In the last quarter of the twentieth century, dozens of books and articles on witches and witchcraft were published,amounting to a sort of second witch craze. These publications addressed the topic in general and in specific times and places, witchcraft, witch-hunting, images of witches, witches in art, literature, popular culture, new religious movements, witches in the past and the present,and witches in almost every imaginable connection,with the single exception of the one topic that is most foreign and most absurd to modern readers, students and scholars alike: ideas and knowledge about witches. Of the many dozens or hundreds of modern works on witches,scarcely any have addressed the single necessary precondition for the prosecution, torture and execution of witches: the certainty, to those who did the hunting, that witches existed.

In 1997, Stuart Clark published the first monograph since the time of Jules Michelet to focus on pre-modern ideas about witches.¹ Clark takes early modern ideas about witchcraft seriously; indeed, he devotes his first chapter to the language of witchcraft and the need to take ‘belief’ seriously as a motivating factor. However, Clark and the scholars beginning to follow his lead have retained the language of belief, thus implying sceptical distance in the sense that the ideas in question, while sincerely held,were not only fallacious but also entirely subjective.We do not mean to suggest that Satan-worshipping witches actually flew across the early modern European skies. Rather, we concur with anthropologist Gilbert Lewis: ‘The very word “belief” often implies, in its use, a judgement about the uncertain truth or reliability of that which has been asserted; “knowledge” does not convey the same doubt. “Knowledge” is legitimate; “belief” only questionably so.’² When early moderns held something to be true,they seem generally to have known it, and (religious) belief, in post-Voltairean usage,is not an adequate conceptual analogy.As Edward Muir has argued, sixteenth-century Europeans held magic and witchcraft to be real.³ That certain early modern authors occasionally expressed doubts as to the certainty of their religious knowledge, for instance, need not be any more unsettling to the paradigm we are proposing than the readiness with which physicists admit to doubts as to the nature and workings of atomic structure: despite all such doubts, what is/was known about atoms or witches induces people to act in confident ways, setting off atomic bombs or burning human beings (note the similarity).When we refer to religious and ideological knowledge as belief,we are not only distancing ourselves from the content (a natural enough desire in this case),we are introducing an alien distinction between what appear to us to be different elements of early modern Europeans’ world view. The language of belief in witchcraft studies betrays an anachronistic,modernist and dismissive approach to a mental universe quite different from our own.We do not understand our role as historians to include a duty to judge the veracity of past ideas. As Ian Bostridge has argued, ‘historical explanation of beliefs’ requires the historian to attempt to be neutral as regards their truth content.⁴ We have omitted the word ‘belief’ from our vocabulary in order to present our subjects’ ideas without undue anachronism.

Female and male witches

Research is always most exciting when the data do not quite fit the established paradigms and explanatory models. An entire body of literature, mainly by avowed feminists,has argued that witch-hunting was in essence woman-hunting, despite the fact that many of those executed for the crime of witchcraft were men (see chapter 2).Various attempts have been made to explain away this feature of witch-hunting (see chapter 1), but none will really do, in part because they cannot make sense of areas like Normandy, Estonia or Iceland, where the vast majority of witches were men. A fundamental problem with previous interpretations of male witches is that they treat their subject as anomalous, even impossible. This assumption has led to the virtual exclusion of male witches from witchcraft studies,especially from those concerned with issues of gender. We have set out in this book to make the male witch visible – to construct him as a historical subject – as a first step toward a deeper understanding of the functions and role of gender in pre-modern European witch-hunting and ideas about witches. Our focus is, necessarily, not on a discrete body of court cases,local events or a specific period; rather,we begin with a critique of the very considerable historiography, based on our readings of learned demonology,statistical evidence and certain cases drawn from wide geographical and temporal spans.

