Truth and the languages of accusation
Pieter Brueghel the Elder, The Tower of Babel (1563)
SOME PHILOSOPHICAL PRELIMINARIES
One of the greatest barriers to understanding both witchcraft trials and demonology is the insistent asking of the question, "Is it true?" Almost inevitably, the answer will be, "No, it isn't!" This does not get our enquiries much further, because it is based on our assumptions about how the world works and how truth is established. Such a question thus disables us from understanding the past, because we are too busy judging it for failing to be the present, and it disables us from understanding ourselves, because it privileges so absolutely those things which we currently happen to believe to be true.
At the root of our difficulties with truth lies the problem of language. We might be prepared to concede that when people in the past described some of their neighbours as "witches", there are features of the description that we could recognize. However, we assume that the curses and charms could not have had any physical effect, although we might allow a little scope for what we would call "psychosomatic" effects. Thus, we transfer the blame to the credulous accusers, while permitting ourselves a little scorn for the impoverished crone who brought the accusation upon herself. As for the more extravagant claims, of transmutation, devil worship and flight to the sabbat, we take it for granted that there was a radical mismatch between the language and the world it purported to describe. Our language, on the other hand, describes a real world full of real phenomena.
However,
philosophers from both the European tradition and the Anglo-American tradition
have now spent many years questioning the notion that there can ever be
a perfect match between language and the world. The facile belief
that truths are there to be discovered by heroic individual thinkers, as
depicted in Rodin's statue, and then simply described in transparent language,
has given way to the idea that truths are made by language-users and supported
by those members of a speech community who find them good to use and think
with. "Natural" languages and dialects support perceptions and ranges
of expression that are peculiar to them, and so do intellectual traditions
and academic disciplines, with their distinctive jargon, rhetoric and standards
of proof. Interestingly, just such issues vexed early modern philosophers,
who were rarely so befuddled as to imagine that the creation and establishment
of knowledge was a matter for individuals, rather than a social process
which involved carefully acquired credibility and the use of rhetoric and
language. Nominalism was more widespread than realism, although there
was a frustrated search for the way to regain a universal language which
would have a direct connection with the natural order of things.
The image of the Tower of Babel was invoked to underline not only the radical
divergences between current languages but also the absence of any direct
connection between language and reality, in contrast to the words spoken
in Eden: "whatsoever Adam called every living creature, that was the name
thereof."
The realist model of language, when it was turned on the history of witchcraft, implied that one could look at the statements of participants, determine whether there was any acceptable truth in them by reference to the world, and then dismiss them as the product of disease, conspiracy or deranged psychology. This model lies behind attempts to show that "really" witchcraft beliefs were caused by outbreaks of ergotism or hysteria, the promotion of patriarchy or statebuilding, or whatever happens to be the current explanation. Witchcraft beliefs are explained away, as symptoms of something else. We often try to explain away the supposedly mistaken beliefs of others, such as our religious or political opponents, but we rarely feel the need to perform this procedure on people in the past, explaining away their belief in a geocentric universe as if it were caused by some epidemic of glaucoma. Witchcraft clearly arouses our passions and self-defences, threatening our certainties about the world.
Most of the past theories about witchcraft, whether Romantic, rationalist or sociological, have assumed that what mattered was the reference of the language to reality, although they have arrived at rather different answers to this puzzle. The Romantics wanted to see the past language as describing real people doing real things, while exaggerating a little. The rationalists insisted, on the contrary, that there was no correspondence between the language and reality, so the whole affair had to be dismissed as the product of fanatical delusions or disease. The later "social science" approach saw the language as mistaken in its overt sense but as expressing all too real social stresses and strains of one sort or another. In England, this took an anthropological turn, with village accusers seen as displacing their guilt over refusals to give alms. In Germany, the unfortunate association of folklore studies with the Nazis led to the social science approach remaining largely statist and economistic, until fairly recently.
