,.,r . F w. , M Realizing the Witch  r s i g tefanos Geroulanos and Todd Meyers, series editors ~1 1 1' Realizing the Witch Science, Cinema, and the Mastery of the Invisible Richard Baxstrom and Todd Meyers FORDHAM UNIVERSITY PRESS NEW YORK 2016 Frontispiece: Svensk Filmindustri poster for Hdxan (1922) Copyright @ 2016 Fordham University Press All rights reserved. No part of this publication may be repro- duced, stored in a retrieval system, or transmitted in any form or by any means-electronic, mechanical, photocopy, recording, or any other-except for brief quotations in printed reviews, without the prior permission of the publisher. Fordham University Press has no responsibility for the persis- tence or accuracy of URLs for external or third-party Internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Fordham University Press also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Visit us online at www.fordhampress.com. Library of Congress Cataloging-in-Publication Data Baxstrom, Richard. Realizing the witch : science, cinema, and the mastery of the invisible / Richard Baxstrom, Todd Meyers. pages cm. - (Forms of living) Includes bibliographical references and index. Includes filmography. ISBN 978-0-8232-6824-5 (hardback) - ISBN 978-0-8232-6825-2 (paper) 1. Hdxan (Motion picture) 2. Witchcraft-Europe- History. 3. Witches-Europe. 4. Christensen, Benjamin, 1879-1959. I. Meyers, Todd. II. Title. BF1584.E85B39 2015 133-4309-dc23 2015oo8857 Printed in the United States of America 18 17 16 5 4 3 2 1 First edition for.Julia for Fanny  CONTENTS Introduction: What Is Haxan? 1 Part I. The Realization of the Witch The Witch in the Human Sciences and the Mastery of Nonsense 17 I. Evidence, First Movement: Words and Things 32 2. Evidence, Second Movement: Tableaux and Faces 62 3. The Viral Character of the Witch 87 4. Demonology 103 Part II. A Mobile Force in the Modern Age 1922 133 5. Sex, Touch, and Materiality 145 6. Possession and Ecstasy 168 7. Hysterias 187 Postscript: It Is Very Hard to Believe. . . 205 Acknowledgments 213 Notes 215 Benjamin Christensen's Cited Source Material 249 x Contents Filmography 253 Bibliography 257 Index 275 I N T R O D U C T ION What Is Haxan? Witches always claim that they do not believe in spells, object to the discourse of witchcraft, and appeal to the language of positivism. -JEANNE FAVRET-SAADA, Deadly Words (1980) It is not living life for an art form to serve merely as an echo of something else. -BENJAMIN CHRISTENSEN, "The Future of Film" (1921) To think is always to follow the witch's flight. -GILLES DELEUZE AND FELIX GUATTARI, What Is Philosophy? (1991) The Wild Ride. The Sabbat. Child sacrifice. Diseases, ruin, and torture. The old hag. The kleptomaniac. The modern hysteric. Benjamin Christensen took the threads of phantasm and wove them into a film thesis that would not talk about witches, but would give the witch life. Haxan is a document, an amplified account of the witch insistent on its historical and anthropo- logical qualities, presented through excesses so great that they toyed with his audience's skepticism as much as their sensitivity. Christensen created an artistic work filled with irrationalities that not only made the witch plausible, but real. By the time Benjamin Christensen (1879-1959) began filming Haxan in 1921, he had already spent nearly three years conducting research for his film and securing a studio in Copenhagen to accommodate his costuming and elaborate set designs. Haxan was not the Danish actor/director's first foray into filmmaking, but it would be his most ambitious. The silent film was the most expensive ever produced in Scandinavia.1 The Swedish film produc- tion company, Svensk Filmindustri, provided Christensen with funding 2 Introduction: What Is Hixan? (in addition to buying back the director's own studio facility from credi- tors) in 1919 to make what Christensen called "a cultural-historical film in seven acts." With Swedish funding came a Swedish title for the film (the Danish word for "witch" is heksen). The film was shot in Copenhagen in 1921-22, and premiered in Stockholm in September 1922. Despite laborious planning, seemingly endless trouble with censors, and the unprecedented scale of his project, Haxan was only the first of a trilogy imagined (but never realized) by Christensen-the other films in his unfinished series were tenta- tively titled Helgeninde (Saints [feminine]) and Ander (Spirits). Before filming, Christensen obsessively gathered historical and contem- porary sources for his "lecture in moving pictures."2 His "witch" would be a visual account with direct reference to original writings, art, and literature on witchcraft and witch trials from the Middle Ages through the Reforma- tion and beyond-materials which he considered alongside and through his reading of the modern sciences of neurology, psychiatry, anthropology, and psychology. Taking up an argument made by a long line of eminent scientists-a line dominated by the preeminent psychiatrist and neurologist Jean-Martin Charcot-Christensen intended to make a case for witchcraft as misidentified nervous disease and to highlight the incompatibility of superstition and religious fanaticism with modernity and science. What re- sulted was something very different; it is this productive gap between intent and outcome that forms the core of what follows in this book. Perhaps with the exception of a select group of Scandinavian film experts and enthusiasts, Benjamin Christensen is regarded-when regarded at all as a weirdly interloping figure, much in the way his masterpiece is seen as wild and unmoored from other parts of cinema. Born the last of twelve chil- dren to a bourgeois family in Viborg, Denmark, Christensen had been active on the Danish arts and theater scene since 1902, after having trained first in medicine and then as an opera singer. Although highly praised for his singing ability, Christensen developed a nervous illness that resulted in the loss of his voice, a debilitation that led to a brief career in commercial sidelines (most notably the sale and distribution of champagne for a French importer), but this did not end his artistic career. Breaking in as a film actor in 1911, Christensen quickly moved to the director's chair; his first film, The Mysterious X, also known as Sealed Orders (Det hemmelighedsfulde X), was re- leased in 1914. The Mysterious X foreshadowed his work in Haxan and was Introduction: What Is Hixan? 3 highly praised for its innovative command of cinematic technique; it was also a worry for its producers, as the film ended up costing four times its origi- nal budget. The Mysterious X was followed in 1916 by BlindJustice (Hevnens nat), which was celebrated for Christensen's intense performance in the tragic lead role of a man wrongfully jailed and separated from his child. Christensen demonstrated an impressive command of filmmaking well in advance of his contemporaries. These early successes established for Christensen a reputation as an ambitious, innovative, and commercially re- liable director within Scandinavian film circles. Now a proven auteur, Chris- tensen signed with Svenska Biografteatern (renamed Svensk Filmindustri shortly after) in February 1919 and received full artistic control over his proj- ects.3 Rewarded for his previous efforts with unusually generous support, Christensen found the means to pursue the pioneering, bizarre, and lavish project that is the subject of this book. Christensen's post-Hdxan struggles are hardly surprising in retrospect. After previewing an uncensored print of Haxan in 1924, Metro-Goldwyn- Mayer studio head Louis B. Mayer asked, "Is that man crazy or a genius?"4 Banking on "genius," Mayer offered Christensen the chance to make films in Hollywood, which Christensen accepted in 1926. As was often the case for European directors who moved to the major American studios in the 1920s, Christensen found that both his creative freedom and, crucially, his monetary resources were severely limited within the Hollywood studio system. After working on three lackluster projects for Metro-Goldwyn- Mayer,5 Christensen moved to First National Pictures in 1928, directing a number of largely forgettable films, including The Hawk's Nest (1928), The Haunted House (1928), House of Horror (1929), and with minor praise, Seven Footprints to Satan (1929). While at times these films display some of the spark of Christensen's previous work, they are hardly remarkable and are today remembered only because the director of Haxan stood at their helm. Christensen, in uncharacteristically quiet fashion, left Hollywood behind in 1935 and returned to Denmark. After a period of reassessment,6 he en- tered filmmaking again, this time for Nordisk, producing four sound fea- tures between 1939 and 1942. While several of the "social debate" films Christensen directed at this time were critical and commercial successes,7 the failure of his last project, the spy thriller The Lady with the Light Gloves (Damen med de lyse Handsker), brought his filmmaking career to an unceremonious 4 Introduction: What Is Hixan? close. Not even the feted rerelease of Haxan to theaters in Denmark in 1941 could offset this final, humiliating disaster. After losing studio backing for an adaptation of Tove Ditlevsen's novel A Child Was Hurt (Man gjorde et Barn fortred), Christensen "retired" to manage the Rio Bio cinema in Co- penhagen until his death on April 1, 1959. Haxan stands as the filmmaker's lasting contribution to the history of cin- ema. And yet, it is fair to ask what precisely this masterwork is, because of its troubled and often clandestine status in the years since its initial release, and also because it resists easy characterization. It is typically associated today with the horror genre and simultaneously with proto-documentary or nonfiction film, but neither designation fully captures the film's unique character. Its reemergence in the middle of the twentieth century as a "cult classic" has only worked to further obscure its place within the history of cin- ema. Haxan is, perhaps more than any other work in the medium, a singu- lar film, but for reasons that require a detailed look at Christensen's project that goes beyond the narrow lens of its seemingly strange subject matter. In the following pages we attempt to think alongside Christensen, through his vast library of source materials and within the structural logic of his film thesis, to better appreciate Christensen's project. And with this apprecia- tion come questions-questions about the relationship between the image and the word, questions about the kind of new and unexpected visual histo- riography the film seems to produce, and finally, questions about what types of conclusions we are meant to glean from Christensen's cinematic vision. Christensen's vision for Haxan was quite radical. Eschewing typical no- tions of "drama" and "plot," the director described his project as follows: "My film has no continuous story, no 'plot'-it could perhaps best be classified as a cultural history lecture in moving pictures. The goal has not only been to describe the witch trials simply as external events but through cultural history to throw light on the psychological causes of these witch trials by demonstrating their connections with certain abnormalities of the human psyche, abnormalities which have existed throughout history and still exist in our midst."8 In unambiguous terms, Christensen expresses his thesis and his method in this short statement prior to the release of the film. Even if this were "all" the film did, it would still represent a significant work within the history of early nonfiction filmmaking. Yet there is so much more in Haxan as Christensen struggles to realize his thesis. As an enthusiastic Introduction: What Is Hixan? 5 scholar and an unusually innovative artist, he puts cinema to work not only in explicating the witch and the hysteric, but along in bringing their shared power to life. Hdxan exists as one of the most innovative films to emerge during the silent era. The film also affords a fascinating view into wider debates in the 1920s regarding the use of film in medical and scientific research, the evolv- ing study of religion from historical and anthropological perspectives, and the complex relations between popular culture, artistic expression, and sci- entific ideas. Hdxan therefore bears a unique relation to all these areas and yet is not reducible to any one of them. Christensen spent years gathering classical and anthropological sources on sorcery and religion to form the narrative background of Haxan. Our study of Haxan simultaneously offers an analysis of the scope and influence of Christensen's remarkable work and an examination of the sources that made Haxan a "living" cinematic tableau. Future Forms Like every other artist, film artists must display in the future their own personality in their works. Benjamin Christensen, "The Future of Film" (1921) The formal characteristics of cinema as a medium of expression provide Christensen the means to ultimately exceed the power of his historical source material-to breathe life into his subjects. We argue that this excess makes Haxan's status complicated. While cinema has its own limitations in con- veying depth and analysis in relation to reasoned arguments, it is not lim- ited by what James Siegel has called "social[ly] constrained thinking" or a discursive mode of analysis that is, by definition, incapable of comprehend- ing certain situations that lie outside the circle of reason. As Siegel point- edly argues, witchcraft is precisely one of the domains where scientific reasoning finds its limit; this limit is not automatic in regard to cinema, however, and Christensen exploits this fact to the fullest in Haxan.9 Most understandings of cinema would attribute this tension between reason and its absence to film's ability to bring fantasy to life. We would 6 Introduction: What Is Hixan? certainly not deny this claim, but it is important to note that Christensen does not conceive Haxan as a fantasy film. The director begins from the position that he will bring errors of belief in the fifteenth and sixteenth cen- turies into our view in order to show just who witches (and later hysterics) really were. Christensen's strategy is therefore more blunt than those de- ployed, for instance, in the human sciences, in that it requires the violent, erotic nature of this "error" be emphasized rather than suppressed under the signature of a theory of society as a knowable, structural entity. Jeanne Favret-Saada states it best: To say that one is studying beliefs about witchcraft is automatically to deny them any truth; it is just a belief, it is not true. So folklorists never ask of country people: "what are they trying to express by means of a witchcraft crisis?," but only "what are they hiding from us?" They are led on by the idea of some healer's "secret," some local trick, and describing it is enough to gratify academic curiosity. So witchcraft is no more than a body of empty recipes (boil an ox heart, prick it with a thousand pins, etc.)? Grant that sort of thing supernatural power? How gullible can you be?10 Christensen's thesis echoes Favret-Saada's words loudly; he is concerned (gripped) with abnormalities, events, and causes. The director claims, in other words, to provide a diagnosis, which is precisely the case when measured against the film that resulted. Yet in the details of his chosen form of ex- pression, and through the marked intensity of his personal engagement, Benjamin Christensen does not so much unmask the witch or the hysteric as he brings these figures and the power that animates them to life on the screen. Favret-Saada's sarcasm highlights the standard objection that bringing a figure to life in this way would, by definition, violate standards of objectivity. In her ethnography of modern witchcraft in the France, the people of Favret-Saada's village insisted on the ignorance of witch- craft's power, and yet that power dictated their action and movement-to be "caught" (pris) or bewitched was an awareness and a risk, and as a social scientist deemed "expert" on witchcraft, Favret-Saada was an un-witcher at once caught and catching (contagieux) (concepts we explore in detail in our chapter on the viral character of the witch)." Christensen's answer, like Favret-Saada's, is that witchcraft is difficult to study because of the inacces- sible materiality of the witch to the anthropologist and the believer alike. Introduction: What Is Hixan? 7 The observable epiphenomenon signaling the presence of a witch does not yield any objective, total proof when segmented and scrutinized at an ana- lytic distance. And yet, in the face of the failure by the human scientist to objectively signify her "true" meaning, the witch remains. In short, bringing the witch to life "objectively" is a contradiction in terms or, to put it another way, nonsense. Christensen therefore only really succeeds in his aspiration by being caught by the power of the witch-caught up in the contest of objectivity, reason, and its necessary and instrumental nega- tion. As we will show in detail in the introductory section to Part I of the book, there is a lengthy tradition in anthropology claiming (even in the same breath as disavowing it) that what lies at the heart of witchcraft's operations, its resistance, and ultimately the discourses of its depiction and explanation, is precisely this unattributable power.12 The fact that cinema uniquely brings such attributes (among others) to life has led us to pursue an analysis of Christensen's film as both an anthropological and cinematic object.13 Hdxan reveals the logical conundrum Favret-Saada defines. Any claim to knowing the witch implies being caught by her; any claim to objectivity in the face of the witch requires her disavowal as a "real" entity. This operational paradox drives Hdxan-and as we shall suggest, it drove efforts to locate and combat witches, and it served to shape similar aporias between sense and distance imbricated in the invention of modern hysteria in the nineteenth century and the "discovery" of the "native's point of view" by anthropologists in the early twentieth century. Taken in this context, Hdxan is perfectly empirical and is able to express the multiplicitous character of the witch, the hysteric, and the institutions that sought to formulate knowledge regarding the real character of such entities. Hixan: A Film Thesis The idea of Hdxan is relatively straightforward: in light of innovations in psychoanalysis and the biological sciences, Benjamin Christensen advances the thesis that the appearance of witchcraft in Europe during the late me- dieval and early modern periods was actually due to unrecognized manifes- tations of clinical hysteria and psychosis. Lacking the scientific knowledge and insight of the present age, the spectacular symptoms of hysteria (most 8 Introduction: What Is Hixan? often identified in women) were misattributed to the power of Satan and the condition of being in league with him. Deftly weaving contemporary scien- tific analysis and powerfully staged historical reenactments of satanic ini- tiation, possession, and persecution, Haxan creatively blends spectacle and argument to make a deeply humanistic call to reevaluate both the under- standings of witchcraft in European history and the contemporary treatment of hysterics and the psychologically stricken. In doing so Christensen takes on an anthropological disposition, offering Haxan as an expression of his own creative trials and as an empirical visual thesis to be tested in the world. While we believe the above synopsis to be accurate, it only begins to char- acterize the complexity, innovation, excessiveness, and influence of Haxan. As concerned as the film is with expressing a particular idea regarding the relation between witchcraft and hysteria, it differs from many of the docu- mentaries that come immediately after it, largely due to Christensen's ex- plicit understanding that any idea communicated in a film must be expressed cinematographically. Quite unlike the sober documentary ideal that John Grierson formulated at the end of the 1920s and elaborated through the next decade via the influential British social documentary movement, Haxan does not conflate expressing "the real" cinematically with simple "communica- tion." Displaying an affinity with scientists of the mid-nineteenth century (in particular, Jean-Martin Charcot, who himself claimed only "to record images"),14 Christensen very clearly aspires to "make nature speak." What distinguishes Haxan is the fact that its creator had no expectation whatso- ever that the real will simply "speak for itself." This distinction marks Hax- an's unreservedly singular approach toward conveying a particular truth about the world. Christensen's explicit refusal to either privilege a sober, gen- eralized, and abstracted form of "truthful" visual presentation in Haxan or to divest the film of its serious intent despite the excessiveness of its re- enactments was one of the primary sources of controversy at the time-and has continued to be an issue for the film in the decades since. Drawing freely and openly from a variety of representational strategies (scientific, historical, avant-garde, literary), Haxan's status as a truthful representation was/is therefore entirely unclear. In a fashion more extreme than other ambigu- ously "nonfiction" films released at roughly the same time (including The Battle of the Somme and Nanook of the North), Haxan explicitly demonstrated the highly relational and crosscutting influences that characterized "actual" Introduction: What Is Hixan? 9 cinematic documentation and representation of the world. As such, we ar- gue that one lasting effect of Haxan's reception in the 1920s was to energize a negative, conceptually dogmatic discourse that formed and hardened cin- ematic taxonomies, particularly the now-"commonsense" division between "documentary" (nonfiction) and "feature" (fictional) films. Christensen's own intense fascination-his subjectively fraught relationship to the witch he was so keen to objectify-was unacceptable to nonfiction filmmakers at the time. From our vantage point today, these ambiguous, disturbing, non- sensical elements are some of the strongest reasons for revisiting the film and reconsidering its place within the history of nonfiction and documentary cinema. We conceived the following pages at the interstice of cinema studies, film theory, anthropology, intellectual history, and science studies. While work- ing in this "between," we nevertheless treat Christensen's film as a film, purposeful in its artistic crafting, simultaneously image and object.15 Our approach is to think alongside Christensen as his film unfolds, scene by scene. Therefore, in the book we follow the narrative and structure of the film closely, dividing our text into two parts that retain and closely track the seven "cultural-historical" chapters (and thus the general structure) of the film itself. As the film has, by design, no overarching plot or main char- acters in the traditional sense, we feel this close, formalist strategy permits a full reading of the work. And as the film is unmistakably "biographical," that is, impossible to wrench from the events Christensen's life, these de- tails surrounding its production are discussed. We should be clear-others have studied Haxan in a variety of ways (see specifically the writings ofJohn Ernst, Jytte Jensen, Casper Tybjerg, Arne Lunde, and Jack Stevenson), how- ever none have fully explored the film through the theoretical framework that Christensen attempts to construct regarding the witch's material, invis- ible, mobile force.16 We insist that in order to appreciate Christensen's vision, it is not enough to think about the film, one must think with it-to allow oneself to be ensnared by the witch, as Christensen indeed was. We begin Part I by introducing the historical and epistemological con- texts within which Benjamin Christensen's Haxan emerged, specifically en- gaging the source material that guided the film's treatment of irrationality and "nonsense" to form a conceptual framework as a necessary starting point for our analysis. Turning to the film itself, Chapters 1 and 2 are concerned F-i 0 F V - i L~ ( I 45 i -II, N L Introduction: What Is Hixan? I1 with evidence-words and things, the theological debates that establish the groundwork for Christensen's thesis, and the forms of evidence found in the writings on witchcraft and sorcery at the time used by inquisitors to iden- tify or "name the witch." In these chapters we explore the cinematic forms evidence takes-the still image, the vignette, the reenactment, the facial close-up-which Christensen molds as visual strategies, making them tac- tile, twisting and bending the natural order at the level of the profane in order to show the ways in which the witch is realized. In Chapter 3, we explore the viral character of the witch. In particular, we highlight how witchcraft allegations were given signification in early modern Europe and how the increasingly formalized criteria as the basis of witchcraft evidence tended to multiply and spread, rather than reduce, the number of individuals who were suspected of being in league with the Devil. The epidemic atmosphere of accusation required and was enabled by forms of experimental practice, and it is through the emphasis on experimentation and evidence making that Hdxan finds common ground with other nonfiction films and the production of scientific images during the nineteenth and twentieth centuries. In this chapter Christensen demonstrates how witch trials were structured as an experimental process whereby inquisitors and laypeople would labor to establish proof of something they could not yet see but knew to be present. Here, Hdxan employs the well-established trope of sickness and diagnosis to demonstrate how maleficium was detected and confronted. In Chapter 4, we highlight the traces of Christensen's own demono- logical thinking. It is in this chapter of the film that Christensen begins to draw links between diagnostic strategies for identifying nervous disease and the forms of interrogation used by magistrates outlined in guides such as the infamous Malleus Maleficarum (1487). Christensen takes inventory of the elements that together created the witch stereotype in the early sixteenth century, including the Wild Ride, the pact with the Devil solemnized through sexual intercourse, cannibalism, and the cauldron. We begin Part II with an introductory section titled "1922," where we explore activities in other parts of the arts and human sciences around the time of Hdxan's creation that shed light on Christensen's deep commitment to the power of evidentiary thinking. In Chapter 5 we turn to the concepts and uses of sex, touch, and materiality in order to examine the personalized, 12 Introduction: What IsHixan? elaborated visualization of the witch stereotype realized by Christensen in the previous chapters of the film. In this chapter, primary accounts such as Johann Weyer's De praestigiis daemonium (1563) serve to reveal the complex nature of sensual explorations of the flesh through masochism and exorcism. In Chapter 6, the appearance of possession, ecstasy, and "insanity" re- veals the reach of demonic influence and introduces the important genres of transfiguration and metamorphosis. In this chapter of the film, Chris- tensen for the first time uses testimonies to further his cinematic thesis. The chapter showcases instruments and techniques of torture in order to high- light the highly charged expression of religious ecstasy and self-mutilation as part of his visual tableau. It is here that Christensen draws from his most inexhaustible resource: the neurological writings of Jean-Martin Charcot and his followers, especially the volumes they produced in the Bibliotheque diabolique that explicitly dealt with the relationship between witchcraft and nervous disease.17 In the final moments in his film, Christensen is clearly anxious to draw the threads of his thesis together at the close. He makes the most explicit overtures to show how those who were once identified as witches are now the objects of medical and social concern in modern life. Despite his effort, what results in the conclusion is something quite different. Christensen does not dispense with the witch as an aberration from the past haunting the unfortunate and superstitious in the present, but instead shows the potency of her various forms over time. In our postscript we use the ambiguity of Christensen's suppositions to consider Haxan's innovation and singularity within the silent period, the historical and social contingencies that sur- rounded the film, and its place within the history of cinema more generally. We seek to understand Haxan on its own terms. While this sounds straightforward, it is not without some blind alleys and contradictions. Our narrative is woven in and through Christensen's historical sources and the- oretical commitments, as well as into the technical realities and innovations of his filmmaking. We adopt the existing organizational structure of the film in an effort to show what drives Christensen's thesis, often in ways that per- plex as much as inform viewers. While we certainly seek to clarify rather than confuse, we feel it crucial to allow the film and its thesis to stand. It should be obvious by the end that we regard Haxan to be vitally important Introduction: What Is Hixan? 