II. GENERAL INTRODUCTION
BERNARD ROSENTHAL
In popular and in academic culture, the Salem witch trials have been absorbed into American imagina- tion as few other events have. If assumptions as to what actually happened during that episode vary wildly between rigorous academic scrutiny and a persistently evoked cultural memory of a magic past that never was where witches were persecuted for their beliefs, the most basic beginning point for understanding the event must be the records that survive. Records of the Salem Witch-Hunt includes or references all known extant manuscripts and all known published documents of manuscripts no longer extant. The manuscripts have been transcribed from the original documents using editorial principles described later in the introductory material of this book. The principle of organization has been to present these newly transcribed documents, and previously published ones, in an order reflecting as accurately as possible their chronological occurrence. The goal has been to create a narrative through the documents as to what actually happened, and to identify when possible the people who recorded those documents.
Of the Salem witch trials, much has been written, and disagreements have inevitably been many. This edition will not settle those differences, but if it succeeds it will give the reader the most comprehensive, most carefully and consistently transcribed record ever produced of the Salem witch trials, as well as a chronological ordering of the documents. With this material before them, people researching these records will be able to do so with greater confidence that they can draw their conclusions from reliable data. Those doing very extensive research will, of course, recheck selected manuscripts themselves, but the edition should significantly ease the task of getting accessible, reliable information. At least that is the hope that the editors bring to this edition. With that goal in mind, exploring some key issues of the Salem witch trials, as well as a consideration of how this subject of witchcraft had been addressed previously in Massachusetts Bay and in England, may offer a useful beginning for visiting, or revisiting, some of the issues, historical and linguistic, that people may want to consider when examining the documents, their order, and the notes seeking to highlight pertinent matters.
On most outlines of the event there has been little disagreement. A consensus exists on the following: Sometime in the winter of 1692 Betty Parris, the nine-year-old daughter of Samuel Parris, minister of Salem Village, began behaving in very strange ways as did Parris’s niece, eleven-year-old Abigail Williams, who lived with the family.1 The ages of the two girls are approximate given the casualness with which ages are identified in the surviving records. But they are close enough to be used confidently.
The episode in the Parris household took place in a culture where complaints of witchcraft were not unknown, and where a major case had occurred in 1688, leading to the execution of a woman named Glover. She was accused of afflicting four children between the ages of five and thirteen from a Goodwin family living in Boston who appear to have exhibited behaviors similar to those that later occurred in the Parris household.2 The story of the Goodwin children is best known through Cotton Mather’s Memorable Providences (1689). Mather was instrumental in containing the episode, which claimed no victims other than Glover, convicted of witchcraft and executed. Other precedents occurred throughout New England history, but with a total of executions fewer than the total that occurred in 1692.3 In addition to these, the story of witch trials in Sweden that occurred in 1669–1670 was known in Massachusetts Bay, discussed by Mather in The Wonders of the Invisible World (1693) and referenced by Robert Calef in More Wonders of the Invisible World (1700).4 Sir William Phips, in his account of first arriving in Massachusetts as the new Governor, described the witchcraft matters he encountered in a language closer to what was in the literature about Sweden, which he references, than to what was happening in Massachusetts Bay in May 1692.5
More immediately, two other incidents, not widely noted, if at all, may have offered part of the particular background for more recent witchcraft episodes in Massachusetts Bay. One concerned the case of a woman named Martha Sparks, from Chelmsford in Middlesex County, whose case may have more significance than has been realized. She was accused of witchcraft in 1691 and sent to prison in Boston by Thomas Danforth on October 28, 1691, where she remained till December 6, 1692, two days after a recognizance was granted.6 So even before the first legal actions leading to the Salem witch trials occurred, this woman was in prison on witchcraft charges and remained there till after the last executions. Although she was never brought to trial, it seems likely that the presence of a “witch” in prison would have been broadly known in the region.
Another incident immediately preceding the claims coming out of the Parris household concerned a girl named Mary Knowlton. In March 1692, Thomas Knowlton of Ipswich, in testimony against Rachel Clinton, also of Ipswich, referred to the fits his daughter was having in late December 1691 or early January 1692, probably before the Salem Village behaviors, fits attributed to witchcraft committed by Rachel Clinton. The imprisonment of Martha Sparks in 1691 and the claims of Thomas Knowlton give good reason to suspect that the Salem Village behavior of Betty Parris and Abigail Williams in December 1691 or January 1692 had more recent, known antecedents than the Glover case.7 One can only speculate as to what Betty and Abigail knew about Sparks and Clinton, but this recent “witchcraft” should not be overlooked in considering the origins of the 1692 claims.