Such views expressed a confidence about the permanent validity of post-Enlightenment understandings of reality that fits badly with not only present philosophy and linguistics but also present physics. What matters about a truth-claim, it is now argued, is less its correspondence with an empirically discoverable world than its coherence with other signs and symbols for making statements. This does not imply, as is often said in criticism, that objects do not exist, only that they do not unambiguously present themselves to us for our inspection. [Nor, incidentally, does it necessarily imply moral relativism or indifference, as is also often suggested.] We test our statements and beliefs against the world, but only within the constraints of our systems of language, in the broadest sense. There is no independent test, free of language and categories, of the truths we accept or those which we make afresh. When the realist strikes an object and shouts, "There really is a table there!", hidden assumptions about the nature of sense experience and the transcendental reality of the category, "tables", are being invoked. Such assumptions may work well enough in everyday life, but they will not stand very much scrutiny. Thus, there is no useful way of asking whether witchcraft beliefs "really" did correspond with reality, because such referential games can only be played within an established language system. "They" thought the beliefs did correspond to reality, "we" do not. What counts as "real" has changed.
To understand the debates and the trials that took place involving the crime of witchcraft, or indeed anything else in the past, we need to set aside our obsession with demonstrating the superiority of our own way of seeing the world. All the participants saw the world very differently from the way we see it. We cannot fully enter into their world, because we are always making incomplete acts of translation, from their language to ours, but we can increase our understanding by listening very carefully to what they were saying.
TALKING
WITCHCRAFT
Language was conspicuously at the centre of witchcraft phenomena. Witches muttered curses and sorcerors intoned elaborate incantations. Accusers told stories, discussed rumours, and constructed narratives, in which folk tales turned into statements given in court. The bewitched and possessed would utter in strange voices or languages they had not learned, in front of huge audiences. Demonologists translated the folk tales and accusations into a form that was compatible with learned discourse. Clergymen preached sermons against witchcraft and wrote pamphlets and learned tomes. Torturers asked questions and the accused persons uttered impassioned denials or agonized confessions. Judges and magistrates listened to depositions and testimony, about oral exchanges with the accused or the last words of deceased victims, and then translated the folk tales into the language of learned demonology. The judges' chaplain preached an improving sermon from the scaffold and the condemned witch gave a last speech, confessing to sins or calling for God's mercy. Pamphleteers and learned authors selected and transformed the tales, speeches and depositions into a form suitable for publication. The pamphlets would be read aloud at the family hearthside. Playwrights and actors made plays about witchcraft. Balladeers and broadside publishers put the events into a form suited to the humblest reader, who would recite the broadsides or sing the ballads for the benefit of illiterate friends. And all these people, apart from the condemned, would tell and retell these stories.
To examine all these aspects of the process on this page would be excessive, but it is clearly no exaggeration to say that witchcraft was constructed by language. Those Spanish inquisitors who insisted on the reality of witchcraft in the Basque country pointed to confessions. Alonso de Salazar retorted that there were no witches to be found until everyone started talking about them. However, to say that language constructed witchcraft tells us everything and nothing, because the same is true of all human knowledge. That is how humans make and know any truths, or lies for that matter. It will have to suffice to draw attention to some features of the problem.
TRANSLATION
If we are making incomplete acts of translation, from one language game to another, whenever we read a primary source on the subject of witchcraft, the same was true of people at the time. In some ways, the genius of Kramer lay not just in synthesizing the works of his predecessors but in translating the folk tales of the Alpine regions into a language that was compatible with learned philosophy and theology. He thus made it possible for judges and clergymen to understand what was "really" happening behind the incomprehensible language about Dame Abundance and the evil eye, at which his medieval predecessors had laughed. Thus, the coming of the trials depended crucially on the accusations of the witch's victims being translatable into language and concepts suited to their elite hearers. A "trade language", a pigeon or creole, had been constructed to assist communication between different language systems. Such translation was always incomplete - villagers and judges and theologians had different interests and concepts - and it was therefore apt to collapse. The more different groups there were involved in making a bad reputation turn into an execution, the more hurdles there were to overcome. Whenever there were shifts in elite perceptions or political and religious changes, the relationship between the different systems involved had to be renegotiated.
When elite and popular accusers spoke, quite literally, different languages, witchcraft trials were almost impossible. This situation occurred in occupied countries and in countries where the elite had adopted the language and thought systems of a dominant foreign power, as in the Highlands of Scotland where the clan lairds abandoned Gaelic for the Lowland Scots of the dominant elites in Edinburgh. Ireland also saw relatively few trials, as far as surviving documents can show, except in the "English Pale" around Dublin where there seems to have been a cluster of cases in the 1650s. There were probably too many barriers between different groups to make successful prosecution straightforward, so Irish villagers probably took their own counter-measures rather than resorting to the courts. In the case of Mexico, the native population was not under the jurisdiction of the Inquisition but a separate tribunal, which was just as well, in view of the mental gulf revealed by the demonological tract written by Hernando Ruiz de Alarcón in 1629. He could hear what was said in Nahuatl, but was quite unable to enter the worldview of his informants sufficiently in order to be able to translate accurately.