The Malleus maleficarum, or Hammer of Witches, is the best-known early modern work on witchcraft,infamous for its misogynist statements about women and for its argument that most witches were women. With very few exceptions, modern scholars have taken a one-dimensional view of this treatise, citing it almost exclusively to illustrate the inherent misogyny of witch-hunting. Indeed, very few other treatises are ever cited; thus, in many studies, the Malleus maleficarum has come to represent ‘the’ demonological position on women and witches. It is, therefore, an appropriate starting point for an examination of attitudes toward gender and witchcraft.

The Latin word maleficus, and its feminine variant malefica, came into common use in the fourth century CE and were used widely in the medieval and early modern periods to denote a person who committed evil deeds by means of magic.⁵ Maleficus/malefica is often translated as ‘witch’. In the title Malleus maleficarum, the fact that the word maleficarum is a feminine plural noun would seem to suggest that the authors, Dominican Inquisitors Heinrich Institoris (Kramer, Krämer) and Jacob Sprenger,⁶ believed that all witches were female. In Latin, groups containing both males and females conventionally are represented by the masculine plural, even if there are more females than males in the group. The feminine plural implies an absence of males from a group; therefore, the use of the feminine in the title Malleus maleficarum suggests that all witches are female.⁷

In the text itself, however, Institoris and Sprenger use both masculine and feminine forms of maleficus. Indeed, the first line of the Malleus reads ‘Utrum asserere maleficos esse sit a deo [sic] catholicum quod eius oppositum pertinaciter defendere omnino sit hereticum.’⁸ (Whether it is so very orthodox to insist that there are witches, given that maintaining the opposite of this obstinately is wholly heretical?) Where one would expect the authors to use the feminine accusative plural maleficas, they use the masculine maleficos. Furthermore, there are many other references to malefici (masculine plural) as well as to maleficae (feminine plural), sometimes within the same sections of text. In several instances, the text refers plainly to male witches both as specific individuals and as a group. At first glance, it seems amazing that Institoris and Sprenger, notorious to modern scholars as the primary authors of the ‘witches as women’ paradigm, would write about witches in the masculine at all, let alone in their opening lines.

Almost nothing has been published on this topic in academic treatments of gender and witchcraft, demonology, or the Malleus maleficarum itself. Indeed, male witches in general are hardly to be found in witchcraft historiography, despite the ubiquity of gender (or rather, sex) as an issue in the study of witch-hunting and ideas about witches. There is a kind of hole at the centre of witchcraft studies, to borrow an image from Robin Briggs, into which male witches and learned discourse about them have disappeared.⁹ This absence is especially striking in the work of Stuart Clark, who, in an otherwise masterly analysis of early modern demonology, suggests that witchcraft theorists were incapable of conceptualising male witches. This blind spot in witchcraft historiography requires attention.

Of course, the historiographical gap is a common trope with which to introduce a historical study.But research ought to do more than add a little plaster to existing structures. In order to offer a real contribution, it ought to confront and challenge those structures.In the case of this project, this means confronting a historiography committed to causal explanation and to a polarised, essentialising view of gender and its relationship to witch-hunting and ideas about witches. It also means engaging with a strongly politicised discourse about witches: inside the academy and without, the female witch is a potent symbol of women’s oppression by men and, rather paradoxically, of women’s power.

The overtly political dimension to the study of witches in early modern Europe demands a high level of consciousness and reflexivity regarding language, representation, and meaning. Our goal in this book is to make what is hidden visible: not only male witches themselves, but also the historiographical structures and politics that exclude them as historical subjects. This may seem threatening to some readers, especially to those with a heavy investment in representing witches as essentially female,or in claiming the study of early modern witches as women’s history. We disagree with these viewpoints, but we consider our work also to be a feminist history, in Joan Wallach Scott’s sense: ‘Feminist history . . . [is] not just an attempt to correct or supplement an incomplete record of the past but a way of critically understanding how history operates as a site of the production of gender knowledge.’¹⁰ Though as historians we are clearly in one important (professional) sense insiders, we also consider our perspectives to be those of outsiders, and in the interest of transparency, we should like to point out that an atheist and a Jew have relatively little personal investment in Christian approaches to witchcraft. Our ideological and professional investments are, we trust, clearly enough articulated over the course of the book.¹¹

