Now and Then: False Confessions, Beliefs, and Memories in the Modern and Early Modern Eras

David Begelman

Historians widely assume that confessions of witchcraft across three centuries, from 1400 to 1700, were produced in an atmosphere of extreme pressure. Consequently, scholarly efforts to establish the actual existence of witch cults through the haze of forced testimony usually have been complicated for this reason (Cohn, 1975; Midelfort, 1974).  It is less clear how to account for confessions that were seemingly not created under duress by secular or Inquisitorial courts. These involved a minority of women who stepped forward to admit they were guilty of the crimen exceptum, ostensibly without coercion: the voluntary witches. These accused had some significance for witch-theorists of the era. Whatever residual doubts many of the latter may have harbored about the reality of crimes or heresies for which suspects were rounded up, the fact that some had volunteered to confess without external pressure tended to erase lingering doubts about the veracity of the charges against them. Thomas has provided a flavor of the kind of case in question: a certain butcher, Meggs, who voluntarily traveled ten or twelve miles to be interrogated by the witch finder Matthew Hopkins, and was later “duly executed as a witch” (Thomas, 1971, pp. 518-519).

During the era in question, a growing chorus of skeptics needed to reconcile the existence of voluntary witches with presumptions about their innocence. One route was to question whether unforced confessions were other than actually voluntary, being effects of less visible social pressures. Accordingly, some thought that ostensibly voluntary confessions might have been testimonies born of subtle influences which, while short of torture, were nonetheless similar to it in effect. For example, in the investigation of Basque witchcraft in Logrono, Spain, the inquisitor Alonso de Salazar y Frias in 1614 posited just such tacit influences on witch testimony. He felt that “free” confessions were actually the consequence of subtle inquisitorial approaches. He pointedly accused colleagues in his tribunal of “insinuating ideas into their [witches] minds in various ways” by promising the accused they would be freed if they owned up to their misdeeds (Baroja, 1965, p. 187; Henningsen, 1969; 1980). During the Salem Village scare of 1692, all defendants who confessed to witchcraft were spared execution. It would be difficult to plead voluntariness of confession on the part of such defendants in a modern court of law, especially if it could be established that the accused were aware of the connection between confessing and avoiding the gallows—or being pressed to death, as in the case of Giles Corey (Upham, 1867; Hansen, 1969).

Voluntary and Insane 

Another route to explaining the voluntary witch confession was to question the sanity of the accused, a solution that reconciled the presumed falsity of witch confessions with the absence of any inquisitorial or identifiable influence in producing them. Johann Weyer, author of De Praestigiis Daemonum in 1563, chose this second alternative, one that comported well with the misogyny of the early modern period, since most of the accused were women. Confessions of imaginary misdeeds were consistent with the mental disarray caused by melancholia, as Weyer declared witches to have—with or without the horrific incentives to confess arranged by governing bodies (Mora, et.al., 1991).

While cognizant of the questionable status of statements extracted under torture, Weyer, like other demonologists of his time, took them in the main to be faithful expressions of a witch’s belief about her Satanic powers. In Book VI, Chapter XXIV of De Praestigiis Daemonum, Weyer maintained that his opponents could never show that witches had concluded pacts with demons, in part because of tainted voluntary confessions. One type is “defective because after the fact the Devil impresses it upon the imaginations of these poor women that they have done that which occurred naturally or that which he himself brought about” (Mora, et. al., 1991, p. 549). Elsewhere in Chapter XXIV, Weyer characterizes witches as “stupid, deluded, old women” (Mora, et. al., p. 548). 

One assumption about voluntary confessions was that suspects who produced them actually believed they were servants of Satan. The hunch would be riskier had they confessed under torture. During torture, one could argue that the confessor was only verbally compliant in confirming inquisitorial scenarios of witchcraft. Under such circumstances, the witch gave the inquisitor what he wanted to hear, rather than a concocted tale she actually believed. On the other hand, the historian might maintain that the only possible explanation for the voluntary witch not under suspicion until she freely “confesses” is that despite the falsity of her testimony, she truly believes her story. However, the modern understanding of voluntary confessions makes the matter more complicated than this.

When one compares early modern to contemporary jurisprudence, it becomes apparent that an enormous shift has taken place in understanding the myriad factors that may come to affect testimony. Today, we take for granted the vulnerability of verbal reporting to a wide range of variables distorting it. As a result, the relationship between testimony and the reality it uncovers is for us no longer the straightforward affair it was for an early modern jurist. Tribunals during that time took it for granted that confessions extracted under torture were trustworthy. Today, such procedures virtually epitomize the reasons for distrusting the disclosures they produce. Currently, we know that influences far less egregious than torture can influence testimony. How much less before a point of diminished confidence in it should develop is an extraordinarily complex issue. In the last analysis, only forensic studies can supply the answer, and not overnight.

Voluntariness

 

Complex slants on the analysis of voluntariness have a modern provenance. For example, in Kaimowitz v. Michigan Department of Mental Health, 42 U.S.L. Week, 2063 (Michigan Circuit Court, Wayne City, July 10, 1973), the “voluntariness” of a decision to undergo psychosurgery is abridged merely by the fact that a patient has been committed to an institution by the court. Here, whether or not the act is “voluntary” has nothing to do with the character of specific procedures aimed at ensuring informed consent, the particular state of mind of the patient at the point of agreeing to the procedure, or any other local influences impacting a decision to accept it. The court declared that involuntarily committed patients “live in an inherently coercive institutional environment” (Martin, 1975, p. 172). Accordingly, decisions to accept institutional psychosurgery even when informed can never be legally “voluntary.”

Cases like Kaimowitz strike an interesting note in the concept of the voluntary act. They commemorate junctures at which complexities in its analysis drive a sometime wedge between the psychological and legal definitions of “voluntary.” That is, under Kaimowitz, behavioral science or the social psychology of decision-making is largely irrelevant to whether a decision to undergo psychosurgery is truly voluntary: it is institutional status alone that determines this, according to the ruling. In other words, the decision of an institutionally committed patient to submit to the procedure is always “involuntary,” independently of what we know about the psychology, mental state, or behavior of the individual or his supervisors.

When Kassin and Wrightsman aver that “Voluntariness is a difficult concept to operationalize since it requires inference about the suspect’s subjective state of mind” (Kassin & Wrightsman, 1985, p. 70), their view is plausible only when it pertains to the psychological arena of discourse. In the legal arena, “voluntariness” can never be operationalized, any more than can “due process,” because the concept is forensically open-textured (Waismann, 1963). That is, its meaning will shift as a function of its legal history, often independent of the growth of knowledge in the social sciences. While the latter may be relied on to inform the law, it can never define the latter. As it turns out, the addiction to the probative value of the verbal confession—however produced—is not a bias restricted to the early modern period. It is the hallmark of contemporary police investigations as well (Ratner, 1983; Huff, Rattner & Sagarin, 1996).

           False Confessions

Contemporary experts in the psychology of false confessions are careful to distinguish cases in which the falsity of testimony is obvious from the larger category of cases in which falsity may be entertained only as a plausible hypothesis. The latter supposition springs from our current state of knowledge about predictor variables that have been demonstrated to contaminate testimony, making it unreliable—whether or not a crime suspect confesses or is convicted by a court. In obvious cases, it is only a matter of logic that when over two hundred individuals confessed to the kidnapping of the Charles Lindbergh baby, all but one—or possibly several on the assumption of more than one conspirator—of these confessions were patently false (Note, 1953). Likewise, numerous voluntary but obviously false confessions were lodged in the case of the “Blue Dahlia” murder in which the mutilated body of an aspiring actress was discovered in a vacant lot in Los Angeles (Ellroy, 1987; Gilmore, 1994). More recently, the voluntary false confession of John Mark Karr in the murder of Jonbenet Ramsey was given widespread media coverage. In other types of cases, such as the Central Park jogger case of 1989 (Ryan, 1989; Kassin, et. al., 2005) or the Michael Crowe case in 1999 (Sauer, 2002), false confessions resulted in the arrest or imprisonment of individuals subsequently declared innocent on the basis of either DNA evidence or the coercive tactics of arresting officials in extracting confessions.