In response to the behavior of Betty and Abigail, and the conclusion of a physician that the afflicted were bewitched, Parris was joined by other ministers in a day of prayer that did not stop the behavior of the children.8 The diagnosis of witchcraft in the context of beliefs at that time was not unreasonable. Indeed, even in 1692, when the Salem trials occurred, in neighboring Connecticut witchcraft claims were also raised, although unlike Massachusetts Bay, the cases in Connecticut were treated in the normal, judicious, New England way, with no executions in this instance resulting from the charges and with very few people actually being accused.9 People in New England generally believed in witchcraft, and those who may not have shared that belief had the sense to keep such views to themselves.10 And even as Parris and the other ministers prayed, the legal process of bringing witches to justice had already begun, with the first arrest warrants issued on February 29 and the first examinations occurring on March 1.11 On other occasions, as with Mather in the Goodwin case and Samuel Willard in the case of Elizabeth Knapp, Willard’s sixteen-year-old servant claiming fits, ministers had persisted with prayer and had succeeded in limiting the spread of accusations.12 This time, however, continuing prayer notwithstanding, something different happened, and the heart of understanding the causes of the Salem witch trials rests in finding out why those in power chose to depart from the New England tradition of not encouraging such charges.13
At the same time many other lines of inquiry remain open, primarily those relating to Salem Village quarrels. The paradigm for this line of inquiry was set by Charles W. Upham in his two-volume Salem Witchcraft, published in 1867. No book has dominated the direction of future scholarship as has Upham’s, whose attention to Salem Village issues generally set the course for future studies of the Salem witch trials. Scholars in the nineteenth century followed him, and the most influential scholarly study of the twentieth century relating to that episode, Salem Possessed: The Social Origins of Witchcraft by Paul Boyer and Stephen Nissenbaum, was keyed to Salem Village issues.14 Unlike Upham, Boyer and Nissenbaum did not dwell on moral outrage, but sought instead to assess, analyze, and show the explanatory value of social conflict within the community. Although they touched upon broader issues in Massachusetts Bay, their emphasis remained firmly in the Upham tradition of seeing the event as primarily one of Salem Village quarrels. Among various matters basic to their study was the role of Samuel Parris, the man seen in nineteenth-century scholarship as the major villain. Also basic to their study was the role of Thomas Putnam’s support for Parris in Salem Village quarrels.15
The first legal action that took the matter beyond Parris’s reliance on prayer appears on arrest warrants dated February 29. Justice of the Peace John Hathorne recorded two complaints of witchcraft against two local women, Sarah Good and Sarah Osburn, as well as against Tituba, the “Indian servant” (slave) of Parris. Both were on behalf of four men, one of whom, Thomas Putnam, would play a major role in the events that unfolded well beyond Salem Village. Although at the time, Massachusetts Bay was without a Charter – it would arrive in May – legal proceedings had not come to a halt, and the absence of a Charter did not prevent capital prosecutions.16 Consistent with traditional legal procedures, Hathorne ordered the arrest of the women for the purpose of examining them. Tituba confessed at this examination, thus giving credibility to witchcraft claims. According to Calef, Parris had beaten her into both confessing and accusing, but no independent confirmation of this survives.17
These arrests, however, came without the traditional requirement that bond be posted by the person lodging a complaint, a significant departure from English law. Other than this departure, the pattern of complaint, arrest warrant, examination, imprisonment, grand jury hearing, and trial that followed was consistent with English law.18 A likely but not certain source of legal procedure would be Michael Dalton’s Countrey Justice, 1618, which after various reprintings appeared again in 1690 in an edition that included changes made in 1689 with the ascension of William and Mary to the English throne.19 Whether Dalton was used directly or not, however, what followed legally in Massachusetts Bay usually remained consistent with the legal commentary with the major exception of the bond issue.20
However, the failure of John Hathorne and Jonathan Corwin, Justices of the Peace, to require bond in the first cases made charges of withcraft easier to lodge, and by the time the law began to be followed, as it eventually did, matters may have spread too far. It is surely speculative to suggest that had the law been followed the matter might have been contained, but it remains a speculation worth considering. The first instance of the law being followed by the magistrates occurred on March 29 at an Ipswich court in connection with the case of Rachel Clinton.21 However, this did not change the behavior of Hathorne and Corwin, and it was not until July 19, when they and Bartholomew Gedney were joined by Justice of the Peace John Higginson, that these justices joined Higginson in requiring bond before issuing arrest warrants.22 Higginson was the new variable, and from that date on, the law on posting bonds was followed in every case. But prior to July 19, with rare exceptions, charges of witchcraft made no demands on the person bringing the charge. It was a long stretch of legal violation that began on February 29 with complaints and arrest following, and without bond posted until July 19.
