Introduction: Magic at the Crossroads of Ritual and Cosmology

1

This article focusses on the history of learned magic in late medieval Europe, breaking a period of about 500 years into chronological stages to explore how medieval supporters and critics of magic represented the art and responded to each other’s arguments, then reframed their own in a continuous dynamic entanglement. The engagement with the ideas and practices of learned magic on both sides of the argument can be understood as a conversation between members of a single social and administrative order (notably clerics, monks and physicians) who were reading the same books and pondering the same problems (Fanger 2019). The particular problems that I focus on here are cosmology, the meaning of rituals, and what was acceptable for a good Christian to do. In late medieval Europe, learned magic texts from diverse and syncretic cosmological traditions (primarily Christian, Jewish, Arabic and Greco-Roman) circulated among people familiar with, and emotionally invested in, a great variety of institutional and informal rituals. Learned magic was closely related to more mainstream rituals and cosmologies but also deviated from them. Like other medieval rituals, those of learned magic consisted of ordered and repeatable actions that reflected pre-existing beliefs about the nature of the cosmos and expressed a desire to order people’s relations with the sacred, most importantly with God, angels and demons. Unlike institutional rituals, however, learned magic rituals were not underpinned by any requirements to communicate correct doctrine. This meant that their cosmological frameworks could be syncretic, allusive and pragmatic rather than didactic. In late medieval Europe, learned magic texts possessed a further important quality: in the fluid, creative and unstable context of manuscript culture, occult items acquired the qualities of playfulness and a continuously reworked craftsmanship. Every time a scribe copied a recipe, experiment or complex ritual into a manuscript they had the opportunity to alter, adapt or add to it according to their personal interests, cosmological certainties, access to materials or anxieties about orthodoxy.¹ This means, as we shall see, that scribes as well as authors and practitioners were engaged in the dynamic entanglement of pro- and anti-magic conversations.

The Reception of Learned Magic in the Age of Translation

2

From the late eleventh century, scholars from across Europe travelled to Spain, Sicily and the Middle East to translate works of science, philosophy and the occult arts into Latin. The ensuing extraordinary influx into Europe of magic texts from the Greek, Arabic and Jewish traditions transformed the status of magic, as enthusiastic translators made the first serious attempts in the Latin West to classify magic as a branch of knowledge within mainstream philosophy and science. Translators like Dominicus Gundisalvi, Daniel of Morley and Michael Scot engaged sympathetically with Arabic texts that presented magic as the culmination of human knowledge, the magus as a perfect philosopher and astrology as a framework for talisman-making (Burnett 2019). More than seventy learned magic texts of Arabic origin (that is, translated or adapted from Arabic magic texts) have been identified as circulating in late Medieval Western Europe, primarily falling into the category of astral magic rituals that included invocations to the celestial spirits or the inscriptions of their names on talismans and other objects. This genre had significant influence on the contours of medieval learned magic (Boudet 2020). Another influential category of magic, Solomonic magic (referring to magic texts and experiments attributed to the biblical King Solomon) had its origins in Jewish, Arabic and Greek magical traditions but was adapted to Christian sensibilities. Circulating from the twelfth century alongside astral magic, this loose grouping of texts and experiments had characteristic visual instruments, including pentagons, rings and seals (sigillae) (Véronèse 2019).

3

Some of the cosmological ideas and ritual actions that were characteristic of learned magic texts with Greco-Roman, Arabic and Jewish origins aligned with mainstream science and theology: notions of celestial influences on the Earth, occult properties in natural objects and the power of God’s name were generally not controversial. Others, such as invocations to planetary spirits, animal sacrifices, experiments to generate new life forms and rituals to draw down spirits into talismans (a continuation of the late antique art of vivifying statues) were more discordant with Christian sensibilities and thus provoked scrutiny and suspicion. In light of the Christian translators’ exposure to this cosmological and ritual diversity, it is unsurprising that a distinction between magic that drew on natural powers and magic that required the invocation of evil spirits is apparent from the earliest period of translations (Burnett 2019; Weill-Parot 2002a). The processes of acculturation to a Christian audience involved omissions and adaptations, personal predilections and habits. Translators focussed on learned magic texts with few signs of Islamification and sometimes omitted details of myth and cosmology, retaining only the practical instructions.² A well-known example of ritual omission is the translation of an image magic text, Thabit’s De imaginibus (On images), by John of Seville and Limia (fl. 1133–53) without the prayers to spirits. A more gradual process of acculturation is witnessed in different versions of the Solomonic magic text, the Liber Almandal, in which the jinns and shayātīn of the Islamic tradition gradually become more and more like Christian angels (Véronèse 2012, 5–66). Christian sexual morality could also influence translating practice. Experiments promoting same-sex love were dropped when the Arabic compendium of magic, the Ghāyat al-ḥakīm (955–59), was translated into Castilian in 1256 and thence into Latin under the title Picatrix (Boudet 2011). Although the translator of the Picatrix worked in an intelligent and conscientious manner to make sense of the Arabic text, they also consistently cut or adapted Islamic religious references to the Qurʾān and the ḥadīth (Burnett 2017; Boudet and Coulon 2017).

