Distinction of Popular and Learned Traditions
The first step towards ascertainment of popular tradition is to determine which documents are most faithful repositories of that tradition. For obvious reasons literary texts—treatises on witchcraft, judicial manuals, chronicles, sermons, and so forth—cannot qualify as faithful sources for the beliefs of the illiterate masses. They no doubt incorporate material that derived from popular belief, but the historian has no assurance that they present folk tradition in a pure form, unmixed with distinctively learned notions. Even when an author claims to be representing popular belief one cannot accept his testimony as ideal evidence. Elliot Rose has correctly emphasized that litterateurs such as Jean de Meung frequently claimed to be relating contemporary folklore when they were in fact merely paraphrasing texts from canon law.1 Some of the literature of the high and late Middle Ages was ‘popular’ in the sense that its content was intended for widespread dissemination. This is true of anecdotal collections and of many chronicles, which were written in large part to furnish material for preachers.2 Yet there is no assurance that the stories narrated in such literature had popular origins, or that they did in fact take hold on the popular imagination.
Nor are most of the judicial records a great deal more helpful to the student seeking authentic folklore. In most instances the only extant records of trials are the documents drawn up at the end of judicial interrogation, stating in summary fashion the charges ‘proven’ against the subject. In some cases there are records of questions posed to the accused, with the answers that were in some way extracted from them. Yet even when these documents claim that the confessions were made ‘voluntarily,’ or ‘spontaneously,’ the historian can seldom be sure that the admissions were not elicited through judicial coercion. In many cases there is evidence for confession under torture; in other instances, reference to voluntary confession indicates only that the subject’s admissions were confirmed after torture had ceased, but under threat of resumed torture. One might argue that for present purposes torture did not prejudice the evidence. The confessions elicited from witch suspects may not have been true, but it might be argued that the subjects knew in advance what witches were commonly believed to do, and that when they confessed to these deeds they were merely applying popular tradition falsely to themselves. Yet not even this can be assumed. As will be seen in a later chapter, the information that judges wished to educe could easily be suggested to the subjects through use of interrogatories, or lists of leading questions.3 And even if the suspects did know what their judges expected to hear, they may have learned the proper details not from popular tradition but from inquisitors’ sermons and from reports of earlier trials. Both in general themes and in specific details, judicial records are thus likely to reflect influences other than pure folk tradition.
On the other hand, one should not give up in despair on the assumption that all judicial documents are uniformly subject to corruption by learned influence. In a small number of cases, the ipsissima verba of the accusers are recorded, or at least close paraphrases: the testimony given by the original witnesses, made at the beginning of proceedings, before the formal interrogation of the subjects, and before the application of coercive measures. In a few cases, to be sure, the original accusations were made by members of the educated élite; in a few others, they came from persons already convicted of witchcraft, who were exposed to the judges’ notions of the crime during their own interrogations, and were later asked to incriminate their supposed accomplices. But in most instances the original depositions represent the sentiments of the witch’s neighbours, who sought to secure her conviction. Out of about 500 witch trials between 1300 and 1500, these depositions survive in only twenty-one cases. Fifteen come from Switzerland,4 three from Germany,5 and three from France.6 The preponderance of Swiss trials is not surprising; in general, the judicial records from Switzerland are more abundant than for most other regions, frequently with fuller information for individual trials. And there is enough evidence of various types from other regions that it is reasonably easy to discern those respects in which Swiss cases are or are not distinctive.7 Apart from these twenty-three cases there are fourteen further trials in which the proceedings were initiated by the persons who had been accused of witchcraft. Desiring to clear their reputations, they brought charges of defamation against their accusers. Seven of these are from Swiss towns,8 six from England,9 and one from Germany.10 In this case the high number of records from England is interesting, and can probably be explained as a result of the relatively lenient views of English judges, who were frequently more inclined to try those persons who had defamed others than to punish the witch suspects themselves. In these trials for defamation there was no reason for either the judge or the plaintiff to introduce extraneous, distinctively learned concepts of witchcraft, and the notions recorded may once again be taken as faithful reflections of popular tradition. One might hypothesize various reasons why one or another of these thirty-five trials (including both groups) could have been influenced by originally learned notions, but they are the most reliable texts available, and the likelihood of their contamination is so slight that one may confidently use them as touchstones for analysis of further documents.