After the examination of the accused women, they were imprisoned, with one, Sarah Good, eventually going to trial in June 1692 and being executed in July of that year; another, Sarah Osburn, dying in prison on May 10, 1692, before a grand jury could address her case; and Tituba not being sent to trial by the grand jury that met on her case on May 9, 1693.23
Broadly speaking, two sets of witch trials occurred during 1692 and 1693. The first occurred under the Court of Oyer and Terminer, in 1692, where all the trials were held in Salem, although the majority of people involved were not from Salem.24 In 1692 a plurality of accusations and imprisonments came from cases in the Andover area, but many of these cases did not come to trial until 1693. The 1693 trials occurred under the Superior Court of Judicature at more than one location. The special court that heard the 1692 cases was established by the governor, William Phips, when he came to Massachusetts Bay with the new charter. The charter was published on May 16, and the Court of Oyer and Terminer was established on May 27.25 The court may not have been set up specifically to deal with the witchcraft cases – the word “witchcraft” does not appear in the authorizing document – although the people in prison on witchcraft charges certainly constituted the significant segment, cited in a letter by Phips of those “thronging of the Goals at this hot season of the year; there being no Judicatories or Courts of Justice yet Established.”26
The lack of courts did not mean that the initial examinations were considered illegal, since they were recognized by the Court of Oyer and Terminer as legitimate, and since much judicial business was conducted while the colony was without a Charter. Additionally, on June 15 Phips signed into law an act that kept all laws in force that had been made prior to the new Charter, as long as they did not violate English law.27 Subsequent commentary on this period carries debates over the legality of the Court of Oyer and Terminer, but they are not relevant to how the court was seen at the time. Similarly, judicial procedures that occurred while the colony was without a Charter seem not to have been challenged at the time. For example, on April 28 “Sa¯m Passanauton an Indian” was imprisoned and held for eight and a half weeks, and it is unlikely, given his release, that he was in jail on witchcraft charges.28 However, the arrival of Phips with a new Charter opened the way for reconstituting the entire judicial system so that it would be compatible with the new rule of law. Creating a Court of Oyer and Terminer represented an initial step, and none of the surviving records suggest that there was any controversy over its creation. Although the court functioned primarily in addressing “witchcraft” cases, it dealt with at least one other, as in the example Samuel Sewall gives in his Diary for the meeting of the Court of Oyer and Terminer in Boston on October 10, 1692, for a murder trial.29 Other names appear in the jail lists that have no verifiable connection to the witchcraft cases. Whether this means there was none, or whether one existed and the documents related to that connection do not survive, is a matter for further research. Other times a name appears where the person is almost certainly one caught up in the witchcraft accusations, but where almost nothing about the individual’s case appears. For example no arrest warrants or judicial procedures, except for jail accounts, survive in connection with Mary Cox, who was put in irons two days after the Court of Oyer and Terminer was constituted.30 She was in jail until November 22, 1692.31 Such cases make it extremely difficult to get an accurate count of the people caught in the witchcraft accusations.