1 3 as a film precisely due to Christensen's ability to bring the witch to life while retaining the mysterious core that gives her life in the first place. Like Chris- tensen, we cannot avoid being caught a little by the witch, seeking at some level to only pass along the experience of being rapt by this work of cine- matic art in a manner proper to our own positions as anthropologists and human scientists. i Albrecht Diirer, The Poar Witchcs (1497). Courtesy of British Museum, London. PART ON E The Realization of the Witch  The Witch in the Human Sciences and the Mastery of Nonsense Beloved, believe not every spirit, but try the spirits whether they are of God: because many false prophets are gone out into the world. -I JOHN 4:I There is a largely unacknowledged historical tendency and predisposition within the human sciences with roots in much older practices of defining social facts and the discovery, interpretation, and the production of the real itself. In plain language, it emerges from a method that allows the researcher to sense, interpret, and eventually master forces that appear to be nonsensi- cal and yet are held to be essential to the reality of everyday social life. While such invisible forces have gone by many names, one can track a historical persistence of this epistemological concern with things that cannot be seen or logically interpreted but are nevertheless held to be present.1 One way of tracking this problem of the mastery of invisible forces has been offered by the literary scholar Jonathan Strauss, specifically with re- gard to the notion of the irrational as a privileged space in medical discourses in nineteenth-century Paris. Strauss argues that the role of irrationality and "nonsense" was a "legitimizing force" for medicine in that "the very incom- prehensibility of the mad created a mysterious and extra-social language that 17 18 The Realization of the Witch the rising medical profession could adapt to its own purposes."2 This kind of mastery is of course no news to anthropologists, who have claimed a sim- ilarly privileged space in the late nineteenth and early twentieth centuries through their understanding of the "nonsense" of "the native." The empiri- cal mastery of domains consigned to the illogical realm of human social life and in particular life in distant societies-formed the methodological basis that allowed the fieldworker "to see" unknown forces. From Malinowski forward, ethnology depended on exactly this process, as anthropologists forged a bond with the invisible and irrational as a methodological pillar. Anthropologists thus had to develop a battery of tests that could yield some felicitous information as to the "true" nature of unseen forces and their op- erations within empirical, real-world contexts. The heart of our argument in this book is that Haxan, in its curiously excessive attempt to produce a nonfiction film about the power of the witch, deploys an analogous approach and relies on very similar conceits for citing evidence of what is empirically "real" in the world. The attempt to secure evidence of forces felt but unseen is certainly not an invention of the nineteenth-century sciences of life and man.3 A clear conceptual link exists between the investigative techniques developed by sixteenth-century theologians and Church inquisitors in the face of what was understood to be a vast proliferation of the incredible, unbelievable power of Satan and emergent scientific fieldwork practices in anthropology and other social sciences in the late nineteenth and early twentieth centuries. While the systematic, empirical investigation of strange events, singularities, miracles, and other types of staple phenomena in preternatural philosophy predates Francis Bacon's The Advancement of Learning (1605), there is a method that emerges within the ensemble of human sciences proper to the science of man that is unable to expel these direct, necessary engagements with unseen and empirically unprovable forces.4 Although the credible status of such phenomena as real per se has been detached from these dis- ciplines, the status of these phenomena as dark precursors5 driving the inquiries taken through the signatures of anthropology and science serves as the focus of our engagement here. As such, we argue that anthropology as a science is predicated on rationally mastering invisible, irrational forces. Or, perhaps more precisely, anthropology emerges as a distinct human science from the desire to credibly master nonsense. Well versed in anthropological The Witch in the Human Sciences 19 literature regarding witchcraft, possession, and ritual, Benjamin Chris- tensen, too, demonstrates the desire to bring the invisible and nonsensical into view; although Christensen's medium was cinema rather than more traditional forms of ethnological record, Haxan nevertheless stands as one of the most powerful, unsettling expressions of the aspiration to produce evidence of forces unseen. Myths, Origins, and Methods Following what George Stocking has termed the "Euhemerist Myth" of anthropology6-that is, a rationalizing tendency to interpret mythology as historical event-we argue that the links between Christensen's Haxan and Bronislaw Malinowski's fabled definition of the methodological task of the anthropologist are undeniable. In the ur-text of this myth, Argonauts of the Western Pacific, Malinowski confidently identifies "the final goal, of which an Ethnographer should never lose sight": This goal is, briefly, to grasp the native's point of view, his relation to life, to realize his vision of his world. We have to study man, and we must study what concerns him most intimately, that is, the hold which life has on him. In each culture, the values are slightly different; people aspire after different aims, follow different impulses, yearn after a different form of happiness. In each culture, we find different institutions in which man pursues his life-interest, different customs by which he satisfies his aspirations, different codes of law and morality which reward his virtues or punish his defections. To study the institutions, customs, and codes or to study the behavior and mentality without the subjective desire of feeling by what these people live, of realizing the substance of their happiness-is, in my opinion, to miss the greatest reward which we can hope to obtain from the study of man.7 Although subjected to rigorous critique in the decades since its original publication in 1922 (the same year Haxan was released), Malinowski's direct expression of the desirable method and the underlying aspiration grounding this technique has never been definitively overturned within the discipline. To this day the paragraph quoted above serves as the distillation of method and disposition alike when confronted with the deceptively difficult questions 20 The Realization of the Witch "Who are you?" and "What do you do?" The assertion by anthropologists claiming to have assumed the "point of view" of another, not to mention the resulting ethical disequilibrium, has been rightly subjected to a series of stringent critiques over the years. But the idea that we should fully dis- pense with Malinowski's epistemological aspiration and regard interlocutor others as "Other" remains unthinkable within the discipline as well.8 This inconsistency has generally been resolved by one of two potential displace- ments: the first proposes that we detect the underlying structures framing "points of view," while the second aims to appreciate the meaning of social facts as a substitution for Malinowski's blunt demand to assume the simul- taneous position of the "social" scientist and the object of this science. What grounds Malinowski's claim that the fieldworker must achieve the cultivated, sensed point of view of another is a privileged relation to the un- known. This privileged relation must emerge through experimentation and through the ability to, in some fashion, test what is asserted to be real; in the anthropology of Malinowski's vision this test is a series of subjective trials9 subsumed within the rubric of "fieldwork." In this way, a discipline such as anthropology can legitimately claim kinship not only with other human sciences but also with the "hard" sciences. The tie between mastering what Strauss has termed "nonsense" and asserting scientific authority has strong links to transformations that occurred in the course of the "witch craze" in Europe, specifically regarding the terms of evidence within the overlapping institutional domains of science and law, both dominated by the- ology, to which we will return in the following pages. Certainly institu- tions charged with the task of discerning truth from falsehood have shifted dramatically over the centuries, yet the murmurs of this original theology remain audible in Malinowski and Christensen, even today. Haxan exists as a visual amplifier of these persistent murmurs. Malinowski's method requires certain presuppositions in order to be ef- fective. First, it presumes that the experiential disposition of the analyst is a legitimate and effective way by which one can begin to form an under- standing of a phenomenon otherwise held to be imaginary, fictional, or sim- ply untrue. Second, it turns on the principle that witnessing and testimony can concretely serve as evidence as to the reality of something otherwise beyond the direct experience of the researcher. In seeking to bring the invisible and nonsensical into the realm of ethnographic fact, Malinowski The Witch in the Human Sciences 21 explicitly recognizes the representational nature of this truth; only the testimony of the expert makes belief in such phenomena as real (in any sense) possible. Dan Sperber has pointed out that most religious beliefs follow the same representational logic.10 Since Luther's radical assertion that faith can only be a commitment to the representation of a truth, the explicit nature of this relationship has been a contentious element in Western Christian- ity's own efforts to discern truth and the nature of the world. Malinowski has thus only updated and secularized a much older epistemology dating back to precisely the period Haxan depicts. In the words of Joseph Leo Koerner, "Lutheranism is the original anthropology of 'apparently irra- tional beliefs.'"" As we shall see in Haxan (a quite "Protestant" work in many ways), this overriding "conviction in the utterly invisible" is not solely the concern of either theologians or scientists and hardly limited to the time of the Reformation and the subsequent witch craze. Realizing the Witch In the closing decades of the fifteenth century, it was clear to ecclesiastical and secular authorities in Europe that they were witnessing a crisis in the form of a proliferation of witches.12 The growing number of beings intent on the destruction of Christendom mirrored the growing power of Satan on earth, and for many, indicated an impending apocalypse. In more im- mediate terms for theologians, the seemingly viral proliferation of demonic power beyond the grasp of human experience, intuition, or thought required a radical change in the manner by which authorities could investigate and evaluate situations that involved invisible, supernatural powers. First appearing in 1487, in an atmosphere of fear and grave doubt, the notorious demonological text the Malleus Maleficarum (Der Hexenhammer or The Hammer of Witches) established a logical if not disputed relation be- tween investigative procedures, the constitution of evidence, and the asser- tion of fact during the period.13 Proceeding in a manner explicitly contrary to previous scholastic methods of ascertaining the nature of the real, the assertion of expertise in the Malleus by authors Henry Institoris (Heinrich Kramer) and Jacob Sprenger, while quite radical for its time, echoes to a startling degree much later statements to the same effect, including 2 2 The Realization of the Witch Malinowski's own assertions discussed earlier. The claim to expertise in the Malleus is phrased as follows: We are now laboring at subject matter involving morality, and for this reason it is not necessary to dwell on various arguments and explanations everywhere, since the topics that will follow in the chapters have been sufficiently discussed in the preceding questions. Therefore, we beseech the reader in the name of God not to ask for an explanation of all matters, when suitable likelihood is sufficient if facts that are generally agreed to be true either on the basis of one's own experience from seeing or hearing or on the basis of the accounts given by trustworthy witnesses are adduced.14 Institoris and Sprenger were actively responding to concrete fears of Europeans at the time. Their bold assertion of expertise in matters real but (often) invisible shares much with Luther's reply to the question of how might we see God: "Just as our Lord God is the thesis of the Decalogue, so the devil is its antithesis."15 Nothing troubled the soul of the late-fifteenth- and then sixteenth-century European as much as God's apparent absence in times of great change and strife. Forcing Satan and his followers from the shadows through an interpretive expertise over the concrete, secondary manifestations of his reality was often reassuring, relief for the pious believer on the brink of doubt. Heretics such as the Brethren of the Free Spirit, Waldensians, and Cathars managed God's absence without positing the em- brace of life that the Devil urges in binary opposition to that of the Good, albeit infused with the perilous dogmatism eschatology always brings.16 Most had no luxury to imagine such a world. As Satan's power appeared to grow (at least in the treatises of theologians) the problem of the Devil interfering with the most intimate communications with the Divine became acute.17 How does one know who really hears the prayers and entreaties of the faithful? Moreover, given the Devil's deceit and omnipresence, how does one really know who is speaking when prayer is returned? The paradoxical comfort the inquisitor offered was rooted in questions of theodicy in a world where the trappings of belief are every- where but where there is no incontrovertibly visible evidence of God's ac- knowledgment or answer to the prayers of the faithful. Thomas Aquinas had earlier raised this thorny problem of doubt: "It seems that there is no God. For if, of two mutually exclusive things, one were to exist without limit, The Witch in the Human Sciences 23 the other would cease to exist. But by the word 'God' is implied some limit- less good. If then God existed, nobody would ever encounter evil. But evil is encountered in the world. God therefore does not exist."18 Aquinas refutes his own speculative preposition through his famous five proofs of God's existence;19 demonologists of the fifteenth and sixteenth cen- turies were not so sure. For demonologists such as Institoris, Sprenger, or Johannes Nider, a third figure beyond that of "God" or "man" was required; this figure in concrete terms was the witch.20 Thus, the absent term in this understanding is shifted from God (although most could not claim to have directly seen God) to the witch, the chasm between God and man now it- self functioning as a kind of proof, a reassurance that the evil of the world can be explained through the various iterations of Satan's power.21 The Devil therefore serves to prove God's existence, the polarity reversed toward God's permission for demons to cause evil in the world and away from the nag- ging, perceived void where God is expected to be. As demonologists would persistently claim in the sixteenth century, God must exist because Satan is right in front of me!22 If human beings were slow to recognize divinity compared to malicious beings such as demons (after all, it was demons who first recognized the di- vinity of Christ, long before his disciples came around),23 then how could one confidently recognize the presence of Satan? By definition inquisitors would have taken the reality of witches and Satan for granted, yet the scope of demonic power authorizing these beings concrete reality in the world would have nevertheless struck inquisitors as unbelievable.24 Hearing the name of the witch in an accusation or a confession, bolstered by the details of truly sacrilegious and inhuman deeds, would still have been a shock to them and was very much subject to verification. Put differently: with the interweaving of learned demonology into the fabric of a dominant theology that ratified the sovereignty of God primarily through the worldly evidence of Satan's forceful opposition to that divine power, inquisitors believed that what was reported to them was possible; but it would be a gross misrepre- sentation to argue that inquisitors would not have then sought to empiri- cally verify such claims. Indeed, even within this style of reasoning, it was possible that individual accusations could be found to be spurious or false. The invisibility of the spiritual world was expressed as an essential given, but demonologists and inquisitors at the time still desired proof As doubt 24 The Realization of the Witch arose everywhere around them, the viral proliferation of the witch came to provide that proof. As numerous scholars of the witch trials have noted, the strategy of lead- ing the accused in her testimony was common during interrogations.25 In an effort to prove a particular instance of witchcraft had occurred, inquisi- tors often had to lead, goad, and viciously repeat the torture of "the witch" until a narrative was produced that at least partially satisfied the demands of evidence.26 For the inquisitor or witch hunter, it was never enough to sim- ply "believe." Rather, the interrogation under torture represented an experi- mental form of knowing in crisis.27 It would be absurd to argue that this style of interrogation was later simply reproduced in the more modern contexts of the human sciences or in early ethnographic studies such as Malinowski's pioneering work in the Trobriand Islands. Yet the truth value of a nonsensical confession made sensible has a strong connection to a series of truths regarding human belief, action, and social practices across a much longer historical arc than generally acknowl- edged. This link is perhaps even clearer if we shift our attention from the pragmatic humanism of Malinowski's approach to the ethnographic style of early French ethnographers such as Marcel Griaule. While rejecting the stark ontological difference asserted by Lucien Levy-Bruhl between the nonsensi- cal world of "primitives" and the science of the West, Griaule's own approach to ethnographic research developed in the 1930s betrayed an aggressive belief that "natives" could not (or simply would not) ever be able to produce a "proper" explanation of the forces around them or their own beliefs and mo- tivations in relation to these forces. They would lie, conceal, protect-and so wresting their knowledge from them, learning truth from lie, was essential to representing their reality in order to interpret it in its true picture: The role of the sleuth of social facts is often comparable to that of the detective or examining magistrate. The crime is the fact, the guilty party the interlocutor, and accomplices are all the members of this society. This multiplicity of responsible parties, the extent of the areas where they act, the abundance of pieces of evidence serving to convict appear to facilitate the inquest, but in reality they guide it into labyrinths-labyrinths that are often organized. . .. Not to guide the inquest is to allow the instinctive need that the informer has to dissimulate the most delicate points. . . . The inquest must be treated like a strategic operation.28 The Witch in the Human Sciences 25 Thus, while testimony was an essential tool for ethnographers of this school, the encounter between researcher and subject constituted a series of severe tests by which the researcher could gather the necessary empirical evidence in order to make a felicitous truth statement regarding what was "really" at play. While the nonsense to be mastered had shifted from the demonic, incredible forces at play for the inquisitor to the misguided tall tales of the native interlocutor, the logic of gathering evidence through a series of trials or tests is surprisingly durable between these investigative contexts.29 As authoritarian as Griaule's approach to fieldwork explicitly was, it was also consistent in its recognition of the struggle that lay at the heart of rais- ing testimony to the status of the "really real."30 Haunted by the possibility of deception, Griaule was more explicit in his recognition that any form of testimony (his or an interlocutor other's) requires a test or a trial in order for it to be elevated to the status of a fact. Avital Ronell captures this neces- sity when she writes, "A passion or experience without mastery, without sub- jectivity, testimony, as passion, always renders itself vulnerable to doubt."" Aspiring to an objective form of scientific knowledge that obviates this doubt, Griaule aggressively frames the scene of ethnographic encounter it- self as a kind of antagonistic trial whereby the ghosts and gods of the na- tives are forced out of the shadows and made concretely apparent to the senses of the anthropologist. This approach attempts a more delicate balance than its belligerent tone would lead us to believe. Griaule himself appears to acknowledge that the testimony obtained, while able to generate some understanding, will never resolve itself in a proof in the strict sense of the term. In essence, fieldwork in this context produces knowledge of hauntings that is itself haunted. As inquisitors also tacitly acknowledged, this par- adoxical haunting is what gives testimony its power of fact in the first place. If testimony were truly "certainty" or mere "information," as Jacques Derrida reminds us, testimony "would lose its function as testimony. In or- der to remain testimony, it must therefore allow itself to be haunted. It must allow itself to be parasitized by precisely what it excludes from its in- ner depths, the possibility, at least, of literature."32 The expertise that comes with wrangling the invisible and nonsensical very often is rendered visually (it should come as no surprise that the filmmaker Jean Rouch was one of Griaule's students). Yet strategies of visualization 26 The Realization of the Witch are hardly limited to a "French" approach in this instance, as commentators ranging from Clifford Geertz to Anna Grimshaw have noted the visual qualities of Malinowski's ethnographic writing, with Geertz going so far as to playfully term his output the imaginative result of "I-witnessing."33 Nor are such creative test results solely a phenomenon of twentieth-century so- cial science. The paradoxical necessity of an expressive element within an objective test in relation to what would otherwise be nonsense is evident in many of the examples of sixteenth-century visual culture that remain known to us today. For example, in Franz Heinemann's 1900 Rites and Rights in the German Past (a work that figures prominently in Hiixan),34 a woodcut shows an investigative technique deployed by inquisitors and witch hunt- ers: trial by ordeal. The woodcut depicts a crowd of people surrounding a bound woman as she is nudged away from the shore. Heinemann's image is similar to others, including a detail of a bound, naked woman undergoing the trial by water drawn from Eduard Fuchs's Illustrated Social History from the Middle Ages to the Present.35 The possible outcomes were few: if the woman floats she is clearly able to contravene the laws and God and nature and is therefore a witch or heretic; if she sinks, she has made no such pact with Satan and the judges proceed to thank God for her innocence (though she may have just as likely been fished out before drowning). It is important to note the role of procedural expertise that such ordeals required, as the trial by water here functions as experiment as much as a punishment, de- signed to reveal an otherwise invisible truth. It is clear that concerns about what was admissible as evidence of the real motivated such "trials" and served to frame the possible interpretations of their results. Testimony, experimental results, and expert inquisitorial inter- pretation together came to form an early version of the case study that, in turn, could be synthesized as evidence in service of accounting for variation that exceeded general laws regarding relations and phenomenon in the world. Individual cases came to serve as an effective strategy in providing analytic purchase for phenomenon that were otherwise invisible to even the dis- cerning eye of the expert. It is this act of taking a single, natural object (the case) and abstracting its qualities to describe phenomena that we find in the clinical work of Thomas Sydenham in 1668, something the historian Philippe Huneman further traces through the psychiatry of Philippe Pinel, which we can extend even further in the famous cases histories of Sigmund The Witch in the Human Sciences 27 Freud.36 Following this thinking, close analysis of salient individual cases would make hidden tendencies visible in practical, "natural" terms, a char- acteristic that made the method attractive to artists and scientists alike who were seeking to move away from a reliance on metaphysics.37 It is no acci- dent that in the nineteenth century the clinical photography of Guillaume Duchenne de Boulogne and Jean-Martin Charcot, and the chronophotog- raphy of Eadweard Muybridge and Etienne-Jules Marey, exerted a formal, expressive influence that often exceeded the limited audience of scientific peers.38 Charcot and the Bibliotheque diabolique It is easier for superstitious men, in a superstitious age, to change all the notions that are associated with their rites, than to free their minds from their influence. Religions never truly perish, except by natural decay. W.E.H. Lecky, History of the Rise and Influence of the Spirit of Rationalism in Europe (1865) The physician seeks to fill what he knows with what he sees. He is in search of the manifestation of his nosological concepts. Mobilized by attention, he considers the deployment of a knowledge in the new and visible form of an appearing. In short, he discovers without learning. Michel de Certeau, The Possession at Loudun (1986) There are witch confessions that are insane. This fact was recognized by many skeptics in the sixteenth century who, while acknowledging Satan's unquestioned power, cast doubt on the truth value of unlearned witnesses to this invisible power and the theological frameworks deployed by inquisi- tors validating their interpretations of how such reported acts were consistent with the authoritative discourses of the Church or (in the case of Prot- estants) of the gospels themselves. Possessions set the stage for the explicit medicalization of the mobile, invisible forces that experts had been strug- gling to master, explain, and take measures against-a new mode that is equally didactic and forensic. In such well-known incidents as the posses- sions among the Ursuline nuns of Loudun from 1634, we find an increased medicalization of the invisible that, over the course of a long transition,39 28 The Realization of the Witch reverberates through the medical and human sciences of the nineteenth century. The quote from Michel de Certeau's The Possession at Loudun refers to the physicians called upon for aid in the wake of the Church's failure to exor- cise the demons haunting the nuns at Loudun. Jean-Martin Charcot was of course well aware of the enduring relationship between religious sense and medical knowledge underscored by de Certeau's statement.40 Thus Charcot did not so much invent as inherit a perspective on the relation between re- ligious ecstasy, magic, witchcraft, and "nervous disease." He and his students collected and published historical accounts under the title Bibliotheque diabolique-a series in which many texts and treatises from the sixteenth century were reproduced, including Soeur Jeanne des Anges, superieure des Ur- sulines de Loudun: Autobiographie d'une hysterique possedee,41 Science et miracle: Louise Lateau ou la stigmatisee belge,42 and La possession de Jeanne Fery,43 all of which were accounts contemporary to Johann Weyer's De praestigiis daemo- num et incantationibus ac venificiis (On the Illusions of the Demons and on Spells and Poisons, 1563), and follow the Malleus by nearly a century. The books in the Bibliotheque diabolique indexed as much as clarified the link between witchcraft and hysteria for Charcot and his followers. In Science et miracle (a 1875 book on witchcraft, faith healing, and demonic possession), Bournev- ille begins by warning his readers that the "profound time of ignorance in the Middle Ages" has been prolonged into "modern society,"44 appealing to an appraisal of history in service of a project for scientific modernity. For Bourneville and those producing work in the Bibliotheque diabolique, case studies were meant to demonstrate the precariousness of misrepresentation and the consequences of ignorance. The errors of demonologists and exorcists were rooted in what was char- acterized as the mistaken conceptualization of their object of investigation. Yet, accusations of error and superstition aside, the procedural elements of the investigations collected in the Bibliotheque diabolique bore a startling re- semblance to those undertaken by Charcot and his students, particularly in their studies of hysteria. While it is certain that witch hunting and the ex- orcism of spirits in the sixteenth century were hardly interchangeable with clinical studies of nervous illness in the nineteenth century, the conceptual scaffolding of the emergent science that they were creating bore more than a passing resemblance methodologically to these now antiquated forms of The Witch in the Human Sciences 29 inquiry. More than anything, the