When Witches Take Flight

This article is published in partnership with Sacred Writes: Public Scholarship on Religion, which provides support, resources, and networks for scholars of religion committed to translating the significance of their research to a broader audience. Sacred Writes is funded by the Henry Luce Foundation and hosted by Northeastern University.

Witchcraft and flying are inextricably bound together in the popular imagination. From Harry Potter to Kiki’s Delivery Service to The Witch, the hallmark of the witch is flight, whether raised alone into the air or by broomstick or other contrivance. The humor of movies like Hocus Pocus, where Mary flies on a vacuum cleaner instead of a broomstick, is predicated on a shared cultural understanding of what a witch’s flight looks like. We have an expectation of what cleaning implements witches fly on, and vacuums are not included. We also expect them to fly in groups: this is the famous “coven,” often seen flying across the moon’s silhouette on their way to their black sabbath (or to hunt down wisecracking children in Salem).

The Witches by Hans Baldung, woodcut, 1510.

Many of these stereotypes date from the Renaissance, when artists like Albrecht Dürer and Hans Baldung put into vivid image their understanding of what made a witch. These understandings were based on centuries of debate about witches — what constituted a witch, the kind of activities they engaged in, and even whether they were real at all. Central to these conversations was a text known as the Canon Episcopi, a short description of exactly what witches did during their nocturnal flights. The Canon Episcopi appeared frequently in medieval canon collections (collections of religious law derived from various church councils), though the original dated from the tenth century.¹ The Canon represented an innovation in witchcraft beliefs: while witches had flown since antiquity, they had done so while transformed into owls or other flying creatures. Some witch-figures, like the serpentine lamia, were supernatural creatures or monstrous hybrids in the first place.² The idea of the witch as a human woman who could propel herself through the sky, by her own power rather than by transforming or growing wings, was a creation of the Middle Ages.

I first started thinking about early medieval witchcraft in the spring of 2019, while I was supposed to be working on my dissertation (which was not about witchcraft). Procrastinating on what I should have been doing (and about six months out from an ADHD diagnosis), I fell down a wormhole of hyperfocus. For a month, I read nothing but witchcraft. It’s easy to detour like this in early medieval history — relatively few primary sources means that many have been digitized, allowing me to read ninth-century Latin from my couch with my dog on my lap. Not having to travel to an archive made distraction enticingly easy.

The text of the Canon Episcopi in a manuscript of Burchard of Worms.

I was quickly surprised to find that flight — a defining feature of the modern image of the witch — figured relatively little in the earliest medieval sources. The Canon Episcopi, which most scholars cite as the beginning of the group flight motif,³ was only a single entry in a canon collection completed in 906. The text does not seem to have created much of a stir at the time; the next source which mentions (and expands upon) the Canon came more than a century afterward.⁴ This in itself is not odd for early medieval evidence, where spotty and recalcitrant are the rule rather than the exception, but it did seem strange that such an important motif had its origins in such a forgotten corner of the tenth century. I resolved to explore further.

The compiler of the 906 canon collection was an abbot from the Rhineland named Regino of Prüm. Regino had had an illustrious political career as the abbot of a wealthy Carolingian abbey, but had seen his fortunes decline after a change in leadership among the elite in the kingdom of Lotharingia. Whether Regino authored the Canon Episcopi or merely copied it remains a live issue — his collection seems to attribute it to the Synod of Ancyra in 314, but no such canon was promulgated there. Scholars have debated whether Regino got the Canon somewhere else, then assigned it to the prestigious Ancyrene synod (something he was known to do) or simply made it up entirely (something he was also known to do).⁵ For my own part, I think Regino probably composed the text himself, or changed a previous text so significantly to have functionally composed it — the obsessions and anxieties revealed by the text accord remarkably well with Regino’s biography, in particular indicating a deep distrust for women in political power.  