Our concentration on male subjects may appear to subvert the feminist project of constructing women as historical subjects,or to diminish the importance of the female witch. We prefer to think of it as a logical application of a relational concept of gender, in which men and women are defined and constructed in terms of one another. Andrea Cornwall and Nancy Lindisfarne have criticised the relational concept of gender on the basis that it is too narrow to admit of possibilities other than simple male–female difference. We agree with this view, and our understanding of gender incorporates those possibilities, including hierarchical differentiation within genders.The relational concept,nevertheless,is the core of our concept of gender.¹² With this understanding of gender and its historical construction,male witches are neither irrelevant nor a threat; they are necessary components of a complex phenomenon.Any endeavour to understand the relationship between gender and witchcraft has to take male witches into account and explain how they ‘fit’ within the gendered framework of early modern ideas about witches; without them, we are ignoring one half of the gender relationship and,necessarily,limiting our knowledge about both men and women in early modern Europe. As Caroline Walker Bynum has put it, ‘the study of gender is a study of how roles and possibilities are conceptualized; it is a study of one hundred percent,not of only fifty-one percent,of the human race.’¹³

Willem de Blécourt has articulated a useful gender-based approach to witchcraft and witchcraft accusations in his article ‘The making of the female witch: Reflections on witchcraft and gender in the early modern period’. He argues that witches were ‘made’ locally, except when demonic influence was adduced (as in the ‘witch-panics’); and suggests that ‘[a] witchcraft accusation . . . articulated the crossing of male-designated boundaries rather than being restricted to a specific female space’.¹⁴ This suggestion sees gender as embedded in and structuring social relations, rather than as a free-floating ‘concept’. Steve Hindle uses gender in the same functional rather than reified way in his article on gossip, gender and authority in early modern England.¹⁵ In both instances, the category ‘gender’ contributes to historical examination of women and men, female and male spheres, and their interactions with one another.

This volume is an attempt at such an explanation. Ideas about witches and a number of episodes of witch-hunting serve as windows into gender relations at an especially fraught and vexed juncture in occidental history. The book contains three main arguments: first, that male witches have been excluded from witchcraft historiography and that this exclusion by modern scholars is not congruent with early modern understandings of witches; second, that explanations of the dynamics of witchcraft prosecutions should be applied equally to both female and male witches; third, that male witches could exist within the framework of early modern ideas about witches because they were implicitly feminised. Although the first chapters focus on our disagreements with prevailing views about male witches,our conclusions are not utterly opposed to modern scholarly understandings of the relationship between gender and witchcraft. Indeed, our argument that male witches were implicitly feminised tends to support the view that early modern Europeans correlated witchcraft with women very closely. On the other hand, this correlation was neither exact nor straightforward.

We have attempted to maintain consciousness and reflexivity regarding language, representation, and meaning. The questions that propelled the first stages of this work grew out of a close reading of the Malleus maleficarum, and were as follows: what did Institoris and Sprenger mean when they used the masculine malefici? Why did they sometimes write malefici and sometimes maleficae? Were they alone in following this pattern of grammatical gendering,or was it the usual practice in early modern witchcraft treatises? Most importantly, how much significance should one attach to the language of witchcraft literature,not just to what was said but also to the ways in which things were said?

Renato Rosaldo has written that ‘no mode of composition is a neutral medium’.¹⁶ This is as true of our own writing as it is of early modern witchcraft theorists; if we ascribe significance to their language choices, then we must be aware of the significance of our own. In addition, we must be conscious that we are ascribing significance to the empirical fact that there were male witches and that witchcraft theorists discussed them. As Keith Jenkins points out, ‘facts’ are selected, distributed, and weighted in finished narratives: ‘The facts cannot themselves indicate their significance as though it were inherent in them.To give significance to the facts an external theory of significance is always needed.’¹⁷ In other words, meanings are made, not found. Roland Barthes has argued that creating meaning is the essence of what historians do: ‘The historian is not so much a collector of facts as a collector and relater of signifiers; that is to say, he organizes them with the purpose of establishing positive meaning and filling the vacuum of pure, meaningless series.’¹⁸ It is important, therefore, to be open about the kinds of assumptions and positions that inflect the meanings one creates.