continued fascination with the secondary, visible effects of primary invisible forces demonstrated that the discernment of spirits, like the diagnosis of nervous illness, involved a long-term labor of social interpretation that required the mutation of old categories and the cre- ation of new ones.45 The contentious fragility of these endeavors revealed by these historical accounts served as salutary lessons for Charcot and his followers; the fact that their own conceptualization of their object largely retained its status as the insensible, invisible, outside forces that served as the focus of the work in the Bibliotheque diabolique was an irony that largely escaped comment. It is impossible to overstate the influence the works produced by the students of Jean-Martin Charcot had on the trajectory of inquiry across the human sciences of the late nineteenth century. Charcot himself collected artistic and historical materials on the relation between witchcraft and hys- teria, which he presented under the title Les demoniaques dans Part, published by the Academy of Medicine in 1887.46 Charcot's famous students such as Georges Gilles de la Tourette and Paul Auguste Sollier attempted "to trace the hysteric through history" with "sincerity and veracity." They concerned themselves not only with prevailing social attitudes toward "misdiagnosed" hysterics of the early modern period, but also with clinical attention to the physical manifestation of hysteria found in images and writings on the time. In Gilles de la Tourette's Trait6 clinique et therapeutique de l'hysterie d'apres l'enseignement de la Salpetrire47 and Sollier's Genese et nature de l'hysterie, re- cherches cliniques et experimentales de psycho-physiologie,48 we find detailed index- ing of symptoms such as religious fervor and stigmatization alongside psychosomatic indicators such as blue edema or swelling with local cyano- sis, and "autographic skin" that would appear intensely red after touch-all physical signs of witchcraft attributed to earlier centuries.49 As A. R. G. Owen points out, the word "medicine" finds its etymological roots in sorcery, after Seneca's tragedy of Medea, whose betrayal and revenge leads to the murder of her children.50 Even within the writings attributed to Hippocrates, "the sacred disease" (epilepsy), erroneously perceived as re- sulting from hostile magic, could be reconsidered in terms of individual physiological disorder.51 Yet "hysteria," itself from the Greek for "uterus," seemed to hold a special place in the moral imaginary.52 In the century be- fore Charcot's famous neurology clinic at Salpetriere, nearly ten thousand 30 The Realization of the Witch women were kept there at the second Bastille, La Force Prison. These were destitute women, the insane, "idiots," epileptics, and Parisian society's "least favored classes."53 The special susceptibility of women to witchcraft mir- rored the "feminine weakness" associated with the hysteric, exacerbated by low social status. It was nuns and devoted female members of the Church that raised special concern when "possessed" by unexplained forces of de- monic or psychic origin. Like later diagnostics of hysteria, the discernment of spirits was at its root a discernment of female bodies as such.54 Ulrich Baer points out that what Charcot created was a tableau vivant transformed into a tableau clinique-a hysterical reliving of the original symptom and reframed trauma in an attempt to suspend the two tempo- ralities (real and reimagined) in the same image.55 This "reliving" is pre- cisely what Freud and Breuer meant to produce through hypnosis in their studies on hysteria, to isolate the mechanisms of hysteria and the surround- ing symptoms of catharsis and dementia, their most famous cases being Anna O., Frau Emmy von N., and later of course Freud's own Dora.56 It's no wonder that one element of fascination with hysteria was its "look"-its aesthetic link to forms of possession. Traugott Oesterreich, who published his Occultism and Modern Science in 1923, traced a similar path toward an aesthetic ideal of possession beginning with the Acta Sanctorum in the Catholic Church.57 In countless accounts of possession, we find descriptions of demons speaking through the mouths of girls, as well as possession mani- fested through "external signs" of a new physiognomy, particularly in the face. We also find identical descriptions of voice, personality, and facial change in Pierre Janet's Nevroses et idees fixes.58 In his studies of medical psychology, Janet describes the aneasthesias, amnesias, subconscious acts, somnambulisms, and fixed ideas all associated with possession-including a case of spontaneous abortion brought on by powerful thoughts of a previ- ous abortion.59 Oesterreich's observation that Catholic religious ceremonies to "treat" the possessed worked to heighten the intensity of possession, re- flects the same heightening that Charcot himself considered of the subjec- tive state of the hysteric in his clinical theater.60 In sum, what is critical to note in all these cases is an overriding desire to gain some empirical purchase over forces openly acknowledged to be invis- ible and insensible. The tension animating each of these domains lay in the conceptually arranged chasm between outer and inner states. For exorcists, The Witch in the Human Sciences 31 building on the techniques of inquisitors and witch hunters, possession acts as the bridge across this chasm. Two centuries later, neurologists and psy- chologists construct the same scaffolding. Anthropologists by the time of Malinowski attempt to close this aporia by sympathetically occupying the very inner space of their interlocutors, with the witches, spirits, and demons no longer explicitly the target of the inquiry, but rather fieldworkers as truth-tellers returning from the dark corners of the real. In the midst of these efforts, Haxan, aspiring simultaneously to the status of science and of art, sought to force everything into plain view. As powerful and forward-looking as Haxan truly is, the specter of sheer nonsense as its real object remains. And yet we still today hunt ghosts and witches, fueled by a desire operationalized in a method of being close enough to something to sense it, because even objective scientific mastery demands a closeness to things unseen, unprov- able, indeed nonsensical, yet there. O N E Evidence, First Movement: Words and Things The perfect photoplay leaves no doubts, offers no explanations, starts nothing it cannot finish. -HENRY ALBERT PHILLIPS, The Photodrama (1914) The technical structure of the archiving archive also determines the structure of the archivable content even in its very coming into existence and in its relationship to the future. The archivization produces as much as it records the event. -JACQUES DERRIDA, Archive Fever (1998) In the beginning there is a word. That word is "Hsxan." Benjamin Christensen's biblical echo is intentional. From the first frame of Hdxan, Christensen is seeking to dismantle the conventional cinematic image. This is an image of a word. In light of what is to follow, the formal conventions of the silent film by definition destabilize any easy relation to the object "Hsxan"; it exists multiply. Already reaching into his source ma- terial, Christensen borrows Italian inquisitor Zacharia Visconti's categories of language to show us how the word relates to meaning, expressed in the distance between the thing and the thing signified. Visconti designated this "the language of the voice," the language proper to humans.1 Yet, in a silent film there is no obvious voice. Certainly, a printed word occupies the do- main of language for Visconti, but for this to be formally consistent the word requires the syntax that would allow the reader to insert her inner voice, the memory of a voice, in order to make this so. "Hsxan" appears to lack this syntactic force at this opening instant to be properly a statement. 32 Words and Things 3 3 "Hsxan"-the witch-appears to be an impossible object. In Visconti's schema, this word also appears to speak "the language of the mind." This is a language the inquisitor reserves for angels, a language resulting in non- statements. From the very beginning, there is no claim made about the witch-no question is asked. The witch is simply announced. In an instant, "Hsxan"-the witch-is there and this is all. "Hsxan" is simultaneously a word, an image, and a thing. Benjamin Christensen makes every effort to craft a witch that is real to us. It is a grand ambition. Playing with the ontological fluidity of a cinematic image, the di- rector expresses himself through an image-world that seems entirely of his own creation. "Hsxan" in the opening moment of the film is a monad, con- taining the totality of this world in its most basic element. Not just a word, the Word. Visconti reserved this language, "the language of things" to God alone; yet for scientists and filmmakers, it is the language of things that holds the greatest appeal. To the World a Witch Christensen's first task is to open the world of the witch to the film's audi- ence. He does this by immediately following the word with a preposition, albeit still denying us the calming language of voice that is proper to us. This preposition, denoting both agency and possession, comes in the form of a face. His face. The commanding, scowling face of the director stares out at the camera. Christensen's film will make full use of this art of metoposcopy. Dating back to Girolamo Cardano and the Renaissance, metoposcopy defined the operation of reason as the weaving together of images in the mind. In turn, the expression of reasoning was to be found on the face (a proto- cinematic theory of the relation between image and thought if ever there was one).2 Christensen's face is one of many revealed; these faces-of the old woman, of the ecstatic nun, of the novice sorceress-will be offered as primary evidence of the power of the witch and the logic of demonological thinking. It is telling that Christensen's face is the first shown, not in order to place his seal of authorship, but as a way to assert to his audience that it is his argument that resides in the foreground. This is no ordinary film. It is not merely entertainment. Hdxan is a thesis. 34 The Realization of the Witch Benjamin Christensen, Hxan, film still (Svensk Filmindustri, 1922). After this dramatic beginning, Christensen provides some immediate re- prieve through a scarcely noticeable addendum to the opening title card: "A presentation from a cultural and historical point of view in seven chapters of moving pictures." Claiming a reassuring authority, Christensen now sig- nals that he intends to enlighten us in the manner of a professor giving a lecture. The technology of the motion picture is not simply a medium here; in the service of Christensen's thesis, it is a precise, deliberate method. The title cards that follow identify the director, the cinematographer, and turn the audience's attention toward the list of sources for the film distrib- uted as part of the original program (which has been reproduced in the back matter of this volume). Like any respectable scholar, Christensen indexes himself through his sources. Yet his mode of citation is unambiguously rooted in the formal elements of cinema and the image rather than texts, and is ultimately put to different uses from that of the historian or human scientist; this difference will constitute the focus of our own analysis in this chapter, as we move through his textual materials and the production of his images shot by shot. In short, Christensen makes sure the audience Words and Things 3 5 knows that it took three years to research and produce his visual thesis. As with the word and the face, this is stated abruptly for the benefit of context. More title cards follow, filled with an authoritarian, first-person tenor. Lacking any established provenance for a voice-of-God tone that would only later become standard in the Griersonian documentary mode of the 1930s, Christensen takes it upon himself to invent this voice. The common sug- gestion that Luis Bunuel first generated this instrumentally impersonal tenor in Land without Bread (Tierra Sin Pan, 1933) is off by a full decade, ignoring the fact that silent films were anything but silent.3 The director begins in this voice by establishing the witch as a chapter within a much longer con- stellation of practices, discourses, traditions, and institutions. This is em- pirically correct, as scholars from Gaston Maspero to Stuart Clark have emphasized in their own studies of the witch.4 Among many others, Rich- ard Kieckhefer has demonstrated how the long history of practical natural magic was enfolded into the specificity of European witchcraft in the late Middle Ages.5 These findings have only taken root in the historical debates on witchcraft since the 1970s, which Christensen anticipates by some fifty years. It is at this point in Haxan that Christensen gives us an image of the witch. It is a well-known woodcut that first appeared in Ulrich Molitor's Von den Unholden oder Hexen (1489), at the dawn of the witch hysteria in Europe, de- picting two women feeding a boiling cauldron. Many of the stereotypical visual characteristics of the witch are not yet established: the age of the women is difficult to determine and they are far from the withered old crones we see later in Albrecht Diirer and Hans Baldung Grien.6 Yet they are un- mistakably witches. Their boiling brew evaporates into the air, appearing to cause a storm. Drawing on a trope that would instantly signify "the witch" from Shakespeare's Macbeth to the present, Christensen introduces the viewer to the subjects of his film via a classic example of the maleficium that people greatly feared from witches in the early modern period. Christensen carefully limits what we can see of this image, narrowing the visible edges of the shot into a severe vertical line bisecting the screen. The shot is abrupt, barely onscreen for a few seconds before the intertitles return. Our focus is taken to the statement that primitive men "always" con- front the inexplicable with tales of sorcery and evil spirits. This is obvious 36 The Realization of the Witch hyperbole, but not entirely out of step with the evolving scientific explana- tions of the time regarding the origins of human society. Echoing E. B. Tylor's argument that civilization always begins with the imaginative, supersti- tious responses of humans to a world they do not yet understand, Chris- tensen then shifts to consider the power of belief.7 Haxan at this stage appears to be aspiring to Max Muller's dream of presenting an objective, empirical "science of religion."8 Interestingly, the next image takes us to "imaginary creatures" thought to cause disease and pestilence in ancient Persia. A row of six human-animal hybrids confronts the viewer. Christensen immediately divulges his sources for this claim, citing Rawlinson9 and Masperoo as authorities that trace the European belief in witches back to antiquity. Several shots of monstrous hy- brid demons, drawn from Maspero, follow. Christensen is operating in a firmly rationalist mode here, linking these monsters to "naive notions about the mystery of the universe" held by ancient people. A re-creation of Egyptian astrological notions of the nature of the world immediately follows. This is the first explicit set to appear in Haxan, depict- ing (according to Maspero's information, the intertitle asserts) a world of high mountains, stars dangling from ropes, and a sky supported by strong pillars. A nameless assistant out of frame helpfully draws the viewer's atten- tion to the important details. As with any Universalist approach, Christensen traverses time quickly in the presentation of his thesis. No sooner have we glimpsed this scale model of the Egyptian cosmos than we are catapulted into the folklore of early modern Europe. Perhaps the singular feature of the witch craze in Europe is bluntly stated when Christensen informs us that the generalized evil spir- its of ancient times are transformed into devils by the fourteenth century. Cutting from one to another, four iconic images of devils particular to the period flash across the screen, the film stock tinted an ominous, rusty red to heighten the effect. These devils lived at the earth's core, Christensen tells us, with the earth believed to be a stationary sphere in space surrounded by layers of air and fire. Beyond the fire lay moving celestial bodies, ceaselessly rotating around the earth with the fixed stars far above and, "in the tenth crystal sphere," sits the Almighty and His angels, keeping the whole celestial system in mo- Words and Things 37 "Christ in Limbo," from Evangelicac historiac imagines, after Bernardus Passerus (1593). Courtesy of British Museum, London. tion. Intercut title cards offer explanation before Christensen helpfully re- veals a working model of this cosmology, in this case drawn from Hartmann Schedel's Liber Chronicari71,1" slowly pulling back the iris to reveal the me- dieval universe that he is has described. This moving representation of a terra-centered universe resembles the elaborate wonders found in Baroque wunderkammer meticulously assembled by the German elite at the time. It is an effective use of parallel editing to bring this lecture, delivered in text, to life in a visual manner. While not explicitly designating the scene as such, Christensen is visu- ally marking here what Frances Yates has called, after Cornelius Agrippa's handbook of the same name, "the occult philosophy" of the Renaissance.2 The "rediscovery" of a large literature in Greek attributed to the name "Hermes Trimesgistus" by Marsilio Ficino (1433-1499), Giovanni Pico della Mirandola (1463-1494), Henry Cornelius Agrippa (1486-1535), and most powerfully of all, Giordano Bruno (1548-1600), galvanized a critique of the mainstream Church and represented a strong effort on the part of an elite 38 The Realization of the Witch group of scholars for an energized spirituality rooted in an Egyptian-derived wisdom handed down prior to that of the Old Testament. The core of Re- naissance Hermeticism was a deep concern with astrology and the occult sciences, the secret essences of natural things, and the sympathetic magic that was made possible for those who mastered such essences and their relations to one another.13 In short, the writings of Hermes Trimesgistus progressively provided the foundation for Ficino's relatively mild natural magic, Pico della Mirandola's Christian Cabalist, Agrippa's Christian magus, Tommaso Campanella's (1568-1639) utopian City of the Sun,14 and eventu- ally Bruno's full-blown Hermetic-Cabalist philosophy that sought through the power of astrology and magic to bypass the Church altogether, "oper- ating" in such a way that the skilled magi could reach the Divine directly. The fact that, starting with Ficino's Latin translations of the Corpus Hermeticum,15 Renaissance Hermeticism, while not widespread or necessar- ily revolutionary, was based on a massive historical error in determining the provenance of the texts is relatively unimportant for Christensen's thesis here.16 This does not blunt the historical accuracy of the connection Chris- tensen is making between what was presumed to be "Egyptian" wisdom re- garding the nature of the world and what he understood to be a religious tumult that arose out of this challenge in the decades preceding his witch. A full account of the rippling effects of the Hermetic-Cabalist tradition well exceeds Christensen's purpose here. It is clear, however, that the scrupulously mathematical astrology of Girolamo Cardano (1501-1576)17 and the rigor- ously empirical studies of the natural world demanded by Bruno's attempts to operate as a magus paved the way for the science of Newton and Coper- nicus and for a new metaphysics to emerge that, over time, would come to be credited as the precursor to the modern scientific self.18 The visual refer- ences to this tradition that flash by in Haxan's first chapter argues in its own way for the central importance of this tradition well before scholars such as Yates and others19 revolutionized our historical understanding of this tradi- tion in the middle of the twentieth century. It is also relevant to Christensen's thesis that the violent refutation of the Renaissance magi was a crucial ele- ment in the battle with witches and Satan that demonologists and inquisi- tors took up in the sixteenth century. Jean Bodin's De la demonomanie des sorciers (1580) is an excellent example of the relation, as Bodin excoriates key Renaissance figures such as Mirandola and Agrippa as satanic precursors to Words and Things 39 the scourge of witchcraft that he felt was plaguing the faithful at the time. While often lost by modern scholars "in a hurry" to get to the details of witchcraft emblematic of demonological texts such as Bodin's, it is commonly the case that such anti-witch treatises begin with attacks on Renaissance magic and the Hermetic-Cabalist tradition that authorized it.20 The stakes were quite high, as Giordano Bruno's execution demonstrated. In the six- teenth century, Hermetic magic and Cabalism became associated with the notion of "superstition" for Protestants and Catholics alike; for both reform- ers and counter-reformers by the close of the century, such superstition was a crime.21 What appears elliptical in its presentation in Haxan is Chris- tensen's reference to the twisted relationship between "Egyptian antiquity" and the witch trials in sixteenth-century Europe. Christensen advances the analysis of visual culture in Haxan in the next scene, in an extended set piece devoted to the close examination of a minia- ture from the twelfth-century manuscript Hortus deliciarum. Tacking back and forth between the intertitle lecture and the careful consideration of de- tails from the painting (again, with offscreen "pointers" to direct our gaze), the gory, elaborated reality of hell produced by the nun/artist Herrad of Landsberg jumps to life onscreen. It is clear that Christensen is overreach- ing by attributing a largely cohesive image of hell to a period when the na- ture of hell's location and "topography" was a subject of fierce theological debate. Although the early image corresponds to well-known later depictions of hell in literature (Dante) and art (Bosch) that filmgoers in 1922 could have reasonably been expected to know, Christensen's lecture strategically ignores debates and alternate conceptions of damnation that existed in the fifteenth and sixteenth centuries in Europe. For example, Sylvester Prierias's influen- tial De strigmagarum daemonique mirandis22 argued that fallen angels lived in the air, ungrounded and shapeless, manipulating the physicality of this air to act through witches and the wicked on earth.23 While Christensen gives ample attention to the manipulation of air and environment later in the film (largely through the Wild Ride to the Sabbat), his authoritative approach is occasionally strained in close examinations such as the one he offers regarding the Hortus deliciarum miniature. Despite the serious purpose Christensen asserts for Haxan, he appears to realize that spectacular moments of titillation are also necessary to carry through his visual thesis. This is evident in the following scene where a 40 The Realization of the Witch Herrad of Landsberg, Ilortits deliciarum miniature as it appears in lixan, film still (Svensk Filmindustri, 1922). working mechanical presentation of hell is revealed-the first real scene of movement in the film. The title cards suggest that Christensen "found" this visual machine dating to antiquity. It is unclear if the billowing smoke that at times completely obscures the actual mechanism onscreen (a rare tech- nical misstep) is part of the workings of the apparatus or Christensen's own attempt to heighten the fiery terror of the scene. Either way, it works to am- plify affect more than further analysis, leaving the title cards to offer a ge- neric explanation that "during the Middle Ages, devils and hell were considered real and constantly feared." After lingering on this spectacular hell device, Hiixain returns to unfold- ing the necessary facts of witchcraft to the audience by moving to a shot of a woodcut depicting novice witches signing a pact with the Devil. Correctly following the procedural descriptions of famous witch-hunting texts such as Institoris and Sprenger's Mallens Halefcaruzm24 and Nider's Formicarius,5 Christensen emphasizes the essential agency of individuals in making a pact Words and Things 41 with Satan. The woodcut shows a novice witch being raised into the air by the Devil, foreshadowing the procedural reenactments that follow later in the film. Returning to his scholastic mode, Christensen notes that these im- ages are drawn from Gustav Freytag's "A German Life in Pictures."26 Two additional woodcuts from this source immediately follow; one depicting a witch milking an axe handle and the other showing a woman bewitching a man's shoe. While the ability to manipulate and pervert inanimate objects is empirically consistent with the historical record on witchcraft, Chris- tensen offers no additional commentary or explanation of these bizarre images. Christensen's visual narrative moves to the second essential element of medieval witchcraft: the Sabbat.27 Molitor's Female Witches Acting Together woodcut (ca. 1493) is used to visually illustrate the gathering. Not linger- ing on the Sabbat itself, the film proceeds to witches acting in groups in order to cast spells on a cow, an entire village, and an unfortunate person who has fallen ill, drawn from Bourneville and Teinturier's Le sabbat des sorciers.28 Throughout this sequence of maleficium, Christensen emphasizes the arcane symbols used to ward off spells and the fact that the sick person is shown naked, a practice "that was habitual in the past." This vaguely salacious detail is emphasized further in the rapid succession of images finally revealing a group of nude women dancing around a demon at a Sabbat. Though not explicitly stated as such in the title cards, a palpable sexual dimension has crept into Christensen's thesis. This element will be promi- nent throughout the film, depicted here in images of women "sneaking away" to attend Sabbats. Using woodcuts "passed on" to Christensen by "the French doctors Bourneville and Teinturier," the viewer is then taken through the typical rituals of a "secret satanic rite"-the abjuring of the Church by des- ecrating the cross, Satan renaming his initiates, and the horrific ceremonial banquet of the Sabbat. Christensen notes that the food for these banquets was often prepared using the corpses of executed prisoners (though in fact it was babies), continuing with an image found in Molitor, reproduced in Le sabbat des sorcieres. Saving the best for last, witches (both male and female) are shown kissing Satan's anus as a sign of devotion. The image is a climax of sorts, and Christensen's textual accompaniment simply notes the activity without additional embellishment. 