Because many unknown variables come into play in confessions of crimes, social scientists have pointed out that there is no current way of calculating the incidence or prevalence of false confessions (Kassin, 2001). Unless human nature has changed radically from the early modern time-frame, there is no reason to assume that witch confessions were any less vulnerable to the effects of variables contaminating present day testimony.

The witchcraft historian has one advantage over the contemporary social scientist: all witch confessions of maleficia and diabolism were false, not merely on the basis of how they were extracted, but also on account of their thematic content. Despite their testimony, voluntary or otherwise, witches didn’t fly through the air, shapeshift, or wreak magically sourced harm on neighbors, not to mention possess other supernatural capacities attributed to them. Accordingly, for the historian, unlike the modern investigator, quandaries over the determination of actual guilt or innocence do not nowadays arise as they did in the past. When Thomas insists that “historians who dismiss the whole notion of witchcraft as groundless are mistaken” his remark was not intended to endorse the reality of maleficia, but simply to give credence to the idea that “however fabricated the confessions extracted from them, at least some of the witches felt genuine hatred for those around them” (Thomas, 1971, p. 523). Such witch sentiments, together with the occasional coincidence of imprecation and misfortune, could create an illusion of diabolical power. In fact, the accidental contingency between a ritual and external happenstance reinforced the belief in magic and sorcery, black or white. Without the imagined connection, there would have been no belief in witchcraft, any more than there would be a practice of medicine without ill health [i] .

In categorizing false confessions, Kassin (1997) and Kassin and Wrightsman (1985) have developed a tripartite framework: the voluntary confession (V), the coerced compliant confession (C), and the coerced internalized confession (I). The first of these types involves false testimony of a self-incriminating kind that is supplied without any external pressure, as in the case of the voluntary witch. Thomas’s example of the unsolicited confession of the butcher Meggs is probably such a case, as is the over 200 confessors to the Lindbergh baby abduction. Establishing a true (V) confession provides no guarantee the suspect in question, witch or otherwise, believes his or her own testimony. There are enough contemporary instances of voluntary confessions in which individuals provide testimony they could not have believed was true, and which was disclosed for a variety of reasons other than personal conviction about its veracity. Some of these include craving attention (Wrightsman & Kassin, 1993), the wish to impress friends (Radelet, et. al., 1992), or the wish to protect friends or relatives (Gudjonsson, 1992).

The second type of case, (C), are those confessions extracted under extreme conditions in which the testimony is produced in order to avoid an aversive state of affairs like torture, or to secure some reward or social enhancement, but where the suspect knows he or she is innocent of the charges involved. This type of case may characterize most instances of accused witches during the early modern period, although how many is difficult to determine in retrospect without assessing the percentage involving the (V) and (I) types. The (I) cases are those in which a suspect comes to actually believe the charges lodged against her, incorporating the outlook on her actions adopted by her interrogators.

Kassin and Wrightsman’s classification of cases has been subject to criticism on the grounds that it mischaracterizes the features of confessions over time (Ofshe and Leo, 1997). However, such a criticism may well confuse the usefulness of a taxonomic scheme with the predictable characteristics of the confession types it classifies. Thus, should it transpire that the internalized beliefs of (I) confessions are temporary and fail to persist unchanged over time, this finding does not impugn the legitimacy of the original taxonomy; it only furthers empirical knowledge of its subcategories. Fine-tuning information about a probable course of events within a class is not tantamount to claiming misplacement in a superordinate category.

Coerced Compliant Confessions

An oft cited example of the (C) case is that of Johannes Junius, Burgomaster of Bamberg, Germany. In 1628 this “witch” signed a confession after being put to the question and subjected to thumbscrews, leg vises (Beinschrauben), and eight hoistings on the strappado (Robbins, 1959). This procedure involved hoisting a suspect by the arms tied behind the back so that shoulder joints are wrenched in unbearable agony. The latter was sometimes enhanced by attached weights to the feet before hoisting.

 We can with some confidence classify Junius’s confession as the coerced compliant (C) type because he managed to have a letter he wrote to his daughter Veronica smuggled out of prison. It went on at great length how brutal methods were responsible for his confession, a concocted tale of relations with and baptism by the Devil, attendance at a sabbat (i.e., witches’ conventicle), cavorting with a succubus named “Vixen,” nightflying on animals, and other boilerplate witchcraft scenarios. The note to his daughter admonished her to get out of town for her own safety, informing her that “Innocent have I come into prison, innocent have I been tortured, innocent must I die. For whoever comes into the witch prison must become a witch or be tortured until he invents something out of his head” (Robbins, 1959, p. 292).

Coerced Internalized Confessions 

In the third type of case, the coerced-internalized type (I), the suspect comes to believe in the veracity of the charges against him or her. A risk in documenting this type of witchcraft case centers around complexities in establishing a suspect’s self-belief after interrogation, and the possibility of mistakenly confusing true belief with verbal compliance merely, or a (C) case. The paradigm of the (I) case in the witchcraft literature is widely acknowledged to be that of the benandanti (the “do-gooders”), an alleged agrarian group of peasants whose existence in the late sixteenth century was unearthed by Ginzburg (1991; 1992)[ii]. The group resided in Friuli near Udine in northeast Italy. Benandanti were born with a caul, or amniotic sac they later wore as an amulet. Elsewhere in Europe, the caul as a harbinger of special status took on other meanings. In Slavic countries, it augured becoming a werewolf (Ginzburg, 1991, p. 155).

Curiously, the benandanti, predominantly men, fancied themselves adversaries of witches, predominantly women, although the ranks of the latter were reportedly peopled by minority wizards. Benandanti testified that armed with fennel stalks they rode into battle against the witches at night on the backs of animals. The outcome of these engagements four times annually during the Ember days forecast the following year’s crop yield. A failure of the campaign against the witches would mean low yield and famine.

Ginzburg’s exercise in micro-history has been singled out as the only evidence for a European witch cult in the immense literature on the subject. Midelfort has stated: “Ginzburg’s benandanti remain to date the only authenticated witch cult in early modern Europe” (Midelfort, 1974, p. 204). Russell contended that the cult was solid proof regarding the existence of witchcraft (Russell, 1972). Both historians may have since modified this opinion.

Benandanti have been widely characterized as suspects who, at the instigation of the Inquisition, became convinced that they were in fact heretics and witches, and not the heroic brigades they had originally styled themselves to be. If so, they exemplify the (I) confessional type, or testifiers who come to believe the guilt urged on them by their interrogators. However, it is difficult to determine whether benandanti conversion was authentically (I) when it might have exemplified (C) for reasons already outlined. That is, what scholars take to be conversions of the beliefs of benandanti about their status from anti-witch to witch may have been collusion on the part of witnesses to verbally appease inquisitors out of the fear of consequences. When first interrogated, one benandante, feeling the accusation that he was a witch was absurd, laughed in the faces of his interrogators. Some historians would have us believe that just because this suspect subsequently became verbally compliant, he was no longer laughing on the inside.

When Maria Panzona in 1618 admitted that her evening escapades were actually witches’ sabbaths, did she really believe this, or was she merely placating sponsors of regurgitated inquisitorial scenarios forced upon her?  That the latter mind-set prevailed is suggested by the fact that on the verge of seeming to undergo an (I) conversion, Maria recanted, reverting to benandante palaver. One can thus surmise the (C) flavor of her retraction. Might the same have been true for other benandanti were it not for their fear of inquisitorial consequences? For example, one accused benandante, Olivo Caldo, completely recanted his testimony in a private audience with a bishop’s representative after he attempted to hang himself in his cell on January 2, 1645. He insisted everything he said during his trial he said “…out of fear and dread of the law, and that he really thought that in this way he might more easily and quickly free himself from the toils of justice” (Ginzburg, 1966, p. 139).

