Nicolau Eymeric (ca. 1320-1399)
reported in his Directory of Inquisitors (Directorium
inquisitorum) that he had confiscated and burned the
“The Key of Solomon” (Clavis Salomonis).
Also ‘Necromancy’ was
not so broad a term as to cover all varieties of magic that were suspected by
authorities of involving demonic power. Rather, it was a decidedly learned art
involving complex rituals and ceremonies, often patterned on the church's own
liturgical rites, and knowledge of this art was typically conceived as being
contained in books or manuals. For example in the early fifteenth century,
religious reformer Johannes Nider (ca. 1380-1438)
reported that he knew a certain monk in Vienna who, before entering the
religious life, had been a necromancer and had possessed several demonic books.
Some of these texts survived inquisitorial flames and now allow remarkably direct
access into one area of the world of medieval magic. And one example
a fifteenth-century necromantic manual, titled Sworn Book of
Honorius the Magician (Liber iuratus Honorii) contains following illustration:
Since necromancy was
essentially a bookish art, its practitioners necessarily belonged to the small,
educated elite who possessed the Latin literacy necessary to use such manuals.
This meant that necromancers were virtually always clerics. The ranks of the
clergy in the Middle Ages extended down from priests through a variety of more
minor orders, and it was these lower orders that most likely supplied the
majority of necromancers. Medieval schools and universities were religious
organizations, so most students were formally required to become clergy. Thus
virtually all educated people were by definition clerics. After receiving their
degrees, however, these men might have few or no official ecclesiastical
functions.
Much Muslim and
Jewish magical literature furthermore discussed invoking and controlling demons
or spirits in the name of God. Demons could either alter their own forms or
they could affect human perception so that people thought they saw a horse, a
boat, a banquet, or anything else the magician might desire. Aside from
deceiving the senses, demons could also affect the human heart and mind, and so
necromancy could be used to arouse love or hatred, to bring calm or incite
agitation, and so forth. Just as in divination, here too there were debates
about the extent to which human will might be directly affected by magical
practices. (See Richard Kieckhefer, Magic in the Middle Ages, Cambridge
University Press, 1989).
Many authorities
reasoned that demons only had power over human bodies, but that by affecting
the body in certain ways (manipulating bodily humors, for example), they could
induce various mental states, or at least cause people to succumb to such
states more readily. Demonic power over bodies of course allowed them to
inflict physical harm on people and to cause disease. They could also heal,
however, and so through them necromancers also commanded all these powers. And
the belief that interaction with demonic forces was inherently evil and
corrupting had been an essential aspect of church doctrine since the earliest
days of Christianity. From the time of Augustine, Christian authorities had
condemned magic primarily because of the potential involvement of demons in the
rituals and practices that they deemed to be magical. They could hardly now
fail to condemn ritual that was explicitly demonic in nature. That some
necromancers claimed, by virtue of their position as clerics and more basically
by the power of Christ, that they could interact safely with demons and command
them toward positive ends was hardly an effective defense; in fact it
constituted a significant challenge to the longstanding position of the church.
Augustine had noted that most magic was based on evil associations between
humans and demons, and Aquinas had argued that even when a magical ritual did
not com in any explicit submission to demons,
nevertheless a tacit pact might be supposed to exist between the magician and
the entities he summoned. Necromancy, a type of magic contained in Latin
manuals and performed by clerics, was probably the form of magic most familiar
to Christian authorities in the high and late Middle Ages. As concern on the
part of authorities over magic grew in this period and condemnations of magical
practices increased, they were shaped largely by conceptions of elite
necromancy, but applied ultimately to all varieties of magic and superstition.
(See Richard Kieckhefer, Forbidden Rites: A Necromancer-'s Manual, 1998).
By the late
Middle Ages, authorities came to regard magic more seriously, and as a more
serious threat, so that this era was also marked by the increasingly rigorous
and intellectually specific condemnation of many forms of magical practice, as
systematic demonology and awareness of explicitly demonic necromancy fed
longstanding Christian concerns about the potentially demonic nature of all
magic and the demonic threat behind superstition. Legal advances also took
place in this period, above all the use of inquisitorial methods in court
proceedings. These methods allowed more readily for prosecutions and
convictions in cases involving charges of magical practices. In addition,
specially designated inquisitors began to appear who (eventually) brought cases
of heretical demonic magic under their jurisdiction. To support these new
procedures and personnel, advanced legal literature and theory developed, which
defined the illicit qualities of magic more precisely than ever before. Legal
condemnations of magic thus became as profound and encompassing as earlier
moral condemnations had been.