As charges spread, before and after the first trial, so also did the number of those claiming affliction. They were primarily, but not exclusively, females in their teens or younger with charges almost always leading to imprisonment. Prisoners were placed in chains even before Phips arrived and redundantly ordered people accused of witchcraft to be chained. A primary source reveals only one instance where the accusers withdrew their accusation.32 Calef reports another where no primary source survives. In this case, a Boston man accused of witchcraft during the accusations at Andover sent “a Writ to Arrest those Accusers in a Thousand Pound Action for Defamation, with instructions to them, to inform themselves of the certainty of the proof.… ”33 If Calef can be trusted, this episode is significant in supporting the link between young accusers and adult supporters. It certainly seems likely that a withdrawn accusation based on fear of financial punishment would have had its origin in the financial fears of adult supporters who had some money to lose. It is reasonable to speculate that one or more adult supporters told the accusers to back off in this instance. This event, however, was a departure from the basic pattern of accusation, complaint, arrest, and examination. Among other departures was an accusation against Reverend Samuel Willard by one of the accusers during a court proceeding. She “was sent out of the Court, and it was told about she was mistaken in the person.”34
After the accused were imprisoned their fate varied. Those who did not confess usually had indictments presented against them to a grand jury. As the 1692 trials were close to ending, indictments were also drawn against confessors, a few of whom were eventually condemned, but none executed.35 Confessors were indicted for covenanting with the Devil. Although all witchcraft implied such a covenant, those who did not confess were charged primarily for sending their spectres to harm the accusers on the day of the examination, and not specifically for the affliction that precipitated the accusation. There was a good reason for this, since such “tormenting” of the accusers at the examinations could be seen by responsible adult witnesses, of which two were required to support the claim.36 The spectres appeared visible only to the “afflicted.” As accusations spread, more and more people confessed, perhaps to avoid execution, perhaps from family pressure on the same or other grounds. For a compelling description of the pressures to confess, see the declaration of Mary Osgood and others as they recanted their confessions.37 The great majority of confessors such as Mary Osgood had their cases addressed by grand juries in 1693 by the Superior Court of Judicature after the Court of Oyer and Terminer that tried the 1692 cases had been dissolved. Under the 1693 Court evidence of spectral affliction remained embedded in various cases, but no longer remained as central to the outcome of those cases. It had returned to its traditional place as a “presumption,” and the line to a “proof” was no longer crossed. Grand juries overwhelmingly rejected indictments presented to them. Only three cases in 1693 led to guilty verdicts and condemnation – all three, confessors. A semblance of normality had returned. These people, Elizabeth Johnson, Jr., Mary Post, and Sarah Wardwell, received reprieves, however, from Governor Phips. Overwhelmingly, in 1693 the majority were found not guilty and released subject to their paying jail fees.
Starting in October 1692, for the first time in the episode some of the imprisoned were released on bail, “Recognizances,” while they awaited their trial. Many remained in prison. After the exonerations in January and February 1693, most prisoners were released, but some were not, since they did not have the money to pay their jail fees. They remained imprisoned under very harsh conditions, but it is impossible to tell how many of those imprisoned did not survive the winter of 1693 as a result of those conditions. One woman, Lydia Dustin, is known to have died that winter, but there is no way to establish the cause, nor can we be certain that others did not also die in those months. In May 1693 the last cases were heard, and there were no more convictions.
What actually happened during the various stages of the legal procedures is in some cases pretty clearly understood, and in others in need of best guesses. That is, the records of the examinations are numerous enough so that one can read them and get a very good picture of what these examinations were like. Matters get murkier when cases come before the grand jury or go to trial. At the grand jury hearings, it is probable that the attorney general, first Thomas Newton and later Anthony Checkley, continuing into 1693, argued the cases against the accused by presenting depositions to which people swore, or perhaps by testimony without a prior deposition. The grand jury could support the indictment by endorsing it as a “true bill,” meaning it accepted the charge, or not endorsing it by returning an “ignoramus.” When the charge was accepted and the accused stood trial, the person had no legal counsel, other than from the Court. The procedure was very short, the grand jury procedure and the trial sometimes both occurring on the same day, and sometimes with more than one person on the same day being brought before the grand jury and the trial jury. This was not a reflection of “witchcraft hysteria” but was instead consistent with English tradition in trial cases.38