4

As learned magic texts disseminated outwards from the communities of translators, resistance to the idea of magic as a true branch of science was expressed by writers from religious orders, who were used to thinking of knowledge as having a spiritual source as well as being found in books. Herrad of Landsberg (1130–95), Abbess of the Hohenburg Abbey in Alsace, included magic in a diagram of philosophy and the seven Liberal Arts in her illustrated encyclopedia, the Hortus deliciarum (Garden of Delights). But she excluded magicians (and poets) from the true philosophy that comes from God. Magicians are depicted being guided by evil spirits who perch behind their heads and whisper in mockery of the Holy Spirit. Similarly, the Abbess, visionary and author Hildegard of Bingen acknowledged that magic was a kind of philosophy but argued that it engendered sin, hatred and tyranny. In her Liber vitae meritorum of 1158 she personified Magic (maleficium) as a monstrous hybrid creature with the body of a dog, the head of a wolf and the tail of a lion (Carlevaris 1995, 222-3). For Hildegard, the disciples of Magic, Hermes and other philosophers, were wise men (sapientes) who through their investigations of the cosmos had learned how to harness the elements, but she thought that they had acquired their knowledge from evil spirits. This twelfth-century critique of magic prefigures the later trope of demonic teachers of witches in fifteenth-century witchcraft mythologies.

5

A final strand of the twelfth-century reception of magic is the knowledge and influence of classical literature. It is important because it introduced the idea of extraordinary natural skill in performing magic, evidenced in classical workers of magic such as Circe, Medea and Erichtho. Twelfth-century romances influenced by classical sources (romans d’antiquité) described female practitioners of magic (malefica) who drew on natural sources of power, producing extraordinary effects in nature and specialising in disturbing the minds of men. For example, Medea from Benoît de Sainte-Maure’s Le Roman de Troie (c. 1155-60) can perform weather magic, transform into a bird and subvert the flow of water. The sorceress (sorciere) who advises Dido in the Roman d’Enéas, composed in Normandy c. 1160, can revive the dead, predict the future, cause celestial spectacles, and make birds speak, trees walk and water flow uphill. In this period, classical literature evoked the wonders of magical powers, but by the fourteenth century ancient narratives of witches’ activities were used to support theories about the power of the devil and his female co-workers (Montesano 2018).

Critics Respond: Addressative Magic, Idolatry and Delusion

6

A final flourishing of the translation of occult texts from Arabic and Hebrew into Latin and Castilian took place at the court of King Alfonso X of Castile (r.1252-84). Five learned magic texts translated or composed under Alfonso’s patronage—the Lapidario, the Picatrix, the Libro de las formas et las ymagenes, the Libro de astromagia and the Liber Razielis—reveal this ruler’s interest in angelology, astral magic and occult forces in nature. The translating activity at Alfonso’s court also testifies to the attraction learned magic held for those interested in using the occult arts to increase their power, gain political favour or annihilate their enemies, goals richly manifested in the Picatrix. Powerful secular rulers did not usually need to conceal their occult interests—texts to summon demons have been found in royal and aristocratic inventories such as that of Francesco Gonzaga, the Marquis of Mantua—but when magical activities flourished in medieval courts they were invariably accompanied by serious hazards for those from lesser ranks who found themselves caught up in accusations of magical practice (Boudet 2019; Boudet, Ostorero, and Bagliani 2017).

7

In the second half of the thirteenth century, learned magic came under new scrutiny in the writings of the Bishop of Paris, William of Auvergne, the theologian Thomas Aquinas and the author of the Speculum astronomiae, usually assumed to be Albertus Magnus. Three critical approaches to learned magic dominate this period. The first was that ritual instruments, in particular the prayers, invocations, inscriptions and graphic motifs that were distant from the Christian liturgy and iconography, were assumed to be addressed by the human practitioner to another Intelligence (namely demons) (Weill-Parot 2002a). Secondly, the graphic motifs and talismans of learned magic were condemned as objects of idolatrous worship (William of Auvergne [1674] 1963, bk. 2, pt 3, ch 22; Zambelli 1992, 240–7; García Avilés 2019).³ The accusation of idolatry was a recognition of the rich visual nature of learned magic, an expression of unease with the claims of some texts to teach how to vivify images or trap demons inside them, and a defensive response in an age that generally lacked the language or will to censor images (Schmitt 2010). It was also an inevitable critique, given the predominantly pagan origins of Arabic astral magic (Pingree 1980, 1992).

8

The third approach was the sensitivity of some critics to the similarity between the powers that Biblical and hagiographical texts claimed for Christ and Christian saints and the powers that learned magic asserted would be achieved through correct ritual performance; for example to walk on water, to acquire celestial knowledge or to ascend into the company of angels.⁴ Critics rejected the idea that individual practitioners were worthy of such feats on the grounds that these were only given to holy saints by the grace of God. According to this line of thinking, practitioners of magic were therefore either self-deceiving about its efficacy or cynically using their knowledge of natural properties, illusionist magic and rhetorical misdirection to deceive others. Self-delusion is discussed below, and scepticism about the powers of magical practitioners had a continuous if marginal place in medieval thought. The most extensive critique along these lines was by the fourteenth-century French bishop and philosopher Nicole Oresme in his Tractatus de configurationibus qualitatibus et motuum (Treatise on the Configurations of Qualities and Motions) of ca. 1350. This treatise describes how magicians used psychoactive substances, powerful odours, music, dramatic shifts in light and darkness, and tricks with mirrors to create terrifying and discombobulating illusions that their audience mistakenly interpreted as supernatural manifestations (Oresme 1968, 337–61).

9

Nevertheless, some scholars’ classification of learned magic as beneficial knowledge continued to be influential throughout the late Middle Ages. Often when medieval critics categorised learned magic texts as licit or illicit (or some variations thereof), they identified within those texts useful and acceptable practices in contrast to those that were more transgressive. A well-known example is the construction, by the author of the Speculum astronomiae (ca. 1260) and others, of the licit category of ‘astronomical images,’ talismans made with an infusion of natural celestial powers rather than with the aid of spirits (Zambelli 1992, 240–47; Weill-Parot 2002b). Practitioners of magic condemned some practices in order to assert the orthodoxy of others, even if they themselves built on the condemned rituals to create new texts. For example, the Benedictine John of Morigny admitted to experimenting with necromantic rituals and the Ars notoria, though ultimately he rejected their techniques and, in the Liber florum celestis doctrine (ca. 1301–10), presented new ritual approaches to acquiring knowledge and salvation (John of Morigny 2015; Fanger 2015, 116).