Already in the
Symposium, Plato's Socrates claimed that the gods never have direct contact
with humans. Instead, they employ the daemons, beings halfway between gods and
humans, as their intermediaries or messengers. Plato's term daimonion gave the
West its word for demons, his word for messenger, angelos,
the word for angels. When God begins to seem impossibly distant, Western
Christians rediscover angels and demons. If these messengers begin to seem
distant or unreal as well, someone begins daydreaming that somewhere-somewhere
close – other people must be experiencing superhuman reality, physically,
empirically, unmistakably. The thirteenth century was a crucial period in the
development of necromancy as both idea and practice. This was the age of
Aquinas, and there is clear evidence that necromancy was already under
consideration as a way of investigating whether spirits really existed or were
capable of interaction with humans. Around the time Aquinas was born, the
German Cistercian monk Caesanius of Heisterbach wrote his Dialogus miraculorum, or Dialogue on Miracles (1225), which was very
influential in the later Middle Ages. Coming to terms with the
imperceptibility, improvability, and possible nonexistence of the spirit world
happened gradually, in step with a reluctant acceptance of the extraordinary
power of the human imagination.
In Religion and the
Decline of Magic, Keith Thomas asserts that, in the early-modern period,
"the evidence of widespread religious skepticism is not to be underrated,
for it may be reasonably surmised that many thought what they dared not say
aloud." He suggests that "not enough justice has been done to the
volume of apathy, heterodoxy, and agnosticism which existed long before the
onslaught of industrialism. It is generally accepted however that the witch
hunts, magical activities becoming an official matter, was immensely influenced
by the centuries of constant, religious conflicts and threats-usually taking
the form of religious wars between Christians and Muslims or between Catholics
and Protestants.
How strange to see a
Franciscan philosopher-theologian arrive at a conclusion essential to scientific
thinking by asserting the freedom of an omnipotent God to do as he wishes.
Using hindsight, one can see Thomas Aquinas supplying the first part of a
scientific worldview by emphasizing the existence of knowable causal patterns
in an integrated, interdependent natural universe. Aquinas and other
representatives of the via antiqua assumed the
existence of metaphysical "entities" originally derived from
Aristotle, such as a "potential intellect" that permits humans to
receive and store information, and an "active intellect" that permits
them to analyze it. But while Christian theologians could not settle for the
observation that magic simply "worked," their Muslim counterparts
could and did. Muslims believe that by reciting the last several sentences of
the Qur'an following the five daily prayers, they neutralize all evil, forces.
Thus at least one, irony about the European witch-hunts might be that they were
the result of reason and logic applied to a false premise.
Clerical authorities
furthermore linked magic categorically to pagan rites and demonic powers, and
thus they condemned all magical activity as inherently immoral and illicit for
Christians. Of course, these distinctions were never clear-cut or absolute in
practice, as moral judgments were always deeply intertwined with legal rulings.
Even in the ancient world, legal condemnations could become quite general in
tone, and any type of illicit magical practice could be regarded as a threat to
the community's harmonious relationship with divine or spiritual forces. In
Christian Europe, kings and princes relied heavily on clerics and on the church
to buttress their authority. Thus the church's condemnation of magic had legal
effects, and law codes were always based on Christian morality. In the high and
late Middle Ages, however, the church itself became a much more legalistic
entity. Canon law, based on church rulings or "canons," developed
into a science at the new schools and universities that appeared in this
period, and increasingly the church came to define itself and its authority in
terms of these legal codes. Ecclesiastical courts developed to enforce this
law, which applied to all Christians. Ultimately, specialized officials
appeared-inquisitors whose purpose was to root out the worst offenders against
church law, those Christians who denied or rejected essential elements of their
faith or aspects of church authority and so became heretics. Because of their
perceived involvement with demons, people who engaged in many types of magical
practices were eventually included in this category.
The history of the
condemnation of magic in this period culminated in the emergence of the
essentially new category of diabolical, conspiratorial witchcraft. Although
people had been suspected, accused, and prosecuted for performing harmful or
malevolent sorcery-maleficium-throughout the
Christian era, as in antiquity, only in the fifteenth century did the idea
develop that people might engage in maleficium as
members of heretical, demon-worshiping cults, offering themselves to Satan in
exchange for power and acting at his direction to corrupt and subvert all of
Christian society.
