Judicial Assessment of Narratives and Statements
As part of the medieval procedure of compurgation, or wager of law, witnesses made a sworn statement affirming the truthfulness of the claims of the party they supported, and their oath was not subjected to cross-examination.1 The fact-finders in witch trials had to rely almost entirely on narrations. Suspects, accusers and witnesses told their stories, and the judiciary faced the tremendous challenge of distinguishing true stories from false ones.
Different narrators had different knowledge, biases, personalities and perceptual limitations. To reach a decision, the fact-finders needed to assess the veracity and importance of the various testimonies. It was important that they consider not only what these stories were about, but also who told them. The trustworthiness and quality of witnesses were determined both by their individual traits and according to the group of witnesses to which they belonged.
The focus of this analysis, as in other parts of this study, is not on the actual rules of evidence that were used in the courtroom, but on the social attitudes and opinions regarding the question of who should be believed and why. Early modern England saw significant developments of evidentiary rules pertaining to the identity of the witnesses and to the form and content of their testimony. The difficulty of proving witchcraft posed considerable challenges to the assessment of the truthfulness of narratives, and the debate included consideration of evidentiary methods not specific to witchcraft cases (such as credibility, the value of uncorroborated confessions, and so on).
Because testimony was still to be accepted on oath in the sixteenth century, the defendant’s objections could only be to the competency of the witness (to keep categories of potential liars outside the courtroom) rather than to credibility.2 At first, witnesses were filtered through various competency rules that aimed to exclude the possibly biased witnesses. Categories of incompetent witnesses included defendants in criminal cases, ‘spouses of parties, persons with financial interest in the case, convicted felons, irreligious persons’, and so on.3 In Baker’s words, it was ‘rather a law of witnesses than of evidence’.4 Later, the focus shifted from competency to credibility.
This chapter is dedicated to three categories of witnesses in witchcraft cases: children (on the basis of age, normally regarded as incompetent because of a lower level of understanding); accomplices (whose competence was challenged by conflicting interests); and expert witnesses (whose knowledge was a source of authority but also a potential threat to judicial superiority). The debate around these three categories demonstrates major considerations in evaluation of testimonies.
Children were normally regarded as incompetent witnesses. The testimony of children under the age of 14 was not generally allowed. Sir Matthew Hale remarked that the testimony of a child between the ages of 9 and 13 might be admitted ‘in some cases’.5 The rationale, Gilbert explained, relying on Hale and a list of court cases, was ‘the Want of Skill and Discernment … Ideots, Madmen, and Children under the Age of common Knowledge … are perfectly incapable of any Sense of Truth, and therefore are plainly excluded. Children under the Age of 14 are not regularly admitted as Witnesses.’6
Yet, in proving witchcraft, the competency rules were bent. The testimony of otherwise incompetent witnesses, such as children under 13, could be admitted to prove crimes like ‘rape, buggery, witchcraft and such crimes’.7 Henry Goodcole, the minister who tried to compel Elizabeth Sawyer to confess, reported that ‘some children of a good bignesse, and reasonable vnderstanding’ testified in court ‘that they had diuers times seene her feed two white ferrets with white bread & milk’.8 It remains obscure what exactly ‘good bignesse’ meant. Obviously, Goodcole was trying to justify the reliance on testimonies that led to Sawyer’s execution by emphasizing the children’s maturity.
The issue of children’s testimony was even more problematic in many cases where children were required to give testimony against their own parents. James I maintained that children’s testimony was acceptable. One of the characters in his dialogue expressed the dilemma: ‘For it is as great a crime … To condemne the innocent, as to let the guiltie escape free.’9 He therefore concluded that in cases of serious crimes, problematic categories of witnesses (including the children of the accused) should be allowed to testify.10
James I believed that the severity of the crime justified admitting evidence otherwise inadmissible, including testimony of children against their parents.11 Reginald Scot pointed out that Bodin, the renowned French jurist, approved the interrogation of children for statements against their own mothers in cases of witchcraft.12 Scot, however, believed that youthfulness (among other qualities) reflected negatively on the ‘credit’ and ‘weight and importance’ of the testimony. Discussing an account published two years earlier, A True and Iust Recorde, Scot considered childhood among other criteria that damaged the value of witnesses.
