3
THE TERM ‘SHAMANISM’ is one entirely created by Western scholarship, and dependent in all its current public usages on the definitions that this scholarship has made of it. Its utility in everyday language, academic and popular, has largely been due to the fact that those definitions are so diverse that they no longer represent a classification: in the words of one expert, Graham Harvey, they are more of a ‘semantic field’.1 The term itself was coined by German authors in the eighteenth century, and has continued to develop, and expand its meanings, ever since. Although anthropologists have supplied much of the material for the study of it, they have often been wary of it since the mid twentieth century, because of its imprecision, and it has been applied more freely in the disciplines of comparative religion, the history of religion, and religious studies, and by some historians, archaeologists, experts in literature, and psychologists, as well as having a large non-academic currency.2 At its widest application, it is used to describe the practice of anybody who is believed, or claims, to communicate regularly with spirits, as defined at the opening of this book. More often the term is applied to the techniques of a person who regularly communicates with spirits in a traditional non-Western society, and does so for the benefit of other members of it. It is accorded even more frequently to such a person who apparently makes the communication with spirits in an altered state of consciousness, most commonly described as a trance. Quite often, some further refinement is required to meet the definition of shamanism, such as the ability always to control spirits, rather than being controlled by them; or to send out the person’s own spirit into other worlds, leaving the body temporarily; or the use of a dramatic performance to make the necessary contact with spirits. At the most restricted extreme of the semantic field are those who would confine the term to the specific techniques of such figures in native Siberia and neighbouring parts of Eurasia, as it is derived from the word ‘shaman’, used of these people in one of the linguistic groups of Siberia, the Tungus. Some authors indeed employ it to mean the whole religious system of the inhabitants of this region.3 There is no means by which any of these usages of the expression can be objectively judged to be more legitimate than the others: in practice, academic and non-academic authors choose one or another according to the convenience of it for their particular argument. In this respect the ivory tower of professional scholarship has become a Tower of Babel.
Shamanism, and the problem of its definitions, became an issue in the study of early modern witchcraft because of the work of one of Italy’s most celebrated historians, Carlo Ginzburg. Between the 1960s and 1980s, he developed an approach to it based originally on his discovery of an early modern tradition in the Friuli district of the north-east of his country, concerning people called the benandanti, ‘those who go well’. These claimed that at night, when their bodies lay in sleep or some kind of trance, their spirits went forth to do battle with those of witches for the welfare of the community. He saw immediately that this idea corresponded to some of the activities associated with shamans, and detected other traditions of similar figures at places spanning the whole expanse of Eastern Europe. He suggested both that they derived from a common body of ancient ideas, which once covered Eurasia, and that memories of these ideas had helped to create the early modern stereotype of the witches’ sabbath. He was aware of the difficulty of characterizing shamanism, and dealt with it by speaking of ‘elements of shamanistic origin which were by now rooted in folk culture, such as the magic flight and animal metamorphosis’: in other words, people like the benandanti did not necessarily practise shamanism themselves, but drew on old practices which had derived from it or resembled it.4 Carlo Ginzburg’s sense of what shamanism should be bore a striking resemblance to the dominant one during the period in which his ideas were developing. This was a very particular one, articulated by Mircea Eliade, a Romanian refugee who settled in America and became its leading authority on the history of religion: in the 1970s half of all American full professors in that discipline had been his pupils.5 He was also the most influential Western scholar of shamanism during the mid-twentieth century, defining it as a formerly universal and very ancient tradition whereby an elite of warrior magicians sent out their own souls and the platoon of spirits they controlled to do battle with evil forces for the good of their communities.6 The apparent relevance of this model to Ginzburg’s benandanti should be obvious, and Eliade himself perceived and commented on it, proposing that it be extended to other figures found in the folk cultures of south-eastern Europe.7
Support was forthcoming from Hungarian scholars for an association between particular kinds of European folk magician and shamans. The Hungarian people, the Magyars, had migrated westward into their present homeland from the Eurasian steppes in the early Middle Ages, and spoke a language from the same (Uralic) family as some used by western Siberian peoples who practised classical shamanism. Some Hungarians had by the mid-twentieth century perceived a similarity between shamans and a figure from their own society, the táltos, who was believed to have magical powers which were worked to help others, sometimes in trance or dream and by engaging in spiritual battles with evil powers.8 In the 1980s and 1990s two such scholars became especially active and influential in pursuing such parallels and following up Ginzburg’s suggestions, one a historian, Gábor Klaniczay, and the other a folklorist, Éva Pócs. Together they enlarged upon the fact that the different peoples of south-eastern Europe had obtained a range of purveyors of good magic, under different names, who had allegedly possessed the gift of sending out their own spirits at night to work for the benefit of others, often by fighting destructive forces.9 They were as cautious in the equation of these with shamans as Ginzburg had been. For Klaniczay, the similarities consisted only of ‘shamanistic elements’, while Pócs stated roundly that the European magicians and seers in question ‘cannot be seen as shamans in the strictest sense of the word’. They could indeed only be described as ‘shamanistic’, although she still believed that they were ‘vestiges of a European agrarian shamanism’ which had been practised in a prehistoric past, as Ginzburg had proposed.10 By the 2000s, she also acknowledged that Carlo Ginzburg’s own books made ‘great spatial and temporal leaps without sufficient evidence’, while they remained an important influence upon her. She had come to believe by then both that the definition of shamanism was overgeneralized and that figures such as the benandanti should be better related to European and Middle Eastern cults than Siberian shamans. None the less, she persisted in accepting the idea of a ‘European shamanistic substratum’, though she recognized that it had become controversial. Klaniczay had become still more careful, warning against bringing together remote and perhaps incompatible motifs to construct such a hypothetical ‘substratum’, and emphasizing that the imagery of the witches’ sabbath kept on being reinvented as the early modern period progressed. Ginzburg himself acknowledged by the 2000s that he had possibly underestimated the intricacies of the relationship between the different components of his ‘shamanistic’ complex, and that shamanism might be no more than an analogue to the magical traditions of Europe.11
By then, however, the association between European magic and shamanism had been enthusiastically adopted by one of the leading historians of the German witch trials, Wolfgang Behringer, in his valuable case study of the prosecution in 1586 of a folk magician from the Bavarian Alps. This man had claimed to have gained his powers as a healer and witch-finder on long journeys at night, in which his soul left his body to accompany either a train of nocturnal spirits, or else an angel. Behringer had no apparent hesitation in dubbing him the ‘Shaman of Oberstdorf’.12 At the same time the association was subjected to consistent criticism, and rejection, by an equally distinguished scholar of the witch trials, the Dane Gustav Henningsen, who was expert in both Scandinavian and Mediterranean Europe. He proposed that the benandanti and similar figures in south-eastern Europe differed from the ‘classical’, Siberian, shamans in four key respects: they were not in control of their trances; they were usually alone when they entered them and only encountered other humans in the course of their soul journeys; they held no public position; and they normally in fact did not enter trance at all, instead dreaming of their journeys while sleeping. He proposed that they be placed in a separate category from shamanism, or indeed from any kind of activity, under any name, which depended on apparently sending out one’s soul or spirit from the body in a public performance. He held that this category consisted of private and passive experiences of seeming soul-journeying, sustained mostly when asleep and dreaming.13 Meanwhile experts in the witch trials of Western Europe, especially those from the English-speaking world, have tended until recently to regard the whole debate as irrelevant to their concerns, largely because of the apparent lack of equivalent figures in their nations to the benandanti and their Hungarian and Balkan parallels.14