Of course, because of
their rejection of Christianity, Jews were easily depicted as being in league
with demons. One legend related how the early sixth-century Christian saint
Theophilus had been tempted by a Jewish sorcerer into signing a pact with the
devil in order to gain magical powers. This story became an archetype for later
notions of diabolical pacts associated with magic and witchcraft. Thus
promising a copy of the notorious Sixth and Seventh Books of Moses, an
advertisement in the Allgemeiner Litterarischer
Anzeiger, 28 March 1797, drew considerable attention.
Of course this now was when the 'Magic Media Market' was being formed. A copy
from around 1750 (probably not long after the fake was created) is housed in
the British Library. It consists of twenty-two loose, bronze-coated cardboard
pages measuring 30.5 x 44.5 centimeters. They each have writing on the front
and the back, mostly in blood red characters that are or at least appear to be
oriental. The sections of text that are formulated in German, the headings in
particular, are also written using Latin letters, which are blended with the
oriental signs. During the late 1950’s a German Judge suddenly felt the need to
declare the Sixth and Seventh Books of Moses (by now part of folklore)-- as
“anti-Semitic.”( Hans Sebald, The 6th and 7th Books
of Moses: The Historical and Sociological Vagaries of a Grimoire', Ethnologica Europea 18, 1988,
53-8).
Furthermore an
inquisition, inquisitio in Latin, originally did not
imply an ominous institution, but simply meant a process of legal inquiry.
Bishops and their officials were expected, as a matter of course, to inquire
into any potential errors of faith within their jurisdictions. However by the
twelfth century, as heresy became a greater concern, inquisitorial procedures
became more intense and systematized. Seemingly distant from the process of
inquisition, confession was actually closely related, for it involved, or was
supposed to involve, a personal inquiry by an individual, although directed by
a priest, into his or her own beliefs and moral state, leading to
acknowledgment and repentance of any errors. This was also the underlying moral
goal of an inquisition, and authorities' growing preoccupation with uncovering
and uprooting potentially improper beliefs was a critical factor in the
developing condemnation of heresy, as well as the condemnation of magic and
superstition.
Thus growing
reliance on inquisitions marked the emergence of a new type of legal procedure
in Europe , where suspected sorcerers might be required to grasp hot irons, and
in several days their wounds would be examined to determine whether they were
healing properly (little or no healing was a sign of guilt), or suspects might
be bound and dunked in water to see how quickly they rose to the
surface--floating was a sign guilt. Plus if an entire community regarded
an individual with suspicion, punishment could occur, and it was often severe.
In 1075 for example, citizens of Cologne (now Germany) threw a woman from the
town wall because they believed she was practicing magical arts. In 1128 the
people of Ghent (now Belgium) eviscerated an "enchantress" and paraded
her stomach around the town.
The first evidence of
the legal use of torture comes from statutes of the Italian city of Verona in
1228, regarding the use of torture by secular courts. In 1252 Pope Innocent IV
(reigned 1245-1254) permitted papal inquisitors to use torture to extract
information from suspects. To obtain a conviction for a potentially capital
offence, standard legal procedure came to require either the testimony of two'
eyewitnesses or the confession of the accused. Given the clandestine nature of
most magical practices, eyewitnesses were usually out of the question~
Authorities certainly recognized the potential of torture to extract false
confessions, and they devised methods and imposed limitations intended to
reduce this risk. Nevertheless, especially in cases involving accusations of
demonic sorcery, authorities often set these restraints aside. The unrestrained
use of torture would become a hallmark of most of the major witch hunts of
subsequent centuries.
Inquisitorial concern
about ‘magic’ continued to develop over the course of the fourteenth century,
and theories regarding the essentially demonic nature of most forms of magic
became increasingly elaborate and definitive. And as the fourteenth century
progressed, condemnations of magic came from outside inquisitorial circles as
well. In 1398 two Augustinian monks were executed in Paris after they had
failed in their attempts to relieve the intermittent madness of the French king
Charles VI (reigned 1380-1422) and then accused his brother Louis of Orleans of
having used magic against him. Louis' wife was also accused of practicing
sorcery. After Louis' death in 1407, charges of magic again circulated against
him. Most importantly, in close connection to this web of concern at the royal
court, the theological faculty of the University of Paris, the preeminent
intellectual institution in medieval Europe , issued a broad condemnation of
sorcery, divination, and superstition in 1398. The twenty-eight articles of the
Paris condemnation tended to dwell mostly on the sort of elaborate, ritual
magic that clerical necromancers would perform.