Sée whether the witnesses be not single of what credit, sex and age they are; namelie lewd, miserable, and envious poore people; most of them which speake to anie purpose being old women, & children of the age of 188.8.131.52.8. or 9. yeares. And note how and what the witches confesse, and sée of what weight and importance the causes are.13
To Scot, the meaning of the terms credit and weight was to a large extent an extrapolation from the category to which the witness belonged. Thus, the veracity of the testimony of children under the age of nine was questionable. Although to some limited extent Scot referred to the individual characteristics of the how and what of the testimony, he had not as yet developed the conceptualization of credit and weight.
Gifford, an eminent preacher of Maldon, in Essex, was also critical of using children against their parents:
Yea sundry tymes the euidence of children is taken accusing their owne mothers, that they did see them giue milke vnto little thinges which they kept in wooll, The children comming to yeares of discretion confesse they were entised to accuse.14
Gifford stressed that children could be easily incited and their testimony against their mothers should be looked on with suspicion. The execution of innocent people on the testimony of their own children was another evil ploy of the devil. The evidentiary effectiveness of children’s testimony was a minor concern. The major concern was to resist devilish wickedness.
It is most likely that when Gifford criticized the testimony of children, he had in mind a trial that had been held five years previously in St Osyth, a few miles away.15 In that case, eight-year-old Thomas Rabbeth, the ‘base son’ of Ursley Kemp, was made to testify against his mother.16 Another eight-year-old child, Febey Hunt, deposed against her stepmother, Ales Hunt.17 Henry Sills, nine years old, deposed that his father scolded his mother ‘why thou whore cannot you keepe your impes from my childre~[…] whereat shee presently called it away’. His younger brother, John Sills, under seven, was also examined.18 Annis Dowsing, the seven-year-old ‘base’ daughter of Annis Herd, deposed that her mother kept spirits in a box.19 Clearly, Gifford’s opinion did not prevail, and in many cases children’s testimony, even against their own parents or grandparents, was permitted.20 The acceptance of children’s testimony continued after the publication of Gifford’s book. The ten- and twelve-year-old ‘bastard’ boys of the ‘two lewde Daughters’ of Joan Cunny, delivered ‘great euidence’ against their mother and grandmother.21 Henry Sutton ‘the Bastard of Mary Sutton’, quoted an incriminating conversation between his mother and grandmother, ‘little thinking that his fortune should be to give in evidence to breake the necke of his owne Mother and Grandmother’.22 According to the pamphlet, the boy also admitted that his mother’s ‘Spirits in severall shapes as Cats, Moals, &c. used to sucke her’.23 His mother, Mary Sutton, kept denying the allegations against her, even after she was swum and searched for teats. However, when her accuser told her that ‘her owne sonne Henry had revealed all … her heart misgave her, she confessed all, and acknowledged the Divell had not left her to that shame that is reward to such as follow him’.24 Nine-year-old Jennet Device was reported to have testified against her mother and other family members and ‘to discover all their Practises, Meetings, consultations, Murthers, Charmes, and Villanies’.25 The fact that Alizon Device was accused by her own children, Jennet and her older brother James, did not discredit their story, implied Thomas Potts. On the contrary, ‘although she were their owne naturall mother, yet they did not spare to accuse her of every particular fact’.26 Potts regarded the children’s testimony, coming from the mouths of infants, as God-sent help:
such a young witnesse prepeared and instructed to give Evidence against them, that it must be an Act of GOD that must be the means to discover their Practises and Murthers, and by an infant.27
Children’s testimony was sometimes presented as the breaking point of their mother. Alizon Device was reported to have made ‘a very liberall and voluntarie Confession as hereafter shall be given in evidence against her, upon her Arraignment and Triall’, after learning that her daughter Jennet had implicated her.28 It is possible that in some cases the pamphleteers manipulated the dates to avoid the impression of exploiting the children to obtain the mothers’ confessions. Sometimes the children’s testimony was presented as if they confirmed the mothers’ confessions, whereas, in fact, the children were investigated first.29 In some cases the children’s ages are not mentioned.30 The reason, perhaps, might be the pamphleteer’s wish to undermine suspicions of manipulation.
Michael Dalton’s popular manual for justices reflected a legal descriptive approach to the subject of children’s testimony. Dalton acknowledged the difficulty of proving witchcraft and provided a list of evidentiary rules based on the Lancaster precedent from 1612, including ‘The examination and confession of the children, or seruants of the Witch’.31 Giving references to the relevant pages in Potts’ pamphlet, Dalton cited a few examples of children who deposed against their mothers and later testified in court.32 Dalton’s approach was descriptive, as he aimed to portray the existing legal norm as it was reflected in contemporary cases. However, by uncritically relying on precedents, Dalton supported modification of the general rule against the testimony of young children. By describing the practice of children’s testimony in witchcraft cases as the norm, Dalton advocated setting a lower standard of proof for the crime.