We are fortunate here to have another of Regino’s works, a history of the Carolingian empire called the Chronicon. In the Chronicon, Regino blames the downfall of the Carolingians on sexual sin and female ambition, specifically Lothar II’s attempt to divorce his queen and marry his mistress Waldrada. Indeed, Regino’s career had developed alongside a stark reminder of this sin: Lothar and Waldrada’s son Hugh, who attempted to claim the kingship of Lotharingia only to be blinded and sent to Prüm, where Regino himself tonsured the failed rebel.⁶ The Chronicon also reveals a few of Regino’s personal political grudges: by the time he wrote both the Chronicon and his canon collection, Regino had actually fallen from grace, ousted from the abbacy of the wealthy monastery of Prüm. This was part of a political changeover involving the influence of the emperor Arnulf’s widow Uota – another politically ambitious woman.⁷ All of this biographical context should be kept in mind when we read Region’s description of sinister, powerful women in league with the devil.

Lotharingia in the tenth century, with Trier (where Regino compiled his canon collection) marked. (Made by Helios2019 on Wikimedia under a CC BY-SA 4.0 license.)

The weird thing about the Canon is that it is actually a very uncomfortable candidate as the source of the group flight motif. The text describes “certain women” who believe that they ride through the land at night on the back of strange beasts, following the Roman goddess Diana. The core focus of the text is this belief and denying its reality — Regino says that these women had been deluded by the devil, and assigns penance to those who say that such things can happen. Significantly, the text neither specifies that these women are witches, nor that they in any way fly. These innovations would come with a second version of the text, produced more than a century afterwards in Burchard of Worms’ Decretum. In its original form, however, neither detail is present.⁸

Diana, Simon Vouet, 1637. Diana’s role as virgin goddess of the hunt made her especially transgressive to medieval churchmen, to whom she represented unconstrained — and dangerous — femininity.

So why have historians tied Regino’s text so strongly to the witch’s flight? The answer lies in attempts to understand the religio-magical beliefs described by Regino. For nearly a century, scholars have argued that this and related texts represent an attestation of a hidden substrate of European religion, secret folk beliefs practiced by rural populations outside the gaze of watchful churchmen. These arguments have taken a thousand different forms. The Egyptologist Margaret Murray famously argued that the Canon and similar texts testified to a secret witch cult which existed throughout medieval Europe, a continuation of pre-Christian beliefs stretching even into the early modern period and its famous witch trials. (Few scholars believe this anymore.)⁹ In the 1980s, the Italian scholar Carlo Ginzburg connected Regino’s text to the Benendanti folk religion he uncovered in records from 16th century Friuli.¹⁰ Historians, anthropologists, and folklorists have connected the Canon to everything from the supposed “wild hunt” of German folklore, to the “good women” (spirits who wandered the land at night), to European shamanism, the practice of astral projection, and even the belief in werewolves. Modern feminist histories of witchcraft take Regino and similar texts as evidence that medieval women were able to escape the patriarchal bounds of Christianity, cultivating their own religious practices in secret.¹¹

The Wild Hunt, as depicted in Asgårdsreien, 1872, by Peter Nicolai Arbo.

I don’t mean to disparage any of these theories, most of which are well thought out and reasonable interpretations of the belief recorded by Regino. To my mind, however, they all share a single flaw: they assume that the early medieval abbot, buttoned-up and politically-connected Regino, knew anything at all about the marginal religious practices of medieval women. It is a lamentable casualty of the early medieval sourcebase that we often have to listen to men’s voices to understand the lives of women — while there are good sources recorded by women (Dhuoda’s Liber Manualis, the works of Hrotsvitha, and the Annals of Quedlinburg, to name a few)¹², they pale in comparison to the number attributed to male authors. This fact presents some sticky problems when it comes to witchcraft beliefs: powerful men are famously bad at knowing (or caring about) what women are interested in, especially those outside their social circle. Would Regino, a powerful man who spent nearly his entire life in a monastery, literally cloistered away from the opposite sex, have been any different?

There are counter-arguments to this proposition: what if Regino truly was copying an earlier text recorded by someone with real knowledge? What if Regino had gotten this information from parish priests, those tasked with cultivating the religion in more remote areas of the Carolingian empire? These are all possibilities, and I do not expect to win everyone over with this argument. But I do think that historians need to think a little bit more about taking Regino’s word for it on rural women’s religious beliefs. The text tells us a lot more about Regino’s psyche, and who he blamed for his stalled political career, than it does about transgressive religious belief among rural women.