The stereotype of
witchcraft that emerged in the course of the fifteenth century was not an
absolutely stable idea that once constructed, remained constant and
unchallenged in all its aspects. Most people accepted the potential reality of
harmful magic and feared its power, and basic notions of demonic threat hiding
behind common magical practices and superstitions were widely accepted,
certainly among authorities. Even when authorities disagreed about the extent
of demonic power in the world, they acknowledged· the existence of the devil
and his desire to corrupt Christian souls.
Called in French sorciere (derived from the late-Latin sortiarius,
or diviner and German Hexe, between 1626 and 1630, the central German city of
Bamberg executed around six hundred people for this crime of the mind, among
them the mayor of the city, Johannes Junius.
To be a witch was as
much about a person's essential identity as it was a description of certain
practices, for unlike other perceived practitioners of magic, witches were not
just individual agents of harm or ‘malevolence’ in the world. Instead, they
became members of a vast, diabolical army bent on corrupting and subverting
everything that was good and decent in society. Thus the first major theorists
of witchcraft, writing in the early fifteenth century, described groups of
witches gathering to worship demons, engage in orgiastic sex, desecrate
crosses, befoul consecrated hosts, and murder and devour babies at
cannibalistic feasts. For all the fantastic and monstrous acts authorities
envisioned taking place at sabbaths, however, they still could place them in
fairly mundane and realistic settings. Small groups of witches would gather in
cellars, caves, or other isolated but entirely worldly locations.
These beliefs appear
to have had tremendously deep cultural roots, into which authoritative
constructions of witchcraft inadvertently tapped. In many premodern societies,
individuals, whom scholars now typically refer to as shamans, confronted evil
spirits and sought to protect human communities, working to guarantee
fertility, abundant crops, and successful harvests. The superficial although
inverted similarity of such figures to witches, who were typically accused of
impeding fertility and destroying crops, is clear, and in the 1960’s, the
Italian historian Carlo Ginzburg discovered a remarkable historical convergence
of these beliefs.(For the English translation see Ginzburg, The Night Battles:
Witchcraft and Agrarian Cults in the Sixteenth and Seventeenth Centuries, Johns
Hopkins University Press, 1983).
The presence of one
witch in a region therefore indicated the existence of more, and a captured
witch could reasonably be expected (and frequently forced) to identify others.
This notion that witches were not merely individual malefactors but members of
a satanic conspiracy bent on subverting Christian society led not only
religious but also secular authorities to treat witchcraft very harshly. The
threat posed by witchcraft was seen to be so great that authorities in many
jurisdictions declared it to be a crimen exceptum, an exceptional crime, This meant that normal
legal procedures could be suspended, Rules restricting certain types of
questionable evidence were abandoned, the threshold for proof of guilt might be
lowered, and perhaps most importantly, limitations On the use of torture could
be ignored. Equally defining of witchcraft was the belief, again never absolutely
uniform but certainly very broadly held, that witches were typically women.
The roots of this notion were extraordinarily diverse, but it too can be seen
to be at least implied in the idea of the sabbath, insofar as sabbaths
emphasized the sexual congress of witches with demons and more basically their
submission and subservience to demonic masters and ultimately to the devil, who
was of course conceived as being male.
Biblical commandments
and classical Aristotelian philosophy both were marshaled to prove that women
were inferior to men spiritually, mentally, and physically, They suffered
weaknesses and corruptions in their bodies, which in Aristotelian thought were
imperfectly formed versions of male bodies, and they were spiritually and
intellectually more vulnerable to the deceptions and seductions of demons. Yet
throughout the early modern era, many authorities largely avoided much
specifically gendered theorizing about witchcraft. If "witch" often
meant "woman" in this period, this seems to have been due less to
abstract philosophy or theology than practical reality. That is, far more women
than men were being accused of witchcraft in the courts. Across Europe an
average of 75 percent of witchcraft accusations focused on women, and in some
regions the percentages rose into the nineties. In Siena , for example, of more
than two hundred witches tried from the late sixteenth to the early eighteenth
century, over 99 percent were women. Certainly one powerful reason for these
percentages was the simple fact that women were, on the whole, far more legally
vulnerable than men in this period, often having no legal status apart from
their fathers and husbands. But the harmful magic that characterized
witchcraft, encompassed issues of fertility-love potions and charms for
potency, but also withered crops, withered male members, and murdered children.