If the achievement of translation, from one system of signs and meanings to another, is one way to understand what happened in the fifteenth century, perhaps we can use this analogy to explain later changes. The end of the trials in England, for example, left many ordinary people frustrated at the lack of legal remedy, and many members of the religious and political elite continued to believe in witchcraft. After the crime of witchcraft had been abolished, John Wesley remarked, speaking of the Devil of Mascon narrative, "With my latest breath will I bear my testimony against giving up to infidels one great proof of the invisible world: I mean that of Witchcraft and apparitions, confirmed by the testimony of all ages." One way of understanding what happened is that there was a problem of translation: popular accusations, with their accustomed semiotics of vomited hanks of wool and hafts of knives, their stories of sick cows and convulsed children, could no longer be translated into terms that made sense to judges, magistrates, and grand jury members. ["Semiotics" refers to sign systems, originally in diagnostic medicine but now more usually in languages; I am suggesting that many witchcraft phenomena are best understood as parts of a "language", a system for the presenting of symptoms.] This does not allow us to avoid explaining why some people believed one thing and others believed another, but it seems a markedly more useful way of understanding events than simple talk about "the rise of science", as it was leading members of the Royal Society, such as Joseph Glanvill and Robert Boyle, who were most active in trying to preserve the belief in witchcraft and demons.
SOME IMPLICATIONS FOR MODERN READERS
Historians attempt to tell stories that are true, or at least as true as they can make them. It is not necessary to abandon this desire to come as close to past experience as possible, but it is necessary to accept the limitations imposed by language problems. However hard we try, we cannot enter fully into the language world of the past, so there will always be problems of translation. We cannot entirely hear what people in the past were saying, but we can try hard to avoid making premature translations, by listening carefully and by giving due weight to the terminology employed at the time.
Clergymen and lawyers who dealt with witchcraft were very attentive to shifting meanings. Indeed, demonologists were often more sceptical than their critics. They were, after all, dealing with the Prince of Lies and his earthly agents. Nothing was fixed, nothing was certain in the world of witchcraft. After the Salem trials collapsed, the Puritan ministers scored the contest as a win for Satan, who had successfully deceived the court and the colony as a whole. There might well have been witches involved, but it was impossible to tell who they were.
Nevertheless, many of the documents which historians are inclined, perhaps naïvely, to trust in other circumstances, such as depositions and pamphlets, turn out to be deceptive. The information turns out not to be composed of simple statements, which might be mistaken but had honest intentions, but rather made up of strategic positions of the sort one might find in election propaganda. Just as committed voters tend to believe the statements of their own candidate or party and disbelieve the opponents' statements, so too historians have all too often chosen sides in early modern debates about witchcraft. This tends to mislead the historian, who fails to notice that the "correct" side of a witchcraft debate is taking up positions relative to that of its opponents. Allegations of fraud, and even personal confessions of deceit, are just as much constructed by social circumstances as allegations of bewitchment.
The historian who is reading witchcraft pamphlets needs to pay very close attention to who is speaking and why. Among the English pamphlets, mostly anonymous, there is a significant change in about 1590. Before that date, apart from a few clearly sensational pieces, most of the pamphlets come sources from close to the court where the case was tried. This is especially evident in the case of the St Osyth's pamphlet, but it is true of others as well. Thereafter, we have a wider range, few of which come from such authors as the magistrates or their clerks. Some are fictional, while apparently true, others are highly fictionalized, and yet others come from sources close to an elite family who were the accusers in the case. Some pamphlets, such as the Chelmsford pamphlet of 1589, flick from the dispassionate recounting of evidence to opinionated moralizing. The evidence concerns Joan Cunny, a traditional elderly witch figure, who exerts harm with the aid of her familiar spirits. The editorial comments portray a lewd and immoral woman, a "cony". In consequence, historians have either seen her as a woman brought up in the old religion of Catholicism (Keith Thomas) or as a sexual deviant being persecuted for her dissent from prescribed norms (Marianne Hester). The role of the layered nature of the text in creating these contrasting views has not been much noted, because of historians' implicit faith in the correspondence theory of language. What is being conveyed by the text is not truth, however disguised, but meaning, and the construction of the text betrays that aim. Both stories about Joan Cunny, and several others, can be supported by selective use of the source, but they depend more on the interests of the scholars concerned than on any evidence more substantial than story-telling.