Fairly clearly, the total lack of an agreed definition of shamanism makes it impossible to decide objectively how far it is applicable to European magicians who were believed to send out their souls during trance or dream: some usages of the term certainly comprehend them and some as patently do not. It is equally obvious that there are some similarities between them and the ‘classic’ shamans of Siberia, and some differences, and how far those determine whether the former is analogous to the latter must likewise be a matter for subjective opinion. None the less, a study such as the present one cannot shirk the issue. If its preoccupation is with regional variations in beliefs concerning the witch figure, and their relevance to the early modern trials, then some attempt must be made as part of it to decide whether or not there was a generalized ancient shamanism in Europe, or a shamanic province in it during historic times. To make that possible, the question of definition must now be faced head-on.15 In doing so, the work of providing a global contextualization of European witchcraft beliefs, largely undertaken in the first chapter of the present book, will be completed.
Setting the Terms
Éva Pócs, when puzzling over the problem of terminology herself, perceptively noted that ecstatic visionary experience is ‘widespread, commonplace and non-culture-specific’, and as such was found throughout medieval and early modern Europe, among the elite and the populace and in religious and lay contexts.16 A few examples from very different cultures in the European and Mediterranean world may serve to bring home the point. At the end of the eighth century a historian of the Lombard kingdom, recalling a monarch who had reigned 200 years before, said that he was believed to have the power to send out his spirit through his mouth while asleep, in the form of a tiny snake. In this shape, it had the power to achieve feats such as detecting buried treasure.17 In the 1180s the churchman Gerald of Wales noted the existence in his native country of individuals known collectively as awenyddion or inspired persons. When consulted by clients who wanted to know whether or not to undertake a venture they apparently became possessed by spirits, roaring violently and babbling what seemed nonsense but from which an answer to the question could generally be pieced together. At the end they had to be shaken hard to break their trance and preserved no memory of what they had said. They claimed to acquire their ability in dreams and as a result of profound Christian piety, as they invoked the Holy Trinity and saints before making their prophecies. Gerald found analogies for them in the pagan oracles of the ancient world and in the biblical Hebrew prophets.18 Just over four centuries later, in 1591, a Scot called John Fian claimed that he would ‘lie by the space of two or three hours dead, his spirit taken, and suffered himself to be carried and transported . . . through all the world’. He was confessing (under torture) to dealings with the Devil, and may have been recalling the temptation of Christ, as recorded in the Bible, but could also have been describing an ecstatic experience.19 In 1665, the Jewish sage Nathan of Gaza entered an altered state induced by the singing of hymns by his pupils, in which he danced about before suddenly falling and lying as if dead, with little trace of breathing. In this state, he spoke, in a voice other than his own, the words of a divine entity. He had learned this technique from Hebrew instruction manuals, composed in the sixteenth century.20 Also in the mid-seventeenth century, a friar, Marco Bandini, described a particular class of popular magicians in the Balkan province of Moldavia, then under the rule of the Ottoman Turks. They professed, like service magicians all over Europe, to heal, divine the future and find stolen goods, but did so by going into fits and then falling to the ground to lie motionless for up to four hours. On regaining consciousness they would go into paroxysms again and then emerge from these to reveal the visions that they had received, which would provide what their clients needed. We are not told if they performed these acts in private or before others.21
All of these people, by entering altered states of consciousness in which they contacted spirits, sent out their own spirits or obtained visions, easily fit the definition of shamanism in some of its most commonly applied formulations. There is a clear value, therefore, in cross-cultural, multi-period comparisons that bring out common patterns in human experience and behaviour. Recognizing and understanding those patterns may in turn assist an analysis of specific phenomena such as attitudes to witchcraft and magic at particular places and times. Broad-brush, inclusive definitions of shamanism could therefore indeed be of genuine use to a historian. They also, however, pose problems. One is that the category of altered states of consciousness can cover markedly different kinds of experience, such as trance, dream, hallucination, delusion, dementia and reverie (each one of which is itself a loose category), and pre-modern sources can make it very difficult to distinguish between them, not least because pre-modern people only sometimes did so. To group all together as shamanism, or ‘shamanistic’ behaviour, does nothing to ameliorate this difficulty. It is certainly important to discuss what traditional human societies have in common, and shamanism provides a convenient umbrella term for direct contacts with spirit worlds by human specialists in other than normal consciousness. The danger is that the umbrella may turn into a dustbin. All human societies until the eighteenth century believed that they had to deal with spirits; what is significant is the range of different ways in which they responded to that belief. A levelling and universalizing language may deprive us of our best chances to explain varying patterns in the historical record, and – as the greatest single aim of a historian – to elicit why particular changes happened in particular places at particular times. Another problem is that it is by no means obvious that scattered case studies such as those noted above represent survivals of a widespread archaic ‘substratum’ of shamanism, as opposed to experiences and techniques which are possible in most societies at most times and are aspects of the non-culture-specific visionary activity noted by Éva Pócs. If shamans are defined as experts in communicating with a spirit world on behalf of others, then it must be beyond doubt that they existed in ancient Europe, because every traditional society has had such experts; such a definition cannot in itself, therefore, offer the potential to say anything very interesting about ancient Europeans in particular. The slightly tighter definition, of experts who perform that communication in an altered state of consciousness, is not a lot more helpful, because it seems that most human beings who claim or are claimed to make direct contact with spirits (including deities) seem to do that in such a state, and in European history have done so under the names of sibyls, oracles, seers, prophets, visionaries, saints and mystics. The quest for shamans is therefore in essence one for individuals who made such contact in ways perceptively different from all those figures, and the fundamental problem of the quest for a ‘substratum’ of them in prehistoric Europe is that such a phenomenon can only be inferred from historical evidence. In other words, it can only be demonstrated if practices surviving in historic times can be proved to be remnants of a more ancient and universal tradition, because material evidence alone (which is all that prehistory leaves) cannot testify to the nature of belief or ritual action. As the only certain proof that such historical practices were survivals of a prehistoric tradition would be direct knowledge of what actually was believed and enacted in prehistory (which is impossible), the investigation starts to go round in a circle.