An examination of the word ‘witch’ should provide a useful approach to our policy of disclosure. The term is unavoidable, but it is extremely problematic because of the range of meanings associated with it. To begin with, in modern English usage, the word ‘witch’ almost invariably denotes a female person, a woman or a girl. For example, the Concise Oxford Dictionary defines ‘witch’ in female terms, as ‘a sorceress, esp. a woman supposed to have dealings with the Devil or evil spirits.’ The Harry Potter phenomenon obliges us to comment on the word ‘wizard’, which was almost never used in pre-modern sources as a male synonym for witch. The root word ‘wise’ implies knowledge, and the term was used primarily for learned practitioners of the magical arts (ibid.), not for practitioners of ‘diabolical sorcery’.In the Daemonologie, in Forme of a Dialogue, a work claimed by King James I and VI,the characters discuss the difference between magicians and necromancers on the one hand, and sorcerers and witches on the other. The former are learned men who, by virtue of a pact with the Devil, seem to be able to command him to perform certain acts or services for them, though in fact this is merely a ploy of his to gain possession of their soul; the latter are simple folk who are but the servants or slaves of the Devil.¹⁹ The term ‘warlock’ is used commonly now to denote a male sorcerer, but this usage implies that there is some distinction between witches and war-locks. The Old English root of the word ‘witch’ has two forms: wicca, for a male witch, and wicce for a female. ‘Warlock’ is rooted in a different semantic field: ‘oathbreaker, traitor, or devil’.²⁰ Modern English has lost the explicitly gendered forms of ‘witch’, and attributes the feminine gender to the word implicitly. The modern use of two different words for male and female witches is problematic because it encourages the exclusion of men from discussions about pre-modern witches and by extension, about witch-hunting.

In both Latin and French, two very common words for ‘witch’ are used for both men and women, the only distinction being the gender-indicative endings. In Latin, the most common word is maleficus/malefica; French has sorcier/sorcière. German has Hexe, a feminine noun now conventionally used to refer to female witches; early modern German used a number of variants for male witches (as opposed to learned magic users), such as Unhold, Drudner and occasionally Hexenmeister. There are many other terms as well, including gender-specific words such as necromanticus (or nigromanticus) and pythonissa, which were specialised terms for certain kinds of magic-users.²¹ In general, however, early modern authors employed rather generic words when they talked about magic-users, words that did not distinguish between male and female witches as witches.²²

The closest English equivalent to this kind of non-distinguishing language is ‘sorcerer/sorceress’.In the interests of strict correspondence with early modern style, we considered using these terms instead of ‘male witch’ and ‘female witch’. We decided against it, however, because it could lead to some confusion when discussing other scholars’ work. In addition,it would mean backing off from an engagement with the various interpretations of ‘the witch’ when the point of the exercise is to open up debate on precisely that issue. We use ‘witch’ for both men and women, with the gender specified as necessary for clarification. Despite the problems of modern English usage, this is consistent with early modern categories and usage.

The second problem with the word ‘witch’ is more complex. To whom does the label apply? Given that for most people today, or at least most potential readers of this book, early modern witchcraft was not ‘real’, what does it mean to refer to a historical subject as a witch, without quotation marks around the word? The question leads directly into a thorny tangle of issues, including realism, referentiality, agency, and subjectivity. There is not enough space here to do full justice to each of these topics; nevertheless, it is important to address them and establish our positions explicitly.

