to be rather more sympathetic to the indigenous viewpoint, would accept the hunter’s story by adding an ‘as if’ to his account – so instead of talking nonsense, the hunter is deemed to be speaking in metaphors, constructing figurative parallels between the two separate domains of nature and culture. However, to say that the hunter is talking ‘as if’ animals were persons is to say that his story should not be taken in a literal way but instead seen as a symbolic statement … In this book, however, I wish to reverse the primacy of Western metaphysics over indigenous understandings and to follow the lead of the Yukaghirs in what they are saying about the nature of spirits, souls, and animal persons. Only in this way can we hope to develop a framework that takes their viewpoints on these matters seriously

Willerslev 2007, 2–3

Apparently, however, the Yukaghir view of animals as persons with whom they are essentially interchangeable is not continual but situational:

The personhood of animals and things is … something that emerges in particular contexts of close practical involvement, such as during hunting. Outside these particular contexts, Yukaghirs do not necessarily see things as persons any more than we do, but instead live in a world of ordinary objects in which the distinction between human subjects and nonhuman objects is much more readily drawn

Willerslev 2007, 8

In Willerslev’s view, it seems, any ambivalence about the relation of the Yukaghirs to animals and spirits is an attribute of Yukaghir ontology itself, since, for Yukaghirs, “the world of dreams and waking life are two sides of the same reality, which together constitute one world, and neither is therefore amenable to prioritization” (Willerslev 2007, 176). In this sense, the anthropologist’s nature/culture dualism, based, supposedly, on an opposition between the real world and more or less accurate representations of it, is replaced by the Yukaghir form of dualism, one in which the two opposed categories – waking and dreaming, the living and the dead, humans and non-humans – enjoy equal status (Willerslev 2007, 174–176).

Ghost Stories, as Told by Ulturgasheva and Broz

Conducting fieldwork among the Eveny in the village of Topolinoye between 2003 and 2013, Olga Ulturgasheva focused on young people and on the djuluchen, a “travelling spirit” or “forerunner,” which she describes as a key to “the ontological framework” of Eveny adolescent personhood (Ulturgasheva 2012, 167). But, in this review, we focus on another type of spirit that she encountered during fieldwork: “The village … was built on the territory of a former Stalinist labour camp in the 1970s and is considered particularly accursed, haunted by the ghosts of former inmates” (Ulturgasheva 2017, 27).

In Topolinoye, as elsewhere in Siberia, the dead are said to live nearby in a parallel world, which is an inverted image of the world of the living. What is more, the dead are seen as desiring the company of the living, and, so, they must be appeased with offerings, lest they pull the living over to their side: “Locals … feed the deceased by putting a shot of vodka, tea and food on their graves. By such gestures Eveny facilitate peaceful and non-disturbing behavior of the spirits of the deceased” (Ulturgasheva 2017, 40).

However, the unquiet spirits of those who died in the labour camp and who, having been “cut off from their kinship networks,” did not receive last rites are immune to such ritual appeasement (Ulturgasheva 2017, 40). Therefore, for the Eveny, the Gulag ghosts are equivalent not to their own dead but to the arinkael – “the Even name for malevolent forest spirits” (Ulturgasheva 2017, 34). The only option available to villagers is to avoid such spirits as best they can, either by vacating haunted flats or leaving the village.

Ulturgasheva notes a dramatic increase in the presence of arinkael or Gulag-ghosts in the village from the Soviet to the post-Soviet era:

[D]uring the period of the state’s generous support of the local infrastructure (the 1970s and 1980s) this sense of malevolence was not apparent … Flying back and forth [for example, between the village and the regional capital] … would allow one to stay mobile and not to feel trapped within a single social and geographical space. The collapse of this elaborate infrastructure which occurred in the 1990s generated a quite new sense of abandonment, and isolation from the outside world … [T]he appearance of ghosts in local buildings and of stories related to them only became common in the early 1990s”

Ulturgasheva 2017, 41

In short, the ghosts arrived in great numbers only after the dissolution of the Soviet Union and the disintegration of the infrastructure connecting the village to the outside world. Nevertheless, Ulturgasheva seems to resist the temptation to understand ghost stories as a metaphorical expression of the resulting disorientation and desperation of the villagers – an analysis typical of the anthropology of postsocialism and, more generally, of cultural relativism. Rather, she suggests, matter-of-factly, that when Soviet-era settlers abandoned the village in the 1990s, the ghosts of an even earlier generation of newcomers, that is, of Gulag prisoners, moved into the empty flats. The supply of vacated dwellings corresponded, then, to the demand of Gulag-ghosts for housing – a thesis that corresponds quite straightforwardly to efforts of ontologically-inclined anthropologists to take seriously people’s understandings of their own world, without explaining away apparent anomalies in rationalist terms.