Thus concerns over witchcraft naturally focused on the female domains of
reproduction, childbirth, and nurturing.
While courts could
and did initiate hunts on their own, the vast majority of witch trials
throughout the early modern period responded to charges of simple maleficium brought by ordinary people when, for example, a
cow died or a child sickened unexpectedly. Such occurrences did not
automatically raise suspicions of witchcraft; unexplained misfortune was common
in premodern Europe . Yet if some particular animosity existed between the
victim or the victim's family and another person, and if that person had a
reputation for wielding malevolent magical powers, or if there had been some
direct sign of magical attack, however slight-a muttered curse, a threatening
gesture, or even a baleful stare-then a public accusation might be made.
Authorities, when the accusation was brought to their attention, could add charges
of diabolism, apostasy, and attendance at sabbaths, and wring out confessions
through torture. In 1519, in the city of Metz, in Alsace, the scholar Heinrich
Cornelius Agrippa von Nettesheim (1486-1535) came to
the defense of an old woman accused of witchcraft by the Dominican inquisitor
Nicolas Savini, arguing that the woman was senile and
deluded, not a servant of Satan. Agrippa was himself a student and practitioner
of learned magic. He wrote a major study, De occulta philosophia (On Occult
Philosophy), when he was twenty-four.
The Protestant
Reformation was, of course, the great event of the early sixteenth century in
Europe. It began in 1517, when Luther circulated his ninety-five theses
challenging basic doctrines of the Catholic Church at the University of
Wittenberg. Within a few years he had moved into an open break with Rome, and
winning broad support across much of the German Empire, he permanently
shattered the religious unity of Western Christendom. The profound political,
social, and religious forces unleashed by the Reformation dominated European
history until well into the next century. That the major period of witch
hunting in Europe corresponded almost exactly to the Reformation era has often
been noted. And much ink has been spilled over whether Catholic or Protestant
authorities executed more witches, but in the end the numbers tell no clear
story. When trials began to rise after 1560, they did so in both Protestant and
Catholic lands. And while religious wars focused on external enemies,
confessionalism directed its energies inward.
The next hundred
years saw the most intense witch hunting, at least in central and western
Europe. For example in the fifty-year period from 1580 to 1630. Fully 90
percent of executions for witchcraft in German lands, which were the heartland
of European witch hunting, occurred in these few decades. In 1589 the suffragan
bishop of Trier, Peter Binsfeld (ca. 1540-1603),
published De confessionibus maleficorum
et sagarum (On the Confessions of Witches), based on
a major series of trials in Trier .
Then in 1595 the
magistrate Nicholas Remy (1530-1612) issued his Daemonolatreiae
(Demonolatry), drawing on his extensive experience with witch trials in the
Duchy of Lorraine. In 1598 the Scottish king James VI (reigned 1567-1625, also
as James I of England , 1603-1625) wrote Daemonologie
(Demonology) after he too had some direct experience with witch trials. In 1599
and 1600 the Jesuit Martin Del Rio (1551-1608) published his massive,
multivolume Disquisitiones magicae
(Investigations into Magic). Henri Boguet (ca.
1550-1619), a magistrate in Franche-Comte who personally executed many witches,
published his Discours des sorciers
(Discourse on Witches).
In 1563, Warhafftige und Erschreckhenliche
Thatten der 63 Hexen (True
and Horrifying Deeds of Sixty-Three Witches), recounting a group of executions
at Wiesensteig, a small principality of around 5,000
inhabitants in the highly fragmented southwestern region of the German Empire.
This pamphlet described the first major hunt in what would become the region of
most intense witch-hunting activity in Europe. That same year, the English
Parliament passed a new act making witchcraft a capital crime, and similarly
harsh legislation was also approved in Scotland . Within only a few years, in
1566, Protestant England had its first known witch trial, although hardly a
major hunt as had occurred in Wiesensteig. At
Chelmsford, in the southeast of England, three women were accused and one was
ultimately executed. In 1590 and 1591 Scottish officials in Edinburgh put on
trial the North Berwick witches, so called because they supposedly gathered at
regular sabbaths at North Berwick , some twenty-five miles east of the capital.