The Acculturation of Magic to Christian Ritual Forms and Sensibilities

10

In the twelfth century, texts of Solomonic magic in circulation included works of Arabic, Jewish and Christian origin in which the liturgical or paraliturgical element dominated. That is, their rituals used the language (sometimes directly borrowed from the official liturgy) of praise, supplication and devotion to God. This was never a clearly defined corpus, however, and it evolved considerably between the thirteenth and fifteenth centuries to influence and overlap with other categories of learned magic. Solomonic magic was attractive to readers curious about spiritual experiences and interested in transgressive risk taking (and it still is, as Otto’s chapter on the contemporary practitioner Frater Acher demonstrates, 2023). It focussed, above all, on the ‘domestication of spirits,’ compelling demons with adaptations of mainstream exorcism rituals and language or conversing with angels, with the practitioner even assuming the role of a saint or prophet (Véronèse 2019; Véronèse and Chave-Mahir 2015; Page 2011).

11

The Christian authors, compilers and translators of learned magic texts constructed ritual sequences that had plausible efficacy within the medieval worldview and enough markers of orthodoxy to persuade the reader that a ‘good Christian’ could perform them. Genres that modern scholars now categorise as ‘angel magic’ and ‘necromancy’ took different routes to these goals. Angel magic involved practitioners undergoing ablutions and performing rituals to persuade angels to help them achieve various pious goals, notably the acquisition of knowledge, an increased likelihood of salvation or a vision of God (Fanger 2015). Necromancy was the practice of safely conjuring and gaining control over demons to compel them to perform tasks, such as revealing buried treasure or bringing a desired partner into the practitioner’s presence (Kieckhefer 1998; Boudet 2007). Texts belonging to both these genres involved wholesale adaptation of mainstream Christian practices (such as fasting, meditation and prayer), texts (liturgical prayers, exorcism rituals and sacramental language) and goals (salvation, the beatific vision and exercising control over demons). There were also more subtle Christian sensibilities embedded in the texts: respect for the efficacy of ritual actions and objects, awe of God’s power and the purity of angels, a longing for spiritual experience and fear of the malignity of demons. Many of the authors, compilers and practitioners of angel magic and necromancy were priests and monks with good command of the liturgy, experience of performing the sacraments and knowledge of exorcism rituals. The texts thus related closely to their lived experience and spiritual aspirations.

12

The appeal of Christian authored angelic magic texts like the Ars notoria was that they promised the soul’s salvation as well as comprehensive knowledge, desirable attributes (memory, eloquence and understanding) and ritual shortcuts to the University Arts curriculum and the vision of God, achieved by ascending to the level of the angels. The varied practitioners of this genre—students, physicians and monks—were probably attracted by different goals within it. For practitioners dissatisfied with mainstream routes to the numinous, angelic magic offered a relationship with angels that went beyond simply requesting their aid (like the popular genre of prayers to guardian angels), to friendship, companionship and even a joining of their blessed ranks. A good example is the Liber Razielis, a compilation of seven texts and nine appendices of ritual magic collected and translated under the patronage of Alfonso X, which circulated in different Latin and vernacular versions (García Avilés 1997; Page 2012). Each new owner of the Liber Razielis performed a ritual to acquire an angelic instructor, re-enacting the original mythic encounter between Adam and the angel Razielis, who gave him a book of magic and taught him how to use it. As the practitioner operated with the book he became like one of the prophets or angels of heaven and gained knowledge and understanding of all things.⁵ Unusually for a medieval learned magic text, this prophetic status was supposed to entail a permanent rather than temporary transformation, and it came with responsibilities to disseminate the message of God on earth.

13

Aspiring to be like the angels or to participate in the divine was part of medieval monastic ideology and was reflected in visionary literature and narratives of saints conversing with angels. It meant acting out in this life the pure qualities that demonstrated that your soul would be saved. Angelic magic texts usually offered not only temporary conversations with spirits, ascents to the angelic orders, or divinity on a par with God, but also more permanent transformations of the soul to increase the likelihood of the practitioner’s salvation (Page 2012; Gehr 2019, 243). The routes to salvation in learned magic had various levels of difficulty: the full complement of ascetic preparations and rituals required of the practitioner of the Ars notoria took at least three years to perform, but on a more modest scale, the Almandel claimed that angels of the first altitude will render a man perfect after he has spoken with them only once, so that ‘he will not need to fear eternal damnation.’⁶ And if you wanted to save the soul of a loved one, you could turn to the version of the Liber iuratus in the Catalan philosopher Beringarius Ganellus’ compendium of magical texts (ca. 1346). Its Seal of God (Sigillum Dei) is said to redeem a soul from purgatory (Summa sacre magice IV.1.5 and IV.I.6, Gehr 2019, 237–53). Finally, and most efficiently, the Liber vaccae, a ninth-century Arabic learned magic text translated into Latin in the twelfth century, includes a suffumigation with the power to make a soul good.⁷ Not all scribes were comfortable with this goal, however; surviving versions of the Liber iuratus limit the theurgical goals of the seal to a vision of God, and one of the copies of the Liber vaccae omits this suffumigation.⁸

14

Christian authors composed new learned magic texts throughout this period, acculturating it to Christian sensibilities. But this acculturation did not always imply increased orthodoxy; some scribes rewrote their own versions of learned magic texts that incorporated more, not less, occult material. For example, the insertion of material from the Ars notoria, Liber almandal and Kyranides into a lapidary compilation dating to the second half of the thirteenth century may have ‘disguised’ ritual magic within a more licit genre, but it also gave a work focussed on natural occult powers a richer and more complex set of rituals to achieve its goals (Regan 2018). Similarly, one of the surviving fifteenth-century versions of the Almandal includes insertions from other learned magic texts as well as mainstream Christian liturgy, the result of a scribe adapting the ritual to his own interests and cultural context (Véronèse 2012).

15

Christian necromantic experiments could be very transgressive in their cosmological outlook, ritual instructions and liturgical language. For example, an experiment to catch a thief in a fifteenth-century necromantic manual uses a wax lamina with the name ‘Satan’ in the central part of the figure, a place usually reserved for the name of God or the practitioner.⁹ The practitioner is told that the best time to summon demons is after he has been to Church and heard mass, but that he must act quickly because the ritual will fail if the thief confesses his crime, gives the proceeds of his crime to the poor or priests, or acts in any way for the love of God and the health of his soul. In this experiment, therefore, the performative piety of hearing mass increases the success of a ritual but the sincere piety of the target of the experiment can sabotage it.

16

Arabic astral magic influenced the timing of Christian necromantic operations, the variety of powers the practitioner called upon and techniques of image-making. We can see their influence particularly vividly in rituals for love. The Latin Picatrix has twenty-nine different rituals for love (amor), which emphasise the role of the planet and spirits of Venus. In addition to love between friends and the love of powerful lords, these rituals variously address the desire for long lasting, peaceful, affectionate, robust, marital, virtuous, benevolent, united, faithful and happy love. The type of experiment for love that was particularly influential on Christian necromancy involved making an image of the object of desire, inscribing it with spirit names and manipulating it, for example by inserting needles into it.¹⁰ Spirits were invoked and commanded to infuse the image with life to create a sympathetic connection between it and the real object of desire. The inscribing, piercing, igniting or burying of the image that followed was intended to cause restlessness and suffering in the human target, followed by burning love and desire for the operator.

17

Christian authors of necromantic experiments had to adapt astral magic rituals to a world view in which the spirits they invoked were powerful and malefic demons. They reinforced the practitioner’s control by adding commanding language, sacrifices as negotiating actions and protective rituals such as drawing magic circles. Moreover, perhaps because of the greater agency and materiality assigned to demons within the Christian cosmos, rather than spirits being asked to compel a woman to come to the practitioner, the demons were instructed to physically transport her to him. Christian necromantic experiments for love were also influenced by the medical tradition of pathologizing love sickness as a violent and intense illness and by conventions of courtly love, which held that deep and noble love was accompanied by suffering. Influenced by courtly love, Christian experiments for love magic often assume a particular beloved rather than claiming to work for all or any man or woman. The ritual may even fail if the wrong woman is accidentally targeted or if the woman’s emotion is inflamed but the relationship not consummated. For example, the user of a figure for love that should be inscribed on parchment with bat’s blood on the first, fifth or fifteenth day of the Moon is warned that if a woman who is not the one his desire cleaves to (ei quam haere cupis) touches the magical figure she will die or become insane.¹¹

18

An experiment for love sickness (amor hereos), that was copied, censored, and reconcealed in a fifteenth-century Latin and Middle Dutch miscellany of occult and practical items (Wellcome MS 517) requests the help of Venus, God and the demon Astaroth (Page 2019b). The operator draws on a tile a picture of the woman he wants to attract, burns it to inflame her with love and then invokes God. In a formulation that would be morally complex in any context outside magic—where the point is to call as forcefully as possible on all the powers that might help you—the operator invokes God ‘to bind the demons who bind lovers.’¹² The operator then commands Astaroth and other demons specifically focussed on love to deliver the desired man or woman afflicted by an ardour so intense as to allow no rest, whether asleep or awake, standing or sitting, drinking or eating. Medieval love magic rituals were intended to provoke the physical symptoms of love sickness—dizziness, sighing, insomnia—and use powerful, sometimes grim similes: you should love me, one ritual commands, as a deer yearns for a fountain of water, as the raven desires corpses and as the candle melts.¹³ Another provocative fifteenth-century experiment commands the devil to infuse a mirror with his power and to seduce the woman who looks into it so that she burns with love for the operator like the Blessed Virgin burns with love for God.¹⁴ It was common in medieval witchcraft literature to explain such transgressive language and ritual actions by arguing that men and women suffering from extremes of emotion were particularly vulnerable to temptation. Demons took advantage of their desperate passion and promised to help in their pursuits of love or lust in return for the surrender of the lover’s soul.¹⁵

Critical Responses to Angel Magic and Necromancy

19

The addressative critique of learned magic—the idea that under the guise of images and figures, spoken and inscribed names and invocations to spirits or angels lay a true language of communication with demons—was used against astral magic and angel magic throughout the late Middle Ages. But it was not a particularly persuasive attack against the authors, readers and practitioners who followed their own judgment as to what was acceptable for a good Christian to practice. Some early critics of magic tried to tackle this problem. William of Auvergne argued that the author of the Liber de essentia spirituum, an anonymous revelatory discourse that gave cosmological justifications for magic, was not a true Christian because he lived and recruited followers in the desert, the abode of demons (William of Auvergne [1674] 1963, 1024–09). But the most intensive attempt to undermine the idea that learned magic rituals could be undertaken by a pious Christian was the 1398 condemnations by the faculty of theology at Paris, one of the most important medieval statements of the position of the Church authorities against magic (Boudet 2001).

20

The twenty-eight articles of the 1398 condemnation target the following defences of magic that were especially pertinent to angel magic: that holy prophets used magic arts, that the magic arts could be performed licitly or for a good end; that magic rituals honoured God; that devout prayers and fasts in the magic ritual made it licit, and that magic arts were revealed by God or could be used to achieve a vision of God. Condemnations like this never prevented angel magic texts from circulating widely in secret, but they seem to have discouraged contemporary authors from claiming revelatory origins for their works of magic, especially after the burning of John of Morigny’s Liber florum celestis doctrine at the University of Paris in 1323. The failure of the ‘angelic turn’ in medieval learned magic to convince critics that invoking angels was an acceptable, even pious, Christian practice may also reflect changes in Christian thinking about the roles of angels and demons. In late medieval theology and didactic literature, narratives of angelic encounters became less important, but human-demon interactions took new and prominent forms, ultimately arriving at the pact with the devil so often cited by witchcraft theorists (Mayr-Harting 1998; Boureau 2006). Necromancy plays an interesting role in this shift.

21

Some Christians justified performing rituals to summon demons as a desire for spiritual experiences or for goals they viewed as morally good, or as an extension of their duties as exorcists (Kieckhefer 1994; Véronèse and Chave-Mahir 2015). Critics of magic were, therefore, understandably keen to control narratives of the experience of demons (Herzig 2011). The idea that the necromancer could perceive himself acting as an agent of God when he subjected evil forces to his will is presented mockingly in the English poet John Lydgate’s popular allegory of Christian life, the Pylgremage of the Sowle (1426) as a delusion full of “ffalsehod and ffantesye / and cursyd ymagynacyoun.”¹⁶ The 1398 condemnations rejected the idea that demons could be good or morally neutral and denied that they were truly compelled to obey necromancer or that attempting to command them through learned magic rituals could be pleasing to God. Moreover, the condemnations reiterated that rituals that encouraged relationships of any kind with demons were idolatry because they involved the worship of demons in place of God.

22

Few books that can be characterised as wholly or largely ‘necromantic’ survive, presumably due to the repression of this particularly transgressive kind of magic, although single, or small groups of necromantic experiments survive in larger numbers.¹⁷ But the image of the necromancer holding a book and summoning demons from the safety of a magic circle is the archetypal medieval representation of the art of magic. A particular point in the ritual is usually depicted for dramatic effect: after the necromancer has successfully summoned demons but before they have enacted his commands; sometimes his control over them deliberately appears fragile.¹⁸ A rare example of a representation of a demon carrying out a task for a necromancer is depicted in a copy of Lydgate’s Pylgremage of the Sowle, which shows a small horned demon bringing his master treasure.¹⁹ By contrast, the image accompanying one of the thirteenth-century Cantigas de Santa Maria (Canticles of Holy Mary) represents the moment when a necromancer’s hold over demons weakens, thanks to the intervention of the Virgin Mary who has broken the spell of love magic they cast over a young woman. The necromancer is depicted raising his hands in anxiety as the demons swarm menacingly at the edges of his protective magic circle.²⁰

23

The image of the necromancer was popular not only because many critics argued that all magic was implicitly or explicitly based on demonic aid but also because representations of necromancy were thought to be useful instruments to demonstrate the reality of demons. Monastic, pastoral and scholastic texts showed interest in demon-human interactions and a belief in greater demonic intervention in the world from the early thirteenth century. When the Abbot Caesarius of Heisterbach wrote his work of instruction for novice monks, the Dialogus miraculorum (1220–35), he devoted an entire book to demons and included prominent stories in which hapless laypeople and clerics dabbled in necromancy. The idea that possession and necromancy demonstrated the reality of demons was argued even more forcefully by Thomas Aquinas in the De malo (1269–72) and other works, as if these vivid and extreme experiences of demons expressed the constant presence and malign activity of demons best and would worry at the imagination of those tempted to sin (Aquinas 2001, 16:1).

24

While there was a long medieval history of legends in which characters like Theophilus of Adana made written pacts with the devil in order to further their worldly ambitions, it is only after 1320 that the idea of the pact entered mainstream governmental and legal thinking about malign magic. In 1320 Pope John XXII consulted ten theologians and canonists on whether to categorise of magical practices as heretical. A consequence of this meeting was the promotion of the idea of a strong and heretical pact with the devil, that joined possession and necromancy as particularly malign expressions of human-demon relationships. The pact was a logical end point to the idea that learned magic rituals involved willing practitioners and implicit demonic worship. Moreover, the idea of a formal written contract that involved the complete and explicit submission of the witch to demons resolved the difficulty of how men and women who lacked clerical training and literacy could control powerful, threatening demons. In exchange for surrendering their souls, witches could call on the assistance of demons using only simple gestures or spells.

Demons, Astrology and the Rise of Theoretical Treatises

25

Sympathetic theorists of magic emerged from the thirteenth century writing under their own names rather than assigning a mythical authorship to their texts, which has led some theorists to be categorised as ‘author magicians’ by modern scholars (Weill-Parot 2019a; Boudet 2007; Véronèse 2006). These learned men—physicians, astrologers and monks—were keenly aware of the critiques of angelic magic and necromancy and experimented with alternative approaches to the rituals and philosophy of learned magic. The urgent separation of learned magic from witchcraft by its sympathisers involved giving it an ever more complicated theoretical framework. Some authors turned to new sources of power and authority, as John of Morigny did by selecting the Virgin Mary as his revelatory intercessor. Others integrated their study of magic with more respectable topics to increase its status (if not always its orthodoxy). The self-styled magister philosophus Berengarius Ganellus emphasised that magic was a science of words, while the physician Jerome Torella wrote his Opus praeclarum de imaginibus astrologicis (1496) to promote the category of natural astral magic (Gehr 2019; Weill-Parot 2019b). In the monastery to which he retired in 1490, the physician Peter of Zealand composed his eclectic and candid Lucidarius de rebus mirabilibus, a synthesis of magical theory that draws on multiple references from theories of human physiology and imagination to al-Kindi’s cosmology of celestial rays that permeated human bodies (Mandosio 2019). The works of medieval author-magicians and other synthesisers of magical knowledge demonstrate the positive interest that learned magic attracted among learned readers and the diversity of cosmological frameworks that could be invented to support learned magic in the medieval Latin West.

26

The increasing prominence of demons in medieval Christian thinking, from theology to pastoral literature, and the popular use of images and narratives of necromancy to demonstrate the reality of demons, also encouraged new ways of thinking about the role of demons in learned magic. As I have elsewhere discussed in more detail, four authors of works on magic—Michael Scot (1175–ca.1232), Cecco d’Ascoli (1257–1327), Antonio da Montolmo (fl. 1360–93) and Georgio Anselmi (ca. 1386–ca. 1449)—combined their astrological expertise with an interest in demonic agency to create new cosmological frameworks for magical practice (Page 2019a). These theorists subscribed to the theological understanding of demons as an imminent and powerful threat but thought that their powers could be harnessed by practitioners of learned magic drawing on knowledge from the (relatively) respectable and systematising art of astrology. In particular, astrological techniques enabled practitioners to calculate the locations and times in which demons would appear, understand the tasks they were assigned and invoke their ruling planetary spirits to gain control over them.

27

The approach of these writers was somewhat similar to that of Arabic astral magic and was clearly influenced by its techniques. But the significance of these writers and practitioners for the Christian art of necromancy was that instead of depending upon invocations to God and the saints they were using scientific—that is cosmological—knowledge to anticipate and command demons. The new prominence of astrology from the thirteenth century onwards enabled these writers to rethink the idea of the perfect magus along astrological and physiological lines. In most medieval learned magic texts, the ideal practitioner was usually presented as a virtuous Christian and exorcist acting in the name of God. But Antonio da Montolmo, writing about the occult powers of physicians and magicians, thought that there were men and women whose strong wills, complexions or powerful natal horoscopes meant that they could act on another body, healing or harming by their gaze, voice or touch alone (Antonio da Montulmo 2012, 252–53, 268–71). He further suggested that the powerful confidence (confidentia fortis) of some women led them to command demons with such a strong desire that the women were moved to tears. This entanglement of scholastic medicine, learned magic and folk beliefs, especially the idea of the ‘evil eye,’ was also expressed by writers critical of learned magic (Salmón and Cabré 1998; Delaurenti 2007). For example, Antonio’s contemporary Nicole Oresme thought that some old women were able to inflict harm on others with their gaze because of the complexion of their brains (Oresme 1968, 381–87). In the fifteenth century this entanglement became characteristic of witchcraft literature.

Witchcraft and Humanism

28

Influenced by earlier rhetoric, both critical of magic or supportive (or defensive) of it, the fifteenth-century authors of witchcraft and humanist texts explored the powers and limits of the natural human capacity to do magic in ways that downplayed the significance of ritual but elaborated magic’s cosmological framework. The authors of witchcraft literature had to balance scholastic scepticism about the capacity of humans to perform magical feats against their desire to nurture belief in the reality of witchcraft. Witches were not usually thought able to follow the instructions of complex learned magic texts or to conjure spirits to do their bidding. Their harmful magic was efficacious either because demons (with God’s permission) were the true agents behind it or because demons had taught witches how to do magic, or pretended to do so, in order to give witches an illusory sense of their own agency.

29

Persecutions of witches might be influenced by local beliefs that granted them extraordinary powers, however. In lay writings and accounts of the early witchcraft trials conducted by secular authorities in the Alpine region, for example, witches are often said to transform into wolves, a motif that probably originated in local folklore (Ostorero 2008, 6). Hans Fründ, a scribe in the chancellery of Lucerne and an imperial notary, reported the discovery of male and female witches (hexsen), in 1428–30 in the diocese of Valais, who had confessed that the evil spirit taught them how to turn into wolves. He states that they ran after sheep, lambs and goats in their wolf forms and ate them raw before turning back into men and women when they wanted to (Fründ 1999, 36). In this ambiguous account it is not clear how far the reader should believe that an actual metamorposis has taken place, but the transformation is said to seem real to the witches and those who observe them in wolf shape.

30

A few decades after Fründ’s chronicle, the Dominican inquisitor Heinrich Kramer explained in the Malleus maleficarum how demons took on the appearance of cats with the help of three witches and attacked a labourer in the diocese of Strasbourg (Kramer 2006, 323). Kramer explained that the ‘bodily presence’ of the women was first changed into the forms of beasts by the art of sorcery.²¹ Then demons in the effigies of the cats (daemones in effigiis cattorum) attacked the labourer while the female witches remained at home. When the labourer struck back at the cats, however, it was the women, not the demons, who sustained injuries. This sympathetic link between the witches’ bodies and the animal effigies inhabited by demons had its roots in demonic possession and an interpretation of demonic inspiration or instruction in which the demon entered the body of a human and dwelt there (Kieckhefer 2013). In these cases the blurring of the boundaries of human, animal and demon by fifteenth-century witchcraft theorists endowed the magical practitioner with new powers.

31

In fifteenth-century Italy a new intellectual climate allowed the syncretic and positive attitudes toward magic discussed in the previous section to be underpinned by neoplatonic, Hermetic, and humanist currents of thought that offered philosophical justifications for the human capacity to manipulate the forces of the universe. There was, however, a large degree of continuity between medieval and renaissance magic. The rituals and cosmology of learned magic were not changed fundamentally by the Corpus hermeticum, Neoplatonism or the Kabbalah, by the printing press or new networks, or by a supposedly new ‘subjectivity’ or ‘imagination’ at the centre of writings on magic (Kieckhefer 2006). Medieval theorists both critical or supportive of magic also saw a continuity between the classical past and contemporary magic. Their interpretation focussed on the deceit of demons, a view influenced by Augustine’s emphatic argument that the daimones of Antiquity were evil rather than morally neutral beings.

32

The critic of magic, William of Auvergne, thought that demons cunningly adapted their strategies of war on mankind to suit different religious and cultural contexts. In the ancient world they had taken the names of the planets in order to be thought to be celestial gods and trick men into idolatry; in the present, demons tried to deceive magical practitioners into believing that they had descended from heaven. Antonio da Montulmo, a theorist sympathetic to magic, also placed magic within a historical tradition linked to demonic agency. He thought that the ancients had worshipped both good and evil Intelligences as their gods, a practice he said was reflected in the invocations and prayers recorded in astral magic texts.

33

However, in their speculations about the relationship that learned magic had with its past and future, renaissance theorists offered two striking innovations. They were more interested than their medieval predecessors in reconstructing the real cultural world of the classical magi, and they placed more emphasis on the positive role that magic might play in shaping humanity’s future. This marked a shift away from the primary emphasis in medieval learned magic on the problems and pleasures of earthly life and the fate of souls in the afterlife. In this article I have space for only one example to illustrate these new directions, a historiated initial depicting of three kinds of magic in the deluxe manuscript copy of Pliny’s Natural History owned by Giovanni Pico della Mirandola (1463–94), a humanist famous for promoting (certain kinds of) magic (see figure 1).²² Pico della Mirandola supervised the iconographic cycle of this manuscript, which is based on a careful reading of Pliny’s discussion of ancient beliefs and practices and anecdotes about Greek and Roman historical figures (McHam 2016). The historiated initial accompanying book 30 on magic provides insights into Pico’s interpretation of Pliny’s idea of magic and his own views on the role it could play in society.²³

34

In fifteenth-century Italy, Pliny’s Natural History was part of the literary canon, influential, sought after and widely disseminated (McHam 2013). Pliny’s intention was to show that the bounty of the natural world was what made man’s achievement possible, whether in agriculture, healing or art. His descriptions of the appearance and properties of natural materials and his celebration of the high level of civilisation that Rome derived from these appealed to medieval encyclopaedists, for whom God had placed uses in all parts of his creation, and to renaissance humanists engaged in an idealising reconstruction of Antiquity. Magic was, of course, a much more challenging craft to ‘recover’ or ‘revive’ than classical medicine or art, because Pliny viewed it as fraudulent, corrupt and dangerously seductive—an analysis many fifteenth-century readers would have agreed with. Pliny’s famous definition of magic (Historia naturalis 30.1) declares that the art holds men’s emotions in a three-fold bond of medicine (to inspire the belief that magic promoted good health), religion (to add holiness, power and mystery) and astrology (to lure with the false promise of knowing one’s destiny). Moreover, Pliny’s account of the origins and history of magic is the story of a contagious superstition spread across the ancient world by credulous and greedy travellers, armies and exiles.

35

In view of this extremely negative assessment of magic, the artist of the Marciana manuscript (known as the Pico Master) faced the challenge of designing a historiated initial that was true to Pliny but also alluded to the positive potential in magic we see in Pico’s works. The solution was to incorporate depictions of three different kinds of magic within the image: roman agricultural magic, contemporary learned magic and an idealised magus in the classical tradition. Agricultural magic, specifically the protection of crops from pests and vermin, takes the form of skull and fruit motifs in the top section of the image that are used throughout the manuscript to evoke pagan visual imagery. In the Natural History Pliny proposed placing the skull of a mare on a stake in the garden to repel caterpillars and recommended hanging branches of pomegranates to keep snakes away.²⁴ This use of natural objects connected positively to the scholastic category of ‘natural magic’ and the widespread use of protective objects in households, as well as to Pliny’s wider project of describing natural objects that will benefit human civilisation.

36

In prominent position in the image is a depiction of a contemporary necromancer sacrificing a small black animal (probably a bat) in a magic circle with typical instruments of his art beside him: a vase of oil, a magic book, a bell and a pentagram. The necromancer was part of a relatively standardised repertory of images by Italian miniaturists that accompanied other fifteenth-century manuscripts and printed books of Pliny’s Natural History, but Pico’s artist chose to omit the demons that were the usual feature of this iconography.²⁵ This omission permitted a stronger association with the world of Pliny, in which animal sacrifice was an important part of Roman religious practice. But it is also likely this representation was intended to evoke Pico’s category of goeteia, that is magic performed with the aid of evil spirits, and to be contrasted with the portrayal of an idealised classical magus to the left of the incipit illumination.

37

This magus is dressed in special clothing (notably a stole) and holds a staff or wand over a jar that perhaps holds a portion of the sacrifice. He is looking at the head of a black ram on what is probably a pagan altar in a mountainous location in daylight.²⁶ To interpret this practitioner, I draw on the ‘perfectly black sheep’ mentioned by Pliny at the end of his prologue during a discussion of the deceitful strategies of magicians. Pliny argues that magi made their rituals extremely difficult to perform in practice in order that no one would discover that they were fake; the inevitable failure of the act could be blamed on blemishes in the performer or unsatisfactory ritual objects. But Pliny notes that on these grounds the Emperor Nero, who was passionate to learn magic, should have had success with the art because he had no physical blemishes and had the means to track down and purchase any ingredients he wanted. In Pico master’s image, the elegant, calm and solemn gesture of the magus contrasts with the bent over and anxious-looking necromancer, suggesting that the former represents a positive interpretation of magic. In other words, where the cruel tyrant Nero failed, an unblemished and noble but also skilled and virtuous magus could succeed. The idealised classical magus, whose appearance and ritual entangles elements of classical past and renaissance present, engages in animal sacrifice as a minister of nature, a practitioner of natural magic, according to Pico’s positive assessment of mageia in his Oration.

Conclusion: From Literary Traditions to Practice, Performance and Cult

38

In this article I have shown how medieval sympathisers and critics of magic were responsive to each other’s perspectives and placed particular emphasis on discursive entanglements. In the final section, I will move beyond discursive entanglements towards the question of entanglements of literary traditions and practice, asking what our written sources can reveal about group activities between magical practitioners from exchanging books to performing communal rituals.

39

Medieval critics usually understood practitioners of magic to be working by and for themselves, as distinct from heretics and witches who belonged to ‘sects,’ that is, large organised groups with initiation rites, communal practices and long-term goals that included the overthrow of the established Church.²⁷ Groups of practitioners, clerical and lay, do appear in our sources, but it is difficult to assess how far their shared practice constituted a ‘cult’ in the above sense.²⁸ Some examples suggest that a community of readers and copyists tended to develop around a book or books of magic. In 1383, for example, a group of necromancers in the village of Rugomagno near Siena using and copying the necromantic book belonging to a certain Agnolo di Corso was identified and condemned by a Tuscan Inquisitor (Brucker 1963, 19). In other cases, descriptions of large groups plotting harmful magic seems to be a function of the political importance of their target, for example Pope John XXII, Charles VI of France or Henry VI of England (Kieckhefer 1976; Veenstra 1997; Boudet 2019). Similarly, accusations against prominent men and women magnify the threat they posed to a ruler by identifying occult specialists in their entourages, from female servants to male physicians and clerics. This reflects anxiety about informal influence in court circles as well as the literary stereotypes of female servants and clerical necromancers.

40

Turning to the evidence of learned magic texts, we see that many of these describe rituals that can be performed alone. But others advise the practitioner to have companions, assistants or child skryers. In astral magic the addition of companions adds solemnity to the ceremony, increases the depth of reverence addressed to spirits and creates a community invested in the ritual’s success that is sometimes cemented with a communal meal (Picatrix 143). In a ritual for operating with the Moon in the Picatrix, for example, the practitioner directs the performance but has the help of friends and associates (amici et socii) to lead sacrificial animals and carry the ritual instruments, and they are supposed to take a particularly active role if the ritual’s goal helps them personally (Picatrix 145). The logistics of leading, releasing, and ritually killing and skinning several animals, in this case rams, may also have lent itself to group participation. But there is no evidence that astral magic was practised by Christians in late Medieval Europe in the sense of a formal veneration of planetary spirits, even if this is what some medieval critics of magic suspected or feared. Certainly, the rituals can be very elaborate; some texts required the practitioner to dress in robes dyed with the colours of a particular planet and to suffumigate himself while addressing the planet with prayers at appropriate times,²⁹ and even the inner body of the operator could be attuned to the planet by eating appropriate foods to attain the right humoral complexion.³⁰

41

A rich example of ritual magic activity centred on a family relationship is provided by the angel magic texts of John of Morigny and his younger sister Bridget who was ‘an early user, consulter and colleague,’ even though she was at first an unlearned teenager and later attached to a religious order some distance from where John lived (Watson 2020). John and Bridget shared their experience of visions as they experimented with the ‘science of infused wisdom’ and collaborated as authors. For example, Bridget contributed experiments to diagnose the outcomes of sickness and pregnancy, perhaps based on her own medical activities. Nicholas Watson argues that Bridget’s most important contribution to the Liber florum was to inspire its shaping as a set of operations that were universally available and intended toward a collective good rather than being a specialized clerical activity (2020). The wider social nature of John’s enterprise included asking his friend John of Fontainejean to contribute an angelic mass to the text and circulating his text to followers as he wrote it (Fanger 2021).

42

In necromantic rituals the presence of multiple companions seems mainly designed to guard against demonic force and malice. The magic text De circulis notes that it is acceptable (if not ideal) for the practitioner (artifex) to perform rituals for self-defence or consulting spirits alone, but for all other purposes he should have four companions (personae).³¹ The Clavicula Salomonis recommends between one and nine companions (compagni)—or at least a dog to intimidate the spirits.³² Certainly, a sense of demonic menace dominates the sources for necromancy although it can also be combined with a sense of play and performance that undercuts this. The four participants described in the prologue of the 1398 condemnation of magic who attempted to trap a spirit in a bottle as part of a treasure hunting experiment are said to have circled about the ritual space in their underwear wielding swords and pointing them at the sky and earth.³³ Other learned magic texts describe special crowns, mitres and priestly robes to wear, which contributes to the impression of elaborate ceremonies being a part of this practice. However, when critics of magic interpreted its rituals as acts of worship (dulia or latria) or idolatry, they did not extend the accusation of blasphemous reverence into descriptions of cults or sects of magicians similar to those described in fifteenth-century witchcraft mythologies.

43

Critics and sympathisers tend instead to use the language of schools and scholars (scolares) or societies and companions (socii).³⁴ For critics the practitioner’s intimacy with demons was an object of opprobrium, even horror. Augustine, in a phrase repeated by several prominent medieval theorists of magic, calls it a toxic fellowship of humans and demons (pestifera societate hominum et angelorum malorum).³⁵ But more sympathetic medieval sources on magic reveal a vibrant culture of exchanges of texts between members of religious orders, physicians and lay men, clerics and lay women—a culture of entanglement: discussion, borrowing, critique and adaptation alongside practitioner-client relationships and necessary secrecy and concealment.³⁶