1. Martine Ostorero, Le Diable au Sabbat (Florence: Sismel Edizioni del Galluzzo, 2011), 571–96. For the tenth-century text, see Regino of Prüm, Libri duo de Synodalibus Causis et Disciplinis Ecclesiasticis (906), 2:371, reprinted in Wilfried Hartmann, ed., Das Sendhandbuch des Regino von Prüm, Wilfried Hartmann (Darmstadt: Wissenschaftliche Buchgesellschaft, 2004), 420. For an English translation, see Chris Halsted, “‘They Ride on the Backs of Certain Beasts’: The Night Rides, the Canon Episcopi, and Regino of Prüm’s Historical Method,” Magic, Ritual, and Witchcraft 15.3 (2021): 362–64.
2. Halsted, “‘They Ride on the Backs of Certain Beasts,’” 361–85.
3. Werner Tschacher,“Der Flug durch die Luft zwischen Illusionstheorie und Realitätsbeweis: Studien zum sogenannten Kanon Episcopi und zum Hexenflug,” Zeitschrift der Savigny-Stiftung für Rechtsgeschicte: Kanonistiche Abteilung 85 (1999): 225–76, esp. 226.
4. Burchard of Worms, Decretorum libri viginti (c. 1023), 19.70, 90, 170–71, reprinted in Hermann Joseph Schmitz, ed., Die Bussbücher und das kanonische Bussverfahren: Nach Handschriftlichen Quellen (Düsseldorf: Druck und Verlag von L. Schwann, 1898), 425, 427, 446–47.
5. See, e.g., Jeffrey Burton Russell, Witchcraft in the Middle Ages (Ithaca: Cornell University Press, 1972), 76; Jean Pierre-Poly, Le chemin des amours barbares: Genèse médiévale de la sexualité européenne (Paris: Perrin, 2003), 164–65; and Werner Tschacher, “Flug durch die Luft,” 242–43.
6. Halsted, “‘They Ride on the Backs of Certain Beasts,’” 377–84.
7. Simon MacLean, “Insinuation, Censorship and the Struggle for Late Carolingian Lotharingia in Regino of Prüm’s Chronicle,” English Historical Review 124.506 (2009): 1–28.
8. Regino of Prüm, Libri duo de Synodalibus Causis, 2:371, reprinted in Hartmann, ed., Sendhandbuch des Regino von Prüm, 420, translated in Halsted, “‘They Ride on the Backs of Certain Beasts,’” 362–64; Burchard of Worms, Decretorum libri viginti, 19.70, 90, 170–71, reprinted in Schmitz, ed., Bussbücher und das kanonische Bussverfahren, 425, 427, 446–47.
9. Margaret Alice Murray, The Witch-Cult in Western Europe: A Study in Anthropology (Oxford: Clarendon Press, 1921), 22, 102; Jacqueline Simpson, “Margaret Murray: Who Believed Her, and Why?,” Folklore 105 (1994): 89–96.
10. Carlo Ginzburg, The Night Battles: Witchcraft and Agrarian Cults in the Sixteenth and Seventeenth Centuries, trans. John and Anne Tedeschi (Baltimore: Johns Hopkins University Press, 1983).
11. Emma Wilby, “Burchard’s strigae, the Witches’ Sabbath, and Shamanistic Cannibalism in Early Modern Europe,” Magic, Ritual, and Witchcraft 8.1 (2013): 18–49; Claude Lecouteux, Phantom Armies of the Night: The Wild Hunt and the Ghostly Processions of the Undead, trans. Jon Graham (Rochester, NY: Inner Traditions, 2011); Raven Grimassi, Old World Witchcraft: Ancient Ways for Modern Days (San Francisco: Weiser Books, 2011), 35.
12. Handbook for William: A Carolingian Woman’s Counsel for Her Son, trans. Carol Neel (Lincoln: University of Nebraska Press, 1991); Phyllis Brown and Stephen Wales, eds., Companion to Hrotsvit of Gandersheim (fl. 960): Contextual and Interpretive Approaches (Leiden: Brill, 2013); Martina Giese, ed., Die Annales Quedlinburgenses, in Monument Germaniae Historica, Scriptores Rerum Germanicarum In Usum Scholarum Separatim Editi 72 (Hanover: Hahnsche Buchhandlung, 2004).