Documents used at trials appear to have been more selective than those used at the grand jury considerations, with heavier reliance on the strongest supporting depositions or testimony. Although trials were not limited to repeating depositions or testimony about what happened at the day of the examination, heavy use was made of that. Also, both during the grand jury hearings and during the trials, the “afflicted” were present and behaving as at the examinations. Thus, grand jurors and trial jurors believing that the “afflictions” were not counterfeit – fraudulent – would have confirming evidence. The “afflictions,” as at the examinations, centered heavily on charges that the spectres of the accused were assaulting the “afflicted,” spectres visible only to them. Cotton Mather is ambiguous as to whether the “touch test,” whereby the “afflicted” were brought out of a fit by the touch of an accused person, was used at trial or simply referenced there, but Brattle is unambiguous as to the presence of the “afflicted” at trial, and it is difficult to believe that their behavior there was different from what it was at the examinations. In the “touch test,” the accused would be asked to touch one of the “afflicted,” and people could see that it worked when the touch brought the person out of the fit.39 More confirming as to the behavior of the “afflicted” at the trials, Cotton Mather writes, in connection with the trial of Bridget Bishop, “There was little occasion to prove the Witchcraft, it being evident and notorious to all beholders.” He also tells the story of how Susannah Shelden at the trial of Martha Carrier “in open Court had her hands Unaccountably ty’d together with a Wheel-band, so fast that without cutting, it could not be loosed; It was done by a Spectre.”40 Juries had to choose between witchcraft and counterfeiting. As at the grand jury, the defendant could not have legal counsel, nor could the person have anyone give sworn, supporting testimony. Unsworn supporting testimony was permitted and sometimes given. At the end of the trial, the jury probably received a charge from Stoughton as to how it should consider the case. An account of the charge of Matthew Hale in a 1662 witchcraft case in England may offer an example. According to the account, Hale informed the jury that witchcraft was a reality and gave what appear to be neutral instructions to the jurors. They were to decide whether the accusers were bewitched, and they were to decide whether the accused had bewitched them if they were.41 There is little reason to doubt that the Salem jurors received comparable instructions.
After the accused was found guilty, normally a period of a few days elapsed before the court pronounced sentence.42 The punishment for witchcraft was death, as indicated in Exodus 22:18 (“Thou shalt not suffer a witch to live”), and when the sentence came, that was the punishment ordered and carried out, except in the cases of the condemned confessors, or where the condemned was pregnant, as in the cases of Abigail Faulkner Sr. and Elizabeth Procter.43 One woman, Dorcas Hoar, who had maintained her innocence and had not been a confessor, was given a reprieve the day before her scheduled execution after finally confessing, leading to a successful appeal by four ministers that her execution be delayed.44 While it seems clear that the prosecutions in the Salem witch trials were pursued with rules of evidence that differed from the cautionary advice of the authorities most likely to have been consulted, such as Richard Bernard, John Gaule, and William Perkins, the reasons for that behavior bear continuing examination.
At the same time, the legal records show a scrupulous attention to following systematic, orderly procedures even though the general legal response to witchcraft charges was inconsistent with opinions of learned witchcraft authorities, as discussed below. Although popular images are those of a society in the grip of “hysteria,” there is nothing in the judicial attention to order and detail to suggest the legal authorities behaved that way. Certainly there were disruptions in the court by the “afflicted,” but these disturbances did not change the orderly, bureaucratic handling of cases. That the court partly failed to follow advice from the main authorities on discovery of witchcraft may indicate bad judgment or other motives but does not demonstrate a judicial system out of control, or a society submerged in a state of panic. Even the failure in the early stages to require bond for prosecution was eventually remedied. Indeed, however much “witchcraft” at Salem has captured popular imagination, it is clear from reading the minutes of the Governor and Council that while the issue remained one of concern to the authorities, it was not in any way the primary one. Government officials were worrying about Indian wars, naval and trade issues, matters of taxation, the killing of wolves, and a variety of other matters that took up far more of their recorded time than the witchcraft issue. On June 8, two days before the first execution in 1692 for witchcraft, that of Bridget Bishop, nine acts were passed by the general court, none of which had any connection to witchcraft.45 Reading the diary of Samuel Sewall is instructive as one notes how small a part the witch trials play in his record. Such a conclusion does not reject the idea of anxiety, or in some instances panic, within the regional community, especially among those not in power and those threatened. Surely such responses were inevitable. Yet the anxiety and panic were not so pervasive that people were afraid to sign their names to petitions in support of accused people, such as Mary Bradbury, Rebecca Nurse, and John and Elizabeth Procter, and it is well to keep in mind that there is not a single instance of a person signing a petition in behalf of these people who was subsequently arrested.46