It is certainly possible to try to discern from prehistoric European remains apparent traces of activities or beliefs associated with what anthropologists have called shamanism among traditional peoples in the non-European world. These include unusual burial postures, or grave-goods, or personal ornaments; particular motifs in art or architecture; musical instruments; representations of human figures that show them dancing or in the company of animals; possible traces of the consumption of mind-altering substances; and various other phenomena.22 The checklist of possible evidence is, however, so long, and the evidence itself is so ambivalent, that the quest is ultimately fruitless. Each piece of data can be interpreted in ways that have no connection with shamanism, however defined, and although plenty of such ambivalent data can be found in each epoch of European prehistory, the amassing of it does nothing to solve this omnipresent difficulty. So far at least, archaeology has only provided secure testimony to the nature of ritual behaviour when dealing with societies in which material evidence is combined with that of texts.23 The way out of this complex of difficulties chosen here is to go back to basics and ask what it was that first made Europeans take up the word ‘shaman’, and invent that of ‘shamanism’, and find either of them so interesting. Eighteenth- and nineteenth-century Europeans were familiar with a world of traditional spirituality, in which most people dwelt in small rural communities, were overawed by the forces of nature, feared and negotiated with the empowering entities of those forces, and had local specialists for that work of negotiation. They were also familiar with trance states and ecstatic visions. What they encountered in Siberia still seemed so new and remarkable to them that they had to adopt a native word for its practitioners, to distinguish them from priests, witches, cunning folk, oracles, Druids, prophets, seers, visionaries or any others of the spiritual practitioners already familiar in European culture. By establishing what that alien quality was, it is possible to define with some precision what the essence of shamanism was to the people who first identified it, and then to determine whether that essence was indeed present anywhere in historical Europe.
Classical Shamanism
What struck (and generally shocked and appalled) Europeans who entered Siberia between the sixteenth and early twentieth centuries, was the manner in which specialists among its native peoples contacted spirit worlds to safeguard the well-being of their societies: by a dramatic public performance which commonly included the use of music, song, chant or dance, or any combination of those. It was an impressive piece of drama which held the attention, and engaged the senses and imagination, of an audience. In essence, therefore, shamanism was originally defined as a particular ‘rite technique’, and it was something that Europeans found utterly strange, and for which most of them were aware of no real parallel – contemporary or historical – from within their own societies.24 The nature of that technique, and of its practitioners, may be summed up as follows.
The shamans of Siberia were rarely at the centre of social or religious life, in that they were not usually the political leaders of groups or the people who performed the regular rites in honour of deities. Nor was shamanism a single social institution, as practitioners might serve only their own family, or its neighbours or relatives as well, or a clan or a tribe, or take on any clients. Their most widespread and common function consisted of healing, by driving out or propitiating the spirits believed to cause disease; and indeed treatment of the sick is a major function of service magic worldwide. Next in order of importance in the Siberian shaman’s work came divination, another major aspect of magic globally; and in Siberia it could take the form of clairvoyance, to trace lost or stolen property, or prophecy, to advise people on how best to prepare for activities such as hunting, fishing or migration. In addition to these two big roles, there were others which were only important in some regions, such as conducting the souls of the recently deceased to the land of the dead; or repairing the psychic defences of the community and launching magical counter-strikes against its enemies; or negotiating with the spirits or deities believed to control the local supply of game animals; or conducting special rites of sacrifice.
To effect any of these tasks, shamans generally worked with spirits, as part of a world view which divided the uncanny universe into entities that were naturally hostile to humans and those which were either benevolent or could be forced to serve human needs. Such work took place within the context of local cosmologies which tended to possess three beliefs in common: that even apparently inanimate objects were inhabited by spirit forms; that the cosmos was divided into different levels or stacks of worlds, between which travel was possible (in spirit rather than physical form); and that living beings possessed more than one soul, or animating force. All over Siberia shamans were believed to depend for much of their work on the superhuman powers of their spirit helpers, who were usually regarded as taking the form of animals. This was probably for a practical reason: that it endowed the servitor spirits with mobile forms that enabled them to traverse different environments and deal with different challenges at the greatest speed. The nature of the animals concerned was a highly individual and personal matter which varied from shaman to shaman, and most made their own combinations of different species, choosing or being granted them from a very wide possible range. Some peoples believed that each shaman was assisted by one or two spirits in particular, which functioned as their spiritual doubles. The ability to call on these entities at will generally left a Siberian shaman with no need to transform into an animal in person, or to send out their own soul in such a form; though the matter could be confused by the central Siberian belief in animal spirit-doubles, with whom a shaman’s own spirit might fuse, mentioned above. There seems to be no trace, however, throughout Siberia, of a shaman changing his or her own body into that of an animal in the physical world, as Roman witches (for example) were thought to do. That is why an assumption of humans’ ability to turn themselves into animals as a sign of shamanism appears to be incorrect. The relationship between shaman and spirit helpers varied greatly across Siberia, covering broad spectra of fear and affection, and of voluntary association and coercion. Some shamans were very clearly in absolute control of their invisible assistants, while others were with equal clarity serving the wishes of theirs.
The apprenticeship of a Siberian shaman was generally divided into three phases. The first consisted of the discovery of a vocation for shamanism. This often ran in families, but the hereditary principle was qualified heavily by the fact that in theory the spirits concerned had to consent to the transfer to a new owner, and they often chose individuals with no shaman forebears, especially if those who did have such forebears did not seem as talented. In some regions they were thought to come to a new shaman unbidden on the death of the old one, and their arrival was marked in the person concerned by a physical or mental illness. Once the connection with the spirits was made, the apprentice shaman had to be trained, both by a veteran one, and by the spirits themselves in a series of often terrifying visions and dreams. The final phase of maturation consisted of acceptance as a qualified practitioner by existing shamans and by clients. All three of these stages of development varied greatly in form across Siberia and no one model can be regarded as normative; and even less generalization can be made about the relationships between shamans and the human communities they served, or indeed between different shamans.
Universally, Siberian shamans performed with some ritual costume or equipment, which distinguished them on sight from others of their community. Across most of the region they put on a special form of dress, and in the central areas this was usually ornate, consisting of a heavily decorated gown and elaborate headpiece. In most peripheral areas, however, ceremonial costume was vestigial or absent though shamanism just as important. The employment of special implements or objects was more general, and of these by far the most widespread was the drum, the beat of which usually played a major part in the performance. None the less, stringed instruments replaced it in some places and among southern areas a staff was the main shamanic accessory, and occasionally a rattle. The key characteristic of shamans for European observers, their rite technique, was naturally also varied in character, though it was usually dramatic and demanded considerable performative skills. It consisted in essence of a summoning of spirits and the direction or persuasion of them to carry out specific tasks, by a process of which the possible component parts were singing, dancing, chanting, playing of music (usually drumming), and recitation. Some of the shamanic rites were for the general benefit of communities, and others to aid individuals. Shamans often used assistants, and the audiences at the performances were almost as frequently expected to contribute to them by chanting, or singing refrains: in this manner the shamanic rite technique was often a group one led by the shaman. Siberian shamans were mostly male, but women made up a large minority in most areas and probably predominated in one (the lower Amur Valley), and they served as shamans throughout Siberia: indeed, the role gave them a unique opportunity to wield public power and influence in its native societies. No two shamans, however, performed in exactly the same way, even in the same community or family. Some induced an atmosphere of gentleness, melancholy and reflection, while others were menacing and maniacal and terrified onlookers. Some apparently sent out their own souls on journeys to accomplish the task needed, while others took their spirits into their own bodies and were possessed by, and became mouthpieces for, them in the classic manner of the spirit-medium, and yet others carried on an external dialogue with the entities on whom they had called, eliciting information from them. Some remained conscious and active throughout the whole performance, while others fell and then lay motionless and seemingly unconscious for the central part of it.
It may be seen from this that Mircea Eliade’s choice of the personal spirit journey as the definitive feat of the true shaman was mistaken, and his concern to distinguish shamanism from passive spirit mediumship was needless. Shamanism of this sort was found throughout the whole vast region of Siberia, and in neighbouring regions of Central Asia and parts of South Asia. It was also found in the Arctic and sub-Arctic regions of Alaska and Canada, round to Greenland, the Bering Strait between Asia and North America representing no barrier to communication between the peoples on either side. How far it can be said to extend into other parts of the world, such as East Asia, the Middle East, the rest of the Americas, Africa and Australia, is a matter of protracted controversy among experts, a resolution to which always founders on the lack of accepted definitions of shamanism. Mercifully, it is not a concern here, where the focus is firmly on Europe. What can be emphasized is that whereas a belief in the witch figure, while widespread globally, made up a patchwork in most regions, peoples who feared witches fervently being interspersed with those who did not regard them as a serious problem or did not believe in them at all, the North Asian and North American shamanic province forms a compact mass covering Siberia and the Canadian Arctic and sub-Arctic. However hazy its boundaries may be south of that core area, due to the definitional problem, within it all the native peoples had shamanism of the classic kind described above.
The relationship between the two belief systems, in the shaman and the witch, took a strikingly different form on either side of the Bering Strait. Across most of Siberia, as has been noted before, the witch figure was completely absent, as uncanny misfortune was blamed on hostile spirits in the natural world, or some normally benevolent or harmless spirits whom humans had offended, or occasionally on shamans working for enemy clans, who had sent their spirits to inflict damage as an act of invisible warfare. Among some peoples in the north-east of the region, some slight qualifications were found to this rule. The Koryak believed that some individuals had the power to suck life and good luck out of their neighbours, but this was regarded as innate and involuntary, and they were not thought to be possessed by evil spirits. They were therefore avoided rather than persecuted.25 Among the Sakha, a Turkic people who had migrated up into Siberia from the south-west, it was accepted that some shamans could turn bad and secretly attack the persons and property of neighbours. In those cases the culprits could be punished; but they seem to have been treated as delinquents rather than embodiments of evil, and the usual penalty was a fine.26 By contrast, many of the peoples in sub-Arctic and Arctic North America who possessed shamans of the Siberian sort also feared and hunted witches; and there the shamans used their powers to detect and unmask alleged practitioners of witchcraft in the manner of service magicians across the world, including those in Europe. This is true around the north of the New World from the Tlingit of Alaska to the Eskimo of Greenland.27
One further point needs to be made in this portrait of the ‘classical’ shamanism of Siberia: that it was assembled by selecting certain characteristics from the members of what were often quite complex local assemblages of magico-religious specialists. Thus, for example, a study of the Sakha divided such specialists among them into the oyun, or man who worked with spirits in a trance by means of a public performance, and the udaghan, his female equivalent; the körbüöchhü, or diviner; the otohut, or healer; the iicheen, or ‘wise person’; and the tüülleekh kihi, or dream interpreter.28 Their work clearly overlapped, and it is no longer possible to distinguish them absolutely. To the scholar interested in shamanism, it is the oyun and udaghan who are the figures corresponding to that category, but to the anthropologist interested in the spiritual world of the Sakha, all are important; and across Siberia the individuals identified as shamans often engaged in other kinds of magic as well, such as the use of symbolic natural substances, and incantations. In making comparisons between Europe and Siberia, such complexity is an important factor within European societies themselves. The service magicians who featured in early modern Hungarian witch trials included not only the táltos, the spirit-warrior who has been mentioned earlier, but the ‘healing woman’, the ‘woman doctor’, the ‘herb woman’, the ‘learned woman’, the ‘midwife’, the ‘seer’, the ‘bed-maker’, the ‘smearer’ and the ‘wise woman’. Nor does it seem as if many of these terms could be used as alternatives for the same sort of functionary: the ‘learned woman’ and ‘woman doctor’ were of greater status than most of the others, and most could generally be distinguished by their methods.29 Equivalents found in Finnish records between the seventeenth and nineteenth centuries were led by the tietäjä, who was the most high-grade and usually male, and the noita, who was less respected, less reliably benevolent, and mostly female, but there were also five other kinds of magician of lower status, each with its own name.30
It is easy to see why scholars concerned with an understanding of how the concepts of magic and witchcraft have operated in particular human groups might feel uneasy about cultural cross-comparisons made by concentrating on particular figures and features without reference to the local contexts in which they are embedded. None the less, without some such exercise such comparisons can hardly be made at all, and they seem to have some validity. In the case of the historical debate with which this chapter commenced, it should be readily apparent now why some of the participants should have seen some similarities between Siberian shamans and specific kinds of folk magician in early modern and modern south-eastern Europe. It should also be apparent, however, why Gustav Henningsen placed more emphasis on the differences, and his view may be borne out further by the means by which some of the positive comparisons were made. For example, the Hungarian táltos was held to have the following features in common with the Siberian shaman: being born with a distinguishing physical feature (such as teeth or an extra bone); an initiatory experience in childhood (a convulsive illness, a mysterious period of disappearance, or a visionary dream); the acquisition of unusual powers (such as vanishing at will, changing shape into an animal, or duelling with enemies in spirit flight); and the use of special equipment (such as a head-dress, drum or sieve). The first of those, however, is rarely found in Siberia, and this is also true of the first two of the three unusual powers mentioned. Missing in Hungary is the distinctive shamanic performance and much reference to work with spirits. Moreover, the features of the táltoslisted above as significant do not appear together in descriptions of actual individuals: rather, they have been constructed from different morsels of (mostly modern) folklore to create an ideal type. In the earlier accounts, Siberian features such as the special equipment seem to be rarer or missing.31For this reason the táltos, benandanti and similar functionaries from south-eastern Europe cannot readily be accepted here as part of a historical Eurasian shamanic province, and the different question of whether they were survivors from an ancient one cannot be resolved. What can be attempted more readily is a solution to the question of whether any figures unambiguously similar to Siberian shamans can be identified in Europe at a period for which records survive, and the logical place to seek these is in areas closest to Siberia itself: in Russia and the Arctic and sub-Arctic north.
A European Shamanic Province
One obvious point at which to commence this quest is among northern peoples belonging to the Uralic linguistic and ethnic group, which straddles the Ural Mountains that separate Siberia from Russia and also includes the Magyars as an isolated component. The members of this group who live in western Siberia have practised classic shamanism of the universal Siberian sort, so what of those just on the opposite side of the mountains, in Russia itself: the Mordvins, Cheremises, Chuvashes and Votyaks? Here the information available seems to derive mostly from nineteenth-century folklore collections. The Mordvins had people who specialized in communing with the dead, and old men who put on white robes at festivals to bless food. The Cheremises had diviners who cast beans and gazed into water, and the Chuvashes people who healed with herbs, told fortunes and recited charms to banish diseases. None of these sounds very much like a Siberian shaman, and most are no different from forms of popular magician found across the rest of Europe. The Votyaks, however (a people also known as the Udmurts or Chuds), had a figure called the tuno, who told fortunes, healed diseases and found stolen or lost property, either by prayer or by going into trance. Sometimes the trance state was achieved by dancing with a sword and whip to the music of a psaltery, until he cried out the answers to questions in delirium. The vocation was mostly hereditary, although gifted individuals could take it up without such a qualification, and trainees were instructed at night by spirits. This does sound close enough to Siberian shamanism to suggest that it was an offshoot.32 There is also an isolated medieval reference to a similar practice among the same people under the year 1071 in the Russian Primary Chronicle, which was written in 1377 but drew on an original work from the early twelfth century, which in turn was based on older material. It tells how a man from the major Russian city of Novgorod went among the Votyaks. He paid to have his fortune told by a tribal magician, who called spirits to himself to provide the answers, which came to him while lying in a trance inside his dwelling.33
The great European parallel for Siberian shamanism has, however, been located precisely where, perhaps, it would be logical to look for it, in the Arctic and sub-Arctic where it more or less completes the circumpolar shamanic province which extends across Siberia, North America and Greenland. It has been detected among the Sámi or Lapps, a Uralic hunting, fishing and herding people who occupy northern areas of Finland, Scandinavia and Russia and formerly extended their range far into the central portion of the Scandinavian peninsula and further across the top of Russia. The first record of shamanism among them occurs in the twelfth-century Historia Norwegiae, which described the Sámi as having people who were claimed to divine the future, attract desirable things from a great geographical distance, heal, and find hidden treasure, with the aid of spirits. It included a report from Norse traders of how their Sámi hosts had attributed the sudden apparent death of a woman to the theft of her soul by spirits sent by enemies. A magician then took up residence under a cloth hanging and picked up a drum or tambourine painted with pictures of beasts, shoes and a ship, which represented forms of locomotion for his attendant spirit. He then played this, chanted and danced until he suddenly died himself, allegedly because his spirit-double had been killed in a battle with those of the enemies. Another magician was consulted, and performed the same rite, this time with success, as he survived, and revived the woman.34 In every detail, this could be a description of a Siberian shaman at work, and it is small wonder, if they used the classic, dramatic, shamanic rite technique, that Sámi magicians became famed for their prowess among the medieval Norse. It aided their reputation for arcane powers that whereas the other Scandinavian peoples converted to Christianity around the year 1000, the Sámi remained pagan for more than half a millennium longer. Particular examples in Old Norse literature of magic worked by them testify again to the importance of ecstatic trance to their mode of operation. The thirteenth-century Vatnsdalers’ Saga features a group of three of them being hired by Norse chieftains to trace a missing amulet, and doing so by shutting themselves indoors for three days and nights while their spirits wandered abroad and found it.35 Sámi magic also features as a distinctive, exotic and potent phenomenon in other sagas, though none of these literary sources portrays the Siberian performance rite in connection with it.36 What continued to fascinate other Europeans was its reputed ability to send out the magician’s spirit from its body at will to roam freely about the world. When the German Cornelius Agrippa wrote his major study of ceremonial magic in the early sixteenth century, he discussed this ability with reference to the Greek sages, and commented that in his own day it was still found among many in ‘Norway and Lapland’.37 The expression ‘Lapland witches’ became a commonplace one in early modern English literature for especially potent magicians, being found in Shakespeare, Milton, Defoe and Swift, in addition to lesser writers.38
Among the Sámi themselves, a fresh body of information upon their magical practices was generated in the seventeenth and early eighteenth centuries, when the monarchies of Denmark and Sweden, that had partitioned northern Scandinavia and Finland between them, brought the native population under direct government. As part of this process, Sámi magicians were sometimes accused of witchcraft, and also caught up in the more general process of the enforced conversion of their people to Christianity. This generated further descriptions of their practices, even as the practices themselves were wiped out. The main magical practitioners were called noaidis or noaidies, who attempted to divine, influence the weather and heal. These were mostly male, and generally went into deep trances and lay immobile while their spirits left their bodies to work with attendant spirits, which often took animal form. Sometimes their own spirits duelled with each other for supremacy, and the loser sickened or died. Their main piece of equipment was a drum, usually painted with symbols, and at least at times their rites took the form of dramatic public performances. At some of these they had assistants dressed in special costumes.39
All this sounds like absolutely classic Siberian shamanism. It is true that the data are not quite as good as may be wished. All of these early modern sources seem to be second-hand, and there appears to be no eye-witness account of a performance by a noaidi among them. The drums survive in large numbers, but it is not proven that they were used to induce trance, as in Siberia, although magic was certainly made by chanting or singing while playing them. Still, the detail of the descriptions has been sufficient for most scholars of the subject, from the nineteenth century to the present, to describe the noaidis as shamans.40 Recently, however, a leading expert on north Norwegian witch trials, Rune Blix Hagen, has questioned this tradition, noting that none of the actual testimony of noaidis themselves speaks of sending their souls from their bodies; which he takes (in the Eliade tradition) to be the definitive ability of a shaman.41 He has therefore suggested either that the label of shamanism be withdrawn from the Sámi or that the category of shamanism be broadened beyond spirit-flight. It may be suggested here that the latter course is by far the more appropriate, as the emphasis on sending out a spirit from a human body as the hallmark of shamanism is largely a legacy of Eliade’s work that does not match up, as stated above, to the records of actual Siberian practice. The noaidi is clearly portrayed by the trial records used by Hagen himself as playing the drum in order to contact spirits in order to aid other humans, in a dramatic public performance, and that is enough to fit the Siberian model.42 In addition, however, the chain of detailed reports by outsiders of the performances of Sámi magicians from the twelfth to the eighteenth centuries, are so strikingly like those of Siberian shamans that if they are not representing reality, this in itself is an anomaly remarkable enough to demand explanation.43
A form of shamanism of the Siberian sort was also found among the neighbours of the Sámi to the south-east of their range, the Finns, who were another Uralic people. This was centred on the figure mentioned above, the tietäjä, who is recorded both in early modern trial records and modern folklore. Such people dealt with the same human needs as the noaidi, which were indeed the concerns of service magicians across Europe, and did so by going into altered states in which they sent out spirit helpers in human or animal shapes to do battle with evil spirits. Those states sometimes consisted of dream or intoxication, but mostly of exaltation or frenzied anger, accompanied by chants or songs and attained in front of clients and sometimes a larger audience. Their incantations, like the songs of many Siberian shamans, depicted the topography of spirit worlds.44One of the foremost recent experts on these figures, Anna-Leena Siikala, has stated that it is ‘universally agreed’ that they were descended from shamans; indeed, those who had all of the characteristics delineated above could quite accurately be described as continuing to practise it, as only the lack of special equipment is missing in comparison to the Siberian template.45 The medieval Norse referred to the Sámi and the Finns indifferently as ‘Finnar’, and if both had the shamanic rite technique it is easy to see why they retained such a reputation for magical potency among other Northern Europeans. Modern British and American seamen still believed ‘Finns’ to be wizards of frightening power, especially over the weather.46
A European Sub-Shamanic Province
It may therefore now be proposed that the ‘classical’ Siberian shamanic province had an extension into Europe that crossed parts of northern Russia to Finland and northern Scandinavia. What may be asked next is whether this produced any borderlands in which shamanic features of magical practice were blended with those more familiar across most of the European Continent. Here an answer has been proposed regularly since the nineteenth century: that elements of shamanism were found among the medieval Norse who were neighbours of the Sámi and Finns, especially in the specific practice known to the Norse as seiđr. This has, however, been controversial for most of the time in which it has been stated, and the debate continues to the present. In 2002, as part of an extensive and much-admired work on Viking spirituality, which united textual and archaeological evidence, Neil Price argued that many aspects of its ritual were ‘fundamentally shamanistic’ in nature. He also concluded that the seeresses of medieval Norse literature exactly resembled the shamans of the circumpolar region and that seiđr was a ‘shamanic’ belief system.47 In another major study published seven years later, Clive Tolley responded that virtually all of the sources on which such suggestions rely are works of imaginative fiction produced long after the end of the pagan Viking era and so completely unreliable as evidence for its beliefs and practices. He was also concerned that the apparent elements of shamanism in representations of seiđr and other Norse magic could have derived from other belief systems and cultural contexts. None the less, he allowed that the literary portraits had to reflect notions of what was likely to have existed, and that those notions would in part have derived from tradition; and he concluded that the ‘evidence does, however, support the likelihood of some ritual and belief of a broadly shamanic nature’.48
The present study has the advantage that it can avoid the whole issue of the reliability of the sources and the actual practices of Viking Age Scandinavians, because it is concerned with representations of, and beliefs concerning, magic and witchcraft. In other words, if the medieval Norse conceived of magical practices in a certain way, then it does not matter much to the needs of this book that we apparently cannot tell whether they were actually practised. That the portrayals took one form rather than another is significant enough. It is clear that the authors of medieval Norse sagas and romances often used the term seiđr as one for magic, without going into details with regard to its nature, but also that they had a particular tradition of it: as a form of divination, practised mainly by women who went into trance in a raised seat or on a platform. Such people often roamed around districts being entertained by landowners, who sought their services to find answers to problems or learn their future or that of their household members. The most elaborate and famous description of one, in the Saga of Erik the Red, concerns Thorbjorg, a Greenland seeress with an elaborate and decorative costume of mantle, hood, belt, shoes, staff and pouch (for charms). A retinue of women sat round her as she took her place on the ceremonial platform, and one sang incantations to call her spirits to her so that she could provide answers to the questions asked of her. She was the most famous and long-lived of nine sisters with the same gift.49 All this – the hereditary vocation, the visit to the home, the costume, the assistants, the sedate practitioner and the calling of spirits – is paralleled in Siberia: it is a classic shamanic performance. Only the platform is missing in Siberia, and a special seat or framework is sometimes found there. Arrow Odd’s Saga has farmers invite a similar woman to their homes to tell fortunes and predict the weather; she travels with a retinue of thirty young women and men to make a great incantation about her when she performs.50 Frithiof’s Saga tells of two seiđkonur (women who know seiđr) hired to drown some enemies of their paymaster at sea. They ‘moved onto the platform with their charms and spells’, and sent out their spirits to ride a huge whale and raise a storm. Their intended victim, however, saw their shapes on the whale and sailed his ship at them and struck them. The whale dived and vanished and their bodies fell dead off the platform, with broken backs.51 This associates seiđr with the other shamanic trick of sending the shaman’s spirit out, in collaboration with an animal one, to achieve an effect. In Gongu-Hrolf’s Saga, twelve men are paid to use seiđr to kill a pair of victims, and set up a high dais in a house in a wood to do so, making a great din with their spells.52 Hrolf Kraki’s Saga tells of a king who engages a prophetess to locate two concealed rivals of his; she mounts a tall platform, yawns deeply (perhaps to take in spirits) and speaks verses to disclose the necessary information. Later in the story a queen sits on a seiđr platform, and perhaps animated by a servitor spirit, sends a huge boar to join a battle she is fighting against enemy warriors, and also seems to make the dead among her own followers revive to renew the battle.53
Sometimes such shamanic features are found in stories without any explicit connection to seiđr. A famous case in Hrolf Kraki’s Saga concerns a warrior who sits motionless aside from a battle while a huge bear (either containing his spirit or one that he has summoned) attacks the enemy to great effect; it vanishes when his body is disturbed.54 The god Ođinn was believed to send his spirit to range the world in a variety of animal forms while his body lay asleep or dead.55 A magician in the Saga of Howard the Halt sent out his in fox shape to spy out a hall full of enemies. He had proofed his animal form against swords but was killed when somebody bit out its throat.56 A man in the Vatnsdalers’ Saga could make his friends invulnerable in a fight if he lay down motionless nearby (presumably sending out his spirit to aid them).57 Another much-cited Norse form of magic, ùtiseta, ‘sitting out’, seems to have consisted of sitting in a special place out of doors at night and waiting, perhaps in a trance state, to see the future or obtain other wisdom: as this is also found (forbidden) in medieval Norse laws, it was certainly believed to be practised.58 The staff which features as an important accessory of Thorbjorg appears as a significant magical tool in the saga tradition in general. In Laxdaela Saga, the bones of a woman are found beneath the site of a church, with a large ‘seiđr staff’. It is identified as the grave of an evil pagan magician.59 The Saga of the Sworn Brothers has two women who could send their spirits to spy out the land at night, each riding a magic staff, while their bodies slept.60
In a fine discussion of the place of the staff in this literature, Neil Price notes that it was used for divination by the Chukchi people of Siberia; but the parallel is even closer than that, because it was, after the drum, the most common piece of shamanic equipment among Siberians.61 None the less, none of the examples quoted above, except the first four, which illustrate seiđr, amounts to the full Siberian (and Sámi and Finnish) rite technique. Rather, they represent aspects of that technique which are found without the whole assemblage, just as elements of seiđr sometimes appear in stories without the whole package of that tradition, such as women wandering the land to tell fortunes or one sitting in a chair at a feast to do so.62 The ritual platform is sometimes used by a magician or groups of magicians simply to send out spells or curses from it.63 A queen in one tale ‘moves her spirit’ to bring fierce animals to fight for her.64
The idea that people can send out their spirits from their bodies in a shamanic performance to roam spirit worlds blends by degrees into that mentioned above, in which their spirit goes about in the form of a physically solid animal in the human world, while their body remains at home. It is a short step from that to straightforward shape-shifting, in which a human body turns into an animal one, while retaining a human mind, something not a feature of Siberian shamanism. A female magician turns herself into a walrus in Kormák’s Saga in order to attack a ship, but her opponent sees that it has her eyes and spears it, killing her.65 The Saga of the Volsungs held that seiđkonur had the power to change themselves into she-wolves or to take the shape of other humans; and two characters in it become wolves temporarily by donning enchanted wolf skins.66 A mother in Eyrbyggja Saga tries to hide her son from enemies come to kill him by turning him successively into a distaff, a goat and a boar; her power came from her eyes and was thwarted when a bag was put over her head.67 In addition, there is plenty of magic in the sagas which is not in the least shamanic, and is recognizable across most of Europe, such as passing hands over a person and chanting, or reciting spells, or cutting letters and speaking over them, or walking anticlockwise about a space while reciting or sniffing, or offering sacrifices to spirits with invocations.68The medieval Norse language was replete with a proportionately wide range of terms for different kinds of magic.69
It looks, therefore, as if this literature, virtually all of which was written in Iceland in the thirteenth to fifteenth centuries, represents a hybrid culture which blended elements of genuine shamanism with features of magic more familiar from elsewhere in Europe. Both Neil Price and Clive Tolley therefore seem to be correct: Price that the shamanic elements are both major and significant and Tolley that they may not be fundamental to medieval Norse concepts of magic. It is possible that they were survivals from a prehistoric past, and so native to the Norse, but it could also be that they were the result of contact with the Sámi and Finns who were such near neighbours and made an impact on the same literature.70 This would explain why such strong shamanic features as those manifested by Thorbjorg do not appear in the literature of the mainland of medieval Europe, south of the Baltic. If such an influence took place, then the Norse refashioned shamanism in their own way, dropping the drum, for example, and substituting the staff. The hybrid nature of the result also helps us to clarify the components of medieval Norse magic that are notshamanic, and seem to derive from a different tradition. One of these is the prominence given to women as prophetesses and diviners: as said, they formed a minority of Siberian shamans and a small minority of Sámi noaidis. Following a tradition going back to Jacob Grimm in the early nineteenth century, Clive Tolley has linked the Norse pattern to the ancient Germanic reverence for women as possessors of prophetic wisdom, discussed in the previous chapter of this book.71 He is surely right to do so, and his distinction between the two, that seiđrkonur were figures on the fringes of society and the ancient Germanic prophetesses central to it, does not really stand up. The seiđrkonur were invited into the heart of communities in the sagas, even if their peripatetic lifestyle prevented them from belonging to any, while the only ancient Germanic equivalent of whom we have relevant details, Veleda in the first century AD, lived isolated in a tower, and communicated with the local tribes by messenger.72 The semi-detachment of both seems equivalent. A comparison of Norse and Germanic cultures, however, also reveals distinctions between the two. The Icelandic sagas and romances and Norse law codes show none of the fear of the cannibal female witch, preying on her fellow humans at night, which features in the early medieval Germanic sources. Indeed, there seems to be no actual presence of a witch figure in the early medieval Scandinavian texts: no terror of a malevolent worker of magic concealed within local society who causes misfortune to others because of pure evil. Women and men both feature as destructive magicians, but always as part of the feuding that was a key activity of early medieval Scandinavian society and a major theme of its stories. Magic is one more weapon in the waging of factional and personal violence, though a cowardly and disreputable one except when used in defence against other magic or the supernatural.
There are, however, figures in the Norse literature that can in some respects be equated with witches: women who ride around at night on enchanted physical objects or animals. In The Saga of Gunnlaug, a wolf is called svaru skaer, rendered by its English translator as ‘witch’s steed’.73At times it is clearly their spirits that are making the ride while their bodies remain asleep at home, as in the Saga of the Sworn Brothers. The staff that is the means of transport for those characters is not the only such object used: there is mention in the literature of tunriđur, those who ride on a hurdle, fence or roof. In the famous poem Hávamál, the god Ođinn boasts of his ability to see such people ‘play frenzied in the air’, and thwart them by rendering them incapable of finding their ‘home shapes’, and their ‘true homes’ or ‘own skins’ again, once more suggesting that they need to get back to their everyday bodies by morning.74 A law from West Gotland of the early thirteenth century forbids various terms of slander against a woman, one of which is ‘I saw you ride on a hurdle, with hair dishevelled, in the shape of a troll, between night and day’.75 The ‘troll’ here would be one of the human-like, often malevolent, beings famously thought to haunt wild places in Scandinavian mythology, often with homes underground. The women who send out their spirits to ride a whale in Frithiof’s Saga are essentially performing the same trick in animal form. The reference to such night riders as ‘playing in the air’ suggests that they were thought to assemble for revels, and this is borne out by other sources; but where such assemblies are mentioned, they are not always thought to be composed of humans. In Ketil’s Saga, the hero meets a female troll, hurrying to join a gathering of her kind. It is on an island, where the tale states that ‘there was no lack of gandr rides’ that night: the term gandr could refer to a spirit, or perhaps an enchanted object such as a staff or hurdle.76 In the Tale of Thorstein, that hero follows a Sámi boy in a ‘gandrride’ upon a staff to an underworld, to join a festival of its beings.77 Trolls and other underground beings were thought to possess, and to confer, magical powers, and twelfth- and thirteenth-century Norse law codes forbade the ‘raising up’ of trolls to gain such powers: the standard term for witchcraft in all Scandinavian languages later became trolldromr, and witches trollfolk.78 The night riders could be dangerous to humans as well as helpful to them: in Eyrbyggja Saga a boy is attacked while walking home alone after dark and wounded on head and shoulders. A local woman, suspected of being a ‘hurdle-rider’, is accused of having used him as her steed, though she is acquitted when twelve neighbours swear to her innocence.79 The literature, however, does not attest to any great fear of them, or belief that they affected other people as long as the latter kept to their homes at night. It seems, therefore, as if in some accounts of seiđr, spirit-projection of the Siberian and Sámi shamanic sort was being combined with a native Norse one of nocturnal revels, generally of superhuman or non-human creatures, to which humans could ride or fly in spirit using objects, animals or other people as steeds. It may therefore be concluded that there is good evidence for an extensive and compact province of ‘classical’ shamanism in the northern hemisphere, covering not just Siberia and adjacent parts of Central Asia, but the Arctic and sub-Arctic zones of North America, and extending into Russia and Scandinavia. In addition, the influence of this shamanism can be detected in a ‘sub-shamanic’ zone covering other parts of the Nordic world, such as Norway and Iceland. It is entirely legitimate to propose the existence of shamanic elements in magical practices elsewhere in Europe, if only because of the lack of any generally accepted definition of shamanism; but such a proposal cannot be made conclusively, and the evidence for it can be read in alternative ways.