42 The Realization of the Witch Molitor's Female Witches Acting Together as it appears in Hixan, film still (Svensk Filmindustri, 1922). The Vicissitudes of Truth Telling in Early Cinema Christensen's peculiar strategy in opening Il/xan becomes apparent as the film progresses. Grounded in the realism of nascent nonfiction filmmaking, the director establishes his authority on the basis of citation. Inverting the typical structure of the monograph where the notes and sources would come last, H/ixan visually grounds itself in citable evidence from the start. There are historical reasons for why this is done. Although hailed from nearly the mo- ment of its invention as an instrument for "recording reality," the value of cinema as a vehicle for "telling the truth" about the world was increasingly regarded with suspicion by professional historians and social scientists in the early 1920s. Dominated by the "actuality" film, where the provenance of the images and the "historical tracks" of the observer were often obscured or even erased, purportedly nonfictional visual media were increasingly being judged inadequate to the tasks and protocols of the serious scholar at the Words and Things 43 time of Haxan's release.29 Michael Chanan has summarized this period in the history of nonfictional filmmaking as follows: When documentary was not yet documentary (but then fiction wasn't fiction yet either), when the medium was mute and each film ran only a minute or two, moving pictures hardly amounted to more than a miscellany of visual tidbits, which made no demands on literacy and thus spread easily and rapidly far and wide. The world on the screen exerted a magical attraction but remained anecdotal and predominantly iconic. In terms of public discourse, it was practically inarticulate, other than to reinforce already stereotypical images or create some new ones; in short, intensely fascinating but apparently ill-adapted to serving intelligent purposes.30 It is not as though scientists, journalists, and others devoted to making nature speak did not give filmmaking a try. In the waning years of the nine- teenth century, anthropologists such as Alfred Cort Haddon, Walter Bald- win Spencer, and Frank Gillen were already using the new technology to fashion, with mixed success, proto-ethnographic films. Charles Urban founded the Unseen World series in 1903, merging the technologies of the microscope and the cinematograph to attempt to unlock the secrets of nature at its most minuscule level. Films such as Attack on a China Mission Station (1900), Hunting Big Game in Africa (1907), and With Captain Scott, R.N., to the South Pole (1912) sought to bring the immediacy of news headlines to life onscreen. State-sponsored war propaganda generated during the First World War, including The Battle of the Somme (1916) and With Our Heroes at the Somme (Bei unseren Helden an der Somme, 1917) mutated the desire to see far-off contemporary events through visual meaning-making machines that demanded not only attention but belief. The fact that these films nearly al- ways made this demand by staging, as real, reenactments of purportedly real events only added to the early suspicion of cinema's ability to convey unvarnished, objective facts.31 Even for films not surreptitiously staged, the reliance on actualities of iconic cliches, giving the viewer what they largely expected to see, proved to be a serious problem for those who wished to con- vey the complexity and depth of the world and of nature.32 The issue, widely discussed well before John Grierson's proclamation of the "documentary value" of Robert Flaherty's Moana in 1926, concerns the relation between a fragmentary visual artifact drawn "from life" and the 44 The Realization of the Witch truth value of any such fragments. Ultimately, this issue hinges on mimesis. What sorts of filmmaking practices can felicitously mimic life as such? Grierson's own elaboration of documentary recognizes this in asserting that the filmmaking form is the "creative treatment of actuality." Grierson was not the first to conceptualize the matter in this way, as Brian Winston shows that the Polish writer Boleslaw Matuszewski stated the issue in these same terms as early as 1898.33 Crucially, mimesis was not only permissible for writ- ers such as Matuszewski and early documentarians such as Edward Curtis; it was indispensible in the creation of valuable documentary works. Thus, a film such as Curtis's In the Land of the War Canoes (a.k.a. In the Land of the Headhunters, 1914) adhered to prevailing standards of expressing the real not despite its status as a reenactment but because of it. The truth of Kwakiutl (Kwagu'1) life is evident through the spirit of Curtis's expert cinematic ex- pression of what that life is, just as the reality of war was only truly evident to viewers through gaining a sense of the fighting as re-created in otherwise opposing accounts of the truth in the British and German Somme films.34 Later accounts by film historians positing "fact" and "fiction" as oppositional binaries arising out of the earliest approaches to filmmaking were further exemplified by pitting the "realist" Lumiere against the "fanciful" Mdlies within a crypto-structuralist origin myth that falsely represented what "doc- umentary" meant to pre-Griersonian filmmakers.35 The "ahuman" witness of the camera is not enough, as this merely produces a blind sight that can- not, on its own, educate, enlighten, or even fully record the real in any ideal manner. This is not the first time that the gap between witnessing and the real has erupted in European history. As Hiixan demonstrates, the question of evidence occupied inquisitors and theologians long before the invention of cinema. Playing on the fact that, while the traces serving as evidence are quite different, a larger ontological issue binds them across the centuries, Chris- tensen takes the unique tack of assuming the role of the art historian in this opening section of the film. This is a risky strategy, particularly given the static nature of the materials on display, but it does allow Christensen to shift the locus of the empirical to the materiality of images accepted as histori- cal. Taking up this position in the opening chapter of Hiixan also allows Christensen to have it both ways, in that he can simultaneously confront the viewer directly in the manner of an earlier cinema of attractions while Words and Things 45 also preparing viewers for the "diegetic absorption" that was coming to dom- inate the grammar of cinema in the 1920s.36 Given the impossibility of filming witches several centuries "after the fact," Haxan creates present-day empirical images from artifacts of the time. Yet this analytic position does not guarantee that the images will be "brought to life" in any way. To the contrary, the vivisection of the historical image would tend to produce the same outcome that any vivisection would: death or deformity. Thus the risk, quite evident throughout the first chapter of the film, is that the presumed pastness of these images, their "deadness," will subvert the appearance of life that distinguishes cinema from other visual forms such as photography, painting, and printmaking. How well Christensen is able to elide this dead- ness is open to debate; undeniably, many viewers experience the opening minutes of the film as a plodding exhibition of "pictures of pictures." This reaction notwithstanding, the strategy of "reimaging" is methodological and intentional, an acknowledgment on Christensen's part that for a very long time "knowledge" in European terms consists first and foremost of "recita- tions of the known."37 While the opening chapter of Haxan may test the patience of the viewer, the logic of Christensen's visual strategy in this section becomes clearer as the film progresses. The director is laying a foundation for what comes next, though he is quite sensitive to the fact that a visual thesis demands a different relation to its sources. Thus, the parade of classic visual works in this opening section provides the ground not only for the arrangement of a the- sis but also for the creation of new images, constituting its own evidence for what is at stake. Christensen accomplishes this by continually triangulating between paintings and woodcuts, photographs, and cinematic dramatiza- tion. This movement between formally distinct media at times more firmly aligns Christensen with those who affirm that "nonfiction" is a designation determined by techniques of presentation rather than simple content, includ- ing art historian Aby Warburg, filmmaker Chris Marker (particularly in reference to his famous 1962 "film of photographs," La Jetee), and the recent photography of Jeff Wall, Cindy Sherman, and Hiroshi Sugimoto, much more than with his own contemporaries in the cinema of the 1920s.38 There are also echoes in Haxan of the creative displacements effected through Soviet montage and the use of fragments of found footage to assemble a sin- gular work, with Esfir Shub's film The Fall of the Romanov Dynasty (1927) 46 The Realization of the Witch being the most obvious example.39 Hiaxan, not having access to archival footage for obvious reasons, nevertheless re-presents the documents of the visual archive of the witch in a manner recalling the methods of Shub and other Soviet filmmakers such as Dziga Vertov. In formally similar films like Harun Farocki's As You See (Wie man sieht, 1986) and Images of the World and the Inscription of War (Bilder der Welt und Inschrift des Krieges, 1989), the "truth" gained by the reproduction of archival images is unlocked only through their mobility in the context of their new use.40 As with Farocki, Christensen does not seek to embellish such visual artifacts in citing them, but rather empties them out, expressing through their preestablished frame a meaning that was hidden, resisted, or not even invented at the time of their origins. Understood in this way, the disconcerting effect of the opening chapter becomes more plausible, as Hiixan disrupts what the audience can expect from the film. While the medium of expression is undoubtedly mod- ern and allows for these uniquely moving images, the method Christensen deploys helps to cultivate a position that draws authority from an expertise based on the interweaving of the artistic and the scientific rather than an ideal "scientific self" premised on the polarization of the two.41 Visual Strategies: The Wild Ride Two themes prefigured in this first chapter and foregrounded later in the film deserve treatment in terms of the visual strategies they employ: the Wild Ride and the hysteric. Our claims as to the methodological element of Christensen's image-making practices become clearer if we temporarily skip ahead to Hiixan's depiction of the violent moral disorder of the Wild Ride of the witches to their Sabbats. This scene appears in Chapter 4 of the film and is presented as a visual account of the old woman Maria's confes- sion to the "crime" of witchcraft. We will fully analyze the density of this scene in the corresponding chapter of the book, but for now we will focus only on Christensen's complex use of works of art that originally appeared in fifteenth- and sixteenth-century texts by Hans Vintler and Johann Geiler42 in the course of creating new cinematic images in Hiixan. Christensen's presentation of the Wild Ride is thrilling by any standard.43 Making use of the special effects available to him at the time, the fury and Words and Things 47 terror of Haxan's female wild riders stands out as one of many highlights of the film. By the early sixteenth century, the Wild Ride had become a stan- dard element of both demonological and popular literary accounts of the ac- tivities of witches, folding older legends of wild hunters, the restless travels of the dead at night, and tales of the Furious Horde into the standardized script of the Ride. Particularly strong in what is today southern Germany and Switzerland, variations on the myth of the night people retained their durable immediacy deep into the twentieth century.44 Charles Zika claims that in its various tellings the Furious Horde consisted of "cavalcades of de- monic spirits and souls, especially of those who died before their time and enjoyed no peace-soldiers killed in battle, young children, victims of vio- lent acts, and so on."45 Folded into the exegesis of the ninth-century text Canon Episcopi, regarding the power of demonic illusion to deceive women into imagining that they could travel great distances at night, often in the company of the goddess Diana, the Wild Ride violently collapsed a multi- tude of characters and beliefs into a particular time and a singular image of the witch in sixteenth-century Europe. Christensen's own image of the Ride compels the same collapse, though one that assumesfidelity to empirical evi- dence in the time of the witch hunts. This is characteristic of Haxan's cine- matic naturalism. There are many classic examples of images of the Furious Horde and the Wild Ride; two in particular stand out in relation to Haxan's own visualiza- tion of the spectacular event. First is a clear correspondence between a woodcut from Hans Vintler's Buch der Tugend titled Wild Riders on a Wolf Goat, Boar, and Stool (1486) and the special effect of Christensen's image of his witches flying through the air as part of Maria's confession in Chapter 4. This woodcut reflects its origins as a portrayal of Waldensian heresy (the subject of Vintler's text), depicting the riders, men, and vehicles as mostly animals.46 While Christensen's image substitutes iconic objects such as brooms and cooking forks for beasts and reflects a discourse of the witch (found in Kramer in 1486) as being almost singularly female, it neverthe- less takes direct inspiration from the classic woodcut in its perspective, its positioning of the riders in the frame, and the emphasis of the subjects that suppress depth of field against the void of an empty background. Vintler's woodcut, modified naturalistically to mirror the seemingly unnatural and impossible Wild Ride of the witch, moves in the film. 48 The Realization of the Witch Christensen also modifies and brings to life characteristic representations of the Furious Horde, a supernatural band that was not originally associ- ated with witchcraft at all. Again, this conjoining of witch image to demo- nological discourse reflects an empirically verifiable invention in the late medieval period and the Renaissance. In particular, Christensen's long shots of the witch's Sabbat, unfolding in the twisted chaos of the deep forest, re- calls the woodcut The Furious Horde that appears in the 1516 version of Jo- hann Geiler's Die Emeis. As with the echo of the Vintler woodcut in the Wild Ride, the perspective, framing, and composition of the image of the Sabbat in Haxan updates and transforms The Furious Horde, much as demonologists transformed the meaning of the Horde in the invention of the sixteenth- century witch. Again, Christensen is not only "inspired" by Geiler's image; he has in his creative activation of the image simultaneously produced an effect that corresponds to the empirical evidence of the witch's coming into being and exhibited what Charlie Kiel has termed "the oscillating value of the non-fictive."47 Documentary elements can support, contradict, or even wholly become the narrative in early cinema; Haxan in this sense is consis- tent with other contemporary works in the oscillating value of its discrete artifacts. Visually, Haxan offers innovation to the representation of demons that were commonly circulated in woodcuts, broadsheets, and paintings at the time. While the depiction of various lesser demons and fallen angels was quite common, they tended to be rendered as smaller versions of the horned Satan or as hybrid human-animal creatures with each "natural" species be- ing traceable within the complete appearance of the demonic creature (such as in the Geiler woodcut just mentioned). Haxan does not simply reproduce these stereotypic images. Instead, Christensen at times broadens his re- gional frame of reference, drawing on works referring to witchcraft pro- duced outside of German-speaking Europe such as Agostino Veneziano's painting The Carcass (ca. 1518-35) in relation to the Sabbat, or images that portray supernatural creatures that appear in negative sixteenth-century "guides" to pre-Christian Norse myth, particularly some of the woodcuts that accompany Olaus Magnus's Historia de Gentibus Septentrionalbis (in numerous printings from 1555), which appear to provide the inspiration for the "demonic children" Maria claims to have given birth to, revealed in her confession. Agostino Veneziano, ib e Carcass (ca. 1520). Courtesy of Scottish National Gallery, Edinburgh. Salbbat in Il1&van, film still (Svensk Filmindlustri, T922). 50 The Realization of the Witch Maria's confession in Chapter 4 of the film provides additional examples of the breadth of Christensen's visual assemblage of the witch and her ac- tivities. As with the discourse of the witch in the early modern period, figures from antiquity such as Saturn and Circe are also alluded to in the representation of the Sabbat in Haxan. In order to clarify our argument here, it is necessary to briefly analyze Christensen's composition of a se- ries of brief shots in the Sabbat that refer to sixteenth-century represen- tations of Circe and the link they made between the Roman goddess and witchcraft. In Maria's confession, Circe is indirectly named as "Satan's grandmother."48 Images associated with games of chance, gambling, tricks, slight of hand, and illusion were often part of Circe's repertoire. The logic here was that such games, seemingly minor performative elements of popular tricks and entertainment, were actually rooted in the same demonic power of illusion as more obvious forms of maleficium. Elements of Christensen's image here appear to be directly referring to a number of well-known visual represen- tations of Circe in the sixteenth century, particularly a woodcut from the workshop of Michael Wolgemut and Wilhelm Pleydenwurff and tentatively attributed to Albrecht Diirer that appeared in the Liber Chronicarum, titled Circe and Her Magical Arts Confronting Ulysses and His Transformed Compan- ions (1493). Although the literal confrontation depicted in this woodcut be- tween Circe and her assistant on the shore and Ulysses and his companions on a boat is absent in Haxan, the flowing beauty of Circe herself is echoed in the film's image and the table cluttered with instruments of chance and magic directly corresponds to the association Christensen is intending to make here. Other surviving images from the time echo Haxan's meaning here as well, albeit less directly. These would include the 1473 woodcut Circe with Ulysses and His Men Transformed into Animals from Giovanni Boccac- cio's Buch: Von den hochgeruemten frowen, and the pen-and-ink drawing The Children of Luna from the Housebook Master or Master of the Amsterdam Cabinet (1480). The hybrid animal-human forms of the demons dancing around the "grandmother," the woman's surprisingly young and beautiful appearance, the wind whipping around her, her elevated position in frame as if she is floating in the air (she is actually positioned on a ledge, but this is very difficult to discern until the "grandmother" is shown in medium shot, entering a door), and the array of objects and instruments she wields all Giovanni Benedetto Castiglione, Circe Changing U~lysses ' Mena inato Beasts (c. 165o). Courtesy of 13ritish Museum, London. Close-up of Circe's instruments in Haxan, film still (Svensk Filmindustri, 1922). 52 The Realization of the Witch point to the refiguring of the mythological figure of Circe as a powerful witch in the service of Satan.49 Visual Strategies: Hysteria A key component of Haxan's thesis is that the power of the witch is reani- mated in modern times through the signature of hysteria-something fore- shadowed in the first section of the film, and reinforced in Christensen's complex strategy of tacking between painting, photography, and moving cinematic image in Haxan. The "historical framing" in the longue duree of his thesis in the first chapter is carried throughout the entire film. There are a number of scenes in Haxan that activate unconscious associations in the viewer between melancholia, witchcraft, and possession. For example, while none of these paintings is explicitly displayed in Haxan, Christensen appears to have taken direct inspiration for a number of his shots from Lucas Cra- nach the Elder's famous Melencholia series of paintings. Produced between 1528 and 1533, these four paintings that depict the supernatural environ- ment haunting a female melancholic bear many similarities with elements that Christensen brings to life in Haxan, including the Wild Ride, terrify- ingly unnatural children, and a general sense of sexual and societal disorder swirling around a placid, passive female protagonist.50 It makes sense that Christensen would evoke Cranach as the paintings reflect an empirical strain of the discourse of the witch that highlighted the susceptibility of the mel- ancholic to the Devil's illusions and hence to witchcraft and especially pos- session. In the Melencholia series Cranach composes the face of his female subject as a mask, the swirl of activity around her signifying what lay behind her placid, deceptively beautiful faqade. Interestingly, there are several points in Haxan where Christensen self- consciously composes similar faces, simultaneously concealing and reveal- ing the turmoil that lay behind them. In particular, later in the film we find Brother John's troubled reverie in the face of his repressed, possessing de- sire for the Young Maiden and the mask/face of the unnamed hysteric that is the subject of most of the film's concluding chapter. In both cases, Chris- tensen draws a link between these carefully framed faces and possession, a mo- bile element moving between the pact of the witch and the obsessed state of Words and Things 5 3 the hysteric. Reversing Aby Warburg's assertion that donning a mask con- stitutes an active attempt "to wrest something magical from nature through the transformation of the person," Christensen's figures invert this polarity by appearing to be worn by the mask.51 Thus, the re-membered face of Cra- nach's melancholic in these shots works as a relay between Christensen's moving images of the witch/hysteric and unseen, but obviously present, iconic images of Charcot's hysterics. This is entirely consistent with Char- cot's belief that artistic works of demonic possession and melancholia were reliable evidence of hidden and misdiagnosed mental disease. As Avital Ronell has put it, "The scientific imperative, the demand in the nineteenth century for an epistemological reliable inquiry in the nature of things, derives part of its strength from the powerful competition represented by fascination for the freak and the occult, which is always on the way to technology. "5 By formally constructing "the witch" through a cinematic iteration of metoposcopic naturalism, Christensen could not agree more. Although left un- said in the opening chapter, the imperative Ronell cites is progressively etched on the face of the images the director produces, be they explicitly "photographed" icons or evoked as echoes and memories. Using a strat- egy similar to that famously deployed by his closest filmmaker contempo- rary, Carl Theodor Dreyer, Christensen will build from the elements of this opening chapter to a complex, expressive interplay of face and tableau in order to bring the witch to life in Haxan. What Is This Thing? An image is strong not because it is brutal or fantastic-but because the association of ideas is distant and right. Pierre Reverdy, "L'image" (1918) The first chapter of Haxan draws to an abrupt close, its tone descending from the overwhelming affective force of images of explicitly sexual acts with Satan. Christensen actively avoids taxing the audience with any further ex- planation or lecture. We find images of witches flying (this time "returning home" after a "merry dance") as a final set of title cards blandly state that ~*1 i 1 *1 Lucas Cranach the Elder, Die Melancholic (Mclancholy) (1528). Courtesy of Scottish National Gallery, Edin burgh. Words and Thin gs Brother John in Haxan, film still (Svensk Filmindustri, 1922). images such as the ones the audience has just seen "are often found on fa- mous witch Sabbath pictures from the Middle Ages and the Renaissance." Three more dense images then flash in secession (it is unclear if they show Sabbats, hell, or some combination of these on earth) and then a final title card appears and is held several beats longer than those immediately preceding it, having the effect of a door held for a moment before slamming close this chapter of the film. Hdxan gets off to an undeniably peculiar start. In our view this is due to the formal, methodological ambition of the work, particularly in regard to the conscious triangulation of ontologically distinct image-objects arising out of paintings/woodcuts, photographs, and cinema. Christensen is try- ing to make the power of the witch real in a way that seems impossible through a film. Invading the domains of the human sciences, particularly those of the art historian and the ethnographer, Christensen will not remain content to faithfully reproduce traces of the past, devoting the remainder 56 The Realization of the Witch ~. N ~1 Augustine, 'Attitudes passionelles: The Call," from Desire-Magloire H3ourneville, Paul Regnard, Jean-Martin Charcot, and Edouard Delessert, Icon ographie photogr-aph iqiie de la Sa/petnL're: Service de ;1I. Chalrcot, 3 vols. (Paris: Progres Medical, 1877-80). of Hiixani to willing a new life into texts and images. The director's "atlas of images at work" strategy is strikingly reminiscent of the methodological in- novations of Aby Wa rburg, particularly in relation to Warburg's uninished M'nemnosynie project.'' It is worth quoting Philippe Alain-Michaud's summary of W~arburg's scheme at length: Words and Things 57 Hysteric hears the call in Il/fxan, film still (Svensk Filmindustri, 1922). In Mnemosyne, photographic reproduction is not merely illustrative but a general plastic medium to which all figures are reduced before being arranged in the space of a panel. In this way, the viewer participates in two successive transformations of the original material: different types of objects (paintings, reliefs, drawings, architecture, living beings) are unified through photography before being arranged on the panel stretched with black cloth. The panel is in turn rephotographed in order to create a unique image, which will be inserted into a series intended to take the form of a book. The atlas, then, does not limit itself to describing the migrations of images through the history of representation: it reproduces them. In this sense, it is based on a cinematic mode of thought, one that, by using figures, aims at not articulating meaning but at producing effects.54 Heightened by the effect in cinema that everything in frame appears to be alive, the strategy will prove to rupture the very perceptions of "deadness" or "pastness" that allows the modern viewer to evade the power of the witch that Christensen will forcefully assert is still with us. The time of the witch, 58 The Realization of the Witch in all its multiplicity and exigency, will be brought out of the past and into the present by appearing to register the form of life itself on film.55 Thus, the alienating distance of both the objects presented in Haxan's first chapter and the characters they refer to is necessary to begin with, as the task of the film now becomes the closing of this distance between the two-dimensional surfaces of photographs and celluloid and the three- dimensional sense of lived experience. Similar to Robert Wiene's The Cabinet of Dr. Caligari (1920) in this respect, Hiixan shows an affinity with the Cub- ist art contemporary to its release in the tension it strategically heightens by ignoring or contravening the perceptive "rules" of formally distinct im- age artifacts.56 In later decades, artists and filmmakers such as Gerhard Richter (Atlas, 2006) and Jean-Luc Godard (Histoire[s] du cinema, 1988-98) have taken up Warburg/Christensen's methodological logic in their own attempts to link the dimensions of the image with life. Within the arc of this movement in Christensen's film, the objective knowledge of witchcraft is opened to the perception of otherness in the witch, the demonologist, the hysteric, and ultimately the scientist by way of a visible unity of the senses unique to the director's method. "The ethnographic surrealist," wrote James Clifford, "unlike either the typical art critic or anthropologist of the [192os], delights in cultural impu- rities and disturbing syncretisms."57 We are not claiming that Hiixan is ethno- graphic in its formal approach, yet Clifford's description does echo the links we are drawing here between radical approaches to the image in art and subversive methods deployed in documenting the real that were roughly contemporary to the film.58 The transgressive approach to the archive, to classification, and to expression that the film exhibits also is akin to meth- ods deployed in the journal Documents (1929-30) nearly a decade later. Ed- ited by Georges Bataille, Documents willfully transgressed institutional genres through its "subversive, nearly anarchic documentary attitude," an attitude that Christensen plainly shared.59 What distinguished Documents from Warburg's Mnemosyne and Hiixan is that the former seizes cliched objects and then systematically empties them out in the course of its own expressions. Bataille and his contributors sought to defamiliarize the cliches, disturbing the placidly deceptive surface of the mundane in their fragmen- tary, juxtaposing methods of critique and presentation. In contrast, War- burg and Christensen begin by collecting mythological, figurative givens Words and Things 59 seemingly quite distant from the "really" real. Starting at radically different places, the outcomes of these projects converge on the same nodal point unsettling distances between myth and the everyday that in turn produce expressive works that are themselves quite unsettling. It is obvious in light of this shared methodological aspiration why the surrealists would take in- spiration from Haxan, brazenly (and unfairly) advocating Christensen over Dreyer as the Scandinavian filmmaker of note in the 1920s.60 David Bordwell groups Haxan, along with Carl Theodor Dreyer's Leaves from Satan's Book (Blade af Satans Bog, 1921), Maurice Tourneur's Woman (1918), and Fritz Lang's Destiny (Der miide Tod, 1921) within a tradition of "episode films" in the classical period of silent cinema.61 This is consistent with our argument regarding Christensen's film, as all of these cinematic works weave together episodic fragments in order to draw parallels and cor- respondences across situations and characters. More explicitly than the others, however, Haxan also deploys the techniques associated with War- burg's Mnemosyne and Bataille's Documents for purposes of affectively em- phasizing the dark, chaotic forces that lurk under the smooth surface of the everyday. The parallels Christensen draws are therefore not simply between characters or situations but across domains of sense that cut across time. Thus, the episodic structure of Haxan not only allows characters seemingly out of a dead past to live again, it also draws the phenomenology of the hysteric and the work's own contemporary time to the surface. Shadowed by the specter of an everyday fractured by mechanized global war, Haxan in turn brings its witches, inquisitors, and hysterics alive in the haunted now of the film's reception.62 In short, Haxan is promiscuous. It is neither wholly artistic nor scien- tific. It aspires to seize a quality Ulrich Baer granted only to photography when he wrote, "Films fail to fascinate in the same way as photographs do, because they invite the viewer to speculate on the future-even when irresistibly tempted to do so-only on the level of plot or formal arrange- ment. Photographs compel the imagination because they remain radically open-ended."63 Haxan calls Baer's assertion into question. The opening chapter does not offer a speculation as to the future. It