In his Guide to Grand-Iury Men, Bernard listed possible concerns regarding the credibility of types of witnesses. Children belonged to the category of less credible witnesses:
some are fearfull, superstitious, or children, or old silly persons, whose testimonies are to be heard, but not easily credited, as being persons in such a case as this is, very much subiect to mistaking.33
Like Scot, Bernard associated credit with the category the witness belonged to. Unlike Dalton, he expressed uneasiness about the truthfulness of children.
A later legal author relied on both Potts and Bernard when he provided signs for the discovery of witches, including:
the examination and confession of the Children (able and fit to answer) or servants of the Witch; especially concerning the first six observations of the party suspected; Her threatnings and cursings of the sick party; her enquiring after the sick party; her boasting or rejoycing at the sick parties trouble: Also whether they have seen her call upon, speak to, or feed any Spirit, or such like; or have heard her foretell of this mishap, or speak of her power to hurt, or of her transportation to this or that place, &c.34
Children’s testimony was permitted as long as they were ‘able and fit to answer’. The topics they could testify to were based on precedents. Their testimony, an aberration of normal procedure, was to be accepted, but with caution.
We can speculate whether the lack of competency limitations in witchcraft cases helped to develop criteria of credibility. If everyone was competent to give evidence, it was necessary to find a method to assess the value of each testimony. The following chapter discusses the concept of credibility as it was developed in the second half of the seventeenth century.
Accomplices were another category of problematic witnesses. Many witchcraft suspects implicated others, saying they had bodily marks and imps in various shapes that they fed and suckled and whose malicious services they enjoyed.35 The multiple-defendants’ cases inspired by Matthew Hopkins involved multiple allegations of witchcraft by the defendants against each other.36 Suspects in a serious crime who incriminate others through their own confessions might have other interests than the uncovering of the truth.
As objections began to be raised as to credit and weight, the motives and character of accomplices started to be taken into account. At the beginning of the seventeenth century, confessions of accomplices were admitted against each other and even considered as ‘specially cogent evidence’.37 Accomplices’ testimony clearly contained a potential bias, especially when it consequently benefited the accomplice by granting immunity. By the middle of the eighteenth century, the testimony of an accomplice who turned king’s evidence and incriminated his confederate for immunity or some other benefit could be admitted without corroboration.38 In contrast, an accomplice’s testimony in favour of the prisoner bore little weight.39
The crime of witchcraft was committed in secrecy, and the knowledge of the accomplices was invaluable. In his Daemonologie, James I included the rhetorical question: ‘For who but Witches can be prooues, and so witnesses of the doings of Witches.’40 The confederates were the best source of information. This logic led both lawyers and clergymen to support the use of accomplices’ testimony.
From a legal point of view, the possibility of proving a case by the accomplice’s testimony was legitimate and even applauded. Thomas Potts explained that accomplices’ testimony was especially probative in cases of witchcraft.
Who but Witches can be proofes, and so witnesses of the doings of Witches? Since all their Meetings, Conspiracies, Practises, and Murthers, are the workes of Darknesse.41
Potts, who quoted James I without due reference, found his logic convincing. Witchcraft, by nature, was a crime without witnesses. The accomplice was the partner who shared the secret crime and was an excellent source of information.
Bernard, in his Guide to Grand-Iury Men, asserted that the testimony of a fellow witch proved the crime. Interestingly, Bernard believed that the statements of cunning men (the ‘good witches’) ought not to be admitted. Supposedly, the cunning men possessed supernatural powers and could recognize other witches. Bernard did not articulate an explanation to his distinction between good and bad witches. We may speculate that the difference between accomplices and unrelated witches was their possession or lack of direct knowledge of the crime. The accomplices reported to have actually seen the witches:
with their spirits, or that they haue receiued spirits from them; that they can tell, when they vsed Witcherie tricks to do harme; or that they told to do harme; or that they had done; or that they can shew the marke vpon them; or that they haue bin together in their meetings, and such like, as the Lancashire Witches gaue testimony one against another of these things.42
It was not the supernatural capabilities of the accomplices that rendered their testimony sufficient for conviction, but their direct observation of the crime. Bernard also relied on Potts’ account as a precedent of legitimate evidentiary use of accomplices’ testimony.