In her account of relations between Eveny villagers and the Gulag-ghosts, Ulturgasheva encounters the problem that worried Laidlaw (2012) in his critique of Pedersen (2011), just as Willerslev did in his work on the Yukaghirs: Is she talking about another world, one that is different from “ours,” or is she talking about another understanding of the world that “we” all inhabit? Or, to rephrase the question: Is she talking about spirits or about spirit beliefs? In a way that is quite original, Ludek Broz (2018) reflects on just this point, that is, on the difficulty that ontologically-inclined anthropologists have with “ghosts, spirits, gods or demons.” After relating ghost stories from his Altaian field site, he notes that the “various anthropological schools have differed only in the … way in which they have analytically substituted [that is, replaced] entities that are ontologically dubious from the point of view of Western science” – such as ghosts – with “allegedly more respectable” explanations, viewing them, for example, as symbolic expressions of social tensions (Broz 2018, 7). But, he adds, for anthropologists adopting an ontological perspective, ghosts present a real problem, insofar as those who are committed to “offering the vision of a multiplicity of worlds” are suddenly under great pressure to reduce those worlds to theirs (Broz 2018, 7–8). “Messing with spirits and ghosts is dangerous” for Eveny and Altaians, on one hand, and for anthropologists, on the other (Broz 2018, 10). Among the former, the living run the risk of being lured over to the side of the dead; while, among the latter, an “open-ended look at ghosts that does not a priori rule out their existence” may have a similarly transformative and destructive effect, namely, “the potential discrediting of our work in the eyes of our colleagues” (Broz 2018, 8). Broz wonders – half-seriously, it seems – whether that effect as well may be attributed to the ghost’s agency.

Critique of the Anthropology of Ontology in Siberia – on the Example of Using Data from Bogoras and Jochelson

Any thorough review of ontologically-oriented approaches to Siberia would also include, ideally, reflection on the degree to which common criticisms of the anthropology of ontology are applicable to research in this region. Just as some commentators have criticized the anthropological ontologists for neglecting colonial history and the political and economic situation of Indigenous people today (e.g., Heywood 2017, 8; Ramos 2012), so might one take exception to the relative dearth of references, in the corresponding research in Siberia, to the effects of Christianization or of Christian influences during the Tsarist era (e.g., Lambert 2007–08, 2009), to sedentarization and collectivization under the Soviets (e.g., Slezkine 1994; Ssorin-Chaikov 2003), and, starting in the 1990s, to economic privatization and the withdrawal of state services (e.g., Stammler 2005), the hindrance of political mobilization among ethnic minorities (e.g., Gray 2005), and unofficial efforts of officials to prevent Indigenous people from taking advantage of rights that exist largely on paper (e.g., Donahoe 2009).

Similarly, just as some critics have focused their attention on the tendency of the anthropological ontologists to exoticize Indigenous people, while oversimplifying their characterizations of “the West” (e.g., Brightman 2007; Eidson 2019, 145–150; Graeber 2015, 33; Scott 2013), so might one ask whether Siberianists taking an ontological stance do the same. Indeed, comments that some Siberianist ethnographers have made regarding the work of their colleagues, point in this direction. Agnieszka Halemba (2006, 143) has warned against “creating a coherent ‘cosmology’ out of bits and pieces of information” gathered during fieldwork; Willerslev and Ulturgasheva (2012, 50) have criticized those who assume that Siberian communities may be categorized in terms of “the predominance and exclusiveness of one ontological model:” while Vera Skvirskaja (2012, 147) has called on her colleagues to go beyond “the ‘one ontology’ paradigm” and to acknowledge the diverse influences and heterogeneous orientations in the lives of the people under investigation.

While each of these critical perspectives on ontological approaches in Siberia calls for extensive discussion, we choose in this necessarily brief contribution to focus on one particular illustration. The way in which two of the authors featured above – Descola and Willerslev – use data from classic ethnographies of the early twentieth century may serve as an example of tendencies to view Siberian people as if they existed outside of history and to lend their ontologies undue coherence.

Anthropologists gain access to Siberian ontologies – or to Siberians ontologically – not only through fieldwork but also through the older ethnological literature, especially the literature stemming from the time preceding massive Soviet interventions into the lives of Indigenous people. In this regard, those working among the Chukchi, Koryaks, and Yukaghirs have an advantage over others, insofar as they can draw on the classic ethnographies of Bogoras and Jochelson.