In the first chapter of his book Thinking With Demons: The Idea of Witchcraft in Early Modern Europe, Stuart Clark tackles the problems of realism and referentiality head-on.He states that the witchcraft beliefs of the past have been assumed by modern scholars to be utterly wrong about the possibility of supernatural events occurring because scholars are generally committed to a realist epistemology, even if they do not acknowledge it. This model views language as a simple and unproblematic reflection of an outside reality and judges utterances to be true or false as measured by the accuracy of their description of objective things. Ideas about witches, for example the idea that witches flew to Sabbaths, do not, generally speaking, correspond with ‘the real activities of real people’; in Clark’s terms, they lack referents in the real world. According to Clark, this lack of reference to an empirical reality has led scholars to dismiss early modern ideas about witchcraft as irrational, or to explain them away as the secondary consequences of some genuinely real and determining condition: that is to say, some set of circumstances (social, political, economic, biological, psychic, or whatever) that was objectively real in itself but gave rise to objectively false beliefs. Clark objects to these two approaches on the grounds that they make it impossible to interpret witchcraft beliefs as beliefs in terms of either their particular meaning or their ability to produce concrete actions.²³

What is needed, he argues, is a more useful (Saussurean) approach, according to which language is not constituted by reality,but rather itself constitutes reality.This understanding of language suggests that success in communicating meanings depends upon relationships within the language system, not on relationships between the system and things external to it. Were one to subscribe to this view of language, one’s focus of enquiry would shift away from putative external determinants of ideas about witchcraft to the meanings of those ideas within their specific frameworks. This does not mean that the historian would have to accept pre-modern ideas about witchcraft; rather, the whole issue of the truth or falsity of those ideas would become irrelevant. As Clark puts it, ‘Witchcraft’s apparent lack of reality as an objective fact would simply become a non-issue, and the consequent need to reduce witchcraft beliefs to some more real aspect of experience would go away’,thus freeing historians to concentrate their interpretative endeavours on the ways in which the ideas made sense to those who held them.²⁴

Clark’s approach to ideas about witchcraft and language provides some of the major underpinnings of this book. Our interest in male witches is precisely in elucidating how they fit into an ostensibly misogynistic framework, or, to put it another way, into an early modern web of ideas about witches, men, and women. Behind this interest lies the assumption, which we share with Clark, that ideas that survived for nearly three centuries ‘must have made some kind of sense’.²⁵ Witchcraft ‘beliefs’ do not have to be true in terms of our epistemology (our ‘web of beliefs’) in order to have seemed rational and coherent to early modern Europeans.²⁶

In regard to the labelling of early modern men and women as witches, a non-realist methodology allows one to bypass the process of deciding whether or not each individual is ‘really’ a witch, since there is no way of determining this outside their frame of reference. For the purposes of this enquiry into the early modern web of ideas about witches, if a person was understood as a witch in the past, then he or she was one and will be referred to as such.

This leads us to problems of subjectivity and agency. By adopting, without qualification or quotation marks, the label witch, we leave ourselves open to the charge that we are replicating injustices perpetrated against those accused of witchcraft. For some modern authors, those accused and executed were victims, and should be discussed as such. These authors tend also to identify with the (female) ‘victims’. Anne Barstow, for instance, who employs the rhetoric of victimhood to great effect, dedicates her book Witchcraze to ‘those who did not survive’.²⁷ One could argue that it is legitimate to refer to early modern witches as victims if one is interested primarily in the perspective and experiences of the accused. On the other hand, there are several problems with this approach.

First, the identification of modern feminist scholars with early modern witches seems, as Diane Purkiss has suggested, impertinent. Barstow’s dedication assumes a commonality of experience with the witches, an assumption concerning which Diane Purkiss’ remarks are particularly apt: ‘In the face of a degree of fear and suffering which most of us cannot even imagine, a more humble and less eager identification might be advisable.’²⁸