A later legal scholar relied on both Potts and Bernard when he included among the signs for discovery of witches:
the testimony of other Witches, confessing their own Witchcrafts, and witnessing against the suspected, and that they have Spirits, or Markes; that they have been at their meetings: that they have told them what harme they have done, &c.43
The testimony of the accomplice was found to be valuable evidence, not only by legal scholars, but also by several clergymen. Cooper, a cleric who advocated severe measures against witches, held:
The Accusaion of a fellow Witch, either at examination, or at the day of death is not to bee neglected, because now Authoritie hauing seized on hir, though she may lie before she be discouered, yet now hauing confessed herselfe, she is an Instrument of the Lords Iustice, to satisfie Authoritie, and cleare the innocent, by speaking truth, &c. (though otherwise shee would not) to accuse the delinquent.44
For Cooper, the accomplice’s testimony was a divine miracle in which the suspect was transformed from an instrument of darkness into an instrument of divine justice leading to the discovery of truth. Gaule, the vicar of Great Staughton, in Huntingdonshire, agreed that testimony from others, ‘whom other notorious Witches have impeached to be as ill as themselves’, was an ‘infallible and certain sign of witchcraft’.45
Their confidence was not shared by all. Gifford rejected the statements of those who were already convicted and about to be executed. Even if such statements were sincerely made, the convicted witches might be ‘vtterly deluded’, that is, under the illusion of having received imps from other women.46 The devil could ‘set a strong fantasie in the mind that is oppressed with melancholie, that such or such a matter was, which indeéd was neuer so’.47 A century later, the prominent New England Puritan clergyman Increase Mather expressed similar ideas in an address given in the aftermath of the Salem trials, and later published as a book in Boston and then in London. Mather argued that suspects who confessed and implicated others in witchcraft were ‘not such credible Witnesses, as in a Case of Life and Death is to be desired: It is beyound dispute, that the Devil makes his Witches to dream strange things of themselves and others which are not so.’48 Mather’s disapproval of the accomplice testimony was part of his general criticism of the evidentiary techniques used in witchcraft trials, most famously spectral evidence. The unreliability of accomplices’ testimony derived from its spectral nature.
Perkins, the leading English Puritan thinker, ranked accomplices’ testimony as a lower-rate proof, a ‘presumption’ sufficient to warrant an examination, but not enough to support conviction.49 The testimony of ‘a Wisard Witch or cunning Man’ (who were not accomplices) he similarly considered to be of ‘lesse sufficient proofe’. Perkins did not want the jurors to rely on a witness who might actually be the devil who ‘comes in the likenesse of some knowne man, & tells them the person in question is indeed a Witch, and offers with all to confirme the same by othe’. Under no circumstances, Perkins made clear, could the testimony of a wizard serve as a basis for conviction.50 Theoretically, similar logic could apply to disqualifying the testimony of the accomplice. The devil might implicate innocent people through allegations made by another witch. However, Perkins did not explain if this was the reason that he considered the accomplice’s testimony as a mere presumption, which was insufficient for conviction.
Some legal thinkers also expressed concern about the use of accomplices’ testimony. Filmer, who opposed Perkins’ evidentiary methods, simply cited Perkins to demonstrate that the testimony of ‘a fellow Witch’ was not good evidence.51 Filmer, although he himself disapproved of accomplices’ testimony, depicted Perkins as opposing James VI & I, who regarded accomplices’ testimony as valuable proof.52
A major concern missing from this debate is the emerging practice of crown witnesses (accomplices who became prosecution witnesses in return for receiving benefits). Judge Hale (d. 1676) stated that:
it would be hard to take away the life of any person upon such a witness, that swears to save his own, and yet confesseth himself guilty of so great a crime, unless there be also very considerable circumstances, which may give the greater credit to what he swears.53
Hale’s opinion did not prevail, and the practice of prisoners’ avoiding prosecution or pleading to a lesser or clergiable charge continued to develop, eventually resulting in what Langbein called ‘the corroboration rule’ or ‘the accomplice rule’.54 Under this rule, as it was articulated in the 1740s, an accomplice’s testimony was to be excluded unless it was corroborated by other evidence. The rule was mollified in the 1780s, when jurors were requested only to be cautious.55 The rationale is obvious – the incentive of the accomplice to lie in order to save his neck was an invitation to perjury. Although some evidence suggests that the plea-bargaining practice started to appear around 1575,56 there is no known evidence that it was employed in any of the witch trials. Some trials ended in the conviction of the defendant of lesser and non-capital charges, but there was no indication that it was the result of plea bargaining (although the possibility of such a practice cannot be ruled out). Witchcraft was, under statute, a non-clergiable offence, which limited the bargaining possibilities.