Bogoras’s book, The Chukchee (1904–09) is the only source that Descola (2005, 2013) consults regarding this people, which occupies a pivotal position in his scheme. Willerslev supplements his own fieldwork among the Yukaghirs and Chukchi by drawing heavily on Jochelson (1908, 1910–26) and Bogoras, respectively (Lykkegård and Willerslev 2016; Pedersen and Willerslev 2012; Willerslev 2004a, 2004b, 2007, 2009, 2011; Willerslev, Vitebsky, and Alekseyev 2015). Given the degree to which Descola and Willerslev depend on these authoritative sources, it is worth asking how they have used them. To this end, we focus on a particular ethnographic puzzle, namely, the nature of spirits and the quality of human-spirit relations among the Chukchi.

Both Willerslev (2011) and Descola (2013) focus on the distinction between two Chukchi terms designating types of spirits, ke’le and va’irgin. The ke’le, in their view, is a spirit that is typical of animism, insofar as it is an intentional entity that interacts with humans in ways that range between malevolence and ambivalence (Willerslev 2011, 511–517; Descola 2013, 368–371). Va’irgin, on the other hand, is the exceptional category within this contrastive set, and both authors build their theories on the basis of their understanding of it.

Willerslev accepts Jochelson’s (1908, 24) gloss of va’irgin as the word for “the Supreme Being,” though he concludes that it designates not a personal god but something comparable to “orenda among the Iroquois,” “mana among the Maori,” and “the Creative Spirit of the Nuer, whom Evans-Pritchard … described as beyond contact and comprehension yet the giver and sustainer of life” (Willerslev 2011, 517–518). Taking this understanding of va’irgin as his point of departure, Willerslev seeks to mediate the old debate between Edward B. Tylor and Andrew Lang by arguing not for the priority of either animism or monotheism but for their interdependence – for what he calls “the animist perspectival dependence on a Supreme Being” (Willerslev 2011, 520).

Descola’s understanding of va’irgit – a term he always used in the plural – is rather similar to that of Willerslev. He describes va’irgit as “impersonal and localized manifestations of cosmic vitality,” which are “totally distinct” from ke’let (Descola 2013, 370–371). Ke’let and va’irgit are said to differ in the following ways: the relation of the ke’let to humans is ambivalent, personal, interactive, and based on exchange; while the relation of the va’irgit to humans is benevolent, impersonal, non-interactive, and requires sacrifice. The distinction is important to Descola, because it supports his theory that the Chukchi occupy an intermediate position between the animist North Americans and the analogist Exirit-Bulagat of southwestern Siberia: just as the ke’let are a type of spirit that is consistent with animism, so the va’irgit are a type of spirit that is consistent with analogism.

There are several passages in Bogoras that support the interpretations of va’irgin/va’irgit that Willerslev and Descola present. Indeed, Bogoras sometimes seems to promote this interpretation by contrasting the ethereal, benevolent va’irgit (Bogoras 1904–09, 303, 314) and the earth-bound, demonic ke’let (Bogoras 1904–09, 291–299, 311). However, if one considers the full range of data in Bogoras (1904–09; Bogoraz 1900), the distinction between ke’le and va’irgin all but vanishes.

Curiously, neither Willerslev (2011, 518) nor Virginie Vaté (2007, 231) heard the term va’irgin during fieldwork among the Chukchi. This word – in contemporary orthography, vagyrgyn – is composed of the Chukchi verb vak (-k marking the infinitive form), meaning “to be,” “to live,” or “to be located,” and the substantive suffix, -gyrgyn; so that the combination of these two morphemes gives the meaning of “that which exists,” “that which is,” “life,” “being,” or “entity” (Vaté 2007, 223; Weinstein 2018). Va’irgin is, then, a generic expression, which, in combination with other expressions, can have widely differing meanings. For example, atal or etel va’irgin [atalvagyrgyn, from ataltann’ytan – Yukaghir] means “the … Yukaghir disease,” that is, syphilis (Jochelson 1908, 418); ina’n-tam-va’irgin [ynantamvagyrgyn] refers to the “best thing (in the world),” that is, sexual intercourse (Bogoras 1904–09, 571); and akau’ka-va’irgin is “the usual Chukchi expression for trouble, trespass, crime” (Bogoras 1904–09, 662) [akavky vagyrgyn, misfortune; akavkêvagyrgyn, crime, offense]. So, in and of itself, va’irgin is a general term for something that exists, with neither positive nor negative connotations.

Descola insists, however, on the distinction between ke’let and va’irgin, illustrating it by contrasting the ke’le spirit called Pičvu’čin, who is the master “of wild reindeer and of all land-game” (Bogoras 1904–09, 286), with a va’irgin, the “Reindeer Being” (Qo’ren va’irgin in Bogoras 1904–09, 315, or Qorên vagyrgyn), whom Descola identifies as the “nonhuman protector” of domesticated reindeer. When hunters kill a wild reindeer, they must give Pičvu’čin tobacco, sugar, flour, etcetera, in compensation. In contrast, “no kind of interaction is possible” with the va’irgit, including the Reindeer Being. “The sacrifices of domesticated reindeer to the va’irgit are therefore not transfers from one group of individuals to another, which call for reciprocation – as in the case of the food given to the master of wild reindeer” (Descola 2013, 370).[7]

Descola draws a sharp distinction between Pičvu’čin and the Reindeer Being, because, in his understanding, the former is a ke’le and the latter is a va’irgin. But these two spirits have something very important in common: both are e’tin [êtyn]. E’tin means “owner” or “master.” Among the Chukchi, almost everything is thought to have a spirit master – animal species, types of trees, forests, lakes, and so on (Bogoras 1904–09, 285). Bogoras states explicitly that Pičvu’čin is a master, and Reindeer Being is called both Qo’ren va’irgin and Qo’ren e’tin [Qorên êtyn] (Bogoras 1904–09, 286, 315; Bogoraz 1900, viii–ix). Once we learn that Pičvu’čin and Qo’ren va’irgin are both e’tin, the border between ke’le and va’irgin blurs.

Descola’s contrast between the ke’le Pičvu’čin, as a partner in exchange, and the va’irgin Reindeer-Being, as a recipient of sacrifices, is contradicted by information that Bogoras provides. In more than one passage, Bogoras (1904–09, 290, 296, 370) states that people sometimes do make sacrifices to k’elet. This is also true specifically of Pičvu’čin, of whom Bogoras writes the following: “Sometimes he may be seen passing the entrance of a house in the shape of a small black pup. An inspection of his footprints [which look like those of a mouse] will reveal his identity. Then the people must immediately offer him a sacrifice” (Bogoras 1904–09, 286–287).

In a section of The Chukchee in which Bogoras provides translations of incantations, one finds the following incantation that may be used to request of the Being of the Zenith that he help in capturing a wild reindeer buck:

‘O Va’irgin! do not despise my demand. Let me get possession of him [the wild reindeer]! I will give you in exchange something equally worthy of desire’

Bogoras 1904–09, 497

This contradicts Descola’s claim that the va’irgit are not intentional entities who can interact and enter into exchange relations with humans. Therefore, it also relativizes the distinction between ke’le and va’irgin.

One final example leads us to the conclusion that va’irgit and ke’let are by no means “totally distinct” (Descola 2013, 370). In the iaranga, the dwelling of the nomadic Chukchi, “house-spirits” or ya’ra-va’irgit [iaravagyrgyt, from iaran’y – the tent, the house + being] live peacefully with humans, as long as humans “feed” them and obey the behavioural rules of everyday life (Bogoras 1904–09: 318; see also Vaté 2007, 231). In contrast, members of the category “Earth-spirit” or “Ground-Being” – the Chukchi term is Nota’s.qa-Va’irga or Nota’s qa-Va’irgin [notasqyvagyrgyn, from nutêsqyn – earth, ground + being] (Bogoras 1904–09, 293; 1910, 23–25) – are “ke’let who live in deserts” and who are thought to be “very dangerous” (Bogoras 1904–09, 293, 318; 1910, 23). One sees that, in this case, a ke’le is referred to with a complex expression containing the word va’irgin, showing that, in this context, the words are interchangeable. But the main point is this: Bogoras tells us that, if a human dwelling is abandoned, “the ‘house-spirits’ turn into very dangerous ‘earth-spirits’” (Bogoras 1904–09, 318). If, as this example shows, beings that are routinely called va’irgin can, under changing circumstances, become beings that are routinely called ke’let, then the distinction between these two categories cannot be called “total.”

What are the implications for Descola and Willerslev of this review of the meaning of the terms k’ele and va’irgin? Willerslev, who identifies va’irgin as the “animist high god,” may have to rethink aspects of his theory of “animist perspectival dependence on a Supreme Being,” which he develops en route from Siberia to the Grotte des Trois Frères in southern France (Willerslev 2011, 517, 520). Descola, in turn, may have to reformulate parts of his argument regarding the transformation from animism to analogism in the passage from North America to Buryatia (see also Sahlins 2014).