More importantly, however, the use of victimisation rhetoric introduces methodological problems. Referring to witches as victims imposes a single perspective on a multitude of actors. There is no way to employ the rhetoric of victimisation in a sweeping fashion and at the same time offer meaningful and nuanced interpretations of those who accused, tried, tortured, executed, or just wrote about witches: the frames of reference are incompatible. Indeed, it is doubtful that calling witches victims is always consistent with the perspective of the witches themselves. Diane Purkiss, Lyndal Roper, and Malcolm Gaskill have produced studies that suggest witches possessed agency and, in some cases, represented themselves as witches deliberately.²⁹

Nevertheless, it is true that our adoption of witch-hunters’ and witchcraft theorists’ categories and views of witches objectifies to some degree the individuals caught on the receiving end of witch trials. This is regrettable,even distasteful,given the ordeals many of the witches suffered; however, it is unavoidable in a study devoted primarily to the ideas of those who thought witches were real and dangerous. There is very little room for manoeuvre on this point; either one respects historical actors’ categories or one does not.

The existence of male witches, and particularly their presence in demonological treatises, raises many questions. Can the male witch be assimilated into discussions framed by ideas about patriarchal oppression? Is the male witch a figure of the earlier witch-hunting phase only? Were male and female witches believed to be fundamentally different? Were male witches somehow gendered ‘female’ by witchcraft theorists? How can gender theory help us make sense of the conceptual relationship between male and female witches?

We submit that the male witch was not, conceptually speaking, different from the female witch. Early modern writers refer to male witches throughout the witch-hunting period, not just at the beginning. The male witch was not feminised in the ways one might expect; that is, the male witch was not assumed to be, or normally described as, homosexual or effeminate. He was, however, connected with female witches, and femaleness, via the medieval and early modern sense that it was primarily the weak-minded (especially women) who could be duped by the Devil into becoming his servants. The male witch suggests that biological sex was not, at the conceptual level, the primary characteristic of the witch; gender was. The primary affinity between male and female individuals with witchcraft was related to their status as (womanly) ‘fools’. Women were by pre-modern lights more prone to weak-mindedness, but men were by no means immune; and, like women, foolish men represented threats to the (patriarchal) social order. Male and female witches thus have more in common than the majority of participants in the witchcraft and gender debate suggest.

The reality of witches

The question ‘Were there really witches?’ might seem the most obvious and natural one to pose after encountering the grim evidence of late-medieval and early modern witch trials. Yet for the majority of early modern Europeans,there could be no serious doubt about the existence of witches, especially for those involved in laying the theoretical and legal groundwork for the prosecutions, for those who accused people of witchcraft, for those who did the torturing and executing, and probably also for those accused of witchcraft. For almost all of them, the existence of witches and of evil magic was a foregone conclusion, as self-evident as the earth’s orbit around the sun is to us. They had every reason to ‘believe’, indeed to know that witches existed. Ideas about witches had been authoritative long before the witch trials of the fifteenth to the eighteenth centuries and continued to exist in Europe and Britain for some considerable time after the witchcraft laws had been repealed.³⁰ Yet scholars have dismissed ‘belief’ as a universal, a constant that explains nothing³¹ and gone looking for social, economic, even biological and meteorological causes for ‘witch crazes’, to the point that a widespread, consistent and authoritative discourse about diabolical witchcraft is made to seem less important than the individual, quite disparate and often seemingly random triggers that sometimes unleashed a major persecution, and that sometimes had no effect at all.

The question whether or not witches really existed is a modern invention. It has little or nothing to do with ‘actors’ categories’, the thought-worlds and views of late-medieval and early modern Europeans, Britons and Americans.³² We have asked the question for a few hundred years now, ever since the witch trials ended in the eighteenth century. Scholars first answered ‘No’, in the rational spirit of the European ‘Enlightenment’, then ‘Yes’, in the spirit of Romantic and medievalist folklore. Since the beginning of the twentieth century, these opposing positions have defined the debate: those to whom even the idea of a witch or witchcraft is dangerous, superstitious nonsense; and those who believe that there is ‘more in heaven and earth’ than reason and empirical observation can ever discover, and who think that there really were people who knew themselves to be witches and who knew that they practised witchcraft. Some, such as Margaret Murray, have gone so far as to suggest that witches were members of a secret and ancient ‘pagan’ religion that was practised uninterruptedly all through the Middle Ages.³³ In a search for origins and legitimacy characteristic of new societies and new social and religious groups, early proponents of the modern witch-religion Wicca founded their belief system on the idea that witchcraft is an ancient religion that was suppressed for a long time, but never extinguished; though there is very little evidence, if any, for this idea.³⁴

