1
Liminal Fairies
The Anthropological Paradigm
The modern image of an elf is a spectacular one, beaming with radiance, power and nobility. Tolkienian elves have taken us far away from notions of shabby household spirits of the brownie sort and elevated the vision of fairy royalty to heights of splendour matched only by the loftiest of descriptions found in medieval romance. It is in this sense that Tolkienās elves are arguably Celtic (Green, Elf 5), for their image capitalizes on ideas of fairy illustriousness, majesty and beauty characteristic of Irish beliefs in particular. While such a vision of fairy regalityāone readily associated with Celtic provenanceāproved to be particularly amenable for writers of medieval romance, this should not nonetheless blind us to the fact that fairies are a āpanāEuropeanā phenomenon rather than one originally restricted to the fringes of occidental civilization, and āthe questions they raise should not be quarantined to the margins, either geographical or cultural, of medieval societyā (Green, Elf 7). Richard Firth Green vehemently dismisses āthe Celtic fallacyā (Elf 6) regarding the origin of fairy belief and sees it as seriously detrimental to scholarship in that it hampers efforts to recognize the ubiquity of ideas concerning fairies across medieval Europe.1 His point is that even when some elements of local lore do resemble Celtic sources, they should not be taken to constitute a case of borrowing without firm evidence, and there is no reason to suppose such borrowings to be common given the vibrancy of medieval fairy lore on the continent (Elf 6ā7). Thus, while the Celtic material is of great importance and its general import must be recognized by any conscientious historian of fairy belief, one should not lose track of the fact that fairies are not a uniquely Celtic phenomenonānor, indeed, a distinctly insular one, since a whole array of similar entities has for centuries peopled the imagination of Europeans across the continent. This observation will serve as an important context for the theoretical discussion to follow.
Greenās medieval examples often come from France or Italy, but beings analogous to fairies can also be identified in various other cultures and areas. The fairy creatures in Britain and Ireland can often seem wildly dissimilar from one another, and their counterparts in other areas in Europe, too, exhibit a great deal of variety, the fundamental thing that connects them all being a manifold sense of liminality. In Scandinavia, we find the huldrefolk (āthe hidden peopleā), a supernatural race living in parallel to their human neighbours in an underground realm (Davidson 26).2 Also known as the underjordiske (āthose under the groundā) or the haugfolk (āpeople of the moundsā; Bringsvaerd 11), the huldrefolk āstrengthen their gene pool by marrying humans and stealing childrenā (Gilbertson 200) and are a constant threat to be reckoned with, being, āto the Norwegian mind, always lurkingā (Gilbertson 200). These Norwegian fairies illustrate the principle of fairy dependence exceedingly well. To the south of Scandinavia, among the Western Slavs, similarities to the British tradition are less obvious. On a superficial level, British fairies seem to have very little in common with tiny gnomes with red caps known among the Poles as krasnoludki (āthe fair little folkā), but the name of the latter betrays an essential connection by combining two well-known apotropaic expressions: the fair folk and the wee folk.3 Polish folklore actually has a subcategory of āriverbank demonsā (DÅŗwigoÅ 170ā173), and their liminal association with the place where water meets land is matched by their inclinations to steal children and copy human activity. They may not resemble British fairies much, with their long, saggy breasts that hang about their necks and onto their backs, but these bogunki exhibit forms of behaviour that we can immediately recognize.4 Their name translates as demigoddesses (or goddess-like beings) and dates back to the sixteenth century (BrĆ¼ckner 34), which is incidentally when the English literary tradition began to acknowledge on a grander scale than before the conflation of classical demigods and fairies (Cooper, English Romance 177). Bogunki not only leave their boginiaki (ādemigodlingsā) as changelings but also tend to ape their human neighboursāso much so that one of their folkloric names is maÅpy (āapesā): āwhen they see someone comb their hair, they too comb their hair; when they see someone wash clothes, they do the sameā5 (DÅŗwigoÅ 165). Once again, this is fairy mimicry at its most prominent.