The pre-1590 pamphlets usually derive from the court material, and depict the familiar tale of refused charity and muttered curses coming to fruition, which allows the historian to transfer the guilt back onto the now long-dead accuser, by saying that the transfer of guilt from accuser to witch was what was happening at the time. This point was made at the time by Reginald Scot and George Gifford, who parodied the stories of the accusers, and it was picked up in the functionalist analysis of the early 1970s, advanced by Keith Thomas and Alan Macfarlane. Although this view was heavily criticized by anthropologists, who had already abandoned the functionalism of Max Gluckman and other founding fathers of social anthropology, it remained influential among historians, in part because new full-scale studies of English witchcraft were not undertaken. This then is the basis for the "English stereotype", even more influential among historians than it was at the time.
Many of the post-1590 pamphlets, such as the Throckmorton children's case or the Lord Roos case, derive from sources close to the victim's family and are intent on showing the victims as entirely innocent, so the origin of the witch's grudge is wholly or partly effaced. Thus, we have a new witch stereotype emerging, in print at least, wherein the witch is senselessly enraged against those who wish her no harm. At times, an original motive peeks out of the narrative, such as the dismissal of a servant, denied in the story of the Lord Roos case but revealed in the examination of Margaret Flower, printed with it. Since the original depositions do not survive, we have no way of checking just how much erasure has taken place. Where the old story survives into the seventeenth century, it is in pamphlets such as Thomas Potts's large piece on the Lancashire cases of 1612, which mainly reproduces the depositions. [These should not be confused, in English or New England cases, with the evidence given in the trial, which might well have been quite different.]
TRUTH,
TRUST AND ORAL NARRATIVE
Yet if we distrust the second category of stories, and following Scot's analysis we doubt the honest intentions of the stories about curses coming to pass, why should we trust the body of the narratives, whether in court or in pamphlets? Margaret Murray carefully edited out of her material all the supernatural elements in order to tell her tales of misunderstood pagans. In setting aside the magical element but trusting the social aspects of the narrative, perhaps we commit the same error, albeit less conspicuously. We could follow Montague Summers and trust everything in the pamphlets, but it might make more sense to distrust everything, or rather to accept all the elements as part of a story rather than only those we happen to disbelieve. The stories about children falling ill, and butter failing to churn, and dead men's last words are often many years old by the time they get to court or pamphlet. They have been told and retold, restructured and confabulated, first to make a narrative of life that made sense to the teller, then to make a convincing accusation, and finally to appear in a court record or pamphlet. There are stereotyped elements that derive from other cases or from folklore, motives that show why the events occurred and actions or conversations that betray those motives.
Nobody in such cases needs to be engaged in conscious deceit, although some may be. In any event, even deceitful stories have to be credible to listeners. The participants are simply telling positioned narratives with which they are comfortable and which are designed to achieve desired results. Some elements may have been created or reshaped in the informal processes of village gossip, others in the process of questioning by an investigating magistrate. It was not only the lifestories of confessing witches that were restructured by this process. The whole community was telling stories of one sort or another. Thus, the problems of language, semiotics, and narrative are everywhere. We simply have to accept them as part of the task, both in reading the texts and in writing our own stories of witchcraft.
This can be disturbing for historians, who are accustomed to reading documents produced at a time very close to the events described. One possible approach would be to take the work of oral historians, who have to contend constantly with the remaking of memory, and apply some of their approaches to the construction of depositions in witchcraft cases, many of which recount events of fifteen or twenty years earlier. When Alessandro Portelli was doing an oral history of a small working-class Italian city in the 1970s, he became puzzled when his subjects repeatedly made factual errors or even related events that had never happened. For instance, when talking about the death of a worker named Luigi Trastulli, who had been killed in a clash with the police in 1949, the people Portelli interviewed all insisted that the event had occurred during demonstrations in 1953. He eventually recognized that the original demonstrations, against the Italian decision to join NATO, had lost their meaning, whereas the later ones against factory closures had a continuing resonance. "I realized that memory was itself an event on which we needed to reflect," he said in a recent interview at the University of Rome. "Memory is not just a mirror of what has happened, it is one of the things that happens, which merits study." Instead of judging oral accounts merely by standards of truth and falsity, historians now look for themes and structures, the symbolic material out of which people construct their lives and memories.