However, there is rarely a simple answer to so simple a question, as the newspaper editor was forced to admit when he answered ‘Yes, Virginia, there is a Santa Claus.’ Both witches and Santa Claus exist(ed) not because there really were or are people who ride around on broomsticks or in magical sleighs, but because some people think/thought these things to be true. Ideas can be more important than ‘reality’ – even (perhaps especially) if we concede that there is such a thing as reality outside what we perceive. It is unsettling to contemporary world views, formed by trust in science, empirical methods and experimentation, to be confronted by what seems to be nothing but ‘belief’, but which was firmly proven knowledge centuries ago.

Perhaps more unsettling is that there probably is a certain amount of historical ‘hard fact’ behind both stories. The North American Santa Claus³⁵ is a version of the traditional Christian figure St Nicholas (feast day: 6 December), a bishop born around 280 CE in Asia Minor who is said to have been especially generous: whence the tradition of giving presents at Christmas-time. Carlo Ginzburg has argued that European ideas about witchcraft are based on actual remnants of folk or ‘pagan’religion,³⁶ and most historians will agree that folkloric medicine, herbal cures and traditional knowledge about the natural world helped shape what came to be dangerous ideas about diabolical witchcraft.There is (or ought to be) a large distance between recognising historical roots, or the nuggets of truth behind a legend, and thinking that the legend itself is true. If we assume that a lack of evidence is simply the result of oppression or suppression, then we must also believe in Atlantis, the extra-terrestrial origin of the pyramids, or the existence of werewolves and vampires.There are better and simpler explanations for all these notions and any underlying phenomena than their literal existence.

Historians not only study the past; we also make it.Much of what we think about witches comes from the above-mentioned traditions of Enlightenment and Romantic scholarship. To both camps, the witch provided a useful test case, a means to prove a broader set of ideas and values using an unquestionably important phase in the common western past. If pre-Enlightenment Europeans were so fatally wrong about the existence of witches, then the rest of their world view must also be suspected of the worst inconsistencies, irrationalities and errors. The project of ‘enlightenment’was to prove just that,to depict an irrational ancien régime so decayed, so perverse and dangerous, especially to the powerless and the marginal, that only a new and rational world view could replace it. The Romantic reaction was to insist on the value of tradition, of mystical and spiritual perception and experience,especially that of the common people,putatively unspoiled by industrialisation and urbanisation; and to rescue the non-material, the numinous, from the sceptical gaze of the deists and atheists. Tenacious herbal lore or even pagan survivals were in the Romantic age not part of a decayed and intellectually bankrupt old world, but precious evidence of the value and richness of folk tradition.

Therefore, a historian’s intellectual and ideological background and constitution play a strong role in determining how he or she will see both ideas about witches and the witch trials themselves. As Diane Purkiss has argued, the academic ‘discourse’ of historians studying witchcraft, especially in England, is essentially a male product, mainly about women, of scholars uneasy with self-consciously theoretical approaches of the sort that might illuminate gender, ideology or ‘belief’, who describe themselves as ‘sceptical empiricists’³⁷ and act accordingly when confronted with other people’s thought-worlds and mentalities: ‘Rather than trying to understand how witch-beliefs were structured for and by the believer, historians have often bent their energies towards explaining witch-beliefs away.’³⁸

The underlying reason for this, according to Purkiss, is that ‘History’ is essentially an Enlightenment discourse, one of those that ‘gradually displace[d] the supernatural in the seventeenth century’³⁹ and thus was programmed from the start to stake out claims to truth on territory previously occupied by interpreters and practitioners of the supernatural. While many contemporary historians have engaged with the post-structuralist and postmodern challenges to ‘enlightenment truth-claims’,Purkiss points out that most scholars of English witchcraft have not,and the result,until quite recently,has been stagnation and conceptual constipation.⁴⁰ No new insights into witchcraft can be gained so long as it is part of a hegemonic male academic discourse that brushes off the rather different ideas and subjectivities of people who lived a long time ago under often cruel conditions.