Farther south, in the Balkans, the beliefs of the Southern Slavs are not much different. Their fairy creatures can act in both benevolent and malicious ways (Kropej 144). Local fairies, known as vile, can easily bring prosperity to the household and just as easily take it away when they take offence. They are generally thought of as a source of wealth and good luck, but one who interrupts their dances in fairy circles may end up paralyzed or even dead (Kropej 147). They are also known to interact with human beings in countless ways, including the substitution of changelings for human children (Kropej 145) and sexual liaisons with human lovers. Slovenian folklore knows stories of fairy wives imposing taboos on their mortal husbands and leaving them as soon as the taboos are broken (Kropej 148ā149) that are little different from analogous tales in medieval romance. Vile also live in a kind of parallel dimension, for while they are usually seen engaging their human neighbours or roaming remote and wild places such as forests and mountaintops, the latter are believed to provide entry to a āhidden paradiseā of their own, often accessible through cracks and hollow rocks (Kropej 147). The usual image of vile is that of beautiful young girls, and in this they resemble the Greek nymphs. Nymphs are themselves liminal beings through and through: they āare caught between the world of gods and the world of menā (Pur-kiss, At the Bottom 39), native to the transitional space ābetween heaven and earthā like the vile (Kropej 125). Imagined as young women on the brink of adulthood and motherhood, they are āāstuckā in a particular phase of lifeā (Purkiss, At the Bottom 39), seen as having āfailed to pass from a transitional phase to a phase of completionā (Purkiss, At the Bottom 35), that is, from girl to woman. Diane Purkiss makes much of the fact that no stories are ever told about them, only about the mortals they abduct for lovers or otherwise interact with: ā[t]heir story is to have no story. They are forever young, forever on the brink of love, marriage and commitment. But not there yet and so nowhereā (At the Bottom 40). Like British fairies, they may superficially seem to be associated in the popular imagination with groves, lakes or hills, but they are not pure spirits of nature. ā[E]ven modern Greeks believe that nymphs are connected not with the untamed natural wilderness, but with the housewifery that turns nature into cultureā (Purkiss, At the Bottom 41), and thus, like their counterparts from across the continent, they occupy the conceptual space of transition, threshold and ambiguity. Purkiss adds a few examples from beyond Europe, looking to the Middle East for more liminal demigods, and her survey of ancient demons such as the Mesopotamian child demon Kubuāātrapped between states of beingā (At the Bottom 15) by virtue of being unborn or stillbornāreinforces the impression of liminality being a defining feature of a whole class of entities intimately familiar to a variety of cultures and historical periods.
Indeed, while not all such beliefs can confidently be traced back to the Middle Ages and serve as valid points for detailed comparison with the medieval British tradition, this only proves that the idea of a class of beings characterized by a manifold sense of liminality has persisted across both space and time. To understand British fairies, one must take this broad perspective into account, whatever the permutations found in local lore and its literary representations, and the anthropological paradigm inaugurated by Arnold Van Gennep is where such a universal theory of liminality has emerged.
Liminality, Danger, Pollution
Alongside liminality, another key phrase in the intellectual tradition of Arnold Van Gennep is undoubtedly ārite of passage.ā It is to numerous examples of these that he devoted his Les rites de passage, published in 1909, and while his study attempts to offer a broader analytic look at the phenomenon in its brief introduction, it mostly comprises a body of evidence on cultural practices that strikes the reader as quite repetitive in many ways but ultimately succeeds in adducing the bookās main point. The argument illustrates that despite much variation, there is a universal pattern to superficially dissimilar ritual activity pertaining to, and performed in, a whole variety of situations by people across cultures. Perhaps the most interesting part of the bookās legacy is Van Gennepās vision of society that underlies his claims:
He saw society as a house with rooms and corridors in which passage from one to another is dangerous. Danger lies in transitional states, simply because transition is neither one state nor the next, it is undefinable. The person who must pass from one to another is himself in danger and emanates danger to others. The danger is controlled by ritual which precisely separates him from his old status, segregates him for a time and then publicly declares his entry to his new status.
(Douglas, Purity and Danger 97)
While most summaries of Van Gennepās ideas begin with the division of rites of passage into phases of separation, transition and incorporation (Van Gennep 10ā11), it is just as important to stress that the scheme serves a particular purpose, which Van Gennep identifies as making the passage possible, or facilitating its success: āthere are ceremonies whose essential purpose is to enable the individual to pass from one defined position to another which is equally well definedā (Van Gennep 3). To understand the logic behind this vision, one must therefore attend not just to the remedy offered by ritual but above all to the danger lurking in the act of transitionāone, it would appear, indiscriminately inherent in all acts of transition. Whereas Van Gennep himself never elaborates in much detail on the precise nature of the problem that ritual serves to alleviate, his insights become much clearer when viewed through the later work of two British anthropologists: Victor Turner, who brought Van Gennep into the spotlight roughly half a century after the completion of The Rites of Passage and continued his work on ritual and liminality, and Mary Douglas, who took a particular interest in liminal danger and its polluting nature.