disorients the viewer, leaving her with the insistent, fundamental question, "What is this thing?" It compresses times past and future into a sequence of cliched images that traverses the 6o The Realization of the Witch steep slope between past and future in the form of an event. This is not a plot. Rather, it is a strategy to "compel" the viewer, although we would not limit this compulsion to the imagination alone. In other words, the inabil- ity to automatically categorize Haxan emerges out of a formal strategy rooted in an epistemic virtue. In science, such virtues demand that the subject know the world and not necessarily the self; Haxan's demand is greater in its own way as it demands both.64 Thus, while Christensen never backs away from his claim that Haxan offers a truthful examination of the witch that can stand up to the test, he also deploys strategies of evidence making that would have been familiar to the subjects of his film. As Joseph Leo Koerner puts it: "In the later Middle Ages, in practices ranging from persecuting witchcraft to meditating on Christ, techniques were developed to draw distinctions among visual phenomena, differentiating, say, physical objects from fantasies, dreams, and diabolical or artful deceptions. Some of the best testimonies of this sorting operation come from artists. This is not surprising given that image-makers specialized in manipulating one thing (their materials) in or- der that a viewer should see something else."6 While Christensen's materials might have been radically different than those of an artist in the late Middle Ages, his aim to manipulate these ma- terials in order to make something invisible visible is consistent with his aims. This description, of course, could also be applied to experimental sci- entific techniques without much alteration to the stated aims of tests taken under the signature of such disciplines. For Christensen, objective knowledge itself has been possessed by the un- canny, rendering "imagination" or "reason" alone inadequate to bringing the witch to life, to forcing her to speak to what is already known in her pathological language of diabolic proofs. The witch must be experienced in her own milieu, a satanic biome that we will presently argue is one that Christensen represents as her state in nature. As it moves from the first chap- ter to the second, Haxan constitutes an extension from the techniques and virtues of Mnemosyne to those of the nature film. In other words, the first chapter of Hiixan is the presentation of a series of cliches-visual cliches and stereotypes of the witch, fragments which were most likely already familiar to the viewer. This is hardly a waste of time, however, as these cliches (what Deleuze terms figurative givens) will not only provide the empirical evidence for Christensen's thesis but will also provide media from which the director Words and Things 61 will conjure the power of the witch. It is important to note that Deleuze dis- cusses figurative givens in reference to painting, not cinema; thus, the con- cept would not seem to readily apply here.66 Yet we suggest that Christensen is attempting to do something quite paradoxical, which is to release the movement of the painting and the woodcut through the cinematic image. Indeed, as we move through the film, we cumulatively gain the sense that Haxan is a living tableau. This is by no means an accident. The film excels in providing the ground for this sense, possessing the spectator through the immediacy regardless of whether the viewer logically knows that the repre- sented event is already in the distance. This quality sets Haxan apart.67 T W 0 Evidence, Second Movement: Tableaux and Faces What was normal [by the seventeenth century], at all levels from the patrician to the plebeian, was the marriage of word and image. That "mutuality" may be tacit: thus heroic painting in the grand manner does not generally carry with it, in so many words, a commentary, keying in the mythological figures; yet artists habitually gave their paintings titles, mottoes, tags and quotations, and their works abound in literary allusions. But very commonly the interleaving of the verbal and visual is quite explicit. -ROY PORTER, "Seeing the Past" (1988) The originary world is therefore both radical beginning and absolute end; and finally it links the one to the other, it puts the one into the other, according to a law which is that of the steepest slope. It is thus a world of a very special kind of violence (in certain respects, it is the radical evil); but it has the merit of causing an originary image of time to rise, with the beginning, the end, and the slope, all the cruelty of Chronos. This is naturalism. -GILLES DELEUZE, Cinema I (1986) An Unnatural Business What better place to begin pictorially than in the underground lair of an old witch named Karna? A hoary, wrinkled woman moves busily in her dark underground room, the scene cluttered with a variety of objects that are dif- ficult to recognize and yet generate an ominous, dreadful sense that some- thing malevolent is going on here. The old woman tends to a pot over the hearth in the middle of the large, dank room; not precisely the cauldron central to the witch stereotype, but certainly close enough. An accomplice, 62 Tableaux and Faces 63 a somewhat younger version of the sorceress, enters the room with a large bundle of straw, roughly tossing the bundle to the side. The bundle falls heavily and the old shrew quickly reveals why; hidden within the shock of straw is a corpse (or part of a corpse), exposed by the old woman parting the straw and drawing a lifeless hand from within the bundle. The meaning of her previous dialogue ("Tonight the stars shine favorably over the gallows hill") is answered by this gesture. Complaining about the quality of the item her coconspirator has pro- cured ("Ugh! What a stench!"), the old woman nevertheless proceeds to examine the hand carefully, suddenly snapping one of the fingers off the decaying hand. Noting that the finger of the thief may be "too dried out" to lend any power to her brew, Karna nevertheless ties a string to it and lowers it into a large cask. Her partner does not respond and blithely stirs the small cauldron boiling on the hearth. In a series of flowing, intercut close and medium shots, Christensen reveals the terrible ingredients of the cauldron-ingredients that (save for a large, writhing snake and a still-alive toad feebly attempting an escape) are unidentifiable in their strangeness. This mildly creepy reveal visualizes a stock cliche of the witch stereotype regarding the ingredients she uses: of nature and yet revoltingly unnatural and unwholesome at the same time. The casual suggestion of cannibalism also references the popular understanding of the witch, something we dis- cuss in relation to the witch's Sabbat later in the book. Someone approaches the entrance to the old woman's home. She is hid- den and nearly frantic to enter before she is seen. A customer. Karna shows the plain, middle-aged woman in. Warily surveying the scene, the customer gets directly to business. She is in need of a love potion to be used to entice "a pious man of the church." Karna, being a savvy entrepreneur, has a range of choices to offer her customer. In succession, she offers a potion of "cat feces and dove hearts boiled in the moonlight" or a stronger brew rendered from "a young and playful male sparrow." As Karna speaks, the customer visualizes the potential outcomes of each potion, Christensen dissolving back and forth from our perspective eavesdropping on the transaction to scenes of the customer administering the potions and their results, first one of ardor, then one of frenzied sexual arousal. Although played for amusement, this sequence serves the purpose of presenting several additional elements of the witch that were not precisely 64 The Realization of the Witch germane to the stern lecture in the opening chapter. First, the conflation and merging of myriad forms of popular magic with the person of the witch is plainly shown here. Classic witch-hunting manuals such as the Malleus sought to associate common magical practices of protection and healing with the witch; this association had to be strongly asserted by elite writers of the time as the connection was not obvious and was never fully absorbed into popular discourses of what constituted a witch.1 Second, the stereotype of a debased and corrupted priest is forcefully introduced at this juncture. The object of the customer's affections, surrounded by wealth and comfort, is a fat, uncouth, and (if one is to "read" the face) vaguely stupid friar who appears to be completely subjugated by his desires. In each short fantasy aside, the friar is shown eating lavishly prepared food in the manner of "an animal," dismissive of the woman serving him prior to administering of the potion he is then instantaneously and completely overtaken by his desire once the witch's concoction is ingested. The question that forms in the mind of the viewer is not about the effectiveness of the potion, but why the customer would desire such a slovenly, corrupt man in the first place.2 Christensen leaves this particular question hanging, although his insult- ingly satirical portrayal of this gluttonous, lustful priest has an antecedent in the strategically vulgar aspects of Protestant discourse against the Cath- olic Church in the sixteenth century. No less than Luther himself was known to slander, scandalize, and offend his adversaries using explicitly vulgar lan- guage against them. Taking cues, artists of the period extended the instru- mentalization of slander through the production of proto-pornographic images of bishops, priests, and the pope engaged in myriad obscene acts. It was not uncommon for the pope to be depicted as the Antichrist in such works. Ignoring Thomas Aquinas's assertion that scandal, either "active" or "passive," is always a sin, Protestant propagandists sought to offend, us- ing explicit rudeness as a weapon in the battles of the Reformation and Counter-Reformation. For them, satirical, injurious images were regarded as expressions of empirical realities; this was not in spite of their obscenity, but rather because of it.3 As empirically grounded as Hiixan's depiction of this persuasive discursive form is, Christensen's fidelity to historical facts did not endear his film to censors. The difficulty Christensen had in getting Hiixan released in Germany serves as a good example of the issues raised in nearly every country outside Tableaux and Faces 65 Childhood of Christ; the Passion; three cuttings from Itinerariun beatae liariae virginis, printed by Johann Reger, Ulm. --- -- - - Courtesy of British Museum, London. of Scandinavia where distributors attempted to show the film. Although several private screenings prior to its public premiere generated praise for Hdxan in the German press, the film's release was denied when first submit- ted to German censors in February 1924. Only after substantial cuts was the film publicly shown in Berlin in June of the same year. The controversy over Ilaxan lingered despite Christensen substantially truncating the work. In January 1925, the Prussian Ministry of the Interior formally petitioned the censors to rescind their approval, citing scenes such as the one described above as expressly intending to offend religious feeling, to threaten public order, and to "brutalize."4 Although the petition was rejected, such efforts severely restricted showings of Hfxan, laying the groundwork for its unfor- tunate status as an "unseen classic." Interestingly, it was quite often scenes such as those depicting the emotional states and desires of the clergy that aroused as much ire as scenes showing nudity, demonic violence, sex, and cannibalism. Christensen ends this lusty introductory scene with Karna, not yet tak- ing her customer's money, suggesting that the best remedy of all may be her 66 The Realization of the Witch II Devils Watc~h while a Jesuit Sodomizes a Young Woman, from Historisce rint en lDicht-'lafereelen, van Jan Baptist Girard, en Jmijfrou M aria CTatharina CTadiere (1735). Courtesy of Bayerische Staatsbibliothek, Muinich. salve. This ointment (which she dramatically displays by carving out a por- tion with a blade) is so powerful that the "pious monk might directly come to [her] chamber." She notes that they would fly together at night in amo- rous bliss. Again, Christensen introduces another core element of the witch stereotype: the salve. Movhre than any other substances, magical salves and Tableaix and Faces 67 Desirous priest in Iiixan, film still (Svensk Filmindustri, 1922). ointments were considered particularly powerful and troubling by demon- ologists, and their use was a sure sign that an individual was engaged in witchcraft.f Haxan stops before showing whether Karna's customer takes her up on the offer. Illxan's thesis is now unfolding. Without warning, the film cuts from the scenes with Karna to a moody, melodramatic scene with a young man hur- rying another corpse through the streets of the village, finally arriving at a door where he is ushered in to the home with his load. Joined by a second young man, the two of them laboriously heave the bundled body down into the basement, at first unnoticed, observed by a woman awakened by the com- motion. Through a series of cuts between the men and the spying woman, the female corpse is revealed. They are troubled and nervous. "Listen brother, shouldn't we pray . . . ?" asks one as the other stands dramatically poised with a large knife, ready to plunge it into the cadaver. Pray they do, beseeching the Holy Mother for forgiveness for cutting open a body in or- der to learn its secrets. The spying woman, overcome with curiosity, walks 68 The Realization of the Witch in on the macabre scene. Realizing what is about to take place, she runs screaming into the streets, shrieking of the desecration brought about by "two witches." This scene is quite dramatic, but also puzzling in its placement in this section of the film. Certainly the evolving practice of anatomical dissection and the development of pathological anatomy as a central element of medi- cine drew objections and accusations of blasphemy and desecration (although in an historically uneven way); what is strange is that this was not a strong element of the witch stereotype in the time period to which Haxan ostensi- bly restricts itself. Dissection and dismemberment of bodies did exist as a religious practice, but tended to be restricted to the securing of holy relics from the bodies of recent dead regarded to be "saintly" or in public displays for the purpose of understanding anatomy and general curiosity.6 One must wait nearly two centuries before the trope of the "body snatcher" in the name of anatomical science really comes into being. While inquisitors and demo- nologists were certainly concerned with abuses visited on corpses, the evi- dence to which Christensen cleaves in the inquisitorial manuals drafted during the early modern period simply does not support the idea that anat- omists or physicians were mistaken for witches. It is clear that Christensen wants practices associated with developing and testing medical knowledge to enter his visual narrative from the earliest moments, even at the risk of contradicting the historical record. Christensen's inclusion of this event, directly following the scene with the lay practitioner of natural magic, is peculiar, as the film clearly depicts Karna as a "witch" while implying that the young men are simply "mistaken" for witches. We can only speculate as to the reasons for this empirical lapse, but the short scene does allow for a strong visual correspondence between misunderstandings of illness and scientific medical practice and the reasons for witchcraft accusations. Haxan, for all its cinematic license and unspo- ken complexity in relation to the power of the witch, never explicitly moves away from its self-positioning as a scientific investigation. As such, the idea that we are waiting to discover the "real" reasons for witches is never far from its agenda, whatever else Haxan actually communicates to its audience. To undercut the idea that Karna is a witch, however, would have severely weakened her scenes, and Christensen's tone here is one of unambiguously Tableaux and Faces 69 positioning her as not only a misguided healer but also someone who ex- plicitly believes she is a witch and acts more or less in the conspiratorial man- ner that is described in the Malleus and elsewhere. Yet also aspiring to have the audience feel the power of the witch, Christensen must constantly pull back from the dramatic outcomes of what Haxan depicts. Thus, pulled be- tween these tensions, we see a dramatic slippage in the scene of the two ama- teur anatomists. Unable to fully negotiate the multiple demands Christensen himself makes of the film, we are presented with a dramatically useful error in the presentation of the evidence for his thesis. Christensen notes at this point (via title cards) that it was common for everyday people at the time to see witchcraft as the source for a wide vari- ety of misfortunes; in order to illustrate this statement, he moves to a scene of a conflict on the street at night between an old woman sleeping on some steps and a passer-by. The man angrily rousts the woman, accusing her of "bewitch[ing] the legs of honest people." Her aggressive reply is to bewitch the man's jaw, forcing his "filthy mouth [to] remain open for eternity." The stricken man, a look of terror on his face, collapses on the very steps from which he has rousted the old woman. This seemingly throwaway scene is interesting in that Christensen's depiction does not appear to illustrate the paranoid delusions of witch-crazed villagers; rather, it is a scene that unam- biguously shows the bewitching of a man. While the audience can specu- late as to potential somatic or psychosomatic causes for the man's sudden affliction, Haxan unequivocally plays the scene as a demonstration of the power of the witch. Unlike the previous scene of misunderstanding arising from the activities of the amateur anatomists, there is no visible cause for the action here except the angry spell of the old woman. The function of this short scene is subtle. As the film unfolds, it is clear that Christensen is aware that the witch comes into being in overlapping vectors between learned discourse and popular belief. This requires some ground for agreement between the two domains; one clear aspect of this overlap is the widely held set of beliefs regarding the "nature" or "essence" of women. Echoing Christina Larner's assertion that witch trials were gender-related but not by definition gender-specific, the rhythmic alternation in this chapter of Haxan between women and men acting according to their supposed natures provides some insight into not only why the accusations 70 The Realization of the Witch against women came to dominate the witch trials but also how it was possible, under some conditions, for the accusation to be leveled at men as well.7 The gender that Christensen invokes here, although perhaps somewhat overdetermined in its reliance on functionalist binaries and contrarieties, is not far from the social-functionalist explanations offered by anthropologists such as E. E. Evans-Pritchard, and more recently by historians such as Stu- art Clark who have been inspired by anthropological attempts to apprehend the figure of the witch.8 Returning to Larner's insight, it important to note that Haxan does not represent the vicious pursuit of witches as a straight- forward instance of misogyny on the part of the Church or sixteenth-century civil authorities. Rather, it was the case that women often came under sus- picion of being witches because they were understood to be particularly susceptible to lust, avarice, and jealousy by their very nature. Women, acting in accordance with this presumed nature (as the mean old woman briefly demonstrates), were therefore "naturally" susceptible to witchcraft. Men could also be witches, but generally only in instances where they were clearly acting against norms. The young proto-anatomists, therefore, have given in to a curiosity that was by the standards of the time morbid and unnatu- ral. They, too, are taken to be witches, but only because what they are doing is not expected of them, a sure sign of unnatural forces at work. By contrast, women could signal the mobilization of these same unnatural forces simply by enacting elements of the nature that they were always already presumed to possess.9 This difference will be important as Haxan progresses through its thesis. The uneasy structural ambiguities carry over into the following scene. "So it happens with witchcraft as with the Devil; people's belief in him was so strong that he became real." This intertitle follows the line Christensen established earlier regarding the error and false consciousness of those who believed in witches and generates a nagging, almost unconscious, reluctance about Satan being made real. This phrasing is consistent with the posi- tions taken by Clark, Roper, and a host of other recent historians that one cannot approach witchcraft or possession from a vantage point in the pre- sent without granting some legitimate status to the ways in which the Devil and witches were not only asserted to be real but were experienced as such.10 Tableaix and Faces 71 Reassiring Visions Hdxani's visual expression of this point constitutes one of the best-known scenes in the film. A priest, revealed later to be Father Henrik, engaged in intense prayer, is suddenly confronted by the Devil himself. Seeming to emerge directly out of the large Bible the priest is reading (in fact, popping up from behind the stand supporting the book), Satan is monstrously inti- midating, leering at the terrified priest who backs away in horror. Played by Christensen himself, Satan taunts the priest and his colleague, who has rushed over in aid. Spreading his terrible claws over the pages of scripture, Satan dominates in even the holiest places (a church) and through things (sacred text), a fact that Christensen renders powerfully in this sequence. This scene is campy by today's standards, in part due to the lasciviousness of Christensen's Satan and the hysterically overwrought reactions of the Satan appears in IIixan, film still (Svensk Filmindustri, 1922). 72 The Realization of the Witch Satan appears in Hiixan, film still (Svensk Filmindustri, 1922). harassed priest. Its power to shock, however, remains intact. Satan erupt- ing forth for the first time during an act of prayer, in a church, and con- fronting a pious believer and instrument of God, visually conveys the terrifying and reassuring sense of power and threat Satan possessed. Al- though often understood as the point in which IlIxan begins to slide into the territory of farcical reenactment, this short scene is among the most em- pirically consistent sequences in the entire film, particularly in relation to an understanding of the power of the witch and the Devil and the sense of these beings that existed in relation to life at the time. Prior to Satan's dramatic entrance, we see the religious trappings, but there is no evidence of God's acknowledgment or answer to the friar's prayer. Yet, despite the obvious shock of the event, Father Henrik's confrontation with Satan at the very moment he beseeches God is shown to consolidate, rather than dissipate, his pious resolve. This resolve in IIaxani reverberates outward toward the implied (visual) term in this powerful image; Satan is Tableaux and Faces 73 indeed right in front of Father Henrik but the witch is not far behind. Encircled by these diabolic figures, the friar can ironically perceive the truth of the words he had just been carefully reading. On its own, the Bible is unable to convey information or simply communicate in a reliable, testable manner." Coming face to face with Satan, at the very moment of the Word's perception in the mind of Father Henrik, the required supplementary proof is given via the concrete, threatening body of the evil one himself. Body and Word conjoined in Christensen's cinematic image; the ritual the scene be- gan with can now speak. The sense that Satan could be everywhere, positioned just out of sight, pulling the strings of his demonic human puppets is not limited to the early modern period. Nor is Hiixan the only film where the notion appears. The figure of Satan as the power behind calamitous events in human history is also used in Dreyer's Leaves from Satan's Book. Like Hiixan, Leaves is an epi- sode film, but in Dreyer's slightly earlier work, Satan's malefic presence not only traverses far spaces but also crisscrosses time itself in what Bordwell has termed "a density of parallelisms."" Composed of four sections show- ing the crucifixion of Jesus, the Spanish Inquisition, the French Revolution, and the then-current civil war in Finland, Leaves portrays Satan as being potentially behind all calamity, driven to subvert humankind as punishment for his rebellion against God. In Dreyer's version, the Devil is awesomely powerful but also somewhat pitiable in that he can do nothing else but dis- rupt and destroy as an enforced condition of God's punishment. This ren- dering of Satan, at times strangely sympathetic, relies explicitly on discourses of theodicy and God's ultimate permission for Satan's deeds, debates reach- ing back well into the fifteenth and sixteenth centuries. Importantly, Dreyer figures Satan as a problem of the present in the last episode of the film, with the Bolsheviks violently engaged in revolutionary struggle being the mani- festation of a transcendental demonic power. There is a great deal of debate as to Dreyer's political leanings in associating Satan with contemporary communists. The disagreement about whether Dreyer was a "conservative" or engaged in explicitly ideological filmmaking with Leaves is important on its own terms, but is not relevant here.13 It is, however, important to note that Satan was a very powerful figural vehicle in Scandinavian cinema at this time and that Christensen's own rendering of the Devil as potentially "be- ing everywhere" would have certainly reverberated with contemporary 74 The Realization of the Witch debates over politics and evil, and with other films such as Leaves that take up a similar expressive strategy. As in the time of Thomas Aquinas, this question of theodicy ultimately cannot help but come around to God's seeming absence from the world; in the wake of the devastation of the First World War, this question for many had never been more pressing. Several scenes building on the displayed power to deceive and bedevil human beings follow the shock of the Devil's first appearance in Hiixan. In sequence, Satan's ubiquity and lustfulness is shown over three scenes: (z) ter- rifying a woman lying in bed at night (it is unclear whether this is a night- mare); (2) enticing a nude female somnambulist out of her home into the forest, where she eventually kneels before a demon who embraces her; (3) appearing at the window of a woman sleeping with her husband and violently "encouraging" her to come with him-she does not immediately succumb, but is shown in several close-up shots grimacing and licking her lips, then in medium shot arching her body in erotic pleasure, and finally returns the Devil's embrace in her bed. While the charged eroticism of this sequence is played for shock and titillation to some degree, it remains consistent with the strategy of giving a powerful sense of the descriptions of Satan that ex- ist in the demonological literature to which Hiixan refers, particularly re- garding the materiality of the Devil and sexual encounters with him. As Walter Stephens points out, proving sexual relations with the Devil was an essential task in many witch trials and served as crucial empirical evidence for Satan's existence.14 It is notable that in this series the targets of poisonous attention are young, beautiful women. While they are described as the "Devil's companions," these scenes are ambiguous as to the will and agency of the young women who become entangled with Satan's erotic power. Moving to the next se- quence, Hiixan's narrative shifts focus to a character that actively seeks the companionship and assistance of the Devil-the viewer is brought back to the figure of an old woman. Venusberg Hiixan now returns to the basement lair where this chapter of the film be- gan. It is apparent that the set is the same one inhabited by Karna earlier, Tableaux and Faces 75 but this time the viewer is introduced to Apelone. Christensen's first-person form of address in the intertitles is jarring as he directly queries Apelone by asking if it is "from the eternal fright of the pyre that you get drunk every night, you poor old woman of the Middle Ages." As the audience "hears" this condescending question directed at the old woman, she is shown shuffling and stumbling around the dimly lit basement, although interestingly she is not shown to be actually drinking. Christensen is forcing the issue some- what, as there is an obvious distance between what is "said" and what is seen. Sarcastically driving a further wedge, doubt as to the director's felicity arises at various stages in Haxan, with this short sequence being a prime example. While it is unclear if Christensen intended to make a self-consciously critical statement, this quasi-humanist magnanimity reveals what Catherine Russell has termed "condescension toward the Other."15 Her critical analysis of Bu- nuel's Land without Bread is also appropriate here: "Surrealist ethnography might therefore be a means of denoting the strategic roles of ambivalence, cruelty and empathy in refiguring the ethnographic relationship in postcolo- nial culture. Bunuel evokes the dangers of the photographic image and its implicit historical structure, marking the deep divide between those 'out there' in the real, and those who watch 'in here,' in the auditorium."16 Christensen is hardly a surrealist ethnographer, but his dreamlike histori- ography of witches is presented with the same double signature, simulta- neously relying on and then disavowing the authority of science and its humanist social iterations. The unsteady trace of this signature become all too apparent in harsh moments such as the director's address to Apelone in Hiixan, revealing Christensen's documentary to exist as an often cruel sec- ondary revision of what is, or was, already in the world. The correspondence between art and science driving this revision is left unmarked, but, echoing Walter Benjamin, it is quite clear that Christensen's art sets out to "conquer meaning."17 This move, made manifest at the expense of a defenseless figure from the past, scarcely distinguishes Hiixan from the science it overtly aspires to or its contemporary approach to the unfortunate hysterics that become the focus of the final two chapters of the film. In this sense, Hiixan is rigorously consistent throughout. Apelone falls into a stupor in the corner; the Devil appears. In the course of rousting Apelone, Christensen's Satan performs some of the most out- landish and obscene gestures of the film. In particular, his frenzied thrusting 76 The Realization of the Witch Satan "churns his butter" in Hdxan, film still (Svensk Filmindustri, 1922). of what appears to be a butter churn positioned suggestively between his legs unmistakably is meant to intimate masturbation. Christensen's wild, onanistic gestures may draw shocked laughs today, but it is an effective, purposeful performance, producing a disturbed affect that ratifies the blas- phemous, obscene experience of being confronted by Satan himself. The shouting, tongue-wagging, powerfully stroking Devil is truly disturbing and grotesquely attractive in the way that only a night vision can be. Apelone, like the audience, appears terrified when she comes to realize what she is seeing. She is nevertheless compelled to follow where Satan com- mands her to go. Satan suddenly flies up in the air and out through a high window; now fully aware, the old woman hurries over to the window and is hurled into the air herself. The film cuts to a shot of a castle, noting to the audience that this is Apelone's "dream castle" (with the potential double meaning of the word "dream" left open to interpretation) and the place where the Devil will fulfill her wishes. Tableaux and Faces 77 Htixan then cuts to a close-up of an unconscious Apelone. Interestingly, although she is shown flying through the air immediately prior, the ride it- self and her mode of conveyance (typically represented as a broomstick or chair in woodcuts at the time) are not shown. A full visualization of the in- famous Wild Ride will have to wait until later in the film. We see gold coins pouring down on Apelone's head, awakening her. The shot widens to show the floor of the well-appointed room where she is now covered in these coins. Stunned, she excitedly gropes the coins, unaware that Satan is watch- ing her. She dumps a pile of coins on the table located in the foreground of the room when suddenly the coins begin to fly up in the air, disappearing. Apelone is alarmed and feebly attempts to corral the vanishing coins. She fails miserably and pleadingly looks up into the ether where the coins have vanished. Hdxan then moves to the door of the room, now open. The coins remaining on the floor in front of the door fly away as well. A close-up of Apelone's frantic, greedy face betrays her confused terror to the viewer. The Apelone's disappearing coins in ihxan, film still (Svensk Filmindustri, 1922). 