But even without the option of plea bargaining, suspects in witchcraft cases had motives for falsely implicating others. After Ursley Kemp had confessed on 20 February 1582, she made another confession the next day, in which she tried to shift much of the blame to another woman, Ales Newman. Kemp is reported to have admitted that she had asked Newman to send her imp to a woman with whom she had quarreled. This version cleared her from managing a direct contact with the devil. In addition, Kemp implicated two other women.57
Other motivations for falsely implicating others could be the wish of a suspect for revenge or perhaps the attempt of a desperate suspect to expose the unbearable ease with which innocent people were executed for crimes in which innocence was impossible to prove. Such may be the case of 1645 in which Anne Leech, being interrogated, implicated three other women. Two of these women, Elizabeth Clarke and Elizabeth Gooding, whom she blamed for having imps and doing harm, were stereotypical witchcraft suspects, and her charges against them did not raise doubts. However, the accusations of Leech against the third woman angered the pamphleteer significantly:
VVHereas there was a Booke (of the Essex Witches) came forth in print, wherein on Mrs Wayt a Ministers wife was nominiated for one, but it was a palpable mistake, for it is very well knowne that she is a gentlewoman of a very godly and religious life, and a very good conuersations: and this was set on purpose to vindicate her: and lay the fault on the Author, in whom it was a great mistake.58
When it came to an accusation against the minister’s wife, the author had no doubt it was false. The dangers inherent in the testimony of a fellow witch might have been more visible when the charges were directed against a reputed person.
The value of statements made by suspects against confederates was under debate. The dispute cannot be clearly drawn along lines of professional affiliation (though it seems that, in general, lawyers were more supportive of the use of accomplices’ testimony, whereas clerics diverged in their opinions). In practice, however, accusations of suspects against others were often made during the criminal proceeding.
Expert witnesses formed another distinct category in dealing with proof of witchcraft. The involvement of experts in witch trials generated a debate in which three main elite discourses, the medical, the legal and the theological, interacted, sometimes competing with, and sometimes complementing, one another.
Overall, the physicians were most successful in positioning themselves as experts in witch trials. Some cases even turned into a battle of experts in which physicians testified for both parties. In some cases a physician was summoned by the bench. Educated physicians strove to distinguish themselves from other contemporary medical practitioners, who included apothecaries, barber-surgeons, uroscopists, herbalists, empirics, midwives, astrologers, tooth-drawers, lithotomists and cunning men and women.59 Physicians used the arena of the witch trials to advance their professional superiority, although their prominent role in witch trials was not obvious. Presumably, clergymen were the best experts on the subject of witches and devils. Yet, the clerics were unsuccessful in establishing for themselves a pivotal position in the process of criminal adjudication.
Were doctors, judges or divines best suited to determine whether an illness was caused by witchcraft? The physicians’ claim to exclusive expertise was received with ambivalence. Some supported the use of expert witnesses – defenders and accusers alike, as well as the judiciary, used experts to reach a determination. Nonetheless, the claim of exclusive ability to diagnose medical symptoms and determine whether they were natural was sometimes disapproved of by judges. They did not want to delegate to the doctors the ultimate power to determine facts and preferred that the testimony of the expert witnesses be just another piece of evidence to be accepted or rejected at the discretion of the fact-finders.
Physicians’ advice was often sought in the preliminary stages of the case, well before it reached court, because those who fell ill sought diagnosis and medical help.60 The magistrate investigating the case sometimes procured the expert opinion of a physician on whether the symptoms were natural or unnatural. The expert’s opinion could also be obtained at the trial itself, either at the request of the bench or of one of the parties. The expert witness at the trial was not necessarily the physician who initially treated the patient. The first known witchcraft case that turned into a battle of experts was in 1602, when Elizabeth Jackson was indicted for bewitching Mary Glover.61 Both the prosecution and defence tried to prove their arguments by summoning medical experts to testify on their behalf. Another notable case was the trial against Rose Cullender and Amy Denny, where the court, presided over by Sir Matthew Hale, used Sir Thomas Browne as an expert witness. Through their role as experts in witchcraft cases, physicians sought not only to determine whether a certain symptom originated from witchcraft, but also to define the boundaries, identity and standards of the medical profession.
It is unclear whether defendants were routinely allowed to call expert witnesses on their behalf. By the mid-seventeenth century, it was common to allow defence witnesses, although this was not yet a statutory right.62 An account of the trial of Joan Peterson, in 1652, who was indicted for killing an 80-year-old woman by witchcraft, listed a few medical experts who testified in her defence: