Absent powers: magic and loss in post-socialist Mongolia

Absent powers: magic and loss in post-socialist Mongolia L H ø University of Copenhagen Across contemporary Mongolia, people feel that precious religious knowledge has been lost during socialism, and that an unbridgeable gap between pre-socialist and post-socialist times has emerged. Instead of viewing this as a passive lack of knowledge, however, this article explores the procreative aspects of absent knowledge. It is argued that such absence is a precondition for the efficacy of certain religious images and practices because it directs people’s attention towards a compelling unknown. Here, religious knowledge is effective only insofar as it is revealed as a concealed but potent presence through magico-religious charms and (post-)socialist imaginations. It is argued that the presence of absent knowledge is a cause that does not contain its effects, and, hence, that relations to the unknown are also unknown relations. When Mongolian villagers and nomadic pastoralists left socialism behind in the early s, they were filled with the certainty that something had now changed – postsocialism had become an ethnographic reality rather than merely an academic invention. Socialism had profoundly altered the cultural and religious conditions prevailing in pre-revolutionary Mongolia, and with the passing of socialism Mongolians confronted the fact that something had been lost and an unbridgeable gap between presocialist and post-socialist times had emerged (cf. Humphrey ). It is this perception of gap and loss that is the focus of this article. However, rather than seeing it as lamentable loss, I will be concerned with the creative aspects of absent knowledge. Speaking in the most general terms, anthropology has been obsessed with comprehending, understanding, and accessing culturally constructed knowledge and alternative ways of knowing. These, it is assumed, exist and can be comprehended, understood, and accessed by the culturally adept person as well as by the anthropologist. The anthropologist claims – explicitly or implicitly – that either it is possible to access this knowledge by using the right methods, or alternatively such knowledge cannot be accessed, thereby still presupposing its existence. The anthropology of knowledge, often indiscernible from a general anthropology of culture or meaning1 (Boyer ; Crick ), then, is concerned with an assumed presence of cultural knowledge that can, at least in principle, be accessed, even if such knowledge is dynamic Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  576 Lars Højer (i.e. it is still there to be followed), unevenly distributed (i.e. some still have it), or based on embodied competence and experience (i.e. it can still be learned). As such, the presence of cultural knowledge might have different roots and dynamics, but it is there to be grasped. Within the past few decades, a number of discussions concerning the construction of knowledge, culture, and cosmology, on the one hand, and the revival, reinvention, or memory of tradition, on the other, have repeatedly addressed this question of cultural knowledge (see, e.g., Barth ; Hobsbawm & Ranger ; Jing ; Keesing ; Watson ). It has been shown that traditions of knowledge – or, alternatively, culture or cosmology – are not easily detachable from their pragmatic and performative transmission through social relations (Barth ; ), and it has been ‘revealed’ how cultural traditions are constructed, that is, made up or transformed, in line with power structures (Bloch a; b; Keesing ) or according to the (instrumental) agenda of elites or a present concern with national or other collective identities (Kaplonski ; Trevor-Roper ). Alternatively, it has been argued that cultural structure is not a side-effect of politics but a potent and invisible presence in human endeavours, and, hence, that one should substitute ‘invention of tradition’ with ‘inventiveness of tradition’ (Sahlins ). For instance, a number of recent studies focusing on Mongolian and Inner Asian cosmologies and practices have shown them to contain their own specific cultural-cum-regional ontology, in one instance labelled ‘transcendental perspectivism’ (Holbraad & Willerslev ) and in another ‘totemism’ (Pedersen ). Apart from assuming the omnipresence of culture/knowledge, such discussions demonstrate that the sociology or anthropology of knowledge (Barth ; Boyer ; Crick ; Keesing ) – or, more generally, an epistemological anthropology of what and how we are made to know – has been the focus of much attention, as has the study of (cultural) ontologies of unique (and existing, i.e. present) ethnographic realities (Holbraad & Willerslev : ).2 The fundamental agreement behind the disagreements – politics versus tradition or function versus culture – is that the presence of culture, knowledge, or ontology is not to be questioned. Less attention, however, has been paid to absence of knowledge as an elusive but effective agent in people’s actions and as a proper anthropological (non-)object of investigation. This does not contain cultural presence – whether in the form of ontology, cosmology, or knowledge – at its centre. The construction of knowledge loss in socialist historicity, for example, and the constructiveness of absence of knowledge in magical practices, as well as the intimate and mutually encouraging relation between the two, is a case in point. It is the aim of this article to remedy such disregard for absence, because certain magico-religious practices in Mongolia (and elsewhere), it will be argued, can only be understood if absence of knowledge – and, more precisely, knowledge of absence – is treated as a powerful presence gaining effects exactly by not being fully known. The argument will proceed as follows. First, the problem about loss of knowledge in the Mongolian post-socialist context is briefly addressed. Next, attention turns to the socialist understanding and fabrication of history. It will be argued that, apart from having literally made persons, things, and religious knowledge disappear, socialism has also made people imagine a radical disconnection between past and present and served to empower what it was meant to suppress. Such radical disconnections with both the past and socialism’s imagined enemies have now become a powerful but also muddled and confused space for magical agency in post-socialist Mongolia. Third, it will be Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 577 shown how a particular magico-religious object is constructed in order to defer full comprehension and thereby highlight the presence of a compelling absence. It is further argued that this mode of empowering is, if not identical to, then at least analogous to the unintended effects of socialism described in the earlier section. Finally, the two strands – magic and socialism – will be brought together in a case study from Ulaanbaatar, the capital of Mongolia. Unknown spirit powers in contemporary Mongolia On the surface of it, the unknown, or simply the lack of knowledge, is a problem in present-day Mongolia.3 Although Tibetan Buddhism was popularized and institutionalized in Mongolian areas in the sixteenth century and emerged as the strongest institution in Mongolian society at the beginning of the twentieth century, when the Manchu Qing dynasty collapsed, all this came to an abrupt end during socialism. While some people secretly continued to carry out rituals at home and kept religious objects in the bottom of their household chests, away from public view, and even though knowledge of religious persons and texts was sometimes retained in ambiguous semipublic discourses (see Humphrey ), it is hard not to conceive of this as a historical rupture and detachment from the past. Most religious practitioners were executed or forced to abandon their religious practices and a huge number of religious objects and buildings were destroyed during the Stalin-inspired purges of the s. Combined with a general ideological campaign against ‘superstition’ during socialism, such destruction has led to a major loss of canonical magico-religious knowledge. One lama4 explained to me that socialism changed the mind of the young generation, who learned to destroy old religious objects with ease. Certainly, the transmission of religious knowledge was risky in socialist times, and now (people say) the old, wise Buddhist lamas have all died. If any potential students of magico-religious traditions were present during socialism, the teachers were – quite understandably – reluctant to accept them, and religious practices in the post-socialist era have often been constrained by the shortage of lamas, at least in the countryside.5 People are therefore acutely aware of ‘loss’ and consider this particular loss of knowledge as a hindrance to the contemporary practice of religion, often claiming not to be strong believers in anything. In addition, people imagine the past as a time when this loss was absent; back then, they would say, people knew all those things. So a common concern among Mongolians seems to be: how should one practise religion in the face of this loss, when one does not know how to do it, or what kind of powers one is dealing with?6 By turning this concern on its head, I will argue that such loss has a productive side to it, and that it is exactly from – and in – this loss that magical technologies and powers emerge. The concern with loss and absence as productive categories necessitates a reconsideration of the native Mongolian assumption that the amount of magicoreligious knowledge and the magnitude and extent of religious practice correlate with people’s strength of belief, as in the often-heard conflation of lost knowledge with lost religion. This quantitative conflation implies that the more you worship and know, the more you are thought to believe, and the more you believe, the more you are thought to worship and know. A further implication is that people will cease to believe if they lose the knowledge of how to worship. Obviously, this suggestion is not easily dismissed. Some people do reject religion, at least on certain occasions, and people in rural areas often avoid relating to spiritually charged places and objects precisely because they do not know how to do it. Many people in Northern Mongolia, for instance, will Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  578 Lars Højer not worship Dayan Deerh, a famous cave cult site, because they are unacquainted with the correct procedures. Similarly, it is not uncommon for people to hand over religious items to a temple (cf. Bareja-Starzynska & Havnevik : ), simply because they do not know how to treat them properly and, as a result, consider them dangerous (which, it should be noted, gives a novel and thought-provoking perspective on the trade in religious objects in post-socialist Mongolia). Yet, it is immediately clear that it is not because they do not believe in spirit powers that people avoid such places and items, but rather they avoid them, and avoid relating to them, because they do believe.7 Hence, the issue is that avoidance arises from a belief in the erratic power of such agencies. They are erratic and uncontrolled because they have come to be imagined as lost objects and places – lost to present-day people, lost to the present (in that they belong to the past), and lost to the past (in that they survived into the present); yet very present as lost. They are what one might call a present absence, because they point to an absence by virtue of their presence8 (and, hence, it is also an absent presence in that it is a shadowy, enigmatic, and incomplete manifestation [cf. Delaplace & Empson : ; Holbraad & Willerslev : , ]). Socialism and the labour of the negative Before proceeding along these lines, it is useful first to explore the context of Mongolian socialism. During the socialist period (roughly -), Mongolian cities became associated with industrialization and development, and the sparsely populated countryside, home to villagers and nomadic pastoralists, came to be considered as a space to be civilized and urbanized (Humphrey & Sneath : ), at least in public ideology. The implication was that ‘city versus countryside’ was socialism versus a backwards, feudal, and traditional way of life and that history was to be written as ‘the progressive nature of the socialist regime’ (Kaplonski : ). In this way, the spatial distinction was simultaneously a temporal evolutionary distinction between pre-socialism and socialism. It was essential to the identity of socialism – and to the realization of this identity – to depart from the past and the periphery, and moving to a new historical condition was to revolt against the past (cf. Žižek ). Things had to be destroyed, and an often specified number of enemies had to be killed (Humphrey b: ; Shimamura : ; Žižek : ); the summoning of imagined enemies, it seems, was more important than the existence of true enemies. In the words of one Mongolian: ‘They showed my father a paper saying he was an enemy of the people, and was to be arrested ... It wasn’t necessary [to explain]. He was an enemy’ (Kaplonski : ). It was imagined and realized that the past – i.e. the enemy – was not the present and the present was not the past, and, hence, that the past was lacking in and lost to the present. The purges of the s were the most violent eruption of socialist modernity and its strategy of opposition, revolution, and historical rupture. At least , people died during the purges (Kaplonski : ), most of them Buddhist lamas; and  out of  Buddhist temples and monasteries were demolished (Baabar : ). As a strategy of radical critique, however, it had the unintended consequence of also bringing into being what it tried to destroy. ‘The labour of the negative’, a phrase coined by Hegel and recently appropriated by Taussig (; see also Højbjerg ), is a notion that captures the dynamics of these unintended consequences; it is also a notion which, eventually, will lead us back to the range of ethnographic phenomena mentioned above concerning lack of knowledge as knowledge of lack. The labour of the negative points Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 579 to the dialectic logic that socialism’s attack on its imagined enemy backfired, because not only did socialism’s imagination of superstition serve to eradicate such superstition, but it did – while eradicating it – bring it into existence as superstition: that is, as something which was important and powerful enough to necessitate destruction. The point is that the force of destruction and the sheer amount of energy expended on superstition mystified ‘mystification’. It fashioned an entity and gave it potential life.9 Much ‘superstition’ was lost, surely, but simultaneously an imagined space of absence was created. People were made to know that certain things existed of which they did not, and should not, know, and the negative came into being and gained power by virtue of being subject to destruction. We need to remind ourselves that many stories are circulating in Mongolia about how the socialist destruction of shamanic paraphernalia backfired and led to the violent and mystical death of the communists destroying it. An old woman in Tsagaan Nuur district in the northernmost part of Mongolia once told me of a man who drowned after burning a shamamic costume. Likewise, a revolutionary in a different part of Mongolia is said to have caused his own death and the death of all his descendants after burning a shamanic tree in  (Yanjmaa : ). While nobody dared to touch the sacred stone cairn at the Dayan Deerh cave cult site in Northern Mongolia, one man, I was told, tried to shoot down a piece of silver from the wooden pole of a destroyed Buddhist temple nearby, and he soon died and so did his wife and children. Nowadays, previously suppressed – and, hence, angry – spirits are (re-)emerging among the Buryat Mongols in Eastern Mongolia to take their revenge for being neglected in socialist times (Buyandelgeriyn ), and some of them – the so-called uheer – ‘were made into outcasts of the socialist state’ and their tormented souls are now returning to make trouble (Buyandelgeriyn : ; compare with the shurkul and barkan spirits of the Daur Mongols [Humphrey with Onon ]). The destruction, so to speak, recharged the battery of the shamanic paraphernalia, sacred trees, and ‘supernatural’ powers destroyed. Now that socialism has been abandoned, people are left with this awareness of absence. What was considered to be a feudal past full of superstition in socialist times is today a potent space for magical possibilities and magical agency. In a recent article, Buyandelgeriyn presents a substantial amount of ethnography which supports this claim, and she – like me – points to the productiveness of the lost Mongolian past when writing that ‘the more the Buryats [Mongols] believe in the loss of that tradition during socialism, the more shamanic rituals they generate’ (: ), but she stops short of attending to the constitutive indeterminate nature and generative capabilities of perceived absence within magical workings. Basing her understanding of emerging shamanic practices and spirit powers on the Buryat Mongols’ attempt to come to terms with historical loss, tragic past events, and existing post-socialist uncertainties, she relies mainly – like many others – on implicit ‘emotional-functional’ explanations such as the Buryat Mongols’ need to find meaning, construct identity and history,10 or alleviate suffering. Shamanic practices are means for dealing with ‘anxieties and uncertainties’ (: ), for making ‘misfortunes meaningful’ (: , , ), for offering explanations (: ), for making ‘sufferings bearable’ (: ), and for controlling ‘the flow of misfortunes’ (: ); moreover, ‘economic anxiety has pushed the Buryats to seek help from shamans’ (: , my emphasis). While Buyandelgeriyn’s analysis does not necessarily contradict the present argument, I am – for now – less interested in ‘actually existing’ historical loss (although this enters our equation) and the real traumas of past and present, or in what people psychologically, Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  580 Lars Højer emotionally, and economically need to do, so to speak, than in what certain magical and modernist technologies do to people by working with – and creating – conspicuous and productive gaps where meaning and knowledge are constantly deferred. While everyone acknowledges that things – and revolutions – certainly did happen in the past (and also brought about traumatic events), the historical gaps take on particularly strong significance when merging with magical domains constituted in – and working with – a strong presence of absence and a constant deferral of meaning; a red sect lama whom we will meet below, for example, is much more inclined than others to evoke and engage with historical gaps and past mysteries. Following from this, absence cannot simply be seen as a product of socialist modernity, but is intrinsic to the magical technologies that will be explored ethnographically in the following sections; like modernity, they can create imagined spaces of productive absence. The aim of the following ethnography will be to show how magical technologies join the modern (post-)socialist space by making absence present, and how they are fuelled by absence and cultivate it. Manufacturing absence in a written Buddhist charm The manufacture of absence is well illustrated by a particular protective written charm created by one of my key informants, a so-called ‘red’ Buddhist lama from Chandman’Öndör district in the northern Hövsgöl province. Although the term ‘red’ is unclear and diffuse when applied to particular Buddhist sects in Mongolia, it is often – and remarkably so in this case – used to refer to more ‘basic’, ancient, and shamanic Mongolian religious traditions concerned with the requirements of everyday life at the periphery of centralized powers, as opposed to more institutionalized and dogmatic ‘yellow’ Buddhist practices. The ‘red’ lama brings to mind Barth’s notion of the ‘conjurer’ (Barth ), secrecy being constitutive of the kind of knowledge he works with and of the way in which it is transmitted. Having been taught secretly in socialist times – or so he told me – by a high-ranking and powerful lama, he has a direct link to a potent past and, like many Mongolian shamans, he had a serious illness in his childhood. Doctors were unable to diagnose his illness or treat him, but one night a lama performed a gürem (ritual of exorcism to repel misfortune or sickness), and three days later he had dramatically improved. From then on he was interested in magic power (id shid) and eager to learn it. At first he was taught the Tibetan alphabet by a ‘yellow’ lama, and then a ‘red’ lama reluctantly accepted him as his only pupil (shav’) at a time when it was illegal to practise any kind of religion. The latter started by introducing him to five religious books and some incense offerings (san), and finally – after five or six years – he was taught a number of secret mantras (tarni), the most secret of which was only ‘orally transmitted’ (amyn jüd be). Along the way, he realized that he had been taught shamanic ‘things’ and his teacher told him that he was to follow the red direction. He used a similar form of secrecy – of revealing things as concealed – to refer to his own abilities and religious education (which took place secretly in a remote area) and when – at other times – he told me about hidden treasures and mysterious happenings. This secrecy is also manifest in the written charm to which we will now turn. This charm is written on a long piece of paper and can be divided into four sections. In the first section, the main text – a mantra – is written in black ink in Tibetan, a written language almost solely understood and read by lamas but nevertheless recognizable as Tibetan scripture by most people in Mongolia. In the second, the names to be protected by the charm form a section between the initial and the final part of Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 581 Tibetan text. They are written in blue ink in Cyrillic, the script most commonly used in Mongolia and therefore easily read and understood by almost any adult person in the country. In the third section, a whole range of destructive or enchaining tools and frightening weapons have been drawn below the main text on the front. The fourth section, on the reverse of the paper, contains a mantra written in Sanskrit, used for the consecration of the device. The written charm was designed to be placed over the entrance to a Mongolian home, making the distinction between host and visitor crucial. From the description above, we would expect both host and visitor to perceive the device as a display of weaponry, inducing a sense of fear, protection, and physical power, associated with an enumeration of familiar names related to the household and surrounded by unintelligible – but recognizable – religious text. The point to be made here is that it is the unintelligible text that empowers the ‘familiar’ device by providing a level beyond complete understanding, making it transcendent and powerfully ‘captivating’ (Gell : -).11 The incomprehensible – i.e. that which is beyond the comprehension of ordinary people – is evoked by the meta-sign of Tibetan scripture. In addition, the importance of the incomprehensible is emphasized by the red lama, who stresses the secrecy of the charm and warns the client against copying it. Also, he claims, the charm – and the mantra – will not be understood by Buddhist lamas from other sects. On the reverse of the paper, the Sanskrit mantra (tarni) for the consecration of the device is written in Tibetan letters. This mantra consists of incomprehensible syllables, which the lama knows by heart but which he cannot and certainly should not translate, because their power derives from their mystical form as words without lexical meaning. As such, they are powerful rather than meaningful – they gain power from being beyond ordinary meaning and communicative understanding. They do not communicate power; they are power. The charm’s materialization of indecipherable ‘magical’ knowledge, which is, however, decipherable as a paradoxical knowledge of the unknown,12 is associated with well-known names (such as members of the household) written in a conspicuous blue colour, and easily accessible symbols of aggression (axe, gun, etc.) and forced pacification (hobble and handcuffs). Its simultaneous embodiment of the known and the unknown makes the known and the unknown draw on the power of the other. The known world is mystified by embedding it in an agency without a known agent. Our imagination is provoked (what is behind the scripture?) and the form elicits a reaction, but not a specific action on our part. We have been acted upon, we have sensed agency, and agency has been elicited.13 In other words, the sign works by virtue of referring to something beyond itself, or rather by containing this beyond within itself. It hides the content while conjuring it up; it creates by mystifying, it reveals by concealing. The charm is designed to provoke absence as such: that is, to put absence to work and to cause reactions. One might say it is less a matter of the particularity of ‘thinking through things’ (Henare, Holbraad & Wastell ), than of something that simply makes us think (and do), much like the shamans’ mirrors described by Humphrey, whose ‘design encourages a progression of thought’ (: ). Indeed, present absence is mirroring the unknown. A sceptic might claim that the lama has simply used the written charm to mystify his client deliberately, but – apart from being unlikely – this claim is irrelevant for at least two reasons. First, the present argument concerns form in itself, and as such deliberately avoids political, economic, and functional issues of strategic manipulation by Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  582 Lars Højer individuals or collectives. The argument concerns what magical forms do to people (cf. Pedersen ), not the other way around. Secondly, it might be argued that the lama is just as mystified as the client (remember the Sanskrit mantra on the reverse of the paper), since the power and ‘meaning’ of the mantra can never be made transparent (even to the lama himself), and that it is, paraphrasing Lévi-Strauss ( []: ), a question of mystification thinking through us. If form, for example the ‘pure’ form of Tibetan or Sanskrit scripture in the charm, is opposed to content (or intention), yet only exists by virtue of content (as form has been created with a purpose by an agent and begs the question of what, or which intention, is behind it), then to create form is to create content or intention. Moreover, the more abstract the form becomes, as when form appears to be mystical, the more abstract, intangible, and pure the intentionality seems to become. It becomes, purely and simply, the elicitation of a reaction. In what appears to be in line with the present argument, Bloch claims in a seminal article that religion and ritual are not about communicating in an ordinary locutionary sense, where a large number of potential choices are available to the communicators, but rather about creating a formalized situation where actors are caught and unable to resist – through ordinary argument – the demands made of them (Bloch c: ), because religion and ritual have ‘no “truth” conditions’ (c: ). There simply is no argument to be made when language is formalized in ritual: It is because the formalization of language is a way whereby one speaker can coerce the response of another that it can be seen as a form of social control. It is really a type of communication where rebellion is impossible and only revolution could be feasible. It is a situation where power is all or nothing and of course in society total refusal is normally out of the question (c: ). Hence, ritual, religion, and formalization can be likened to a tunnel into which one plunges (c: ), extinguishing any possibility of choice or creativity. We can follow Bloch to the extent that an irresistible demand is caused by what is acknowledged as form in the charm, but otherwise his argument appears to be an illuminating inversion of the present line of reasoning. Bloch argues that formalization turns reality into ‘a timeless placeless zone in which everybody is in his right place’ and the elder ‘into an ancestor speaking eternal truth’ (c: ). The charm described above is certainly recognized as a religious form, but, apart from the fact that it is not placed in an expanded and fixed universe or cosmology with ancestors occupying the apex, the whole point about the charm is that it does not contain truth. The ‘minimal’ formalization of objects that are recognized as religious form, but not locked to tightly knit cosmologies, does not place one on a highway to eternal truth. Rather, such fragments of form direct one towards the indefinite, and as such make one open to any reaction. They are in a sense – and as opposed to Bloch – even more creative, as it were, than ordinary language, because they allow any possibility (cf. Lévi-Strauss  []: ). Bloch also concludes that religion is a special form of authority. If the present objects and powers, however, qualify as religion, they are – at least when it comes to the most ‘wild’ fragments – what amounts to the least authoritative phenomena possible, because they define actions minimally. They create a space where politics are minimized,14 so to speak. If power, according to Bloch (c: ), is all or nothing – either you plunge into the tunnel with no possibility of turning left or right, or you do not – then power in the present case is all and nothing. When you plunge into the tunnel, everything becomes possible, at least in principle. This should become clearer as we proceed. Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 583 Effects of absence: indeterminacy and avoidance Let us now turn away from religious objects and their inherent effects to a setting where people and their specific reactions to construed absences are more prominent. This is important because the absence designed in objects such as the written charm above does not have an effect in a cultural and historical vacuum. The above charm, for example, was made by a red sect lama living in the remote periphery of Mongolia and it was made for a young Mongolian man living in the absolute centre of Mongolia, the capital Ulaanbaatar. The latter needed the charm for the protection of family members. He would see the lama as embodying the periphery and a deep pre-socialist past in the form of a particular ‘red’ Buddhist tradition practised on the threshold of an ancient Mongolian shamanic tradition. The lama would ‘do something’ which the young man – in his own words – ‘did not know what it was’. This included a ritual for consecrating (aravnailah) or enlivening the device, and the young man then took it to his home in the capital city. The lama had created a rare charm, an almost one-off creation; the only other instance of it – as observed by me and the young man – was placed over the door of the lama’s own home. In the young man’s own words, he wanted the charm because ‘he hadn’t seen similar things in other places’ and because the lama ‘did it on his own, made it himself ’ (singularity being the ontological premise for shamanic activity [Humphrey with Onon ]). A further reason, the young man added, was that ‘it did not seem to be completely Buddhist’. The lama personified a continuity of the unknown, that is, a singularity; he embodied a lost past in the distant countryside and an esoteric semi-shamanic tradition, whose secrecy – one might argue – was even boosted during socialism, when it was forced to become clandestine and to avoid the use of tangible written material. The effects of the charm, then, are conditioned by such other powerful cultural forms as the socialist and post-socialist retrospective and prospective making of historical gaps and spatial distance. Such conditioning through cultural forms is related to people’s everyday concerns as well. In certain divination sessions described elsewhere (Højer ), an essential indeterminacy is brought about by the evocation of enigmatic and elusive religious objects and by powers that are brought to bear on clients’ present misfortunes. Common to such powers is often their unwelcome and unexpected intrusion into peoples’ lives. Through divination sessions, then, people are suddenly brought into relations with powerful and precarious agencies, such as a hardly known Buddhist lama relative from the old days, a forgotten religious object once owned by someone in the family line, a spirit power in the landscape, or a spell caused by a dispute. These objects and agencies are often not known to exist beforehand by the clients in question, and throughout the divination session they are construed as – and remain – partly unfamiliar, indefinite, and evasive; they are revealed-as-concealed. Now the point is that such unknown powers are a distinctive kind of agency effecting ‘undirected movement’. If present absence, as claimed with regard to the charm above, elicits reaction, then we should note that its specificity, rather paradoxically, is to elicit ‘any reaction’ by forcing people into an exchange relation with the unknown. Such reactive exchange scenarios might be of a very tense and erratic nature, which is not characteristic of more institutionalized religious practices, where agencies are related to in a more formalized and secure manner. When paying respect to Buddhist deities or pictures of deceased family members on a family altar, for example, Mongolians usually know what to do, and at ovoo (stone cairn) ceremonies, where spirit masters of the Mongolian landscape are celebrated, educated lamas officiate at the formalized and Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  584 Lars Højer highly predictable proceedings. While ‘mystification’ does take place on such occasions, this mystification is contained and does not have the potential of the singular, exceptional, and fragmented form of a rare charm. At ovoo ceremonies, reacting to the known (unknown) is ritualized and hence relatively calm, and such occasions, besides being official, formal, and ceremonial, take place in a safe atmosphere. Exchange relations with the (known) unknown, however, are either ‘wild’ and unpredictable or characterized by avoidance. This is conveyed in a story related by a young woman in Ulaanbaatar. The story begins with the tragic suicide of the woman’s younger brother when he was a young man in the mid-s during the closing stages of socialist rule in Mongolia. Staunch believers in the socialist system, her parents were neither ‘traditional people’ nor practising Buddhists. They did, nevertheless, have an image of a Buddhist deity, inherited from her grandparents, at home, but it remained locked in their chest with other belongings and was never displayed publicly on an altar. The death of her younger brother was, of course, a great shock to the parents, and they decided, despite their atheist convictions, to visit a Buddhist lama. The lama told them that the deity in the chest had been lying upside down and next to a pair of shoes. This, they were told, had offended the deity, who had become angry and taken their son. The parents were now afraid of what the deity would do next, and they were unsure about ‘the wants of the deity’ and how to treat it properly. They decided to hand it over to the main Gandan temple in Ulaanbaatar, where Buddhist lamas, they assumed, would know what to do and how to treat it. In turn they received a new deity from the same place. After the death of her son, the mother suffered mental problems and was hospitalized for a while. She was depressed for a long time and her hair turned grey almost overnight. One day, a few months later, the mother was cleaning the ground of the family’s compound (hashaa) on the outskirts of Ulaanbaatar. She was sweeping up rubbish and weeds to burn as waste. Suddenly she became furious and exclaimed that the gods had never helped her or her family. They had taken her son, she continued, and the gods would not help them in the future. She then took the new deity image from the altar and threw it on the fire. For many years afterwards, the family had no deities in their home. A few years ago they acquired a new one, but now the parents have grown old and, as the woman relating the story explained, it is their last chance to ‘do some good’ – i.e. gain religious merit – before they die. The recently acquired deity does not relate to the first part of the story, and – as I have observed on many occasions and been told by numerous Mongolian informants – it is common in Mongolia for people to develop a renewed interest in institutionalized and ‘calm’ religious practices concerned with reincarnation and gaining merit (buyan) as they grow old. The main point of interest in the story is the first part, from which a number of points can be drawn. First, we have a clear case of an unknown power literally emerging after having been locked up for many years. The deity is inherited from a deep Buddhist past, which is radically separated from the contemporary atheist life of the family.15 In the name of socialism the family has, so to speak, eclipsed the past; they have not destroyed the deity but have hidden it, and although the power of past religion has been dismissed, it has also been kept. What has happened, however, is that in creating a distance – in terms of both time and knowledge – to the power in question, it has changed from being a conventional actuality (a deity on the grandparent’s altar) to becoming an erratic possibility (a locked-up power).16 The parents are insecure. What does the deity want? What will it do next? The only way out is to give it away: that is, to sever their relation Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 585 with it. The new deity, acquired from the Gandan monastery and thus emerging from present times, is initially thought to be more manageable, and it does not act in unpredictable ways. Secondly, the story conjures up a fierce ambiguity and change of mind with regard to semi-alien powers and shows that relations with the unknown are also unknown relations: that is, relations that are uncertain, unpredictable, and capricious and can only be settled once and for all through avoidance. The parents are atheists, but Buddhism has remained a possibility; they have kept the deity and decide to visit a lama when their son dies. They give away the god, not because they do not believe in it, but because it is not for them (and their insufficient knowledge) to handle it. They fear it and believe in it because they do not know it. Yet, their relation to religion and religious powers is still unsettled, and when they receive a new deity, the mother decides to get rid of it, but in doing so she still proclaims a kind of belief in impulsively and furiously burning it, rather than just ‘clinically’ throwing it out, and by claiming that the gods have never done anything good for them. Rather than declaring disbelief, she is – it seems – declaring what amounts to war. A constant change of mind, and a radical questioning of one’s own relation to these powers, is intrinsic to such anxious relations with the unknown. But as long as one engages with the powers, they cannot be ignored. On the one hand, they are construed as revealed-as-concealed powers by lamas and diviners. On the other hand, people attend to them through a lens of self-declared ignorance (we have lost the knowledge) leading to both a lack of belief (how can you really believe in what is so unknown?) and an abundance of belief (the excessive fear of the radically unknown). These aspects converge to create a domain where much can be at stake but little is certain. The tragic death of this woman’s brother, and the mother’s reaction to it, speaks of a universe where the radically unknown is made to be imagined, because nothing known, familiar, and close to the family would cause such a thing to happen.17 The unknown has, so to speak, a minimal ability to animate maximally, because we do not know, quite obviously, what the unknown is. Hence, it animates maximally by releasing undirected action; reactions are unconstrained by what they react to, because the ‘exchange partner’ is unknown (even rejection of the agency as agency, i.e. deciding that it is ‘fake’, is a possibility). The unknown’s unpredictable (elicitation of) hyper-activity is also suggested by Shimamura when he writes that among the Aga-Buryat Mongols, ‘ancestral spirits who were unknown to them were more actively seeking (demanding – ug nekhekh) their descendants’ (: ). The fact that they are unknown seems to make them more active, and often also more erratic and dangerous: ‘Shamans make a particular strong point by arguing that the erasure of an origin spirit’s name from the genealogical record is one of the worst causes of misfortune, far worse than forgetting the spirits whose names do exist in the records’ (Buyandelgeriyn : ). If the power of gods and spirits, then, is that they always have a residue of meaning that we seek but can never completely uncover (Swancutt : ), that is, a compelling unknown (cf. Pedersen’s [] description of the ‘pulling’ of the shamanic spirit vessels), then one might also argue – in line with the present argument – that the more unknown and residual the spirit powers, the stronger, the more compelling, and the more unpredictable they become. All such powers – of course – are concealed, but the more manifest the concealment, the purer the power becomes in eliciting uncontrolled reactions. In the written charm, the unintelligible form called forth a mystical content, accentuated by the lama, who emphasized the secrecy of the charm, and by the client, who was sure Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  586 Lars Højer that he had now acquired a particularly powerful object. In a similar manner, the deities from the past conjure up layer upon layer of concealment: they are powers from another world, they are from a deep and unfamiliar past, and during socialism they were clandestine objects not for public view. The charm and the deity, as well as other past and unknown powers that are evoked, are vivid images animating people. At the heart of these agencies is disguise and absence. Obviously, this is not entirely a consequence of socialism, but socialism has created a topography of fragmented religious – and partly empty – forms, such as religious objects whose significance is unknown owing to the (believed) loss of knowledgeable people, places where monasteries and spirit powers were once located, and lost or stored possessions of deceased religious people: that is, places or ‘points’ loaded with the energy of destruction and the vitality of a lost past. Conclusion The anticipation of the unknown is present in the effects of socialism and in magical modalities, and it is gaining renewed momentum through the fusion of magical absence and the radically ‘negative’ effects of socialism. This mode of anticipation is already in itself an effect – not in the sense of creating an explanation or destroying another one as in fights over different cultural ideologies such as Buddhism or shamanism, but rather by virtue of giving itself away to other effects. People are drawn into relationships and worlds, not just in the manner of painting, clarifying and synthesizing new cultural landscapes of knowledge – although this quite obviously also happens – but in the sense of being drawn into landscapes constituted in, and gaining effect from, suspense or in the fact that they are not fully known. In concluding this argument, I was made aware of a comparable argument made by Strathern (). She writes about the ‘procreativity of absence’ in relation to Melanesian ethnography and suggests that the knowledge that they [vehicles for communication] are lost is not, so to speak, lost knowledge, it is knowledge about absence, about forgetting and about an unrecoverable background. That sense of loss stimulates the Baktaman initiators, it would seem, to making present images work ... The important thing is that the gaps are preserved ... It is as if they knew that by insisting on that absence they create their own creativity (: -, emphasis omitted). It is in this sense that I understand the magical workings of certain Mongolian powers as – in Strathern’s words – ‘the “one” form’ doing ‘multiple work’ (: ). Yet, one is then faced with the problem of whether such absences simply concern a condition of being human, be this in Melanesia or Mongolia, or whether they can be actively produced or accentuated through the unintended effects of socialism, by means of religious artefacts, or through divination sessions. I would argue that both apply, because this modality refers equally to the production of images and practices that enhances the sense of absence through their radical dialectical work, and to a way of attending to the world, where this absence is potentially perceived in all images. This enhancement can, as it has been argued, take place in more coherent, institutionalized, and formalized systems of religiosity like those analysed by Bloch and mentioned briefly above, or it can take place in fragments of minimal magico-religious form, as they have been dealt with here. In the former case, enhancing absence is, in effect, turning it to work for the present world as a kind of ideology (cf. Bloch c). In the Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 587 latter case, absence works on its own terms, so to speak, and produces multiplicity. This is not an absolute difference, though, as absence can only be evoked through a known form (Tibetan scripture, Buddhist deities) – often even related to a known presence (the personal names on the charm and its location in specific homes) – and because both contain the potential of the other. Multiplicity can be the starting-point of new ideologies, and existing ideologies can implode and be subverted or torn apart by the potential possibilities of absence as it exists in all magico-religious practices. The difference is well illustrated by two approaches to the avoidance of spirit powers in the Mongolian context. Hürelbaatar (: ) has observed that people in Inner Mongolia sometimes avoid making offerings to ghosts and new sites because, once initiated, these offerings become obligatory and cannot be broken. Such obligatory relations are tiresome, people say, and they often prefer not to establish them at all. In a similar vein, Buyandelgeriyn (: ) has observed that once a spirit is accepted into a Mongolian family’s pantheon, the relation cannot be broken and might turn into a psychological and economic burden. In these cases we are dealing not with avoidance of initially dangerous powers, but with known powers that turn dangerous if already established formalized exchanges are abolished. In this case, you avoid the reciprocal obligation of formalized exchange, whereas in the cases mainly dealt with in this article, you avoid the uncertain relations implicated in exchanges with the unknown. In formalized relations with relatively known powers, the danger lies in breaking the relationship, but in the cases we are dealing with here, the danger lies in establishing it. A transformation from one to the other – from danger to obligation – is conveyed by Humphrey when she writes about the Buryat Mongols that: ‘When the offended spirit has been revealed, the shaman commonly orders the client to go out to the mountain (tree, etc.) residence of the spirit and perform the ritual called alban, which means duty, service or tribute’ (a: ). What I have tried to reveal is a space that points not to the presence of institutionalized cultural knowledge – or not-entirely-present cultural imaginations or anticipations extending such knowledge to new domains (cf. Barth : ) – but to the presence of absence as such: that is, to what is important by virtue of not being there, by virtue of being a powerful centre of gravity where only nothing – or rather no thing – is yet. But the fact that only nothing is in this ‘gap’ does not imply that this absence cannot be produced, or that it cannot be effective. Rather this space might be depicted as the non-cause for effects, or as an effect which is indefinite. The absence contains potential whose effects cannot be contained; it just elicits (re)action.18 And its unattainability – or its being relatively external to relations – is what gives it power. Thus, if cultural knowledge, so to speak, is elaborated inside our (cultural) minds, the presence of absence points to the power of the outside. NOTES I am grateful to Caroline Humphrey, Morten A. Pedersen, Michael Mahrt, and Hildegaard Diemberger for comments and important insights that have helped me to develop this argument. Also, I thank the Editor and anonymous reviewers at the JRAI for valuable suggestions. The research on which this article is based was generously funded by the Danish Research Council for the Humanities and the Danish Research Academy. 1 Crick’s review of the anthropology of knowledge is, by his own admission, indistinguishable from a review of anthropology in general (: ). In the same vein, Barth defines knowledge in very general terms as ‘what a person employs to interpret and act on the world’ (: ), although he claims that knowledge, as opposed to culture, is less ‘embracing’ and not just diffusely shared (: ). Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  588 Lars Højer 2 The debate on the transmission of culture (see, e.g., Ingold ; Sperber ) also seems to be concerned with presence: that is, the presence of psychological dispositions or the presence of a non-detached practical engagement with the world. In contrast, my argument concerns the creativity of engagement with absence or detachment. 3 Present-day Mongolia refers to / and , when the material presented here was collected. 4 Whereas a lama in the Tibetan context is a religious teacher, the term is used here for the Buddhist clergy in Mongolia, although many present-day lamas are not technically proper monks in the sense of having been ordained according to the canonical Buddhist monastic rules (Bareja-Starzynska & Havnevik ). Accordingly, the term is often used in a vague sense. 5 This picture is now changing in certain areas with the education of a new generation of young Buddhist lamas in Ulaanbaatar and abroad. 6 It should be stressed that most of the ethnography presented here is based on my field experience from Chandman’-Öndör district, Hövsgöl province, in northern Mongolia, where religious practitioners are still few and where elaborate religious knowledge of cosmological ideas is rare and has a strikingly fragmented form. Anthropologists like Shimamura, Buyandelgeriyn, and Swancutt have worked among Buryat Mongols in the easternmost regions of Mongolia, where a particularly strong revival of elaborate shamanic practices can be observed. A similar revival of shamanic practices has not taken place in other parts of Mongolia (see also Shimamura : -). While this might account for my stress on ‘present absence’ and the lack of references to coherent and elaborate cosmologies, I still believe that most points made in this article, even if slightly more pronounced in this region, are still highly relevant for studies in other parts of Mongolia. 7 In this sense loss is not opposed to belief but is rather its precondition. It is not implied, however, that people keeping religious objects are non-believers; rather, in this case the loss or absence has taken enough shape for it to be dealt with safely. 8 One reviewer suggested that I use Derrida’s notion of ‘spectre’ () rather than the dialectical expression ‘present absence’, which points to the iconoclastic power of what ‘always remains’ or the failure of iconoclasm ‘if any fragment escapes destruction’ (Pietz : ; I thank Victor Buchli for this reference). I have, however, chosen to retain the latter expression, since a dialectical expression, I believe, much better conveys the importance of difference or split as potentiality – whether in (post-)socialist history, (post-) socialist historical narratives, magico-religious summoning of concealed powers (present and absent at the same time), or magico-religious images (where meaning is deferred). While the notion of spectre – as something which simultaneously is present and not present, inaugurates and reveals, and repeats and initiates (Derrida : , ) – might capture all that, it runs the danger, like so many other single hybrid terms, of moving attention away from their split and paradoxical nature. Even spectrality, it seems, can only be described through the use of oppositions (present/not present, inaugurate/reveal, repeat/initiate). 9 Interestingly, the labour of the negative is strongly reminiscent of the Mongolian notion of hel am, which concerns, among other things, the danger of ‘a discourse of binary knowledge, where the effects can just as well be the opposite ... because opposites are integral to each other’ (Højer : []). Hel am concerns the fact that extreme and definite statements (‘she is good/bad/beautiful’, etc.) are dangerous because they act on people and are emotionally evocative. Stating things too clearly – as socialism did – evokes opposites/enemies and emotionality. 10 While also stressing the importance of historical rupture, the socio-psychological aspect is even more pronounced in an article by Shimamura () whose focus is almost solely on the role played by shamanism in reconstructing the ethnic identity of the Aga-Buryat Mongols. 11 The principle is comparable to Gell’s notion of ‘captivation’ (Gell : -), although Gell introduces this term with regard to the effects of artistic virtuosity: Artistic agency ... is socially efficacious because it establishes an inequality between the agency responsible for the production of the work of art, and the spectators ... Captivation or fascination – the demoralization produced by the spectacle of unimaginable virtuosity – ensues from the spectator becoming trapped within the index because the index embodies agency which is essentially indecipherable (: ). 12 This can be compared to Houseman’s notion of ‘avowed secrecy’ (: ), where the excluded party is aware of the existence of a secret and of his/her exclusion from it. 13 Cf. Strathern’s notion of ‘elicitation’ (: ) and Gell’s notion of ‘abduction’ (). 14 This is not meant to suggest that the charm – using symbols of aggression and being placed to protect a household – is not political or used politically, but that its way of working with layers – and, hence, being a form that simply elicits the anticipation of an intention – is minimally political. Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 589 15 Inside the Mongolian household chest are things that have been detached from people at moments of separation and transformation (Empson : ). Through such things you keep only ‘partial connections’ (Empson : -) and ‘difficult relations are maintained as open possibilities’ (Empson : ). We might then – and in line with the notion ‘present absence’ – characterize the emerging religious powers as ‘partial connections’ which allow people ‘to cross boundaries’ (Empson : ). And if the partial relations inside the household chest are concealed by – but necessary to – the visible agnatic relations displayed on the chest’s surface, as argued by Empson, then we might also say that the hidden deity (i.e. supernature) is the concealed – but necessary – inside of socialism. We might even further speculate that socialism has served to ‘shamanize’ magico-religious practices, since it seems to be in the nature of the sacred artefacts/spirit powers of shamans, such as the ongon (spirit vessel), that they should be hidden from public view (see also Pedersen : -, , ). 16 In line with this, Empson writes that it is as if things kept at the bottom of the Mongolian household chest have a ‘potential to move but must be contained’ (: ). 17 It is also possible to interpret this story in psychological terms as a family trying to come to terms with a tragic suicide. While this is a possible explanation, it does not necessarily contradict the present one. The explanatory coming-to-terms (mistreatment of a deity), as well as the grief and anger, could quite obviously not be projected onto anything, but had to be perceived in a form that contained the possibility of agency. 18 In continuation of Gell (), we might define this (spirit) potential as the abduction of abduction. REFERENCES B (B-E B) . Twentieth century Mongolia (ed. C. Kaplonski; trans. D. Sühjargalmaa, S. Burenbayar, H. Hulan & N. Tuya). Cambridge: White Horse Press. B-S, A. & H. H . A preliminary study of Buddhism in present-day Mongolia. In Mongols: from country to city (eds) O. Bruun & L. Narangoa, -. Copenhagen: NIAS Press. B, F. . Cosmologies in the making: a generative approach to cultural variation in inner New Guinea. Cambridge: University Press. ——— . The guru and the conjurer: transactions in knowledge and the shaping of culture in Southeast Asia and Melanesia. Man (N.S.) , -. ——— . An anthropology of knowledge. Current Anthropology , -. B, M. a. The past and the present in the present. In Ritual, history and power: selected papers in anthropology (ed.) M. Bloch, -. London: Athlone Press. ——— b. The disconnection between power and rank as a process: an outline of the development of kingdoms in Central Madagascar. In Ritual, history and power: selected papers in anthropology (ed.) M. Bloch, -. London: Athlone Press. ——— c. Symbols, song, dance and features of articulation: is religion an extreme form of traditional authority? In Ritual, history and power: selected papers in anthropology (ed.) M. Bloch, -. London: Athlone Press. B, D. . Visiting knowledge in anthropology: an introduction. Ethnos , -. B, M. . Dealing with uncertainty: shamans, marginal capitalism, and the remaking of history in postsocialist Mongolia. American Ethnologist , -. C, M.R. . Anthropology of knowledge. Annual Review of Anthropology , -. D, G. & R. E . The little human and the daughter-in-law: invisibles as seen through the eyes of different kinds of people. Inner Asia , -. D, J. . Spectres of Marx: the state of the debt, the work of mourning and the new International (trans. P. Kamuf). New York: Routledge. E, R. . Separating and containing people and things in Mongolia. In Thinking through things: theorizing aretfacts theoretically (eds) A. Henare, M. Holbraad & S. Wastell, -. London: Routledge. G, A. . Art and agency: an anthropological theory. Oxford: Clarendon Press. H, A., M. H & S. W (eds) . Thinking through things: theorizing artefacts ethnographically. London: Routledge. H, E. & T. R (eds) . The invention of tradition. Cambridge: University Press. Hø, C.K. . Inner iconoclasm: forms of reflexivity in Loma rituals of sacrifice. Social Anthropology , -. Hø, L. . The anti-social contract: enmity and suspicion in Northern Mongolia. Cambridge Anthropology , -. ——— . Entertained by the unknown: divination in postsocialist rural Mongolia. In Unveiling the hidden (eds) A. Lisdorf & K. Munk. Submitted for publication. Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  590 Lars Højer H, M. & R. W . Transcendental perspectivism: anonymous viewpoints from Inner Asia. Inner Asia , -. H, M. . The interactive basis of ritual effectiveness in a male initiation rite. In Cognitive aspects of religious symbolism (ed.) P. Boyer, -. Cambridge: University Press. H, C. . The moral authority of the past in post-socialist Mongolia. Religion, State and Society , -. ——— . Remembering an ‘enemy’ – the Bogd Khaan in twentieth-century Mongolia. In Memory, history, and opposition under state socialism (ed.) R.S. Watson, -. Santa Fe, N.M.: School of American Research Press. ——— a. Shamans in the city. In The unmaking of Soviet life (ed.) C. Humphrey, -. Ithaca, N.Y.: Cornell University Press. ——— b. Stalin and the Blue Elephant: paranoia and complicity in postcommunist metahistories. Diogenes , -. ——— . Inside and outside the mirror: Mongolian shamans’ mirrors as instruments of perspectivism. Inner Asia , -. ——— with U. O . Shamans and elders: experience, knowledge, and power among the Daur Mongols. Oxford: University Press. ——— & D. S . The end of nomadism? Society, state and the environment in Inner Asia. Cambridge: White Horse Press. H̈, A. . ‘Tradition’ as differently practised by the Buryat-Mongols of Russia and China. Unpublished Ph.D. thesis, University of Cambridge. I, T. . From the transmission of representations to the education of attention. In The debated mind: evolutionary psychology versus ethnography (ed.) H. Whitehouse, -. Oxford: Berg. J, J. . The temple of memories: history, power, and morality in a Chinese village. Stanford: University Press. K, C. . One hundred years of history: changing paradigms in Mongolian historiography. Inner Asia. Occasional papers of the Mongolia and Inner Asia Studies Unit , -. ——— . Creating national identity in socialist Mongolia. Central Asian Survey , -. ——— . Blame, guilt and avoidance: the struggle to control the past in post-socialist Mongolia. History and Memory , -. K, R.M. . Anthropology as interpretive quest. Current Anthropology , -. L́-S, C.  []. The raw and the cooked (trans. J. Weightman & D. Weightman). New York: Harper & Row. ———  []. Introduction to the work of Marcel Mauss (trans. F. Baker). London: Routledge & Kegan Paul. P, M.A. . Totemism, animism and North Asian indigenous ontologies. Journal of the Royal Anthropological Institute (N.S.) , -. ——— . Talismans of thought: shamanist ontologies and extended cognition in Northern Mongolia. In Thinking through things: theorizing artefacts ethnographically (eds) A. Henare, M. Holbraad & S. Wastell, -. London: Routledge. P, W. . The sin of Saul. In Iconoclash: beyond the image war in science, religion, and art (eds) B. Latour & P. Weibel, -. Cambridge, Mass.: MIT Press. S, M. . Two or three things that I know about culture. Journal of the Royal Anthropological Institute (N.S.) , -. S, I. . The movement for reconstructing identity through shamanism: a case study of the Aga-Buryats in postsocialist Mongolia. Inner Asia , -. S, D. . Anthropology and psychology: towards an epidemiology of representation. Man (N.S.) , -. S, M. . The gender of the gift. Berkeley: University of California Press. ——— . Partial connections. Walnut Creek, Calif.: AltaMira Press. S, K. . Representational vs conjectural divination: innovating out of nothing in Mongolia. Journal of the Royal Anthropological Institute (N.S.) , -. T, M. . Defacement: public secrecy and the labour of the negative. Stanford: University Press. T-R, H. . The invention of tradition: the highland traditions of Scotland. In The invention of tradition (eds) E. Hobsbawm & T. Ranger, -. Cambridge: University Press. W, R. (ed.) . Memory, history, and opposition under state socialism. Santa Fe, N.M.: School of American Research. Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute  Absent powers 591 Y, B. . Hövsgöl aimgiin Chandman’-Öndör sumyn tüühiin zarim asuudal (Some issues of the history of the Chandman’-Öndör district of the Hövsgöl province). Unpublished master’s thesis (diplomyn ajil), National University of Mongolia, Ulaanbaatar. Ž̌, S. . When the Party commits suicide. New Left Review , -. Pouvoirs absents : magie et perte dans la Mongolie post-socialiste Résumé Les habitants de la Mongolie contemporaine ont le sentiment que des savoirs religieux précieux se sont perdus pendant l’ère socialiste et qu’un fossé infranchissable s’est creusé entre les époques pré- et post-socialiste. Au lieu d’y voir un manque de connaissances passif, l’auteur explore ici les aspects procréatifs de l’absence de connaissances. Il avance que cette absence est une condition nécessaire à l’efficacité de certaines images et pratiques religieuses, parce qu’elle attire l’attention des gens vers un inconnu irrésistible. Les connaissances religieuses ne sont efficaces ici que pour autant qu’elles soient révélées comme une présence cachée mais puissante, grâce à des charmes magico-religieux et à l’imaginaire (post-)socialiste. L’auteur affirme que la présence de connaissances absentes est une cause qui ne contient pas ses effets et que, de ce fait, les relations à l’inconnu sont aussi des relations inconnues. Lars Højer is associate professor at the Department of Cross-Cultural and Regional Studies, University of Copenhagen. He has worked on social, economic, and religious transition processes in Mongolia, and is currently doing research on minority issues related to Western China. Department of Cross-Cultural and Regional Studies, Faculty of Humanities, University of Copenhagen, Snorresgade -,  København S, Denmark. [email protected] Journal of the Royal Anthropological Institute (N.S.) , - © Royal Anthropological Institute 