78 The Realization of the Witch coins completely fly away from Apelone as she chases them, stumbling clum- sily into the next room. The effect of the disappearing coins, generated by running the original shot backward, is nearly as old as cinema itself. The Lumiere brothers' short actuality, Demolition of a Wall (Demolition d'un mur, 1896) is believed to the first film to deploy this technique by simply running the film backward through the projector, showing first a wall being torn down and then the smashed wall miraculously reconstituting itself. If the conceptualization of time as an "arrow" dominated thinking in the late nineteenth century, then the revolutionary potential of this simple technological reversal is ob- vious. These unnatural reversals were in centuries past attributed to the Devil's deceitful manipulation of natural laws or the senses, so it is no stretch to suggest that cinema's early association with magic is a logical one, an as- sociation linked to the special effect in early horror such as J. Searle Dawley's version of Frankenstein (1910). Mary Ann Doane attributes this correspon- dence to the "semiosis of cinema's own technological condition," whereby such conditions are transformed into "legible signs."18 By attempting to bring the witch to life, such signs serve as both subject and subtext. The Vita- graph/Edison film The Artist's Dilemma (1901) is more explicitly a precur- sor to Hiixan in this respect. The short film begins with an artist carefully painting a model in what appears to be a Victorian drawing room. As de- scribed by Doane, a "clown/demon" emerges from the clock and proceeds to "unpaint" the original picture and with rough brushstrokes substitutes his own photo-likeness version of the model, which he then proceeds to bring to life and help down off the canvas. Although used in a much more sophisticated way in Hiixan, the reverse-motion effect in The Artist's Di- lemma bluntly demonstrates the technique's general purpose in both films. As Doane writes about the earlier short, "The parallel between the realis- tic portrait and the film image-both inhabit a frame and emerge out of blackness-demonstrates that [The Artist's Dilemma] seeks to reinscribe the uncanny likeness of the cinematic image as magic, and magic as the under- side of science."19 Over two decades later, Hiixan would deploy the same special effect in its scene of Apalone's visit to Venusberg, despite the technique having gone out of favor with filmmakers as a gimmick of earlier cinema. In this case, the technique's connection to what was even in 1922 a somewhat anachro- Tableaux and Faces 79 nistic cinematic past grounds its use within the subject of the film, allowing it to work precisely because of its association with the past. The alterity of the antiquated method here edges the radical power of the witch a little closer to the viewer. In the next scene, a sumptuous feast is laid out for Apelone. Having for- gotten her lost fortune in gold as quickly as it appeared, Apelone greedily moves over to the table to eat. Before she can even begin, a small demon claws his way through a nearby door, tearing through it sharply. Thinking better of the situation, Apelone backs off and escapes the room through an- other exit, entering a dark room dominated by a large wall painted with Satan's face. The eyes fix their gaze on the old woman, glowing, as the door (positioned as his nose and mouth) opens to reveal a group of witches danc- ing wildly in a circular fashion, darting in and out of sight through the ori- fice. These glimpses of the unfolding Sabbat are intercut with close-ups of Apelone's beseeching face, tears streaming down her cheeks. Apelone charges the door, but it slams shut, Satan's glowing eyes ludicrously crossing as his gaze continues to hold the old woman. On the side of the room a beautiful maiden beckons for Apelone to join her, a slightly older woman observing from immediately behind. We see Apelone's relieved smile, but before she can move, the scene abruptly changes; the old woman starts awake, back in her dark basement, the moon shining through the open window. We now see a man, slumped in a chair, holding a trumpet. He moves toward the gauzy light of what appears to be dawn filtering through his window and blows the trumpet. Other trumpeters answer the call in the twilight. Apelone, in profile, stares out the window toward the sound of the echoing horns. Through Apelone's night visit to Brocken, Christensen has broadly in- troduced two more elements of the witch stereotype: the Wild Ride and the legend of "Venusberg" as a gathering place for the Sabbat. These elements return throughout the film. As the Wild Ride itself is not shown in any great detail in Christensen's depiction of Apelone, our discussion here will follow Hiixan's rhythm and will come to the Ride later. We can at this point, how- ever, say more about the settings of Brocken and Venusberg. Apelone's travel to "Brocken" is either a concession or an error on Chris- tensen's part. While the place may have been recognizable to viewers in 1922 through the famous scene of the Sabbat at Brocken in Goethe's Faust, the seventeenth-century writings of Johannes Prstorius, and Christopher 80 The Realization of the Witch Marlowe's attack against conjurors (and Jews) in his circa 1593 play Doctor Faustus,20 the setting would have been completely unknown to the people depicted in Haxan. The best-known meeting place for the Sabbat was the Heuberg ("Hay Mountain") in southwestern Germany. Sometimes also called "Venusberg," this remote site was suffused with myth well before the emergence of the witch in the late Middle Ages. Believed to be the peak where the goddess Venus convened her clandestine court, the Heuberg was known far beyond its local region, as evidenced by Nider's mention of the place where witches assembled at the Council of Basel in 1435 and the fact that the site is directly named in trial transcripts from the 1520s. Even in the case of the rare male accused, Chonrad Stoeckhlin, well-known today through Wolfgang Behringer's careful study of the case, the Heuberg was named and played a prominent role in the mobilization of the witch stereotype.21 Of course, the fact that Institoris gained, by serving as an inquisitor, the "ethnological" experience with witches that provided the basis for his writ- ing of the Malleus Maleficarum in this region is also quite significant, given the inspiration Christensen took from the book.22 Dreamtime Apelone's night flight in Haxan serves to place the audience in the milieu of "dreamtime" of the sixteenth century as described so clearly by Behringer.23 The presentation of rigorous empirical details pertaining to this dreamtime will come later in the film. Christensen is well aware that the time of the witch craze will appear naive, strange, and distant to a viewer in 1922; his task at this stage of the film is to decisively close this distance. The task is trickier than it seems, particularly given Christensen's own indecision. On the one hand, setting the mood of the film through the context of Apelone's apparent hallucination serves the need of affectively positioning the audi- ence for what follows; this is hardly revolutionary. Yet the steep slope of Christensen's naturalism in Haxan edges into view. Recalling Deleuze's de- scription of naturalism in cinema, at this point in the film we do not yet have a proper sense of either the witch or the demonic source of her power. What we do know, or more precisely what we can sense, is the originary world from Tableaux and Faces 81 which these figures come. Apelone making a quick trip to the fantastic Heu- berg or Satan lasciviously mocking the Church through the startling deni- gration of its servants are essential elements of what Haxan aspires to, but it is only a beginning. Now building on top of the repertoire of images shown in the first chapter, Christensen is visually rendering oral and written sources as elements of the image. This is parallel, but is not identical to, what the original artists were doing in creating the woodcuts and paintings seen ear- lier. Christensen is now working cinematically, seeking to create affective conditions that differ from those of painting or drawing. Again, this differ- ence, this naturalism that is now properly cinematic, relates to time. Thus, we cannot yet see the witch (Apelone hardly appears to qualify as one), but without Christensen's efforts to affectively shift the sensory world of the viewer as he does in this chapter she would effectively remain invisible to us. This was not the situation for Direr, Baldung, or Cranach. Artists at the dawn of the Reformation sought to represent the void as a figure;24 Chris- tensen, veering away somewhat from the obvious Protestant influence on his art, seeks to coax the figure of the witch out of this profound void. In a film as indebted to painting as Haxan, the impulse to ascribe to it a label such as "Romantic" or "Expressionist" is great. Such a move is not entirely without merit. The Romantic oscillation between the macabre and the lyrical appears to be one obvious correspondence. The attention paid to the extremes of mundane life and the strategies by which Expressionist painting sought to externalize states of mind do at times appears to be an- other. Read in this way, Haxan can be understood as being very similar in its modeling to the other great masterpiece of early horror cinema released in 1922: F. W. Murnau's Nosferatu (Nosferatu, eine Symphonie des Grauens). Like Christensen, Murnau took direct inspiration from the visual art of cen- turies past, composing his scenes in a manner that reflected the influences of a diverse set of painters, including Arnold B6cklin, Giorgio De Chirico, and Caspar David Friedrich.15 While some film scholars have challenged the usefulness of the label "German Expressionist cinema," it is nevertheless un- deniable that Murnau and other German directors such as Wiene drew heavily from the visual strategies and creative energies of works associated with these movements in painting.26 Considering the close ties that existed between Svensk Filmindustri and the German studios-not to mention 82 The Realization of the Witch Christensen's own strong connection to this film industry-the idea that Haxan can be placed among films such as Nosferatu, The Cabinet of Dr. Ca- ligari, and Faust (F. W. Murnau, 1926) is plausible. While we acknowledge the reasonableness of this grouping, our own analysis demonstrates that one should be wary of Hiixan fitting neatly into given categories. Hiixan does not play well with others. Rather than defend- ing or refuting questions of categorization, our own approach has been to emphasize how Hiixan corresponds with a variety of traditions without seek- ing to assimilate the film fully within one over another. It is true that we claim that Hiixan exists as a naturalist film, but this claim is intended to mark a relation rather than a rule. Faces, Tableaux In the context of our refusal to wholly associate Hiixan with any single des- ignation, our concern here shifts from whether Hiixan is indeed an Expres- sionist film to how Hiixan expressively operates in relation to other films contemporary to it. Specifically, we return to questions of tableau and face, as it is along these two poles that one can discern similarities with some (Dreyer) and differences with others (Murnau). Bordwell has identified these two elements as essential to understanding Dreyer's early films, particularly Mika l(1924), and while the interplay of these two elements produces a much darker outcome in Hiixan, they are nevertheless similar.27 In particular, the stillness and fixity of tableau-like shot composition that is evident in the works of both filmmakers and distinguishes them from nearly all their con- temporaries. Implying a closed system in such shots, the affect is often one of a suffocating organization. Dreyer takes this principle to new heights in films such as Master of the House (Du skal sere din hustru, 1925) and particu- larly The Passion ofjoan ofArc (La passion de Jeanne dArc, 1928). Christensen himself had already deployed an evolving version of this logic in his previ- ous films The Mysterious X (Det hemmelighedsfulde X, 1914) and especially in Blind justice (Hevnens nat, 1916). Hiixan, too, puts this principle to work, al- beit in correspondence with very specific countershots of great mobility and freedom. Crucially, Christensen ruptures the tableau element of Hiixan in order to visually express the lively, mobile power of Satan; specific ex- Tableaux and Faces 83 amples include the bewitched priest chasing his servant, Satan's initial erup- tion before the praying friar (both in Chapter 2), and of course the extended scenes of the Sabbat (Chapter 4) and the possession of the nuns in the con- vent (Chapter 6). Thus, only the upsurge of Satan's power can break the im- mobility of the tableau, which we find well into the sequences regarding possession and hysteria. This visual strategy bears a precise relation to how demonologists conceptualized the workings of the Devil's power in practi- cal terms. The face in Haxan also disturbs the tableau element of the shot. As with the tableau, the film resembles but is not identical to Expressionist art or cinema in this specific regard. Deleuze summarized Expressionism as the play of light and darkness, with the mixture of the two producing an effect that suggests either "fall[ing] into the black hole or ascend[ing] towards the light."28 In Deleuze's analysis, the face concentrates this series, elevating what may be symbolically rendered as "light" or "dark" to a power or a qual- ity.29 In Dreyer, the viewer finds that the face allows for a perspective that, in its suppression of depth of field and backgrounds generally, makes this affective power mobile along the lines of Deleuze's meaning: mobile as spiri- tual in its effect. Although perhaps not as finely developed as in Dreyer's later films, this formal characteristic aligns Haxan with Dreyer's work in the 1920s through to Vampyr (1932) and serves to distinguish Christensen's use of tableaux from that of Murnau.30 In Murnau the tableau frees the viewer for introspection regarding nature in a kind of emotional, spiritual release. The close-up is almost never deployed in many of Murnau's German films, as it would structurally disrupt the affect he was seeking in works such as Nos- feratu; compare this with the disruptive pathos the close-ups generate in the German director's The Last Laugh (Der letzte Mann, 1924). In Haxan the tab- leau grounds the uncontrollable forces at work on the faces of those con- fronted by the power of the witch, constituting an intensive rather than introspective power in the shot.31 The functional interplay between tableaux and faces in the course of grounding Hiixan in a naturalist cinema of the demonic is only faintly reg- istered at this specific point in the film. At the conclusion of this chapter, the audience is comforted by the thought that this is all a delusion, not yet realizing that they have not simply witnessed Apelone's "hallucination," but are in fact themselves being drawn into someone else's dream. Living among 84 The Realization of the Witch the restless dead and in the specter of an increasingly active evil, the dream- beings for Apelone would not have presented themselves as the harmless images of an overactive subconscious; there is no question of what is "real" in her dreamtime. Setting us up in this way, it is increasingly clear in Haxan that the dreamtime of the witch is not as harmless or as distant to the mod- ern viewer as we would like to believe. We are not yet fully held by the witch at this point in Haxan, but she is starting to move closer. Haxan presents itself in the formal procedure of a progressive unfolding of the material world through the style of a lecture. This form would have been quite familiar to those members of an educated, literate public in 1922 interested in the apprehension of the world through humankind's chief in- strument: science. A hierarchy emerges, and thus art and religion are subjected to the scrutiny of scientific proof. Drawing force from the near-messianic belief in the perfectibility of man, Haxan's opening chapter invites viewers into a narrative of the witch and an exploration of the wonders and "errors" of the past. But Christensen is not simply addressing an assembly of experts; rather, he is trying to draw a spectating public into the zone of the witch. Then, as now, a filmed lecture would not generally qualify as a satisfying film-going experience for anyone but the most dogmatic viewer. Continuing in this way may not even qualify Haxan as a work of cinema. Haxan thus abandons the rather overbearing didacticism of the opening by moving directly into the "underground home of a sorceress in the year of the Lord 1488." Now the static images presented earlier come to life on the screen. The full force of this reanimation only becomes apparent as the film moves forward-Haxan gives an affective form to the otherwise ab- stract, myth-like notion that witches were widely believed to be real and powerful in the early modern period. Christensen is composing images of a figure that is already present. The witch of the opening chapter was a perceived thing held at a distance, a set of ready-made circuits of recognition and association that nearly any viewer would instantly recognize. From here on out, Haxan progressively moves away from the cliched figure, bringing us dangerously close to the real power of the witch in the process. Our use of the hazardous term "real" is meant quite precisely, as Haxan proves to be a film based on a magnified form of realist cinema. More specifically, Haxan is rooted in a naturalist impulse. Through an assemblage of fragments from this basis in its formation of a Tableaux and Faces 85 cumulative image of the witch, Haxan will, in Deleuze's words, make ap- parent "the invisible lines which divide up the real, which dislocate modes of behavior and objects, are supercharged, filled out and extended."" Chris- tensen not only marks out an image of the present in Haxan but also con- jures the aura of a seemingly timeless origin myth, collapsing the distance between them in the violent multiplicity of Haxan's surfaces and figures. Christensen's witch is not only here now, it has always been here: a figure of nature. Demonologists of the fifteenth and sixteenth centuries explicitly as- sociated their work with the "advancement" of natural knowledge.33 In a de- liberately perverse style, Christensen appears to agree with them. It is a testament to Christensen's skill that the viewer is seized by the witch despite the nagging suspicion that one should "know better." Like everyone else, the viewer is gradually ensnared. Terms like "capture" or "seizure" denote physicality, a manual activity that lay at the heart of Christensen's method. This seems paradoxical con- sidering the virtual nature of images, but it is important to remember where Haxan begins-with woodcuts, drawings, and paintings that originate from an act of touch. While it is questionable if the cinematic image can ever achieve the tactility of the painting or woodcut, Christensen aspires to sur- pass the commonsense division between the tactile and the optical in order to generate for cinema viewers what Deleuze in a different context called a "haptic vision."4 This corresponds precisely to the tactic in the opening chapter of the film, to present figurative, cliched givens as they establish the ground by which Christensen can transform these figures from virtual giv- ens to haptic modifications through the remainder of the film. The eye and the hand work together in Haxan to mold the image. This is possibly due to Christensen's ability, working expertly with cinematographer Johan Anker- stjerne and set designer Richard Louw, to correspond the movement of cin- ema in line with what are more commonly painterly images. In this sense, Haxan bears a close relation to the silent films of Carl Theodor Dreyer in that the "plane-ness" of the image, the negation or perversion of depth of field (particularly through the close-ups of faces), and the occasional use of eccen- tric, disorienting continuity editing (eye-line mismatches, violation of the 18o-degree rule, etc.) produces a molded, affective, tactile quality that compels the viewer to grope for the image.35 In the case of Haxan, the "clay" from which Christensen will mold these haptic images-the tactile substances 86 The Realization of the Witch by which one can touch (or be touched by) what is happening onscreen has already been literally shown to the audience and will continue on occasion to recur throughout the remainder of the film. Haxan seizes (the audience) and is seized (by the witch), establishing a formal cinematic strat- egy that parallels the very problems of seeing and touching virtual beings (such as devils) that transformed what constituted evidence for the pres- ences of the witch. In Chapter z, Haxan began with image fragments, narratively held to- gether through an almost belligerent narrative logic. In Chapter 2, Chris- tensen empties out and "paints over" these figurative givens in his own assemblage of the witch. In moving to the live action of "the underground lair of a sorceress," the film is now utilizing the repertoire of oral tradition: "old witches tales." Now firmly in the creative mode of composing images on top of given visual surfaces, Haxan will move progressively from tales to theology, and finally, diagnosis. It is a rigorously logical structure that works simultaneously to throw the fissures of the real into relief while also boldly expressing the tangible singularity of the power of the witch. T H R E E The Viral Character of the Witch In witchcraft, words wage war. Anyone talking about it is a belligerent. -JEANNE FAVRET-SAADA, Deadly Words (1980) In the third chapter of Haxan, Christensen returns to his didactic mode of presentation. Through Franz Heinemann's Rites and Rights in the German Past,' he presents to the viewer a common investigative technique deployed by inquisitors and witch hunters: trial by water. Heinemann's image is in- tercut with another image, a detail of a bound, naked woman undergoing a similar trial, which Christensen tells us is drawn from Eduard Fuchs's Illus- trated Social History from the Middle Ages the Present.2 The title card explains the scene: "If she floats, she will be pulled up and burned. If she sinks, the judges thank God for her innocence."3 The scene is not a "trial" but rather an experiment. It needs to be clear, however, that the aim is not to identify "the seed" or origin of a particular evil, only evidence of its presence. As we will show in this chapter, such trials operate though a form of non-knowledge that comes from the mastery of non- sense. In lieu of direct, unmotivated proof, it was common for inquisitors to undertake such experiments in order to obtain evidence. These procedures, 87 88 The Realization of the Witch when understood as adjacent to witnessing, display the growing concern felt by authorities at the time regarding their ability to prove what they knew in advance to be true. Experiments such as "trial by water" demonstrate not an indifference to the truth and a retreat into superstition, but rather a deep (if not misguided) appreciation of cause-and-effect relations, as well as the forces at work in the natural world. Christensen presents this experimental practice as barbaric, but he also allows its logic to remain clear. After all, only a witch could manipulate the properties of a natural substance like water in order to avoid drowning in this experiment. These practices concerned cases suggestive of and empirically linked to general laws. The case said some- thing about the world-and once a case was established, it would spread like a contagion. Making a Case Despite the powerful illustrative quality of the case study, there is an obvious aporia opened when the individual provides "evidence" of something different from what is understood as a collective phenomenon. In practice, the indi- vidual is useful insofar as she represents an ideal type.4 The particular necessity of the individual to perform this role is already problematic, as psychoanalysts and anthropologists can certainly attest. In nonfiction cinema, the distance between the individual case and the ideal type is nearly impossi- ble to close, clearly reflected in the famously controversial proto-ethnographic film also released in 1922, Robert Flaherty's Nanook of the North. Aspiring to document the life of the Inuit through the reenacted case study of Nanook (real name: Allakariallak) and his "family" (also actors, of sorts), Flaherty's film expresses a certain truth regarding the total social environment. There is great irony in the fact that Flaherty can do this only by undercutting his claim to objectivity, as the film affectively brings case study examples alive as sub- jects. At the level of a cinematic artifact, this effect does not by definition sug- gest failure, as the force of Nanook's life not only provides empirical evidence as to his mode of living but also allows for a reflection on "nature," "human- ness," and "modernity" rooted in the haptic qualities of Flaherty's images.5 Although more fragmented as a work, the middle chapters of Hiixan move in a direction similar to Flaherty's film through the presentation of a single The Viral Character of the Witch 89 reenacted witch trial. No specific, historically documented trial is men- tioned, suggesting that Haxan will present for the viewer a "typical" proce- dure. It is a case study populated by ideal types drawn from similar facts and events. Haxan also enters the case study through a scene of illness and diagno- sis. In light