A CAUTIONARY NOTE
There is a temptation to deploy a very simple explanation: language creates the perceived world, so the use of the terminology of witchcraft creates prosecutions. Clearly, without a popular discourse of witchcraft there would be no accusations, and without a learned discourse there could be no prosecutions. However, on this basis, we would expect Catholics and Protestants, who constantly used the language of witchcraft and demonic rebellion against one another, to prosecute their opponents at every opportunity. In fact, they did not, although we can find places where intense religious conflict appears to have resulted in displaced accusations, onto associates rather than members of the despised group. Christine Heyrman has suggested that friends of Quakers were vulnerable at Salem, and Urbain Grandier at Loudun was perhaps too friendly towards Protestants. If anything, the over-use of the language of witchcraft in the context of religious politics appears to have destabilized the discourse, making it too partisan to command universal assent. Peter Elmer has suggested that the use of witchcraft language against the Quakers and other sectarians, during the 1650s in England, contributed to the undermining of prosecutions thereafter.
SHORT BIBLIOGRAPHY
Hernando
Ruiz de Alarcón, Treatise on the Heathen Superstitions that Today
Live Among the Indians Native to this New Spain (Norman, OK, 1984)
F 1219.76 .R45 R8413 1984
Martin Elsky, Authorizing Words: Speech, Writing, and Print in the English Renaissance. (Ithaca, 1989)
Alessandro Portelli, The Death of Luigi Trastulli, and Other Stories : form and meaning in oral history (Albany, N.Y., 1991) D 16.14 .P67 1991
The
series of three volumes of essays edited by Pierre Nora, Les Lieux de
Mémoire (translated as The Realms of Memory) is particularly
interesting on mythmaking, although it deals with a larger community, the
French nation.
Richard Waswo, Language and Meaning in the Renaissance (Princeton, 1987)
A new collection of essays offers a particularly pertinent introduction to some of these problems: Languages of Witchcraft: Narrative, Ideology and Meaning in Early Modern Culture , ed. Stuart Clark (Basingstoke, 2001) BF 1584 .E85 L36 2001 The contents of this book are as follows:
PART 1: HISTORY AND STORY IN WITCHCRAFT TRIALS
Peter Rushton, "Texts
of Authority: Witchcraft Accusations and the Demonstration of Truth in
Early Modern England" [PDF file] this file includes the
table of contents and the index of the book
M. Gibson, "Understanding Witchcraft"
M. Gaskill, "Witches and Witnesses in Old and New England"
D.Purkiss, "Sounds of Silence: Fairies and Incest in
Scottish Witchcraft Stories"
PART 2: CONTEXTS OF WITCHCRAFT
P. Elmer, "Towards a Politics of Witchcraft in Early
Modern England"
D. Wootton, "The Religion of Reginald Scot"
J. Barry, "Hell Upon Earth or the Language of the Playhouse"
PART 3: HOW CONTEMPORARIES READ WITCHCRAFT
R. Briggs, "Circling the Devil: Witch-doctors and Magic
Healers in Early Modern Lorraine"
M. Tausiet, "Witchcraft as Metaphor: Infanticide and
its Translations in Aragon in the Sixteenth and Seventeenth Centuries"
T. Robisheaux, "Witchcraft and Forensic Medicine in Seventeenth-Century
Germany"
K. Hodgkin, "Reasoning with Unreason: Visions, Witchcraft
and Madness in Early Modern England"
Katharine Eisaman Maus, "Proof and Consequences: Inwardness and its Exposure in the English Renaissance", Representations, No. 34. (Spring, 1991) 29-52
Elizabeth Hanson, "Torture and Truth in Renaissance England", Representations, No. 34. (Spring, 1991) 53-84
Brendan Dooley, "Veritas Filia Temporis: Experience and Belief in Early Modern Culture", Journal of the History of Ideas 60.3 (1999) 487-504
Brendan Dooley, The Social History of Skepticism: Experience and Doubt in Early Modern Culture (Baltimore, 1999)
There is a substantial literature on the social roles involved in establishing credibility. A useful review of recent work is Adrian Johns, "Identity, practice, and trust in early modern natural philosophy", The Historical Journal 42 (1999) 1125-1145 [PDF file]