The universality of Van Gennepās notion of rite of passage emerges quite clearly when one realizes that the passage in question occurs between āstatesā of numerous kinds. Turner defines āstateā as āa relatively fixed or stable conditionā pertaining to ālegal status, profession, office or calling, rank or degreeā as well as āecological conditions, or ā¦ the physical, mental or emotional condition in which a person or a group may be found at a particular timeā (Forest of Symbols 93ā94):
A man may thus be in a state of good or bad health; a society in a state of war or peace, or a state of famine or of plenty. State, in short, is a more inclusive concept than status or office and refers to any type of stable or recurrent condition that is culturally recognized.
(Forest of Symbols 94)
Furthermore, Turner is apt to point out that āVan Gennep himself defined ārites of passageā as ārites which accompany every change of place, state, social position and ageāā (Forest of Symbols 94).6 This works both on the level of the individual, at moments of āculturally defined life-crises,ā and with regard to the entire community, as when āa whole tribe goes to warā or holds a harvest festival to mark the seasonal āpassage from scarcity to plentyā (Forest of Symbols 94ā95). Any change-of-state scenario invites ritual, and thus, the problem of āpassageā proves to be of fundamental importance to the workings of culture.7
One reason why the legacy of Van Gennep exerted considerable influence in the second half of the twentieth century is the proto-structuralist character of his method. Structuralism, which had its roots in Ferdinand de Saussureās (1857ā1913) approach to problems of language and linguistics, was later adopted by other fields of scholarship, and the structuralist method came to dominate anthropology and frame much literary theory from around the 1950s. Notes taken from Saussureās lectures, published by his students in 1916 under the title of a Course in General Linguistics, were an inspiration for much innovative work done within both domains. While the finer aspects of the structuralist methodology of linguistics did not always translate smoothly into the exigencies of anthropological or literary scholarship, Saussure formulated a number of insights that profoundly affected the development of twentieth-century thought, and Van Gennepās understanding of culture and society reveals a fundamental agreement with them. What overlaps here is the concern of both thinkers with the underlying system that subtends cultural practices, and a consequent methodological focus on deep-level structures rather than surface-level phenomena. As structuralism has it, while reality may be a continuum, the way human thought organizes it is necessarily in discrete and binary terms. Thus, in order to conceptualize anything, thought breaks it down into discrete units, and the meaning of any such unit is not inherent in it or derivable from its inner essential nature but rather emerges out of the differential interplay of all the elements in a particular system under scrutiny. According to Saussure, the meaning of any given word in a language is purely relational and system-dependent, and no element in a language has substantial value on its ownāwhich is what makes it possible for the same word to have different meanings in different languages.8
Saussure was predominantly interested in language and the meaning of linguistic signs, but his theories lent themselves easily to broader applications. One of them was the idea of linguistic determinism, which suggested that ā[s]ince the actual concepts which are available for speakers to encode and communicate are determined by the particular language they speak, thought itself must be dependent on languageā (Chapman 149). The entire conceptual grid at the disposal of any human being came thus to be seen as dependent on his or her language, an approach that effectively equated the systemic configuration of concepts operative within the human mind with the way different languages broke down the totality of experience into discrete units of meaning. This approach stressed to a great degree the differences between various languages and invited much criticism on this account (Chapman 151), but on a more fundamental level, it suggested that however much particular languages or their concurrent conceptual grids may differ, each system is invariably structured in terms of discrete binary oppositions. In the words of Jonathan Culler, ā[s]tructuralists have ā¦ taken the binary opposition as a fundamental operation of the human mind basic to the production of meaningā (Structuralist Poetics 15). Structuralism thus went beyond the study of language to make claims about the structure of human thought as such, and the laws that defined the operations of language could now be transposed to the fabric of conceptual thinking.
It was a similar line of thought, and his understanding that the deep-level structuring of culture was crucial for his scholarly enterprise which allowed Van Gennep to argue that ritual accompanies any transition between conditions taken to operate on the level of the system. This can be best illustrated by his reflections on puberty rites. While the term puberty may seem to indicate a well-defined period with essential biological characteristics, from a structuralist perspective what determines its meaning are ideas framing it on both sides, that is the periods of human development understood to precede or follow puberty in a given culture. Structuralismāwhether linguistic or anthropologicalāinquires into the system rather than the undifferentiated continuum, since it is on the level of the system that it locates the structural base for all operations ...