More generally, the examples of the k’ele and va’irgin show that caution is advised when using secondary sources, even in the case of such a grand achievement as Bogoras’s book, The Chukchee. For researchers working among the Chukchi or in Siberia generally, Bogoras’s massive volume is a godsend. Still, in his discussion of the various types of spirits – va’irgin and ke’le are only two of many – Bogoras provides a wealth of information, which, however, often seems to be “indefinite” or “unclear,” as Willerslev (2011, 518) himself notes. The reader sometimes gains the impression that Bogoras struggled with inconsistent information gathered during fieldwork, developing elaborate typologies in an attempt to resolve contradictions (e.g., Bogoras 1904–09, 291–293 on the ke’let). While Bogoras sought coherence in Chukchi accounts of the spirit world, the people Bogoras met were not necessarily concerned with this sort of coherence – as Willerslev (2004b, 399) himself notes. On one occasion, Bogoras asked someone who was sacrificing a reindeer to whom the sacrifice was directed, and the response he received was: “Who knows! To the va’irgin, to the ke’le” (Bogoras 1904–09, 290). Similar experiences led Vaté to summarise the situation as follows:

Reading … authors … such as Bogoras or Vdovin, … one gets the impression that in Chukchi representations many spiritual entities have well-defined outlines and are given fixed names. My experience in the field, on the contrary, has led me to conclude … that, for the Chukchis, spirits are flexible and ambiguous categories of which there are many regional and familial variations … [F]or Chukchis the essence of spirits is ambivalent and fluid, in much the same way as is the relation that humans establish with the reindeer and the land

Vaté 2007, 219–220

Conclusion

Contemporary anthropological research in Siberia covers a broad range of topics – human-animal relations (e.g., Anderson et al. 2017; Stépanoff 2017); Indigenous rights (e.g., Gray 2005; Donahoe 2009); the anthropology of Christianity (e.g., Vaté 2009); extractive industries (e.g., Wilson and Stammler 2015); and climate change (e.g., Lavrillier and Gabyshev 2018). To date, however, the anthropologists of ontology seem to have had most success in moving Siberia from the margins toward the centre of discussion in our field.

Descola (2005, 2013), a major figure in international anthropology, has no doubt drawn attention to Siberia by integrating it into the comparative scheme of his widely read works, while Pedersen (2001, 2011), a leading proponent of the ontological turn, has contributed to this project as well. Similarly, Willerslev and Broz have brought Siberian materials to major journals, stimulating debate and enlivening discussion (e.g., Broz and Willerslev 2012; Ingold 2015; Pedersen and Willerslev 2012; Willerslev 2004a, 2004b, 2009, 2011; Willerslev, Vitebsky, and Alekseyev 2015). Some monographs based on fieldwork in Siberia have gained recognition far beyond the restricted sphere of Siberian or Arctic research – Willerslev (2007) is a good example.[8] As a result, or in a parallel development, Amazonia-Siberia has become an important axis of research on animism and perspectivism, thus occupying a central place in the anthropology of ontology (e.g., Brightman, Grotti, and Ulturgasheva 2012; Vitebsky and Alekseyev 2012, 447). The question remains whether the shift of Siberianist research toward the centre of anthropological discussion is also indicative of an improvement in or refinement of anthropological theory and method. On this point debate prevails.

Curiously, the anthropology of ontology has gained a reputation for theoretical and methodological innovation, while renewing interest in the themes of over a century ago – animism, totemism, primitive monotheism, and what used to be called “primitive mentality.” Clearly, however, the simple opposition between the animists of Siberia and the naturalists of the West will no longer do. Rather, the people of Siberia, Indigenous or otherwise, must be viewed within the historically constituted world or worlds that they occupy. As Broz notes, moving beyond oversimplified understandings of those espousing “animist” and “naturalist” orientations to life would be salutary but might also blur the distinction between ontological and non-ontological approaches in anthropology:

Even if we take such difference [between animism and naturalism] as an analytical starting point we should recognise that the ‘incompatibility’ is bridged in dozens of ways … which deserve to be studied more closely and accommodated in the theoretical framework. It is possible that the clear distinction between perspective/ontology and view/epistemology, which constitutes the difference between perspectivism and relativism, might not hold that firmly

Broz 2007, 306

Broz’s critical observation also seems to suggest the possibility of integrating ontological concerns within projects devoted to the investigation of contemporary issues of grave importance to Indigenous people. In this sense, one might monitor with interest the work of researchers in the project Cosmological Visionaries: Shamans, Scientists, and Climate Change at the Ethnic Borderlands of China and Russia, funded by the European Research Council for six years, beginning in September 2020, with Katherine Swancutt as head of the China team and Olga Ulturgasheva as head of the Russia team.