On the subject of English witchcraft, Thomas has indicated that even innocent suspects might come to believe their guilt under pressure of interrogation (Thomas, 1971, p. 526). The conversion on occasion perhaps occurred. However, it is one thing to allow for this possibility and quite another to prove it happened. We need more than verbal compliance to confirm this; we need evidence he or she actually believed in the testimony—or disconfirming indications of dissembling under pressure, as was the case for (C) victims like Johannes Junius and Olivo Caldo.

On the contemporary scene, the problem of distinguishing (C) from (I) confessions is also rife. For example, Barthel (1976) cites the case of Peter Reilly, an 18 year-old who found his mother murdered after returning home. Suspicions of matricide on the part of the police led to interrogations involving deceptive procedures assuring the suspect of his guilt. Reilly eventually acceded to the pressure to sign a confession, although subsequent evidence proved his innocence. Kassin and Wrightsman classify the case as an (I) one, presumably on the basis of Reilly’s making such statements as “Well it really looks like I did it” (Kassin & Wrightsman, 1985, p. 78). However, the classification as a coerced internalized confession appears to rest on, inter alia, interpretations of Reilly’s responses as (I) rather than (C)-tinctured. That is, Reilly might have capitulated only verbally, and in order to end an otherwise aversive questioning session. Moreover, the appropriate categorization of Reilly’s confession has implications for Ofshe and Leo’s (1997) criticism of the Kassin and Wrightsman tripartite classification scheme. As was mentioned, they fault the scheme in part because of the tendency for (I) beliefs to degrade over time, reverting to pre-interrogation beliefs. However, the latter process might be indicative of the fact that many, if not most, (I) beliefs may have been miscategorized in the first place. In short, if (I) confessors subsequently revert, it may have been the case that they were (C) cases to begin with, rather than cases in which self-belief was altered in the direction of the allegations in question. As was mentioned, Ofshe & Leo’s theory of (I)-case reversion has no implication for the legitimacy of the tripartite scheme. Even in the hypothetical event (I) confessions were actually non-existent, being covert (C) types, the scheme would still be a valid one—on the assumption (I) confessions could be categorized as such should they in future occur. The problem here is not the legitimacy of a taxonomy, but the precise determination of confession types thus classified under it.

Altered States

Benandanti, when questioned by the Inquisition, testified that the night battles were on the spiritual, not the physical plane, a distinction with some importance in the wider literature on witchcraft, especially in attempts to reconcile night flights to sabbats over enormous distances. Excursions in dream states are a culture-bound pattern around the world quite familiar to anthropologists. Because the night battles were reportedly conducted in an “ecstatic” state, commentators have hastily concluded that this meant that benandanti were subject to catalepsies (Cohn, 1975), a somewhat shaky inference. The supposition requires a more unique or sharply defined psychological state: suspended animation, in which the voluntary control of musculature ceases.  However, it is difficult to assess whether the night battles were (a) simply a culturally reinforced myth among a loosely connected group of individuals unified through the possession of cauls to which they attributed magical properties; (b) an actual trance-state or altered state of consciousness given narrative substance by the shamanistic folklore surrounding it, or (c) accounts heavily colored, if not often created, by the inquisitorial process. While (c) cannot be reconciled with the fact that inquisitors began their interrogation of benandanti only after receiving rumors of their activities, it may be naïve to suppose that details of such disclosures to inquisitors were unaffected by close questioning over repeated occasions.  

That the first of these hypotheses is more plausible than the second is suggested by the fact that the combined motif of the night battle, animal transformations and transvection, cauls, predominantly male combatants, and ecstatic states was a pan-European theme extending far beyond the boundaries of the Friuli region. As a related albeit different motif, Wilson (1963) and Bourguignon (1972) reference the African  Nyakyusa tribe, where the counterparts of benandanti, the abamanga, battle witches in dreams. While such isomorphisms may indeed be the result of “semi-erased historical relationships” (Ginzburg, 1991, p. 195), it may be a stretch to imagine independent cultural groups fantasizing quite similar things because of the effects of hallucinogenic properties of ergot or the mushroom claviceps purpurea. Ginzburg broaches this theory, only to aver it lacks confirmation due to the absence of etymologically loaded derivations (Ginzburg, 1991, p. 303-307).

 If benandanti accounts of ecstatic night battles were obviously not reality-based, it begs the question to assume that battles on the “spiritual” plane impose obligations on scholars to unpack them as catalepsies—or any other obscure psychiatric condition—as if the spiritual state once reified requires grounding in an exotic psychological reality other than shared folk belief, dreaming, or reverie, or was drenched in more veracity than nightflying itself. An alternative assumption is that surviving folkloric tales are remnants of systems with largely ideational, rather than experiential components. Contemporary Christian evangelicals may give accounts of being “saved” without any transformative experience necessarily grounding such ideas. Once the questionable assumption of an altered state of consciousness is granted, it becomes easy for the scholar to fetch around for exotic neurological categories to rise to the task of connecting the dots. On the subject of the witch, Ginzburg asserts: “It seems reasonable to suppose…that if not all, at least some of the confessed witches, used unguents capable of inducing states of hallucination and delirium” (Ginzburg, 1966, p. 17). On a somewhat different note, he insists that the riddle of the benandanti cannot be resolved by pharmacology or psychiatry (Ginzburg, 1966, p. 18).

On the contemporary scene, a modern variation of the witchcraft theme appeared to characterize the confessions of Paul Ingram that he had sexually abused his two daughters in a ritualistic manner (Ofshe, 1992; 1994; Wright, 1994). The testimony of Ericka and Julie Ingram contained wildly implausible allegations against their father, including accounts of incest, group sex, cannibalism, and infanticide at Satanic rituals. The allegations mirrored older scenarios, since they were identical to those surfacing in early modern witch narratives based upon confessions we know to be false. The case has stirred up considerable controversy as to Ingram’s actual guilt or innocence, with social psychologists emphasizing the latter, and judges, police, and juries siding with the former verdict.

Several commentators, in arguing a case for Ingram’s innocence, maintain that his false confession was in part produced by a “hypnotic state,” “dissociative state,” or “trance” his interrogators induced in him (Ofshe, 1992; Wright, 1994; Kassin, 1995). The assumption may have served to explain why an ostensibly (I) confession was produced so readily. However, given the atmosphere of pressured questioning, the social context, and the implicit task demands of the interrogations, there may be a more parsimonious explanation of the Ingram confessions without recourse to altered states differing from ordinary principles effecting belief change.

Ingram’s confession that he forced his son and daughter to have sex together was a fiction implanted through a suggestion by Richard Ofshe as an aspect of an experiment to demonstrate Ingram’s suggestibility. Yet Ingram produced the confession in written form when alone in his cell, when he was not subject to the artificial influences presumably responsible for his supposed “trances.” Are we therefore obliged to assume that Ingram went into states of self-hypnosis whenever “trances” were needed to account for his confessions?

False confessions produced under pressured police interrogations are rarely attributed to “hypnotic,” “dissociative,” or “trance” states in suspects. It is therefore puzzling why the Ingram case should rely on them any more than do other confessional scenarios. In this connection, prior to questioning Ingram was of a religious persuasion that not only acknowledged the existence of the living Devil, but his capacity to deceive humankind in devious ways. He also indicated he believed implicitly in the honesty of the two daughters who accused him of crimes against them, and was probably aware of tales of Satanic cult activity and alleged victims who recovered repressed memories of horrific abuse at its hands. With these loaded preconceptions, together with an interrogation process by police, pastors, and psychologists forming an insular information-processing system, one need not resort to altered states of consciousness to explain why a subject asked to imagine what he might have done to his children, starts confusing his imaginings with actual events.

In the Paul Ingram case, the confusion between true memory and confabulation may not have been enabled by the failure to discriminate differences between the two. Tellingly, when Ingram’s wife, Sandy, began producing similar “Satanic” narratives, she indicated she could differentiate the sensorily charged imaginings of these narratives from the ordinary memories of her past. She told Ofshe she could “see” and “smell” things in the narratives, unlike the mundane memories of her life (Wright, 1994, pp. 141-142). Wright’s account of the Ingram affair in Remembering Satan possesses a running reality eerily out of keeping for a reporter supplying information at a considerable distance from the events in question.

At any rate, the assumption of an altered state of consciousness in Ingram was inferential, not demonstrated. This remains so despite the inclusion of “relaxation” techniques and “guided imagery” brought to bear on eliciting information during his interrogation. The latter technique was found in one study to be among the armamentaria of procedures in fully one quarter of psychotherapists polled (Poole, et. al., 1995). The remaining complication for altered state theories is the necessity of distinguishing underlying differences among “hypnosis,” “trance,” and “dissociation.” Some investigators have even held that there is little rigorous evidence such “altered states” embody principles any different from ordinary states of expectation, task demand, or relaxation (Spanos, 1987). 

Cults

There are several other facets of benandanti confessions deserving of commentary. One is the assumption that benandanti formed an agrarian cult. The term “cult” is not precisely defined by Ginzburg, although its use would seem to imply organized sect membership in which votaries have connections to one another and, possibly, a hierarchy. There is a world of difference between tales that undergo thematic deformations from benign fertility mythology into stereotyped inquisitorial scenarios of witchcraft, and the existence of actual cultic entourages involving ceremonial activities. Folk beliefs, however novel from the standpoint of inquisitors forever on the lookout for material to redact as heretical practice, were prevalent across the European landscape. It is to Ginzburg’s credit that he has traced their character and unusual affinities with ingenuity. Nonetheless, his evidence of an agrarian “cult,” however defined, is not always compelling. One wonders whether Ginzburg’s use of the term implies much more than he is willing to grant. In a reply to one critic, he demurs from having stated that benandanti “actually assembled in ecstatic nocturnal gatherings,” and that there was no evidence for actual rituals (Ginzburg, 1991). Much of his data depends upon hearsay accounts of people who, with several exceptions in The Night Battles, are marginally identified from 1570 to 1650.

Testimonies, often reportedly extracted in inquisitorial contexts brimming with auras of pressured questioning, mark a shifting and inconsistent picture of benandanti belief, ritual, and practice. One problem is whether the testimony represents a body of historical data from which a unique sect of believers can be inferred. Cases are trotted out that are a patchwork of similarities and differences. For example, in 1692, a Livonian by the name of Theiss made a confession in court that bears striking similarities to benandanti tales of night battles with witches over crop fertility. Except Theiss fancied himself a werewolf (wahrwolff) who made journeys to hell to take on the Devil as well as  witches in battle. Yet Ginzburg does not classify Theiss as a benandante, unlike Maria Panzona of Latisana, whom he does—despite the fact that her account included homage to a woman called “the abbess,” who sat on the edge of a well in the field of Josephat.  Maria’s talents did not stop there; she had the capacity to turn herself into a black cat, and when younger, accompany her godfather in the form of a butterfly to precincts of heaven and hell.  Nor does Ginzburg deny cult membership to a cowherd from the same region as Maria, Menichino, who, unlike other benandanti, ventured out at night in the form of smoke to battle the witches. If smoking benandanti and butterflies are entitled to class membership, why aren’t hairy, growling creatures with similar portfolios granted the same inclusion?

One suspect assured inquisitors that the banner of witches is of red silk, with four gilded black devils; another insisted the banner was yellow. Benandanti were committed to defeating witches in battles at night during the Ember season, but death tolls were never mentioned—possibly because attacks by fennel stalks or sorghum stalks wielded by witches led only to minor contusions. Benandanti and witches, declared enemies, seemed unusually chummy on return visits from battle, described by suspects as a combination of play and war. Benandanti expressed fear of divulging names of witches for fear of being beaten black and blue, an apprehension that curiously evaporated on the battlefield. Carousing between benandanti and witches seemed to be the order of the day on returns from combat. Witches were reportedly more parched than their adversaries, and if pails of water were not left at doorsteps to slake their thirst, they invaded cellars to spill wine—or, as one might expect from a perverse lot, urinate in the casks.

Of some 850 accused between 1580 and 1634, only two benandanti, Paolo Gasparutto and Battista Moduco, come to life in Ginzburg’s The Night Battles, and we are able to glimpse stretches of their testimony, including verbatim transcripts appended at the end of the author’s tract. The transcripts bear all the signs of the pressured tactics familiar in contemporary research on police interrogation and confessions.

Diagnosing Confession Types

There is a complacent assumption on the part of certain historians that they are able to spot the difference between the three types of confession through a cursory—and inevitably distant—scrutiny of the context in which they were produced. But the contemporary scientific challenge, as many psychologists have come to realize, is to question this discernment. The problem in distinguishing (V) from (C) confessions is making sure that any factor, however subtle or unnoticed, does not contaminate voluntary testimony, pushing it into another category. The problem in distinguishing (C) from (I) testimony is the knotty problem of establishing actual self-belief, so that misclassifications are avoided. For example, Cohn (1975) quotes Crawford (1967) as insisting that:

Any suggestion that the alleged witches were forced to confess is, I think, belied by the very nature of the confessions themselves. Forced confessions are generally grudgingly made and retracted as soon as the threat is removed. The confessions in [this case] were not of this nature and were first made in front of a definitely skeptical European policeman and then repeated or admitted in front of a European magistrate and later a European judge. As far as one can tell there is no reason to suppose that the police brought any pressure to bear on these people (Cohn, 1975, p. 222).

Crawford was not privy to the confessions, so he was in no position to rule on variables influencing the testimony—or the actual beliefs—of the suspects in question. In all likelihood, he identifies the idea of “force” with thumbscrews or strappados (or the threat to use them), and the alleged “skepticism” of policemen with, possibly, their official reassurance they always question witnesses with an “open mind.” Contemporary research has shown that a policeman’s approach to questioning and even the characteristics of the room in which a suspect is interrogated, not to mention many other overlooked factors, like the promise of a light sentence, ethnic status, intellectual limitations, and sympathetic questioning can produce false confessions. Benandanti Paolo Gasperutto and Battista Moduco were given suspended sentences after confessing to “witchcraft” on the promise of “being received and treated with mercy” (Ginzburg, 1992, p. 153).

The concept of “force” underwent intriguing deployment during the early modern period. It was often identified with extreme forms of torture, as in the German persecutions (Midelfort, 1990), whereas in England “torture” was forbidden, although witch suspects were routinely subject to interrogations involving prickings, sleep-deprivation, blindfolding, and binding suspects cross-legged on stools (Hole, 1947; Briggs, 1966). The sometime English slant on witch confessions that they weren’t “forced” because “torture” was repudiated is transparently fallacious in the light of modern research findings.

The Problem of Belief

Several historians feel that the reliance on the confession of guilt stemmed from the fact that witchcraft was an invisible crime, so that a correspondingly greater emphasis was placed on the verbal testimony of the witch as proof of her malfeasance. This viewpoint may have been overstressed. In important ways, witchcraft wasn’t usually a hidden crime. With the exception of such diabolical intimacies as the pact with the Devil, its publicity was highlighted by victims who testified to their bewitchment, possession, or spectral visitations by the accused witch. Second, across centuries there has been an underlying social assumption that the royal road to determining inner belief is its verbal expression. Accordingly, the fancied connection between spoken words and guilt is taken to be so close, it overrides all other considerations. On the basis of mock jury experimental research, investigators report that people commonly find it difficult to believe anyone would confess to a crime he or she did not commit (Kassin & Wrightsman, 1980; 1981).

On the modern scene, the emphasis laid on the probative value of the verbal confession during police interviews prevails over other types of evidence—even when crimes are not so invisible. Where physical evidence is available, it is often not gathered and subject to forensic analysis. Horvath and Meesig (1996) reviewed studies of criminal investigations and found that gathering physical evidence was achieved in less than 10% of reviewed cases. Even at that, the evidence thus gathered was subjected to analysis in only about half those cases.

The vagaries of determining actual belief are not limited to suspected witches from whom confessions were extracted. Questions also arise about beliefs harbored by inquisitors, especially during the task of interrogation. On the one hand, witch theorists of yesteryear attributed to witches far-reaching magical powers to destroy, wrack with pain, blind, bewitch, make impotent, cast glamors (i.e., make penises disappear), sicken, and otherwise possess innocent victims. On the other hand, inquisitors went about their business of interrogation as if they were virtually immune to these fearsome blights. Yet there is ample evidence the witch-theorist was not completely insensitive to the pitfalls of approaching suspects. For example, the principal witch-treatise of the fifteenth century, the Malleus Maleficarum (Part III, Question 15), cautions inquisitors to avoid the gaze of suspects lest their power to sway the judge from his holy task prevail. It also recommends witches be led into interrogation backwards for the “protection” of inquisitors. There are other minor precautions recommended in the manual, such as avoiding physical contact with the accused, and carrying consecrated salt and herbs enclosed in a wax seal stamped with the Agnus Dei. It remains a mystery why the cautionary approach to interrogation stops there, if the witch is by reputation invested with so many dangerous powers. Were these ignored by inquisitors approaching alleged mine-fields of diabolical perversity, or was actual inquisitorial belief about the witch’s power to afflict them during interrogation half-hearted, contradictory, or non-existent? It is one thing to be unwisely complacent about being struck blind, and quite another to have little apprehension this is possible. Stephens (2002) has conjectured that a crisis of belief in the actual reality of witchcraft on the part of witch-theorists as notable as the authors of the Malleus Maleficarum was the driving motivation behind their preoccupation with proving the reality of carnal relations between demons and witches 

Inquisitorial Style

At the trial of the wife of Jean Rose of Villars-Chabod in Savoy in 1477, the suspect was interrogated on September 9, and accused of attending sabbats, an allegation she denied. She was thereupon given a monition followed by two others before being returned for questioning on September 14. She was told she must admit her crimes on that date, or the inquisitorial process would move forward. An audience was held on September 15, when she was confronted with the names of five other suspects whom she knew. She asserted they were of good reputation, and she had no antagonism towards them. She was then presented with the name of Massetus Garini, reportedly a suspect who had been executed for heresy, and who named her as an accomplice. She stated she knew him, and there was no enmity between them. Put to another question, she denied she ever cured children and animals by magical charms. The fiscal (i.e., inquisitorial attorney) present asked the tribunal to condemn the witness to torture, after which she was taken to a chamber and hung by strappado for half an hour. On October 20, the witness was again brought before the tribunal and pressured to confess. She once more denied the charges. The fiscal in attendance again demanded reinstatement of torture, which was granted. After a short duration of this treatment, the suspect exclaimed she would confess if the torture was stopped. When hauled before the same tribunal, however, she once more denies the veracity of the allegations. Her case was continued to the next day. When brought before her accusers, she confessed to attending a witch’s sabbat where a mixed company of heretics were dancing backwards. Other infamies admitted were paying homage to the Devil, renouncing the faith, receiving a pot of money from the demon as well as a pole 18 inches long and some ointment which, when rubbed on the pole, could transport her to witches’ conventicles. Her confession was reconfirmed on October 22 along with supplying the names of 13 others, and in requestioning on October 23 she named yet four more accomplices (Lea, 1957, pp. 238-239).

The foregoing inquisitorial process was characteristic of many confessional scenarios. It also captures the inquisitorial presumption of guilt, in sundry trials one overriding any considerations of innocence standing in the way of demonstrating a preordained truth. The inquisitorial model was presumed to cut through the verbal camouflage thrown up by the devious ways of the Devil. And here we have one take on the connection between confession and guilt or self-belief not only driving the conviction of many early modern inquisitors, but contemporary police interrogations as well. It is a theory that single-mindedly sweeps away as trivia all facts deflecting attention away from the crowning significance of what a suspect is eventually brought to say about a crime—irrespective of context. The verbal confession as a criterion of guilt or inner belief is the smoking gun, however extracted (Kassin, 2001).

There were, of course, broad exceptions to the picture of free-wheeling intentions to convict a witch at any cost. Witch-manias saw as many variations on approaches to demonstrating guilt as does our own era. The historical distortion that most, if not all, accused witches were eventually executed does violence to the actual record, both in America and abroad. Even during the most ferocious European witch panics, the percentage of convictions out of the total number of accusations and subsequent trials for witchcraft averaged only around fifty per cent (Levack, 1987; Katz, 1992). In Salem Village, 162 persons were accused of witchcraft. Of these only seventy six were tried, but only 30 convicted (Levack, 1987, p. 162). Godbeer has shown that the juridical approach to investigating witchcraft in Stamford, Connecticut, during the same time-frame as the Salem Village, Massachusetts trials in 1692, was much more skeptical and experimental in nature than the latter (Godbeer, 2005).

Deliberations of many juries across the Continent and in America considering cases against alleged witchcraft involved sifting carefully through the evidence for the crimen exceptum. This was not always the uncomplicated affair we might imagine it to have been. The approach was often judicially painstaking and systematic, minute attention being given to distinguishing natural causes from demonological influence. Jurists often wracked their brains over determining which of the two was playing the essential role. Hence, there grew to be an obsessive reliance on “signs” that would clinch the issue and remove uncertainty. These included the Devil’s mark, the supernumerary nipple, the rejection by water, the testimony of neighbors, an accused family member or members (Midelfort, 1990), and most decisively, the confession—however extracted. Considering the amazing array of approaches to witchcraft, it appears to be a historical distortion to collapse the era in question uniformly into a mindless juggernaut of terror and unthinking superstition. It is also ironic that while many contemporaries profess horror at the barbarity of inquisitorial procedure, they appear to share with their early modern counterparts an enthusiasm for extracting confessions of guilt—at whatever cost to innocence or justice.

False Memories

The phrase “false memory” is ubiquitous in modern commentary on questionable testimony, therapeutic disclosures, and confessions. One wonders whether it has overextended its reach due to a conceptual sprawl that often obscures necessary distinctions among psychological processes or mechanisms. It would appear that in several arenas of discussion, distortions that creep into accounts of past events may have nothing to do with vagaries of memory, but alterations of belief.  Examples might suffice to illustrate the confusion.

Possible misattributions of “false memory” in court cases was apparent in the case of John Demjanjuk, a Ukrainian born retired automotive employee who had been working in Cleveland, Ohio. He was deported to Israel, tried and sentenced to death for the destruction of hundreds of thousands of Jews at the Treblinka death camp in Germany. Demjanjuk was identified by individual survivors of the camp as “Ivan the Terrible,” a brutally sadistic Nazi war criminal. Demjanjuk’s attorney, Mark O’Connor, felt that there were several features of his client’s case that might strengthen defense arguments, including the fact that “false memories” might be playing a role in the allegations since survivor memories of Ivan were thirty five years old. Reflection will show that the case did not revolve around any issue of “false memory,” much less around an issue that survivor memory was even being questioned or presumed to be faulty.

Five out of twenty-eight survivors of Treblinka identified Demjanjuk as Ivan. However, such “identifications” would depend upon comparisons drawn between the person survivors knew thirty-five years ago (Ivan), and the look of the individual they saw before them in court at a later date (Demjanjuk). Accordingly, the identification of Demjanjuk as Ivan even if erroneous does not necessarily involve mistaken memories of what Ivan looked like thirty five years prior to survivor testimony. Where, then, does the risk of false memories surface?

The problem in the Demjanjuk case is not the possibility that what the memory of what Ivan looked like thirty five years before is faulty, but only the belief that Demjanjuk thirty five years later looks sufficiently like what Ivan should look like in the contemporary time-frame to validate the positive identification. The error risk, then, is not necessarily a memorial one; it is the failure to appropriately connect two imaginings, or pictures: one of Demjanjuk in the here and now, and the other of a future correct projection of how Ivan should look thirty-five years after Treblinka. Consequently, the risk is a false positive identification based upon a cognitive mistake in comparing a visual image-projection from the past with the look of a present defendant. O’Connor’s case thus hinged on the integrity of a visual time-projection, not on the possibility of “false memories” contaminating testimony. That is, the time projection, a cognitive operation, could have been faulty while the older memory was faithful. The exception here would be the inability of the five accusers to identify Ivan as he looked 35 years prior to his trial by whatever faultless procedure such identification is achieved. The foregoing discussion is not meant to diminish the importance of myriad factors that have been shown to contaminate eyewitness identifications (Brainerd & Reyna, 2005, p. 268)

            Parties to current debates over so-called “recovered memories” may sometimes fail to distinguish between (1) claiming a (defendant, witch, suspect, patient) falsely remembered an event and (2) falsely believed an event occurred. In the former case, we recover something justifiably classified as a “memory,” whereas in the latter case, we merely suppose an event occurred through cognitive recasting of material already available. Did the patient actually recover a “false memory” when, at the prodding of her psychotherapist, she comes to reinterpret her current distaste for bananas to past occasions of forced oral sex? Or is the manipulation in question only one of belief, not memory distortion? The question arises in contexts in which an individual may aver: “I don’t remember the episode, but it must have occurred all the same.” Are the actual, but inchoate memories of dark figures around a crib a “false memory” of a Satanic ritual when this interpretation is placed upon them, or only a questionable reinterpretation of what a real memory actually signifies?

To complicate matters, a distinction can be drawn between a patient or suspect whose cognitive mistake consists in misnaming a mental event (e.g., in mislabeling as a “memory” what boils down to merely a fresh interpretation or slant on a past event) and misidentifying an interpretation as a true memory (e.g., coming to mistakenly believe that one’s interpretation of memory fragments was actually a freshly recovered memory). The metamorphosis of benandanti tales into witch narratives was not a process by which the former were mutated content-wise, only reclassified. Even as some scholars insist that the confessions were of the (I) type, they were not this because of being materially transformed in content, but because of reassignment to another category of concern: heresy. Here, two kinds of errors are possible: holding a (C) confession to be an (I) type, and claiming the conversion in question involved thematic variation. 

 Despite these complications, once having distinguished memory and belief and their combinatory forms, one is still obliged to establish an evidential basis for applying a concept to a particular case. Distinctions are one thing, but clinching the case that what therapists occasionally do is to manipulate beliefs in contrast to facilitating memory retrieval is quite another. In fact, we are not entitled to maintain any position on the matter before studying individual cases. Current waves of sloganeering over such concepts as repression—especially in advance of the outcome of laboratory research and forensic studies—do a disservice to fair minded attempts to investigate individual cases.

Return of the Repressed

While many commentators single out Freud’s concept of repression as the chief mischief-maker in the false memory controversy (Stannard, 1980; Ofshe & Watters, 1994; Crews, 1995; 1998), it is also possible to view it, troublesome or not, as a secondary inferential move. The primary error may have been the assumption of a connection between assumed historical events and current clinical patterns held to be their representative effect (Tversky & Kahneman, 1982). In the absence of any memories of such events in patients who exhibit certain patterns, the concept of repression rides piggy-back on the connection. The notion of repression evaporates when there is no warrant for positing the causal connection in question. For example, if it is not assumed that a patient with bulimia or anorexia nervosa was sexually abused as a child, there is no need to attribute the lack of any memory of such abuse to an unconscious process responsible for the amnesia. Indeed, in an era in which sexual abuse has been documented to be tragically frequent, the chance association between it and any number of clinical syndromes may come to be perceived as causal, when it is only adventitious.

Courts in recent years have adopted the stance that testimony based upon a theory of repressed memories of child sexual abuse and their recovery is forensically, because theoretically, problematic. Jurisprudence on the inadmissibility of testimony involving the recovery of repressed memories of sexual abuse was instanced in Justice Groff’s ruling in State v. Hungerford and Morahan, State of New Hampshire Nos. 94-S-45 through 47 and 93-S-1734 through 1936, September 14, 1994. Needless to say, some researchers have held the skepticism surfacing in legal precedents does not go far enough; it is the very notion of repression itself that still awaits validation through scientific scrutiny (Holmes, 1972; 1974; Holmes & Schallow, 1969). [iii] 

Vagaries of statistical association may result in imaginary connections between childhood abuse and later clinical patterns held to be its markers. The lesson to be drawn is not merely an academic one; during the late eighties and early nineties in the United States a cottage industry sprang up based upon the idea that current symptomatology is a hieroglyph of early patterns of sexual abuse. Such manuals as the Incest Survivors Aftereffects Checklist (Blume, 1991) were disseminated in order to alert the otherwise troubled patient to symptomatic signs of early sexual abuse, from “absent memories of early years” to “noise aversion.”

Actually, the Ingram allegations of sexual abuse contain Victorian, as well as  early modern thematic strains. It is commonly supposed that intrafamilial sexual abuse, especially father-daughter incest, was originally uncovered by Freud, but subsequently rejected by him in favor of a theory of infantile fantasy. However, paradoxes abound in Freud’s reportorial style. For example, despite Freud’s references to “reports” of early seduction by his eighteen hysterical and obsessional patients, the actual character of those disclosures remains unclear to this day. Freud wavered between claiming he had actually obtained reports of infantile abuse from patients and claiming they resisted the interpretations he attempted to force upon them (Schimek, 1987; Macmillan, 1991).

Doubts about the nature of the Freudian data-base of his thirteen (and subsequently eighteen) patient caseload aside, there are other troublesome inconsistencies in the canon (Thornton, 1983; Vetter, 1988; Israels & Schatzman, 1993). For example, after his momentous reversal on the reality of seductions in 1897, Freud subsequently reported that one of the reasons for his newer stance on fantasy was that he hadn’t realized in the 1896 papers that seduction was only a necessary, not sufficient condition for hysteria. In other words, Freud later denied what he had presumably known earlier: that while all hysterics were sexually abused as infants, not all persons so abused developed hysteria later. The denial is belied by Freud’s comments in The Aetiology of Hysteria, where he explicitly took pains to emphasize his conviction that early trauma is sufficient to produce hysteria.

In his later Autobiographical Study (Freud, 1925, pp. 33-35), Freud finds retrospective support for the so-called “father etiology” in the 1895 cases he treated with Breuer (Breuer & Freud, 1895) as well as the 1896 caseload. Paternal incest was therefore only later declared to be the pathogen in the development of hysteria. However, there is little basis in his earlier identification of perpetrators to justify the retrospective attribution. Accordingly, while commentators like Masson (1984) have raised concerns about Freud’s “betrayal” of the victims of paternal incest in abandoning the Seduction Theory, no such pattern is referenced in the original three seduction papers. In the thirteen patient caseload of the first of those papers, over half the perpetrators are brothers. In the last of the papers, most perpetrators are hired servants. In addition, one stranger and two “close relatives” are cited (Esterson, 1993). While some critics have insisted that Freud disguised the identity of fathers to protect Victorian sensibilities (Gay, 1988, 94-95), the identification of seven brothers as sexual predators suggests that maintaining veneers of social respectability was not uppermost in Freud’s mind. The so-called “father etiology” hinted at by Freud only surfaces within a year after the delivery of the last seduction paper, as well as in subsequent letters to Wilhelm Fliess. 4  

Witchcraft as a Transformational Exercise

Forms of witchcraft in the early modern period dropping out of accounts of contemporary Satanic cult activity, or SRA, are those that would nullify plausibility had they been incorporated. Accordingly, recent cult narratives have been conveniently updated. For citizens of the older time-frame, assignment to novice witches of magical toads who served as their familiars was not an unbelievable notion. But nothing of the kind survives in contemporary accounts of Satanic cult activity.

Broadly, differences between early modern and contemporary cult themes are those associated with the limits of plausibility fixed by separate historical eras. Yet criteria of plausibility may differ even within a given age. Examples are cases of belief systems within contrasting world-views in contemporaneous cultures. For example, when the existence of a demonic entity is part and parcel of religious belief, Satan himself may put in an appearance at covens. This was the case in the dramatic accounts of Michele Smith (Smith & Pazder, 1980) and Lauren Stratford (Stratford, 1988). For those outside the religious subculture, however, testimony involving fiery embodiments of the Devil, his occult and sophomoric poetry, or tell-tale stigmata of his marks on victims bodies (another early modern theme) are greeted with skepticism—and amusement. For nonbelievers, such reports are canards, or else regarded as transparent attempts at back-door proselytizing for versions of Christianity in little danger of being demythologized (Bultmann, 1957).

For those who take reports of contemporary cult involvement seriously, testimony of cult victims occasions opportunities for salvaging perceived levels of reality. This is accomplished by transforming the implausible layers of older thematic material into updated narratives. Recasting the manifest layer into one accommodating the belief system of the investigator is common in present day exegeses. Thus, witches who transvected during the early modern period now “fly through the air” at day care nurseries like those in the McMartin case (Wright, 1994, p. 74). Or else sectaries improvise “harnesses” or “trapezes” for the purposes of casting doubt on the testimony of children. Cattle mutilations without a visible spoor around the carcass of the dead animal are contrived not through aerial transvection, but through “helicopter drops” (Kanaher, 1988) chartered by the witches (whose ingenuity in leaving us perplexed over the mechanics of the kill is rivaled only by the matchless stupidity of the logistics involved).

Reconstructions of the “real” trauma experienced by patients with Dissociative Identity Disorder a.k.a. Multiple Personality Disorder (Putnam, 1989) whose alters regale us with tales not handily assimilated to the world-view of the therapist are likewise attempted. For example, Michele Smith’s revelations, explicitly because they strain credulity for so many not easily convinced of a living Devil, become for them intra-familial sexual abuse transmuted from the archetypal layers of her account. On this formulation, the creative authorship responsible for the metamorphosis of mundane abuse into mythological content is the dissociative process.

Transformational slants are hardly new. They have striking counterparts in traditions of witchcraft scholarship of which Pedro de Valencia’s de-Satanization theory is but one example (Henningsen, 1980). Pedro was a humanist and classical scholar who was invited by the Inquisitor General of Spain, Bernardo de Sandoval y Rojas, to contribute his views on the alleged outbreak of witchcraft in the Spanish Basque during the early seventeenth century. One of his three hypotheses was that the lurid tales of witch cult activity amounted to nothing more than a gang of sociopaths indulging their appetites for the outré, merely. The hoary motif of the pre-Christian, non-satanic fertility cult pace Murray (1921, 1934) and Gardner (1954) is another example of the tradition. It survives as the benign “neo-pagan” or Wiccan witchcraft belief system, an organized religious movement based upon a suppositious lineage (Cohn, 1975). So mythologies have a way of accommodating to the convictions of mythologists, not the other way around. Exuviating layers of implausibility in order to reconstruct plausible historical scenarios is the warp and woof of some witchcraft scholarship, especially that of the anthropological school (Frazer, 1922).

In all of this, there is no unwavering standard of plausibility from era to era. Believability of narrative depends ultimately upon the vagaries of broader believe-systems, and what they can absorb. What was believable for the early modern citizen is the stuff of fairy-tales today. And what is a real, malevolent being for contemporary fundamentalists is a product of whimsy or superstition for the nonbeliever. Issues over the reality of witchcraft in yesteryear may as they do today transcend issues arising from the inherent ambiguities of experience. They also derive from vastly different ways of looking at the world, which might be why Satanism never goes away—nor, like comets, stay the same on return visits.

*Dr. Begelman is a clinical psychologist and neuropsychologist in private practice in Connecticut. He is the past theater critic for the Danbury News-Times and drama and film critic for The Citizen News of New Fairfield.

REFERENCES 

Barthel, J. (1976). A death in Canaan. New York: Dutton.

Bourguignon, E. (1972). Dreams and altered states of consciousness in anthropological research. In F. L. K. Hsu (Ed.) Psychological Anthropology, Cambridge, Massachusetts: Schenkman Publishing Company, pp. 403-434.

Brainerd, C. J. and Reyna, V. F. (2005). The science of false memory. Oxford: Oxford University Press.

Breuer, J. and Freud, S. (1895). Studies on hysteria. In J. Strachey (Ed.), The standard edition of the complete psychological works of Sigmund Freud, Vol. 2 (pp. 19-305). London: Hogarth Press and the Institute of Psycho-Analysis.

Bultmann, R. (1957). The presence of eternity. New York: Harper & Row.

Blume, E. S. (1991). Secret survivors. New York: Ballantine.

Cioffi, F. (1974). Was Freud a liar? The Listener, 91, 172-174.

Cohn, N. (1975). Europe’s inner demons. New York: Basic Books.

Crawford, J. R. (1967). Witchcraft and sorcery in Rhodesia. London.

Crews, F. (1986). Skeptical engagements. New York: Oxford University Press.

Crews, F. (1995). The memory wars: Freud’s legacy in dispute. New York: A New York Review Book.

Crews, F. (1998). Unauthorized Freud: Doubters confront a legend. New York:Viking.

Ellroy, J. The black dahlia. New York: Mysterious Press.

Erwin, E. (1980). Psychoanalytic therapy. The Eysenck argument. American Psychologist, 35, 435-443.

Esterson, A. (1993). Seductive mirage: An exploration of the work of Sigmund Freud. Chicago: Open Court.

Frazer, J. G. (1922). The golden bough. London: Macmillan.

Freud, S. (1925). In James Strachey (Ed.), An autobiographical study. The standard edition of the complete psychological works of Sigmund Freud. Vol. 20. (pp. 7-74). London: Hogarth Press and the Institute of Psycho-Analysis.

Gardner, G. B. (1952). Witchcraft today. London: Rider & Company.

Gay, P. (1988). Freud: A life for our time. New York: Doubleday.

Gilmore, J. Severed: The true story of the Black Dahlia murder. Los Angelis: Zanja Press

Ginzburg, C. (1991). Ecstasies: Deciphering the witches’ Sabbath. New York: Pantheon Books.

Ginzburg, C. (1991). The witches’ Sabbath. The New York Review of Books. Volume 23,13,45.                                                                                                                                                                                 

 

Ginzburg, C. (1992). The night battles: Witchcraft and agrarian cults in the sixteenth and seventeenth centuries. Baltimore: The Johns Hopkins University Press.

Godbeer, R. (2005). Escaping Salem: The other witch hunt of 1692. Oxford: Oxford University Press.

Grunbaum, A. (1984). The foundations of psychoanalysis: A philosophical critique.  California: University of California Press.

Gudjonsson, G. H. (1992). The psychology of interrogations. London: Wiley.

Hansen, C. (1969). Witchcraft at Salem. New York: George Braziller.

Hole, C. (1947). Witchcraft in England. London: B. T. Batsford, Ltd.

Holmes, D. S. and Schallow, J.R. (1969). Reduced recall after ego threat: Repression or response competition? Journal of Personality and Social Psychology, 22, 145-152.

Holmes, D. S. (1972). Repression or interference: A further investigation. Journal of Personality and Social Psychology, 22, 163-170.

Holmes, D. S. (1974). Investigations of repression: Differential recall of material experimentally or naturally associated with ego threat. Psychological Bulletin, 81, No. 10, 632-653.

Horvath, F. and Messig, R. (1996). A content analysis of textbooks on criminal investigation: An evaluative comparison to empirical research findings on physical evidence of a crime. Journal of Forensic Sciences, 41, 963-969.

Huff, C., Rattner, A. & Sagarin, E. (1996). Convicted but innocent. Thousand Oaks, California: Sage Publications.

Kanaher, L. (1988). Cults that kill. New York: Warner Books.

Kassin, S. M. (1997). The psychology of confession evidence. American Psychologist, 52, 221-223.

Kassin, S. M. (2001). Confessions: Psychological and forensic aspects. In Smelser and Baltes (Eds.) International Encyclopedia of the Social and Behavioral Sciences, pp. 2516-2521.

Kassin, S. M. and Wrightsman, L. S. (1980). Prior confessions and mock juror verdicts. Journal of Applied Social Psychology, 10, 133-146.

Kassin, S. M. and Wrightsman, L. S. (1981). Coerced confessions, judicial instructions, and mock juror verdicts. Journal of Applied Social Psychology, 11, 489-506.

Kassin, S. M. and Wrightsman, L. S. (1985). Confession evidence. In S. M. Kassin & L. S. Wrightsman (Eds.), The psychology of evidence and trial procedure. Los Angeles, CA: Sage Publications.

Kassin, S. M., Meissner, C. A. and Norwick, R. J. (2005). “I’d know a false confession if I saw one”: A comparative study of college students and police investigators. Law and Human Behavior, 29, No. 2, 211-227.

Katz, S. (1992). Convicting witches. New York Review of Books, 34, 65.

Lea, H. C. (1957). Materials toward a history of witchcraft,  Vol 1. New York: Thomas Yoseloff.

Macmillan, M. (1997). Freud evaluated: The completed arc. Cambridge, Massachusetts: The MIT Press.

Martin, R. (1975). Legal challenges to behavior modification: Trends in schools, corrections and mental health. Champaign, Illinois: Research Press.

 

Marwick (Ed.), Witchcraft and sorcery (pp. 174-189). London: Penguin Books.

Masson, J. (1984). The assault on truth: Freud’s suppression of the seduction theory. New York: Farrar, Straus, and Giroux.

Masson, J. (1995) the complete letter of Sigmund Freud to Wilhelm Fliess: 1887-1904. Cambridge, Massachusetts: Harvard University Press.

Midelfort, H. C. E. (1974) Were there really witches? In R. M. Kingdon (Ed.) Transition and revolution: Problems and issues of European Renaissance and Reformation history. Minneapolis, Minnesota: Burgess Publishing, Pp. 189-205.

Midelfort, H. C. (1990). The social position of the witch in southwestern Germany.  

Mora, G., Kohl, B., Shea, J., Midelfort, E., & Bacon, H. (Eds.) (1991). Witches, devils, and doctors in the Renaissance: Johann Weyer, De praestigiis demonum. Binghamton, New York: Medieval & Renaissance texts and Studies.

Nagel, E. (1959). Methodological issues in psychoanalytic theory. In S. Hook (Ed.), Psychoanalysis, scientific method, and philosophy. New York: New York University Press. Pp. 38-56.

Note (1953). False confessions: A neglected area of criminal investigation. Indiana Law Review, 28, 374-392.

Ofshe, R. (1992). Inadvertent hypnosis during interrogation: False confession due to

dissociative state, misidentified multiple personality and the satanic cult hypothesis.

International Journal of Clinical and Experimental Hypnosis, Vol XL, No. 3, 125-156.

Ofshe, R. (1994). Making grossly damaging, but avoidable errors: The pitfalls of the Olio/Cornell thesis. Journal of Child Sexual Abuse, 3, (3), 95-108.

Ofshe, R. J. and Leo, R. A. The psychology of police interrogation: The theory and classification of true and false confessions. Studies in Law, Politics and Society, 16, 189-251.

Poole, D. A.,  Lindsay, D. S., memon, A. & Bull, R. (1995). Psychotherapy and the recovery of memories of childhood sexual abuse: U. S. and British practitioners’ opinions, practices and experiences. Journal of Consulting and Clinical Psychology, 63, 426-437.

Putnam, F. W. (1989). Diagnosis and treatment of multiple personality disorder.  New York: Guilford.

Radelet, M. L., Bedau, H. A., and Putnam, C. E. (1992). In spite of innocence: Erroneous convictions in capital cases. Boston: Northeastern University Press.

Rattner, A. (1983). When justice goes wrong: Convicting the innocent. Unpublished doctoral dissertation, Ohio State University.

Robbins, R. H. (1959). The encyclopedia of witchcraft and demonology. New York: Crown Publishers.

Russell, J. B. (1972). Witchcraft in the middle ages. Ithaca, New York: Cornell University Press.

Ryan, N. (1989). Affirmation in response to vacate judgment of conviction, indictment no. 4762/89. New York: Manhattan District Attorney’s Office.

Sauer, M. (2002). Confessions in the third degree: Court TV film revisits police grilling. San Diego tribune, December 13.

Skinner, B. F. and Ferster, C. B. (1957). Schedules of reinforcement. New York: Appleton Century Crofts

Spanos, N. (1987). Hypnosis research: Paradigms in conflict. Behavioral and Brain Sciences, 10, 525-530.

Stannard, D. E. (1980). Shrinking history: On Freud and the failure of psychohistory. Oxford: Oxford University Press.

Stephens, W. (2002). Demon lovers: Witchcraft, sex, and the crisis of belief. Chicago: The University of Chicago Press.

Tversky, A. and Kahneman, D. (Eds.) (1982). Judgments under uncertainty: Heuristics and biases. London: Cambridge University Press.

Upham, C. W. (1867). Salem witchcraft, 2 volumes. Boston: Wiggins and Lunt.

Waismann, F. (1963). Verifiability. In  A. G.  N.  Flew (Ed.), Logic and language (First Series) pp. 117-144. Oxford: Basil Blackwell.

Wilson, M. (1963). Good company: A study of Nyakyusa age-villages. Boston: Beacon Press.

Wright, L. (1994). Remembering Satan: A case of recovered memory and the shattering of an American family. New York: Alfred A. Knopf.

Wrightsman, L. S. and Kassin, S. M. (1993) Confessions in the courtroom. Newbury Park, California: Sage Publication

 

 

---

[i] The belief in causal connections between a ritual and external happenstance is the foundation of all magical practices. The latter may be maintained by a partial reinforcement effect (Skinner & Ferster, 1957), accounting for the survival of such practices in the sometime confirmation of an apparent connection. Amulets, talismans, cauls, astrological signs, incantations, prayers, magnets, relics, and crystals thus continue to be relied on when their occasional seeming success in bringing about a desired effect overshadows the greater frequency of their failures. The latter, of course, are largely ignored in the flush of statistically random hits.

 

[ii] The Night Battles, originally published as I Benandanti: Stegonerie e Culti A e Seicento in 1966, was translated into English in 1982 by Anne and John Tedeschi.

 

[iii] In recent years, there has been an outpouring of broad scale attacks on the entire psychoanalytic theoretical edifice. (Nagel, 1959; Cioffi, 1974; Erwin, 1980; Grünbaum, 1984; Esterson, 1993; Crews, 1986; 1995; 1998; Macmillan, 1997). Curiously, these have been authored by humanities scholars some of whom have been doubtlessly soured on their former partialities to this kind of theorizing. Experimental psychologists, on the other hand, have been less vocal about the enthusiasm, having largely renounced psychoanalytic theory from its inception. The disappointed are quicker to remonstrate than the wearied.

 

4  Mention of paternal incest occurs in the 1896 letter of 12/6, and the 1897 letters of 1/3, 1/12, 1/24, 2/8, 4/28, 5/2, 5/31, 6/22, and 12/12 (Masson, 1985). Its introduction—as well as its later repudiation--in all likelihood coincided with theory-driven surmises, not actual “discoveries” Freud made in the Seduction Theory days of 1896.