Carlo Ginzburg has also attacked the English tradition of witchcraft studies, focusing on the naive empiricism and terminology of Keith Thomas and Alan Macfarlane,among others,whose approach to ‘belief’ was influenced by anthropological functionalism and thus uninterested in its symbolic dimension.⁴¹ Ginzburg’s critique of Macfarlane is harsh:

In this book we assume that there were some people, in some times and places,who thought 1) that they were witches; and 2) that they practised (for good or for ill) magical arts. This has been demonstrated by Carlo Ginzburg, among others. However, we insist that the majority of those accused,tortured and even burned as witches did not think of themselves as such, at least not before their interrogation. They were ‘made’, as de Blécourt puts it,but not only by (gendered) village dynamics.That many were willing to confess to participating in the witches’Sabbath or to practising ‘black arts’ reflects not only the efficacy of the means of persuasion (physical and cultural), but also the predominance and widespread acceptance of ideas about witches in their environment – rather than real events, as Murray claimed. Most will have shared the idea, dominant in their societies, that witches existed, that they used magic to cause harm to people and animals,and even that they served and worshipped Satan; but they did not think, in most cases, that they themselves did so or had done so in the past, until persuaded into saying it and perhaps even believing it under extreme duress.

So the question as to the existence of witches and ideas about witchcraft is hard to separate, then as now. Most contemporary scholars do not think that those accused of witchcraft were in fact witches (or that they even understood themselves to be witches), and most also allow their readers to see that they themselves either do not ‘believe’ in the existence of witches and witchcraft, or consider the question irrelevant. Certain modern scholars of the witch-hunts and of witchcraft, for instance Montague Summers and Margaret Murray, have themselves thought that witches existed but held diametrically opposed views concerning the nature of witchcraft: to Murray, it was a pre-Christian fertility religion; to Summers,a satanic cult that was the root of feminism.But the question is neither irrelevant nor clear; it is ‘badly posed’.

We refuse to allow ourselves to be drawn into ideological debates between representatives of Enlightenment and Romantic ideas about the existence of witches, no matter how important they are to adherents of Wicca and similar new religions, or to agendas concerning the oppression of counter-hegemonic traditions and structures. To engage in such debates is to lose sight of the historical data and ‘reality’, so far as we can discern it,of witch-hunting and ideas about witches.Without such ideas, there would have been no pyres or gallows. But ideas alone are insufficient to explain the events of the period roughly 1450-1750, during which thousands of women, men and children were tortured and executed.Had ideas been enough,the fires would have been lit hundreds of years earlier and continued burning for many decades more.

Although our questions may have no general validity or answer,they are nonetheless relevant in a certain form to any study of historical events, their preconditions and their causes. Medieval and early modern westerners knew that witches and the magical arts were real; and there seems to have been a small number of people who thought either that they were witches or that they practised magic, or both. Very careful distinctions are required if we are to avoid grotesque and useless generalisations.

We offer this book not primarily as a set of original studies designed to bring to light ‘new cases’, nor even as a test bed to try old and established ideas by the light of new evidence. Many of these sorts of books have been published already, without the authors having stopped to consider seriously whether or not the paradigms governing their studies were leading them closer to understanding ideas about witches or witch-hunting, or even closer to understanding our culture’s recent obsession with witch-hunting. Diane Purkiss has addressed the latter question usefully in her book The Witch in History, and we attempt the same sort of synthesis and revision of seemingly well-known sources, episodes and texts with the goal of rethinking the relationship between concepts of gender and concepts of witchcraft.