The trials are particularly famous because these witches were accused of
plotting to murder King James VI by raising storms while he journeyed across
the North Sea . James observed portions of the trials, and they may have
inspired the interest in witchcraft that led him to write his Daemonologie.
The most terrible
hunts, however, were those conducted by a handful of German bishops and
prince-bishops In the territory of Trier over 300 people were executed in the
1580s and 1590s. In Maim, major hunts with victims running into the hundreds
erupted every decade from the 1590s through the 1620s. From these Rhineland
archbishoprics, witch hunting spread east along the Main River to the
Franconian bishoprics of Bamberg and Wlirzburg, each
of which experienced major hunts in the 1610’s and 1620’s, as did the Bavarian
bishopric of Eichstatt along with the associated territory of the abbey of
Ellwangen. The absolute worst hunt took place in the Rhineland in the territory
of Cologne, the third of the great German archbishoprics. Here highly organized
and efficient trials from about 1624 until 1634 resulted in the deaths of
probably around 2,000 people.
Yet such gigantic
hunts, were abnormalities. Despite widespread notions of satanic cults and
diabolical conspiracies, most accusations of witchcraft were rooted in specific
cases of perceived harm believed to be wrought through maleficium.
When trials escalated into major hunts, feeding on their own energies, the
situation was different. Inspired by the initial trial, other people might come
forward to make unrelated accusations, magistrates might become convinced that
more witches were hiding in the community and press their own investigations,
and of course accused witches themselves were pressed to name names. Yet even
in these cases, a hunt might well end of its own accord after a certain number
of trials and executions. Officials, comforted that they had uprooted evil from
their region, might stop pursuing their investigations. The community as a
whole, after an initial fearful wave, might grow calmer, and so accusations
would subside. In other cases, of course, this happy release of tension did not
occur. Instead, as accusations and convictions mounted, fear grew, paranoia
might seize courts, and real panic might grip the entire community. In these
situations, accusations multiplied and grew more indiscriminate. That is,
people who did not conform to the stereotypical image of a witch were accused
and arrested. This was certainly the case in Bamberg when the wealthy, socially
respected (and, of course, male) Burgermeister
Johannes Junius was found guilty of witchcraft in 1628.
1533 account of the
execution of a witch charged with burning the town of Schiltach
(Baden-Württemberg, Germany)in 1531:
To categorize "
Germany", as the zone of witch hunting par excellence is, however,
somewhat misleading. The situation in the Low Countries was exceptionally
complex. The territories comprising present-day Belgium and the Netherlands
initially lay within the German Empire, but in 1555" Emperor Charles V,
who was also King Charles I of Spain, gave them to his son Philip II of Spain
(reigned 1556-1598). Philip did not, however, succeed his father as emperor
(that title going instead to Charles's brother Ferdinand 1), so the Low
Countries became Spanish rather than imperial territory. In the 1560s and 1570s
these lands saw significant witch trials. Then in 1579, the northern provinces
banded together as the United Provinces of the Netherlands, declaring their
formal independence from Spain in 1581. These provinces also became
predominantly Calvinist, while the southern Spanish Netherlands were Catholic.
In 1592 Philip II issued a decree that extended the right to try witches to
local authorities, and not surprisingly the number of trials in the south (in
what is now Belgium) increased.
Unlike the German
Empire, France had been a unified kingdom for centuries prior to the early
modern period. It did not, however, have exactly the same boundaries as the
modern French state. If we use present-day borders, then " France "
had around 5,000 executions for witchcraft. But this figure shrinks
dramatically if we exclude the eastern regions of Alsace, Lorraine, and
Franche-Comte, which were French-speaking and had significant numbers of
trials, but at this time nominally belonged to the German Empire and were in
fact largely independent. Within its early modern borders, the Kingdom of
France saw fewer than 500 recorded executions for witchcraft. Moreover, as in
the German Empire, so within the Kingdom of France there were important
regional variations.
The Key of Solomon P.2: Occult Science
The
Key of Solomon P.3: Magical Revival
For updates
click homepage here