of the film's strategy to present the audience with a typical event, such an entry remains true to its purpose, as concrete forms of witchcraft were found in the real misfortunes that people faced, particularly maladies with poorly understood causes and cures. In such times, the maleficium of the witch was palpable and its physicality highlighted throughout Haxan. Although many other forms of maleficium are cited across primary histori- cal sources (destructive weather magic, assault of farm animals, etc.), by far the most profoundly embodied (and cinematic) type would be sickness and unexplained death. Hiding in Plain Sight The scene of diagnosis in Haxan is beautifully composed, a tableau of fig- ures expertly arranged. A man (Jesper the Printer) lies in his sickbed, mo- tionless, as a male healer (Peter Vitta) and various female members of the household attend to him. Christensen's framing moves between medium shots and close-ups of the attendants, not revealing who they are, but estab- lishing that the man's wife and infant are among those in attendance. Their faces are tense and apprehensive. Scenes of the activity in the bedroom are interspersed with scenes of activity in an adjoining kitchen. Older female servants bustle and silently carry out their tasks. In the next room, a tiny old woman slips into the kitchen unnoticed by the servants, hurrying across the frame. "The power of lead will soon reveal it," Anna, wife of the Printer, is reassured. The male healer moves to a small cauldron at the foot of the bed and then carries out the simple rite, which ends with the reading of melted lead suddenly solidified in cold water. The power of Saturn is invoked in the ritual. The inscrutable physical result of the lead experiment is re- vealed: the cause of the malady is "atrocious witchcraft." The direct reference to Saturn in the ritual is significant to the arc of Christensen's cinematic strategy, as it provides the basis for a chain of possible 90 The Realization of the Witch associations. Although we will offer a fuller explanation of the ancient Ro- man god's link to concepts of witchcraft later in the book, it is important to note that Christensen is using a well-established trope when identifying Peter Vitter as one of "Saturn's children" in this scene. Georg Pencz's wood- cut Saturn and His Children (1531) captures the character of the belief that Saturn serves as patron to social outliers, including the poor, elderly, and disabled, as well as criminals, Jews, cannibals, magicians, and witches.6 Throughout the century, Saturn's mythological violence was increasingly associated with the demonological violence of Satan. Witches were under- stood to be clients of the Devil's patronage much as Saturn's children were bound to the ancient god and to one another. In short, what results is a conjoined figure of Saturn/Satan.7 This con- flation serves as additional evidence of natural magic, illustrating how long- established techniques of practical magic were subsumed into the figure of the witch. Yet, while this is hinted at in the scene of Jesper the Printer's di- agnosis, it is important to remember that Peter Vitter is not suspected of being a witch in Haxan. It was certainly not unheard of for lay healers and diviners to find themselves among the accused, but this is not Vitter's fate.8 In fact, once he has completed his diagnosis, he disappears from the film; interestingly (or perhaps tragically), so does the sick man, Jesper the Printer. This scene in Haxan is significant for its depiction of the steady medical- ization of witchcraft, as well as for offering us clues to the director's own position in his cinematic thesis. Specifically, while it is difficult to determine Christensen's intent, the scene draws a clear line between the empirical in- struments and techniques of the healer and the persistent notion that Satan himself was, in fact, the principal authority of the natural world. Henri Boguet offers one of the clearest examples of this belief in his Discours des sorciers (161o), arguing to the point of dogmatism that Satan was not only fully subject to the laws of nature but was also the foremost master of the knowledge of natural properties and the techniques of their instrumental- ization.9 If Christensen's only intent was to provide yet another visual dem- onstration of people's naivete and misguidedness during the early modern period, he did not have to go to the lengths of this otherwise minor scene. Michel de Certeau observed that every exercise of trained judgment is au- thorized through the dark, ratifying force of theology.10 The association be- tween Peter Vitter and Saturn gives the veiled sense that the healer's powers A A LWurlrn0. G-e org Pencz, Saturn and his Children (1531). Courtesy of British Museum, London. 92 The Realization of the Witch may originate, either knowingly or unknowingly, from dark forces that rat- ify all forms of natural expertise. As Haxan moves from theology to diag- nostics, it is difficult to avoid the thought that the witch's power may be, at its core, identical to that of the inquisitors, exorcists, and physicians who seek to "remedy" her acts. His task accomplished, Peter Vitter packs up and readies to depart through the kitchen. The Young Maiden,11 the sister of the sick man's wife, Anna, is foregrounded, shown slowly turning the lump of lead over in her hands, ner- vously fondling the oblique object. The Young Maiden then rushes to Peter Vitter demanding to know the identity of the sorceress. He answers crypti- cally, "You might see that witch, before you wish to.. ." The Young Maiden repeats these last four words, wandering through the kitchen in a daze. The scene returns to the old woman who entered the kitchen earlier. Peter Vitter is not simply playing a cruel game with the Young Maiden. He has, in effect, done everything possible and expected. As a man who has a certain "touch" when it comes to divining the causes of supernatural mis- fortune, he has identified the mechanism of the Printer's malady. This is not, however, the same thing as having knowledge of the specific source. Confirming the suspicion of witchcraft is, if anything, a form of non-knowledge derived from a mastery of nonsense; it opens a gap in knowing (specifically, who). It would be imprudent and dangerous to go further. It is not up to Peter Vitter to name the witch; this burden is on the Young Maiden and the rest of Jesper's family. Still, it begs the question, why would Peter Vitter stop short of giving a name? In fact, Peter Vitter has given a name; he has announced that a witch is at work here, and this naming is essential to the arc of counteracting the maleficium that has been visited on the household. The act of naming is dan- gerous enough for Peter Vitter, as he has not only laid hands on Jesper but has also come in contact with the (un)natural force that has stricken him. The healer knows that the source of evil is close-hence his cryptic message to the Young Maiden as he departs. He is, we assume, divulging everything he knows in order to turn the family in the right direction, simulta- neously assigning their roles within the progressive structural logic of witchcraft. Who is close enough to Jesper the Printer to undertake this evil spell? Who would have cause to do so? These are questions that the family must answer. Peter Vitter has given them more than enough infor- The Viral Character of the Witch 93 mation for them to carry out their tasks; it is now up to them to move to the next step.12 Peter Vitter has departed. The Young Maiden stares wide-eyed, tears streaking her face. Although it would seem that we are not at a particularly riveting point of the narrative, the jolt of the image effectively mirrors the Young Maiden's state of mind. This is also the first of many shots in Haxan where Christensen employs an extreme close-up of a face. The Young Maiden instinctively utters a "sacramental" phrase and identifies the old woman: Maria the Weaver. It is clear that it is not going to take long for the Young Maiden to put into further motion what Peter Vitter began. Maria begs for a meal and the Young Maiden reluctantly agrees. The dis- dain for the uncouth old woman is evident on her face. She serves Maria, betraying a suspicious look as the woman ravenously eats. Christensen re- turns to tight facial close-ups to convey Maria's animalistic mannerisms and the Maiden's horror. The images move from close-ups of Maria's food- smeared face, to a medium shot of the Young Maiden disgustedly turning away, again to Maria, and finally to the Maiden turning to face Maria, her face contorted with revulsion. The Young Maiden runs into the adjoining bedroom where her sick husband rests, slamming the door shut behind her. The dynamic, tensile relation between tableau and face has reached its height, and continues from this point forward in Haxan. As we discussed earlier, this relation between "tableaux" and "faces" is established from the beginning of the film. As he moves into his case studies, where cliches and particularities collide, the dynamic qualities of Christensen's approach be- gin to overwrite the figural givens in the previous sections. The link to Dreyer is obvious; at this fever pitch, no face in Haxan performs exactly the same formal task or conveys the same affective sense for the viewer. Thus, Christensen's witch, like Dreyer's Jeanne d'Arc, resists the enfolding con- tours of a single "type" in sequence with other equally composed faces. It is in this way that Christensen has made the witch his. Forgetting for a mo- ment the sympathetic logic that makes this so, it is not entirely clear if Chris- tensen recognizes that the polarity of this expressive, metoposcopic force can flow in more than one direction. The Young Maiden can now name the witch. We see it on her face as she calls out to others in the house. They rush in to hear her frenzied discovery. Christensen moves the action along through a quick succession of medium 94 The Realization of the Witch shots of the commotion and extreme close-ups of the women arriving one by one at the same conclusion. He returns to close-ups of Maria wildly shoveling food into her mouth to amplify the sense of what is to come.13 There is only one intertitle dialogue in this sequence, as the Young Maiden warns her sister that Maria "has evil eyes." The women pantomime their troubled reasoning so clearly that no explanation is needed. The viewers are meant to draw the same conclusion. Maria is indeed the witch of whom the healer spoke. We know that she is the one. The Young Maiden prepares to get help. Her mother, whose previous close-up revealed her to be more doubting, surveys the scene with the same look, again shown in close-up. It is unclear if she is suspicious of Maria or of the panic of her daughters. She leans out the door to look at Maria; her face is stone, furrowed and troubled. Another mask. With the opening scenes of the film's third chapter, Haxan has established one line of accusation that was quite common in the early modern period. As nearly all historians of the so-called witch craze agree, the desperate search for the cause of what was otherwise an unexplained illness or mis- fortune was frequently the catalyst for specific witchcraft accusations be- tween friends, acquaintances, and often between family members themselves. This is a very old and well-documented story of the fear of maleficium expressed through violence and killing (as much as forming an alliance or buying her off). What has changed by the sixteenth century is that this violence now bore the sanction of both secular and religious institutions. Thus, the mere fact of misfortune does not account for the specific mode by which this often-deadly reaction would take place. After all, human beings have been suffering misfortune, illness, and death long before the power of the witch was felt during this time. However, now peasant complaints of maleficium were by being heard by experts, who had determined, well in advance, that a satanic conspiracy was underway within Christendom.14 In effect the Church subsumes the role of the un-witcher-a role that someone such as Peter Vitter would have taken on in previous centuries. No longer the personalized, supernatural contest between witch and un-witcher, it is now the Church that, in opposition to its previous position exemplified by the Canon Episcopi, performs the task of engaging the witch. As Haxan powerfully demonstrates in the pivotal scenes that follow, this engagement is no longer characterized by the deployment of techniques to counteract The Viral Character of the Witch 95 individual maleficium, but rather becomes more fully directed toward estab- lishing the witch as a key player in a transcendental drama aimed at de- stroying Christianity itself. Witch as Vector Haxan leaves Jesper the Printer's home where (the witch) Maria remains. The scene is now the interior of an ornate gothic church. A young priest, Brother John, is shown as he approaches a shrine of the Virgin Mary in the foreground. He begins to pray. The Young Maiden bursts into the church, distracting the priest from his prayer; he turns, startled. Although Brother John's reaction is not as tightly framed in close-up as the previous shots, the sequence rhymes visually with the Young Maiden's own shocked reaction earlier. The Young Maiden calls out to the priest and rushes over to him; he slowly turns away, reticent. Again, the face dominates the screen in tight close-ups. The Young Maiden tearfully recounts her story to the young priest. The face of the priest is fixed and without emotion. He slowly turns and looks at the Young Maiden. He reports (via the title card) that, as the youngest inquisitor, he cannot speak to strange women. His fear of impropriety (and its perception) is confirmed as the film cuts to an older priest, revealed in the credits as "Jo- hannes," descending the stairs in the background and observing the en- counter with narrowed, suspicious eyes. The Young Maiden and the fledgling priest are now positioned in the foreground of the frame, with the small figure of the distant eavesdropping friar on the stairs visible in the space between them. The Young Maiden, not taking "no" for an answer, darts forward and grabs the priest's arm, continuing to plead her case. Stunned, the young man slowly looks down toward his touched arm, the Young Maiden's hand grasping his bare flesh under the sleeve of his cassock. This is too much for the spying senior priest and he barks loudly at the couple, causing his younger colleague to reflexively jerk back and spin away from the Young Maiden. They are caught-not only in an act of impropriety, but also by the creeping power of the witch to pervert and pollute. It is no accident that Christensen introduces Brother John by showing him reverently approaching a shrine devoted to the Holy Virgin. Mary was 96 The Realization of the Witch elevated by Catholic demonologists such as Institoris, who belonged to a Do- minican order known for aggressively promoting the cult of the Virgin, to the status of the "perfect" woman. The Holy Virgin therefore provided a stark contrast to the lustful, credulous nature of common women who were often associated with the temptations of Eve, indexed in this discourse as the first to succumb to Satan's subterfuge." Brother John reacts in this scene in Haxan as if the Young Maiden were handing him forbidden fruit, reflecting the ferocious sexual repression demanded of the priest. Not long before, it would not have been uncommon for priests to engage in forms of clerical concubinage. Although such practices fueled heretical movements such as the Waldensians and the Brethren of the Free Spirit, they would not have necessarily drawn scorn from the general community.16 By the time of the Reformation and Counter-Reformation, clerical attitudes regarding sexual desire had hardened to the point of elevating the status of sexual neurosis to a virtue. Clearly, Brother John's repressed distain for the beautiful Young Maiden will make a surging, deadly return later in the film. While these scenes are certainly put in the service of Christensen's unfolding thesis, which directly connects witchcraft to clinical hysteria, they also ring true in their dramatization of primary evidence from the period. The Young Maiden rushes toward the older friar-she will make some- one listen. The force of the Young Maiden's accusation is again conveyed through her face rather than title cards. The relative status and position of each character is emphasized by the camera's dramatic angles. The Young Maiden's close-ups are shot from above as she gazes up at Johannes, beseech- ing him to take action. In turn, the inquisitor benignly, almost dismissively, addresses the girl from his elevated position. A tight smile crosses his lips as she relays the accusation. Recalling similar scenes of Father Henrik's first confrontation with Satan and anticipating the charged confrontation between Maria and her inquisitors, Christensen consciously uses this technique to index status throughout Haxan. The Young Maiden is ushered into the private workspace of the inquisi- tors. The mundane work of the church continues as she enters. Christensen shows this by lingering somewhat peculiarly on a friar having his head shaved to form his tonsure. Father Henrik, recognizable as the priest who was ha- rassed by Satan himself in the previous chapter, rises to receive them. The scene returns to Brother John in the main hall, now with a dumbfounded The Young M\aiden looking up in Il/f vai, ilm still (Svensk Hilmindustri, 1922). Johannes looking down in Haxan, film still (Sxvensk Filmindustri, 1922). 98 The Realization of the Witch expression on his face. He carefully regards the arm that the Young Maiden had only moments ago grasped in her excitement. "How wonderful!" he exclaims. "It felt like fire when the young maiden took my arm." His eyes widen, a look of passion building on his face-he whirls around toward the room upstairs. Meanwhile, the two older inquisitors now remain with the Young Maiden. Father Henrik is seated; he is obviously the senior member of the inquisitorial staff and it is he who will do the talking. He warns the Young Maiden of the seriousness of her accusation and demands that she swear by the cross that she and Maria "are not deadly enemies." Before we have her answer, the film abruptly cuts to Maria, still eating in Jesper's kitchen. The scenes of the Young Maiden in the church are pivotal in Haxan as they reinforce several crucial elements of Christensen's overall thesis. With an economy of shots, the conflicted status of women is directly introduced into the film, particularly through sternly patrilineal visual motifs evident in the shots of the Young Maiden interacting with the older inquisitors. The mixture of condescension and bemusement exhibited by these men works to visually intensify the Young Maiden as a frail and hysterical woman. The priests take her seriously in the end, as they must take all witchcraft accusa- tions seriously. Importantly, while Christensen makes it clear that the priests are critical toward the girl, he is also careful to show the audience that they are, within the realm of their own assumptions about the world, dedicated to investigating and verifying her claims. Contrary to some prevailing his- torical claims being made in the period contemporary to the film, the in- quisitors are not shown to be gullible, fanatical, or overtly misogynistic in the hearing of the Young Maiden's accusation.17 Haxan's witch hunters act in accordance with their own procedures for investigating truth and false- hood and not simply out of malice, fear, or stupidity. While this claim must be taken in light of Christensen's at times explicit condescension, it is clear that the senior inquisitors display a complex and careful attitude toward the Young Maiden and receive her accusation in light of this complexity. The Young Maiden's initial encounter with the inquisitors is dominated by two issues, one that is obvious with the benefit of hindsight, and another which is not directly dealt with in the film but is important to note. First, the manner in which the friars simultaneously display suspicion toward the girl's claim and a marked desire to believe what she is alleging is conspicu- The Viral Character of the Witch 99 ous when the Young Maiden names the witch to the inquisitors. It is im- portant to distinguish between the desire to believe and simple belief. They are not the same in Haxan, just as they were not the same during the witch craze. Despite the unbelievable scope of demonic power, the inquisitors must believe that what the Young Maiden is reporting is possible. Second, while the relative ease by which the Young Maiden is able to make her deadly accusation is presented economically, Christensen skips over some important changes in the legal systems in Europe at the time. Well into the 1400s, proffering a formal indictment against another individual required the plaintiff to submit to an accusatory form of criminal proce- dure. Derived from Roman law, this procedure presumed such offenses as matters between the accused and accuser. Thus, the idea that "crime" was a matter between society and the accused did not exist, making the present- day distinction between criminal and civil complaints meaningless. Raising a formal complaint required the accuser to furnish proof of the allegation and, importantly, to submit to severe penalties agreed in advance if the judge was unconvinced of the complaint's merit. In short, it was complex, expensive, and very risky to enter into this formal framework in order to ad- dress disputes or everyday injustices. As a result, most ordinary people did not do so, choosing instead to pursue local and less formal modes of redress.18 With the emergence of the witch in the fifteenth and sixteenth centuries in Europe, this procedure changed dramatically. As the witch was under- stood as an agent of Satan, the crime of witchcraft came to be understood as a crime against Christendom, and thus society itself.19 Logically, the procedures for discovery and eradication of such rebellion shifted, with an emphasis on the totality of the offense and the responsibility of civil and religious authorities to protect pious Christians from a power that, without the establishment's intervention, would overwhelm the faithful regardless of their individual acts, intentions, or beliefs. This is the logical basis for the hallmark procedures of the inquisitions, as depicted in Hiixan. Further, the responsibility of the accuser shifted from that of providing hard proof to the expectation of merely reporting what they suspected to the proper authorities. The inquisitors, firmly positioned as experts for the first time, would take this suspicion forward administratively. In an impersonal socio- logical sense, this is an example of Weberian rationalization. At the level of individual relations and passions, it meant that the inchoate suspicions of ioo The Realization of the Witch the fearful, the resentful, or the spiteful could now be duly received with- out the formal restraint of a judgment recoiling back on the one giving voice to the charge. Clearly, to suspect someone of maleficium was not new. Rather, the novelty that contributed to the explosion of witch accusations was that authorities were now eager to act on suspicions independently. Thus, accord- ing to its own subterranean expression, Haxan indicates this significantly transformed context of the complaint in the scene of the Young Maiden for- mally incriminating Maria the Weaver.20 In contrast to much that is only gestured toward in the Young Maiden's initial encounter with the inquisitors, Christensen telegraphs the complexity of Brother John's encounter with the girl much more explicitly. He acts prop- erly toward the Young Maiden when he initially attempts to ignore her, and yet it takes only the slightest touch to transfix and overwhelm the young man. His elation in the aftermath of the encounter is clearly meant to index Brother John as ignorant. It is also a not-so-subtle bit of foreshadowing on Christensen's part, as the grammar of the film would not allow such quick, decisive passion to simply dissipate. In this sense, Brother John's passion mirrors the force of the Young Maiden's own feelings in the face of the "witch" Maria, and Christensen skillfully links the affects of each character in the visual expression of their encounter. The meeting between these two young people strongly echoes the pre- viously introduced theme of sexuality. A straightforward reading of this scene implies that witch accusations were often generated by unresolved sex- ual desires and that such passion would be redirected in this pathological, perverse manner.21 Although historians sharply debate this as a causal factor in witch accusations, it is clearly a position that Haxan favors in light of the scenes that follow. The reintroduction of sexuality in this way engages the issue from another angle; the film emphasizes the crucial role sex played in discerning what constituted witchcraft and its status as a knowable category of (malefic) human practice. As Walter Stephens has argued, it was crucial for inquisitors to consider sex in the context of witch trials as sexual desire. The actual act of sexual intercourse was one of the primary empirical means by which witchcraft could be made visible.22 Haxan develops this theme and refers back to the fateful encounter between Brother John and the Young Maiden as the film progresses. The Viral Character of the Witch 101 Laying Hands on the Witch As the third chapter of the film draws to a close, we are witness to Maria's ar- rest under the charge of witchcraft. Although not shown, it is apparent that the inquisitors have found the Young Maiden's accusation credible and they have sent men to apprehend Maria while she is in the family's kitchen, eating what appears to be the largest bowl of soup in recorded history. The magistrates sneak in and roughly bundle Maria into a large sack; still sitting at the table gorging herself, she did not notice the men approaching from behind. The mother enters dramatically from the bedroom; first shaking her fist at the struggling old woman and then spreading her arms wide in a Christlike ges- ture, she warns the men to bind Maria lest her feet touch the floor and her demonic power returns, allowing the old woman to turn them all "into mice." Her warning reflects a long-standing worry among Church and secular authorities that they would themselves become bewitched due to the nature of their work. As Clark has demonstrated, a great deal of demonological thinking was devoted to justifying the fact that civil and Church officials, despite their fears, by and large were not bewitched. In the moment of ap- prehension, the magistrates were immune as they were instruments of the sovereign, understood as an extension of the Divine. As such, a witch's power was drained at the moment of "touch" by the instruments of the Divine.23 Being "but" an ordinary woman, there is no way the mother would be aware of the existence of this immunity. Maria is brutally launched into a small cart that has been pulled into the kitchen. The intensity of the scene is conveyed to the audience, particularly as Christensen returns again to the strategy of paralleling the action with a series of quick, tight close-up shots of the faces of the women. The sister, crying out for vengeance, is shown dramatically double-framed by the boundaries of the shot and her Hapsburg-style headdress. The Young Maiden, who has presumably returned to the house with the magistrates, excitedly witnesses the scene, a look of blood lust palpably dominates her face, rein- forced by her furtive gnawing of her fingernails. Finally, the stony face of the mother appears; she pumps her fists to urge on the men. A title card relays the mother cursing Maria as a "damned mistress of the Devil." We see a close-up of Maria's withered face. She is pleading. 102 The Realization of the Witch Christensen is, through his actors, following Satan's idiom. This is very ambiguous, as it is entirely unclear if one can "imitate" Satan in any sense without in some direct way simply being Satanic. The ambiguity is mirrored in Haxan's tortured relation to "the truth." The question of empirical cer- tainty and reenactment haunts the status of the film as evidence. Although he overtly disavows this association, Christensen has to a significant degree in Haxan sided with the inquisitors, figures who themselves were caught by the witch. He is relying on the fact that the truth of the witch will take its most visible form by acting her out mimetically. Such performance revealed the truth for inquisitors, who by this period did not require evidence of spe- cific malefic acts to make a conviction of witchcraft. It was also sufficient for Flaherty, who knew that the visceral force of Nanook of the North depended on the felicity of his Inuit interlocutors reenacting themselves. In each case, acting the ideal type breathes life into the emptied, cliched figure. It is a tired truism to claim that cinema directors "act like God" in creat- ing their films. However, displaying an intuitive sense of his craft and his subject, Christensen chooses to "play" Satan instead, using his idiom to breathe life into his witch. The interrogation of Maria in the next chapter, the old woman confessing through Christensen's possessive ventriloquism, dis- pels what little doubt remains regarding the truth of the witch in Haxan. F O U R Demonology Trace and aura. The trace is the appearance of a nearness, however far removed the thing that left it behind may be. The aura is the appearance of a distance, however close the thing that calls it forth. In the trace, we gain possession of the thing; in the aura, it takes possession of us. -WALTER BENJAMIN, "The Flaneur" The fourth chapter of Haxan gets to the heart of the matter. It is here where Christensen's skill as a filmmaker, his facility with historical artifacts and documents, and his understanding of witchcraft and the burgeoning fields of neurology, psychology, and anthropology come together in the full arti- culation of the film's thesis. The director brings to life the rich visual cul- ture of witchcraft. He draws links to the spectacular illustrative diagnostics of Duchenne and Charcot and his visual interpretation of Freud's theories of neurosis and the human psyche, which again and again find their way into the film. Playing with an ambiguity inherent to deriving truth from testi- mony (as ambiguous in Freud and Malinowski as in the transcripts of witch trials), Christensen takes the invisible objects of such testimony and flashes them on to the screen. Taking Freud's talking cure a step further, Christensen seems to be offering a cure for the secularized Christian blindness at the heart of positivist human science. Just as the inquisitors are enchanted with the witch, and by her evil, in the same way his audience is pulled in, not as 103 104 The Realization of the Witch misguided inquisitors, but through a lens of science, the details of which we have yet to see. The traces of demonological thinking and the diagnostics of nervous dis- ease are by now discernibly etched upon the surface of the film through Christensen's complex method of inscribing such traces in the composition of images in Haxan. Suggesting that these etchings are "finding their way" to the surface is generally correct, but our meaning is specific. It is not that secrets rise to the surface from the murky depths beneath the image. Chris- tensen plumbs these depths for the viewer from the beginning, placing his labor in clear view. By working with figurative givens of witches and demons, the director formulates his visual thesis. Like Freud's memory traces, the "visual etchings" here are conceptualized as etchings into a material, an act that equally implicates the eye and the hand in the apprehension of the tac- tile, haptic images Christensen produces in Haxan.1 The director is not re- vealing secrets in Haxan; he is carving outlines into the image of figures that have been hiding in plain sight all along, longing not to be seen but rather interpreted. This allows for the power of a figure such as the witch to be perceived, perhaps quite suddenly, in a manner that loosely corresponds to the second term in Walter Benjamin's formulation above: aura. This aura, this sense, distinguishes Haxan from other attempts to empirically docu- ment the witch, following less the logic of dogmatic objectivity than the Freudian understanding of memory that posits it as a trace that is etched someplace in the unconscious and yet simultaneously saturates that very un- conscious with its substance. Where most factual accounts would seek to bury any sense of aura, Haxan blasts it into the open as an uncanny, mobile power. The fourth chapter of the film commences with Maria being "processed" for interrogation by the magistrates. The procedure entails a careful search of her body for "witch powder," carried out by two "honest matrons" who strip the old woman and comb through her hair. The search occurs off- screen, implying that the examination will be much more invasive than can be shown, heightening its perverse thrill. Christensen correctly shows that it was common in sixteenth-century Germany for witches to be completely stripped and searched, not only for "witch powder" or other malefic instru- ments, but also for the telltale mark that Satan was believed to etch on the bodies of his followers. Areas of thick hair growth on the body were of Demonology 105 particular interest, and the head, armpits, and pubic region of women ar- rested under the suspicion of witchcraft were often shaved in the course of the search.2 Maria is at least spared the indignity of having her head shaved. "The suspect's nights are now dictated by the inquisition judges," the title card informs us. Hiixan then moves to a series of images depicting Maria's interrogation. The questioning takes place under torture from the beginning-our first look at the now imprisoned Maria shows her confined to the stocks. Two "honorable men" attempt to elicit a confession from Maria using a vicious pantomime of a "good cop/bad cop" routine. The vio- lence of the men jerking the frail old woman back and forth is emphasized by Christensen's rough match shots, each jerking Maria into frame, bounc- ing her back and forth between them. Maria appears confused and defeated throughout this sequence. The next title card details the procedure: "If she stubbornly denies her charges, they will use a kind of mental torture." In fact, several "sacramen- tal" items are then shown. The old woman would certainly be affected by the objects, not because of their innate power but because of the power of divinity Maria would believe they held. Although often overstated as an in- fluence, Christensen's reliance on the Malleus Maleficarum as a source is quite obvious here, as sacramental objects such as the ones depicted are dis- cussed at length.3 While not particularly concerned with the details of the witch that Hiixan is gearing up to display through Maria's confession, the Malleus offers a great deal of practical, albeit at times dangerously contra- dictory information (such as asserting a great power to the ritual utterances in sacramental rituals while also asserting that God confers no inherent power upon words), as to how witchcraft can be practically counteracted. The focus is not particular to the Malleus, as other famous demonologi- cal works such as Nider's Formicarius also claim an efficacy for sacraments and spend a great deal of effort detailing the proper use of such ritual in- struments. It was common for theologians to distinguish between the sacra- ments and powerful (but lesser) rituals that were nevertheless rooted in the same logic and drew their power from the same divine source. Unlike the sacraments, however, sacramentals such as the ones depicted in Hiixan could be used by laymen and the clergy alike and possessed an unlimited power of iteration depending on the situation at hand. Importantly, sacramentals 1o6 The Realization of the Witch directly counteracted the works of the Devil-they were acts against Satan rather than the works by God.4 To imply, as Haxan does, that the use of such instruments was universal or uncontroversial stretches the historical record, however, as powerful ar- guments were made (particularly by German Protestant intellectuals) that the belief in such "quasi-sacraments" was itself a form of diabolic idolatry and should therefore not be used to combat the power of the witch. Indeed, when compared with the earlier scenes of Karna's use of practical magic, Haxan itself demonstrates the inevitable conclusion that sacramental rituals such as the ones deployed by the inquisitors were ultimately rooted in the logic of magic, plain and simple. For the leading intellectuals of the Reforma- tion, these techniques were superstitious at best and evidence of the idolatry of the Catholic Church promoted through its clergy. Such rituals were formally unnecessary for these Protestants in counteracting the power of a witch under judicial custody, as the inviolability of a sovereign, cascading down the ranks of his judicial machinery, stripped the witch of her ability to act by diabolic means.5 Alternatively, a number of prominent texts, including the influential Canon Episcopi, held that the powers of the witch were really nothing but de- ceptions created by Satan, illusions that dissipated in the face of divine justice embodied by magistrates and inquisitors. Again, in this context, sacramen- tals such as holy words written on consecrated parchment and consecrated wax "as Corpus Christi" were commonly used, but were not a universal practice when interrogating witches. Scripts and Pacts From these relatively minor scenes, Hiixan moves to the centerpiece of the chapter: Maria's interrogation by her inquisitors. In keeping with the previ- ous view on counter-magic and sacramentals, Maria is brought into the torture chamber backward in order to neutralize any residual malefic power. Christensen frames the scene with a beautifully composed tableau of the torture chamber, the instruments of torture balanced across the image. Emphasizing the claustrophobia of the scene, there is a frame-within-a-frame, the stone building supports and low, vaulted ceiling serving as a thick, black Demonology 107 border (just slightly off-center) for the shot. The inquisitors are moving at the center of this limited line of sight, seated in preparation for the formal questioning of Maria. Similar to Dreyer's films of the same period, Chris- tensen is deliberate when using a depth of field, highlighting the solid shapes of doors and arches without creating the sense that this depth signifies an open or free space.6 Rather, his strategy is consistent with the static plane of the tableau. Hiixan cuts to a medium shot of the four inquisitors; they are the same men who were present when the Young Maiden made her accusation in the church. As in the long shot, they are composed in a recognizably classic sense in the frame, visually recalling the style of the woodcuts and drawings that ap- pear in the film's first chapter. Father Henrik, the senior inquisitor, is seated behind the others on a simple version of a bishop's chair; he demands to know if Maria is now willing to confess. Maria is silent, mustering as defiant a look as her frail, withered face will allow. The inquisitor is obviously unhappy with her attitude, dramatically rising and ordering the torture to begin. Christensen shows this first in the medium shot displaying all four inquisitors and then, after a title card, in the original long shot, the senior inquisitor gestures to the executioner who has entered the frame. Hiixan cuts to a close-up of Maria's feet as she is being secured in the stocks. Her face slowly registers pain, again rendered in a close-up. The ex- ecutioner takes a particularly evil-looking pair of tongs off the wall where various torture instruments are hanging-the inquisitors silently watch. This medium shot, utilizing the earlier framing, forcefully expresses the tableau quality Christensen is seeking. The four men are completely still, recalling E. H. Gombrich's description of Vermeer's paintings: "still lifes with human beings."7 The depth of field and static, composed subjects sta- bilize the sequence. The viewer is returned to a close-up of Maria, tears well- ing in her eyes. She asks how can she confess to something that is not true; her question (intertitle) is met with stony silence by the senior inquisitor seated completely still in his bishop's chair. Christensen inserts several match shots between the inquisitor and Maria, emphasizing the suffocating, con- frontational nature of the scene. The old woman's pain escalates; she gri- maces and throws her head back. The inquisitor leans forward, increasingly aggressive in his posture. The totality of this violence is evident in the io8 The Realization of the Witch Maria's inquisitors in Haxlian, film still (Svensk Filmindustri, 1922). composed alterity of the scene's stylistic correspondence between accused and inquisitor. "Let now the evil witch's body sting," says Father Henrik to Erasmus the executioner.8 A quick reaction shot of Father Henrik follows, then the film immediately cuts to a dramatically tight close-up, edged by the black frame of a partially closed iris, of Brother John's troubled face. Eyes closed, it is unclear if he is reacting to the spectacle of Maria's pain or the crude double meaning of his superior's remark. There is a rhyming close-up shot of Ma- ria's face, in the same position and with the same expression as the young inquisitor's. Christensen's visual link between the accused and the accuser is explicit here. The director's disdain for the lead inquisitor is also explicit, as the film then moves to a medium shot of Father Henrik taking a break from baying at Maria to swill a drink. Again visually linking the senior inquisitor to the accused, the shot recalls earlier images of Maria's ill-mannered, wild eating in Jesper's kitchen. By the final cut the viewer is brought full circle to the initial confrontation between Maria and her inquisitors. Demnizology 109 Close-ups in Ilixan, film still (Svensk hilmindustri, 1922). The sequence has shifted the oscillating rhythm between tableau and face that Christensen has established with his method. Dramatic space now re- quires reading movement from face to face, often without the stabilizing term of the tableau. In this scene, Ilaxain most resembles the "flowing close- ups" deployed by Dreyer in The Passion ofjoan ofArc, suppressing perspective and depth of field in favor of a continuous affective movement as expressed in the face. The effect is that even the medium shots that are occasionally interspersed between tight shots of Maria and her inquisitors function as close-ups by virtue of the fact that spatial distinctions fade, leaving it entirely to the spirit or the aura of the shots to carry the narrative forward. Hdxan, is not grounded in a setting here; it is grounded in the forms of life present in the shot. The lead inquisitor then directs Erasmus to allow Maria to "catch her breath" and offers to lessen the intensity of the torture in line with the full- ness of her confession. Again, cutting between close-ups of Erasmus and Maria, Christensen continues to ratchet up the intensity of the exchange. Maria, her will now broken, begins to confess her "evil deeds." Although the timing of Maria's turn is sudden, her fear, pain, and the sense of corporeal alienation from herself are chillingly conveyed. Her reversal recalls Lyndal Roper's assertion that this sense of alterity in relation to one's own body was critical to the "success" of witchcraft confessions.) Able to convincingly inhabit a script not her own, Maria's confession will take her curiously eager inquisitors (and the audience) to the heart of what witches do, regardless of 1 1o The Realization of the Witch the sliding ambiguities of fact and memory. This detail uncomfortably joins Hiixan in a very literal way to the historical scene of a witch's interrogation. Maria's confession is the vehicle by which Christensen's streamlined vi- sion of the European witch comes to life, overlaid onto what are by now the destabilized visual cliches from where the film began. Focused on the Sab- bat, Christensen refers again to nearly all the elements that served to con- stitute the witch stereotype, including the Wild Ride, the pact with the Devil solemnized through sexual intercourse, cannibalism, the cauldron as the locus of the rite, and its location at Venusberg. Although it is obvious that Christensen is selective in bringing these traces to life, the effect is power- ful and largely accurate when judged against the surviving records of the time." The power of the witch, the supernatural drama of the Sabbat, and an overwhelming demonic power is never more real in IIaxan than during this visual retelling of Maria's confession. The "guilt" or "innocence" of Maria in a positivist framework is largely beside the point by this time. Indeed, it is the task of the accused to bring the inquisitors within range of the power of the witch, just as it is the mis- sion that Christensen has set for himself and his unlikely "star" actress1 to do the same for the audience. The safe haven of objective distance is closed off for inquisitor and viewer alike; clearly it is a distance that both desire fervently, albeit unconsciously, to close. Ii xan covers this range through precisely the same means as would have taken place during an interrogation of a witch-through the biographical narrative of an accused woman. What these narratives enable for the inquisitor and the filmmaker are slightly Demozology 1 1 r different, however, as the judicial machinery of a witch trial required evi- dence of criminal acts that by definition could not be witnessed." Witness to Things Unseen As Ha xan demonstrates in bringing Maria's confession to life, the power of cinema to witness exceeds that of the witch hunter. The viewer not only wit- nesses Maria's act of testifying but also witnesses the acts testified to for themselves (already a violation of what is true for dogmatic positivists). In actuality, what Hixan does is "worse" than rigging the truth in that it aligns itself not with a concept of truth or the real but with the power of the witch outside such judgments. Bearing more than a passing resemblance to the emotionally charged, unequal relationship between the analyst and the pa- tient, the complex relation between the accused and the inquisitor is repli- cated in the relation between the film and the audience.14 As is often true in the human sciences, H/ixan works through instruments of knowing rooted in the dynamics of the confession. To this day such instruments attain the status of empirical fact, challenging the conditions by which we can assert that something is real. In light of the position Hdxan starts from, Chris- tensen appears to do this despite himself. Maria begins her confession with an account of sex. This starting point is no accident. We are taken back to the recesses of Karna's dark lair, this time the image tinted with the deep blue of night. Karna is there with her assistants 1 12 The Realization of the Witch helping Maria through a jarringly disturbing birthing. Terrifying, unidenti- fiable hybrid creatures squeeze through the elderly woman's womb, drop- ping to the floor. For the contemporary viewer, these creatures often elicit nervous laughs. In an age of computer-generated special effects, the costum- ing and props in this scene appear hopelessly amateurish. Yet Christensen has carefully chosen the manner of appearance of these demonic creatures. Perhaps clumsily rendered, the wriggling demons reflect an interesting set of variations to the witch stereotype, both ontologically and visually.15 Interestingly, Christensen's account of Maria giving birth to demon chil- dren appears to run counter to nearly every demonological theory on the subject. Drawing heavily from Thomas Aquinas and his theorization of the "virtual bodies" of angels in the Summa theologiae (asserting that "angels do not need bodies for their sake but for ours"), Remy and others argue that demonic intercourse was always sterile.16 Often referencing the same pas- sages in Aquinas, Francesco Maria Guazzo in his Compendium maleficarum acknowledges that such unnatural couplings could produce children, but that the bodily essence of devils would rule out the possibility that these children would themselves be demons.17 The authors of the Malleus agree, claiming that devils themselves are unable to reproduce but can, in the form of succubi, steal semen from men and in turn use that semen to impregnate women as an incubus.18 Less concerned with grotesque offspring, the Malleus takes up this ques- tion as a way of illustrating Satan's attack on the sanctity of marriage and his potential for disrupting and destroying this holy bond. The susceptibil- ity of wives to the Devil's erotic charms and of husbands to witchcraft that specifically hindered the performance of his duties, particularly by "steal- ing" his penis, are given particular attention by Institoris and Sprenger.19 Among the major writings on witchcraft and demons of the period, only Sylvester Prierias's De strigimagarum daemonumque mirandis (1521) agrees with Christensen's position on Satan's ability to impregnate human women and produce offspring such as those depicted in this scene in Haxan.20 It is unlikely, however, that Christensen was relying on Prierias as a source, as the sixteenth-century author was primarily concerned with Satan's ability to manipulate and pervert language (including erotic language) and there is no evidence that the director directly consulted this text. Demonology 113 As with nearly all such instances of artistic license in Hiixan, some re- markable truth is embedded within the scene. Echoing none other than the esteemed pioneer of ethnographic filmmaking Robert Flaherty ("Sometimes you have to lie. One often has to distort a thing to catch its true spirit"),21 Christensen opts to draw the viewer's attention to the fact that fears of the witch often focused on the threat she posed to reproductive processes. The evil of the Devil was practically limitless; the witch, by contrast, directed her attention to very specific domains of life: marriage, children, farming, and health. In an insecure age, it is unsurprising that the danger most often attributed to the witch was a threat to fertility.22 Facing the prospect of a visually tedious cinematic explanation of such a wide-ranging fear, Chris- tensen opts to "lie" in order to catch "the true sprit" of this persistent, and by his reckoning causal, element within the story of the witch. In the span of several seconds this free-floating anxiety is firmly locked in place via an unforgettably graphic image of an elderly woman giving birth to deformed, hybrid demon children. Rapidly recounting her "crimes," Maria suggests a small bargain with her inquisitors: "If I am spared the pain, I will confess that Trina has smeared me with witch ointment." Then shown naked in bed, Maria displays obvi- ous pleasure as Trina applies the salve over her leathery skin, the key light- ing of the scene producing a sharp, reflective sheen. Although seemingly incidental to modern viewers, Christensen's attention to the application of the salve is consistent with that of the inquisitors he is portraying, as the use of the salve was one of the most verifiable signs of the witch. Modern his- torians have speculated that the salve may have been the pharmacological source for the visions and tales of the Wild Ride and the Sabbat, as recipes for these salves (reported centuries after the fact) were said to sometimes contain reference to real narcotics such as belladonna.23 Johannes Nider's famous tale of the woman who attempted to prove that she could fly through the air to a Sabbat by anointing herself with a salve also appears to indirectly support this theory. In particular, the fact that the woman in Nider's story reported that she flew despite the eyewitnesses seeing only that she fell into an insensible trance (of course they beat her while she was under the influ- ence just to make sure) appears to suggest the salve worked as a psychotro- pic. Cohn regards the issue as unsettled. While noting that authors as diverse 1 14 The Realization of the Witch Albrecht Direr, Witch Riding Backwards on a Goat (1500). Courtesy of - TBritish Museum, London. as de Spina, Weyer, and Tostato also report the use of salves with similar results, Cohn's own research found that most salve recipes did not clearly refer to any known psychoactive substances, and he notes that none of the above authors states having actually witnessed such a trance; even Nider's legendary account was told to him by a mentor.24 Christensen appears to be well aware of the ambiguity of what the salve actually does, offering no definitive comment on it. Rather, the visualization of Maria's incrimination of Trina in Haxan is immediately followed by a sharp cut to the outcome, imagined or otherwise, of the application of the salve: the Wild Ride to the Sabbat. High Flying What follows in the film is an extended view of the Wild Ride using the special effect of superimposition to show women from all over the area borne up in the air, propelled toward "Brocken." The speed and fluidity of the Demonology 115 women's flight in this extended sequence, composed of a series of swift track- ing shots of stationary actors giving the effect of high-velocity flight past a stable frame, is as impressive over ninety years later as it must have been upon Hiixan's first release.25 Several motifs associated with the witch that Christensen has not been able to deploy thus far are now on full display, particularly the wild, flowing hair of the women (a common visual metaphor for sexual promiscuity and disorder) and the various "conveyances" (mostly brooms, but benches, stools, and cooking forks also appear to be represented) "known" to be used by witches in flight. It is also here where the overarching element of the witch stereotype is fully displayed: the massed, coordinated,female nature of witch- craft. Wild riding women appear from all points in the frame, multiplying in midair. Christensen's Wild Ride also references earlier source myths of the Furious Horde, the ancient Roman creature the strix, and the associa- tion of night riding with the Roman goddess Diana, who possessed a revived cult status in the late Middle Ages.26 It is important to emphasize that the camera itself does not appear to move in this fast-paced sequence. Even in this scene devoted to the wild abandon of the witch's flight, Christensen maintains the stability of his image by min- imizing depth cues and eschewing techniques that would allow the energy of the event to rupture the image. The shots maintain this flat, tableau per- spective even when demons are visible in the foreground, their movements minimized to balance the swift movements taking place overhead. The cam- era seldom moves in Hiixan, and when it does it completely avoids tracking movements in or out of the image that would disrupt this naturalist unity. Demonstrating Christensen's mastery over his chosen medium, this strategy here stands in contrast to his strategic use of such tracking shots in his earlier film Blind justice. It is clear that the director knew precisely when to move the camera and when to keep it still. The spectacle of Hiixan's special effects should not overshadow the critical importance of the scene in relation to the film's thesis. Cohn notes that, while established beliefs regarding "ladies of the night" flying did not them- selves generate the witch hunts that came later, these much older beliefs nevertheless proved to be crucial in the assertion that witches were numer- ous, highly organized, and under the command of a supernatural leader. Images of cannibalistic night witches were explicitly a demonological variation 116 The Realization of the Witch on long-standing popular conceptions. In the context of such widely held beliefs, it was not difficult for these theorists to plausibly assert that what the Romans called a strix or what peasants would refer to as the "ladies of the night" were actually misunderstood iterations of the witch.27 Using the same tactic that Christensen himself takes up in Haxan, demonologists were able to plausibly join what were until then disparate things, giving force to rationalist, Universalist frameworks, first regarding the existence of the witch, then later vis-a-vis the reality of the hysteric. As Haxan's Wild Ride persuasively (albeit inadvertently) shows, both ideal types relied on the same affective power for their force. Alternating between glimpses of the frenzied gathering in long shot and closer looks at the participants in the orgiastic rite, Haxan moves to an over- view of the Sabbat itself. The scene strongly conveys the energy, tension, and irresistibly horrible eroticism of the rite. The Wild Ride has already ushered Haxan's audience into this affective zone, matching the disordered riding with the perceived sexual inversion and moral disarray such night flights have long symbolized.28 Interspersing shots of demons wildly play- ing brass instruments and various drums around steaming cauldrons, and medium shots of young, beautiful women lasciviously dancing and fondling demons, Haxan is clearly referring not only to voyeuristic and sensational- ist accounts of the Sabbat but also to the inverted time of the carnivals that were very popular in the decades immediately before the outbreak of the witch craze.29 The wild dancing is the key, as it signals the disruption of the everyday, the inversion of what life "normally" had to offer.30 The effect is less one of horror than of desire and abandon. Although this is easily forgotten in the excitement of the images, Maria is still narrating this story to her inquisitors (and to the audience). She makes the curious claim that the "Devil's grandmother was there with all of her witchcraft." The identity of Satan's grandmother is never specified, but the visualization provides some clues to Maria's reference. As Charles Zika has demonstrated, well-known classical tales dating to antiquity were actively modified and put to use in the assertion of the reality of the witch in the early modern period.31 One of the most common literary and iconographi- cal schemas involved the ancient god Saturn and his "children." This asso- ciation is briefly alluded to in Haxan in the earlier scene of Peter Vitter Dernozzology 117 Ltl t-brrea~ II La lx * l/l jw.c atlil r~yi vi meas >iCue~ n i+Fn b raer fteora~rifn J nnba e 'r r y eapceeuetc~ia 0 ibanfrbef ife 2znbf ZictcIbin a erro~i riDetmnff/ titifewne f[de9 elrctnllqe rnlbl an it ilimb t 2(6itlnffiab td flnantd int ert eN y .2! Sofd6rftfit6clmmim/roigvei et Zof t ri 8eff r'lr RIb0) ceffll Itnff/ Z4l aet ttntr uAv~cn i,63if." fteIarute An 1 b Wdny~gti vit/l Zai be, £LD6CII nun qarl/ 4 L)Arillr Ztdgl L rql ir0felr II1L1I ~ bb ffde)rtf'iljflb rfctinl/ !tteiil aii inba l I Zfeuifte~wgc / 2" Nll(ifnlin lii Zei ie Uklllffullrllu / 7ririqQ7 Z r/il 0 te a reriubeQ1¢t~l Xbyfret8 fiurltlttroe ttllltillt. iFa§Ytu~teboIttI