Apart from these thirty-five trials, there are certain categories of trial which taken as classes are superior to others, even though individual trials within these groups may be unreliable. Thus, trials from England, where there was less judicial coercion than on the Continent, are less likely to reflect learned notions, though in certain specific English cases one may suspect the heavy hand of an educated judge in the statement of charges. Likewise, secular courts were in general less prone to transmission of learned concepts than were ecclesiastical courts. Lay tribunals were not wholly immune to ideas that developed in theological and judicial scholarship. There are many cases in which secular authorities maintained close contact with inquisitors, and could easily assimilate ideas that originated in clerical circles.11 Furthermore, some lay judges could read Latin treatises on demonology; and many of these men were certainly familiar with vernacular works which eventually translated late medieval demonology into an attenuated but still vivid idiom.12 Thus, the records of secular courts have no absolute claim to reliability as repositories of popular tradition, but as a class they may be judged more trustworthy than ecclesiastical records. Even within secular tribunals, one must differentiate between records written in Latin and those drawn up in the vernacular. It is seldom possible to ascertain definitely whether there was anyone in the court who had personally read the current demonological literature, or whether there was a written formulary at hand for use in interrogations, but these influences were more likely in trials for which the records were composed in Latin. Hence, the documents in the vernacular form a third class of relatively reliable texts, alongside English records and those of lay courts generally.
It might seem useful to distinguish further between trials initiated ‘from above’ and those begun ‘from below.’ When inquisitorial procedure was introduced into European courts it was no longer necessary for the authorities to wait for accusation from the aggrieved parties before instigating prosecution. One of the main points of inquisitorial procedure was that it permitted judicial officials to act on their own initiative.13 Virtually all the witch trials of the fourteenth and fifteenth centuries seem to have employed inquisitorial procedure, in that there was no formal accuser who took upon himself responsibility for proving the charges. Yet this does not mean that these trials all began ‘from above’ in any absolute sense. Even when the procedure was inquisitorial, prosecution typically began in response to popular complaint. At times the records state explicitly that proceedings began both ex mero officio and fama publica referente.14 Even in those relatively rare cases in which inquisitors made use of a ‘general inquisition’ to uproot witchcraft from an entire community, their own concern was supplemented by that of the witnesses (usually laymen) on whom they relied for names and specific charges.15 It was always possible for inquisitors themselves to furnish the specific data for initiating trials, but evidently this seldom happened, and even when they did act on their own suspicions they no doubt typically solicited and obtained corroboration from the populace. In short, most witch trials appear to have been products of cooperation, in some form or other, between the people and the authorities. (And there is no reason whatsoever to think that the general populace was uncooperative; local authorities resisted proceedings on rare occasions, but even in the celebrated trial at Innsbruck in 1485 the witches’ neighbors brought forth accusations enthusiastically.) In the majority of cases witnesses from the population were heard, whether they came forth spontaneously to state their grievances or whether judges sought them out. Unfortunately, the records seldom indicate how the witnesses got to court. For present purposes, however, the question is of little significance, since the initiative for prosecution did not necessarily have bearing on the content of the charges. So long as the populace played a role, popular beliefs were likely to enter into the accusations; and in any type of trial there was the possibility that popular notions would be overlaid with distinctively learned opinions. It was perhaps more likely that learned tradition would take precedence when authorities manifested their own spontaneous concern with witchcraft, either by holding general inquisitions or by acting on their own suspicions about specific defendants. But such cases were apparently rare, and are generally impossible to isolate on the basis of surviving documents.
Turning, then, to the records which are individually of high reliability, one finds fundamental commonalities. When witnesses’ depositions or charges of defamation survive, the overwhelming emphasis is upon sorcery. Occasionally there were further charges: in a few Swiss trials the defendants were supposed to have ridden on wolves, and in a minority of trials the defendants had allegedly invoked demons for the fulfillment of their wishes.16 But the charge of diabolism—with veneration of the devil, formation of a pact, reception of the devil’s mark, attendance at the Sabbath, flight through the air on route to the Sabbath, ritual repudiation of Christian faith, intercourse with incubi or succubi, and so forth—is altogether absent from these records. If these trials are typical—and further evidence will suggest that they are—diabolism played little or no role in popular belief.
To elucidate the nature of these trials, it may be useful to examine some of the ones for which relatively extensive records have survived. The trial of Dorothea Hindremstein from Lucerne, in the mid-fifteenth century, is in many ways typical.17 After a minor quarrel, Dorothea had allegedly inflicted a sickness of three weeks’ duration on a neighbor’s child; the sorceress had made some kind of vague threat to the child, and the illness that ensued was taken as the result of this curse. Other townsmen tried as best they could to avoid dealings with the woman, for fear that they might antagonize her and suffer bewitchment by her. Yet willy-nilly they came into contact with her, and the resultant conflicts terminated with sundry diseases, cows that gave blood instead of milk, and doubtless many other complaints. Witnesses furthermore swore that Dorothea had used a magical rite to multiply a quantity of cereal, so as to provide enough to feed ten people. The subject herself was not accused of diabolism, but oddly a possessed woman had accused her once of being a witch, claiming to know this on the authority of Beelzebub. In this case, then, even one of the accusers had closer contact with Satan than did the sorceress herself.
When another woman went before the town court several years later, the accusations were more diverse, yet they still did not encompass diabolism.18 The records of this trial are replete with cows that yield nothing but blood, storms produced by maleficent magic, shriveled onions, geese that have gone mad, and so forth, but the devil makes no appearance. The subject is supposed to have caused pain in another woman’s bosom by approaching her one day, grasping her on the breast, and asking why she was so wanton with her breasts. On another occasion, the subject’s children were discussing their respective pets with certain playmates, and divulged the information that their mother kept pet foxes and wolves, which she fed when their father was away from home. Other witnesses told that they had seen the woman riding on a wolf—though neither to nor from a Sabbath. Anyone familiar with the demonological literature of the century would have had ample occasion to suggest the charge of diabolism, but so far as the record shows no one made this transition.
Two trials from the diocese of Lausanne are especially instructive, since they include both the witnesses’ depositions and the later records of interrogation, and they make abundantly clear that popular tradition emphasized sorcery, while the learned tradition of the judges focused on diabolism. In the first instance, in 1464, the accused was a woman named Perrussone Gapit.19 The first two witnesses told of bewitchments they had suffered: one had eaten some pears that Perrussone had cooked, and predictably he had become so ill that he could scarcely arise from bed. The mother of a newborn baby alleged that Perrussone had tried to kidnap the infant. After one thwarted attempt, as Perrussone ran away from the house, the mother tried to call out, ‘Wicked woman, you shall not have him!’ but was mysteriously unable to utter the words. At length, Perrussone succeeded in killing the infant. More than that, one day she placed her hand on another child’s head, causing paralysis in the child’s right side. The judge asked one witness whether he had heard Perrusonne recite the creed, and the man said he had, but only in an unintelligible mutter, ‘in the manner of old people.’ The first interrogation of the subject herself was conducted by a notary and his assistants, and dealt exclusively with the witnesses’ charges. Perrusonne either denied the allegations or gave different interpretations of the events. For example, she admitted having gone to the bedroom of the mother with the newborn baby, yet claimed that her motive was not kidnapping, but merely solicitude for the mother. Soon afterwards, however, an inquisitorial vicar entered the case. Within two days, he had the subject confessing that she had attended diabolical assemblies, had intercourse with the devil (who, true to form, was unbearably cold and ugly), and paid homage to the devil both in human form and in the shape of a bear. On one occasion she had kissed the bear as a sign of homage, and soon afterward the beast had scratched her so roughly on the finger that she feared it had come off; indeed, even at the time of the interrogation it was still sore. In the company of devil-worshippers she had been induced to kill children and raise storms. In these last-mentioned details the inquisitor shared with the original accusers an interest in sorcery. But he was above all concerned with diabolism, and with a modicum of inducement—at least some of the confessions were extracted under torture—he succeeded in eliciting the expected information.
A trial in the same diocese at the end of the century was analogous.20