can hardly be later than the tenth century and may well have been composed in the ninth; moreover, at least one of its manuscripts can be dated to the mid-eleventh century—earlier than the Irish Book of the Dun Cow and considerably earlier than the oldest manuscript of the Welsh Mabinogion.24 One might of course argue that such Celtic legendary material predates the manuscripts that record them, but this is an argument that can be applied to the Germanic materials as well.
As an illustration of the distortion that the Celtic fallacy can cause, we might consider Dorena Allen’s otherwise excellent article “Orpheus and Orfeo: The Dead and the Taken.” Having noted in postmedieval Celtic folklore examples of those ‘taken’ by the fairies (something we shall explore more fully in Chapter 5), she expresses surprise at coming across this motif in a thirteenth-century Belgian writer, Thomas of Cantimpré, and concludes, quite unnecessarily, that “we must have tales of Celtic origin with which, in ways too fortuitous to be discovered, [a Flemish narrator has] become acquainted, and to which [he has] given a local coloring.”25 Thomas spent much of his early life near Cambrai, some fifty miles northwest of the Ardennes forest, an area with a strong fairy tradition: Partonopeu de Blois is set in the Ardennes, and so too is the episode in Reinbrun in which the hero rescues his father’s friend Amis from fairy captivity. Judging by his name, Jean d’Arras, the author of the fairy romance Mélusine, came from a town near Cambrai, and this Jean seems also to have been one of the collectors of old wives’ tales that make up Les Évangiles des quenouilles. There is no reason to suppose that the fairy lore appearing in the Évangiles, or in any of these romances, is anything other than homegrown or that when Thomas of Cantimpré reports stories of neighbors taken by the fairies he is merely giving “local coloring” to imported Celtic material. The single most informative source for medieval fairy beliefs, cited many times in this book, is William of Auvergne’s De Universo, and it is quite clear that this Parisian scholar draws heavily on the traditions of south-central France, where he was brought up. One danger of an overconcentration on the Celtic connection is that fairy allusions from other areas tend to be missed; for instance, the Cambridge lyric “Heriger, Bishop of Mainz,” from early thirteenth-century Germany, is about a fortune-teller who attends a mysterious feast deep in the woods, and yet scholars have rarely noted its obvious fairy associations.26 My point is not, of course, that Celtic traditions were unimportant in this respect (we shall be using the Breton forest of Brocéliande as a test case in Chapter 1) but simply that they were not the fons et origo of all medieval fairy lore.27 The fairies with whom I am concerned in this book are pan-European, and the questions they raise should not be quarantined to the margins, either geographical or cultural, of medieval society.
A further disclaimer concerns the role of folklore. I have not been professionally trained as a folklorist, nor can I lay claim to any special proficiency in this area. In one regard, like, I suspect, many of my medievalist colleagues, I am wary of the use of customs and beliefs recorded in more recent times to throw light on medieval practices. Folklore, as Antonio Gramsci was quick to point out, is far from static, and the notion of a popular culture so deeply conservative that it is possible to treat any given nineteenth-century custom not only as a potential medieval relic but also as evidence for actual medieval practice seems to me highly dubious. On the other hand, the modern folklorist does have one enormous advantage over any medievalist who sets out to construe earlier popular culture: she can question her informants and attempt to expose what it is they think they are doing and why they are doing it. While projecting back the results of such investigations to the Middle Ages can never constitute proof, it does offer us a valuable analogical tool. Valdimar Hafstein, for instance, compares a thousand-year-old vision in the Þáttr Þiðranda ok Þórhalls of “many a hill … opening, and every living thing, both small and large, … packing its bags and moving” in the face of impending Christianization to stories of elves being displaced by new roads and housing developments in late twentieth-century Iceland, and concludes that “urbanization is as anathema to modern day elves as Christianity was to their pagan forebears.”28 Looking at the present through the lens of the past, Hafstein sees elf belief as deeply conservative—rooted in nostalgia for an imagined authentic Iceland threatened by modernization. However, if we reverse the polarities, if we consider medieval beliefs in the light of the modern experience, a further dynamic emerges—one that has more the look of spontaneous resistance than of nostalgic resignation; after all, even in present-day Iceland fear of offending the elves can cause roads to be diverted and housing developments to be relocated.
This view of the political significance of folkloric beliefs is one propounded by the Marxist Antonio Gramsci,29 whose views have admittedly not always been welcomed by folklorists.30 Writing in Mussolini’s Italy, Gramsci evidently regarded whatever challenge contemporary folklore was able to offer fascism as fragmented and incoherent compared with “the philosophy of praxis” (that is, the version of Marxism that he himself espoused), but at the same time he remained an astute observer of popular resistance to the dominant culture particularly as a historical phenomenon, as his account of the nineteenth-century Tuscan preacher Davide Lazzaretti demonstrates.31 As Kate Crehan writes, “while Gramsci could be harsh on the blinkered parochialism of subaltern culture, at the same time he was fascinated by it and believed that much could be learnt from it” (p. 119). My reading of the subversive role of fairy beliefs in the medieval polity owes much to his insights.
Concentration on the political significance of fairyland means that this book makes no claim to provide a comprehensive survey of all fairy phenomena in the Middle Ages. It treats here only in passing, if at all, many of the activities that were commonly associated with fairies: their modification of the weather; their association with great wealth; the trouble they might cause benighted travelers; their ability to induce or ward off sickness; their influence for good or ill on harvests; their skill at prognostication—to name but a few. Diane Purkiss has stressed the way fairies preside at and govern “the big crises of mortal life: birth, childhood and its transitions, adolescence, sexual awakening, pregnancy and childbirth, old age, death … the borders of our lives, the seams between one phase of life and another.”32 Of course in the Middle Ages a very different kind of institution, the Christian church, claimed to have jurisdiction over these areas too, and the last three chapters will examine the consequences of its attempts to discipline rival folkloric beliefs. Not that all ideas about fairyland were heterodox—the church had no argument with those prepared to accept that fairies were demons—but wherever it felt obliged to exercise authority it exposed a further aspect of the operation of what R. I. Moore has characterized as a persecuting society.33 Of course clerical regulation of those who believed that fairies were non-demons was of a different order from its attempts to discipline Cathars, Jews, lepers, and homosexuals. Not only were fairy beliefs so ubiquitous that they could not be quarantined in ghettos and leprosaria or made the targets of self-serving crusades, but they also touched (as we shall see) the higher levels of secular society (and even penetrated the church itself), so that focused persecution was infeasible. However, as Foucault has taught us, social regulation can take different forms, and there are clear signs that many of those who participated in the discourse of fairyland in the Middle Ages felt themselves under surveillance. Moreover, from the fifteenth century onward, as education began to close the cultural gap between clerical and secular authorities, the control of vernacular belief became more and more exacting, culminating in the terrible witch hunts of the early modern period.
Accordingly the last three chapters of this book will examine the church’s attempts to regulate fairyland in three critical domains. Chapter 3 follows its campaign to marginalize popular attitudes to copulation, pregnancy, and childbirth and in particular its demonization of one especially prominent fairy lover, Merlin’s father. In Chapter 4 we see how a motif popularly associated with child rearing, that of the fairy changeling, disturbed the sensibilities of both churchmen and patresfamilias, and how it resisted their attempts to suppress it; here we focus particularly on the representation of the changeling in the mystery