Among those usually not seriously threatened were people in powerful positions. Such people found ways to escape, as did Mary Bradbury after she was condemned, and they were safe once they reached other jurisdictions. Thomas Brattle points clearly to the preferential treatment of the accused Hezekiah Usher and of the failure of the authorities to pursue high-status prisoners who had escaped.47 Ann Dolliver, though arrested and examined on June 6, disappears from the surviving records after that. As the sister of John Higginson Jr., a magistrate heavily involved in witchcraft cases, and the daughter of Reverend John Higginson, she appears to have had protection from further judicial action. Philip English, although at one point hiding in dirty laundry, and eventually arrested, nevertheless managed to get out of the province of Massachusetts Bay.48 High status did not guarantee protection from the judicial system in the witch trials, but it significantly improved one’s chances, and nobody in this category was executed unless one considers George Burroughs to be a high-status person. Puritans were not very different from later Americans who would find judicial outcomes often having some relation to economic or social status. Even some influence peddling may have occurred, as in the use of Mary Gedney’s tavern to put up witnesses and jurors, for which she was reimbursed. As she was Justice of the Peace Bartholomew Gedney’s sister-in-law, it seems likely that some business came her way through the good offices of the man helping to generate the numbers of jurors and witnesses needing accommodation.49 Nothing in this should be construed as meaning that the Salem witch trials were about making money, but rather that, then as now, having good contacts or being in a powerful position made it easier to profit and to achieve safety.50
As a result of the episode, nineteen people were hanged, one was pressed to death, and an indeterminate number, five known, died while in prison.51 What happened to arrested people where further records about them do not survive often remains unknown. Estimates as to the number of people accused and arrested vary, with most scholars traditionally calculating about 150. These are not wild guesses, but estimates based on information from surviving documents. The problem is that various documents do not survive, so the actual number cannot be indicated with certainty as can be done with the number of people judicially executed. A contemporary document indicates a number of 200 accused people, but the accuracy of this remains unknown.52
Agreement exists that the handling of the charges by the judicial authorities strayed radically from traditional colonial ways of handling witchcraft cases and that this unusual behavior by the authorities explains the spreading of accusations, imprisonments, and executions. Similar agreement exists that the contentious issue of “spectral evidence” was allowed by the court until it was disbanded in October of that year, although the trials continued into 1693, not ending until May of that year, these continuing trials conducted without relying on “spectral evidence.” While the subject of “spectral evidence” was controversial in 1692, there is almost no controversy, then or now, that this issue was central to what happened in the Salem witch trials. Spectral evidence was simply acceptance of the claim that a person’s “spectre” – some spirit willingly sent out from the person – harmed people through the agency of the Devil. While convictions were not exclusively based on spectral evidence, no conviction in 1692 or 1693 occurred without it.53
The controversy on this issue in 1692 centered mainly on the question of whether God would allow the Devil to give a human the power to do harm in such a way, or whether it was in fact the Devil afflicting in someone else’s appearance. That the Devil could do this was agreed upon, based upon the interpretation of the “Witch of Endor” story in 1 Samuel 28, where a woman raises the appearance of Samuel at Saul’s request. No disagreement seemed to exist over the proposition that in fact Samuel was not raised, but that it was the Devil who came in the appearance of Samuel. Those defending the proceedings insisted that convictions were based on evidence other than spectral evidence. Those unhappy with the proceedings felt that too much weight was given to this kind of evidence. Accused people often cited in their defense, without success, this Biblical account, and when in 1693 spectral evidence had generally been discredited, the number of convictions dropped dramatically. When Increase Mather wrote his Cases of Conscience, which helped bring the 1692 cases to an end, his title began with “Cases of Conscience Concerning Evil Spirits Personating Men.”54
The context of Mather’s essay, however, spread well beyond the Salem witch trials in that his writing was part of the broader debate in England over the reality of witchcraft, which was vigorously argued by Richard Baxter in his The Certainty of the Worlds of Spirits (1691) in what was becoming more and more a losing fight against those doubting the reality of witchcraft.55 Baxter’s book is not useful for clarifying judicial procedures for discovering witches, but it importantly links New England to the debate in the motherland. Thus, in making his case, one filled with examples of witchcraft, at one point Baxter turns across the ocean to support his views: “They that will read Mr. Increase Mathers Book, and especially his Sons, Mr. Cotton Mathers Book of the Witchcrafts in New-England, may see enough to Silence any Incredulity that pretendeth to be Rational.”56 Although Increase Mather believed in witchcraft, this did not deter him from urging better methods for finding witches, and eventually, he and others objected on theological grounds to the use of spectral evidence, a view that gained in ascendancy. They did not argue against the existence of witches, but they urged better methods of proof and succeeded in diminishing the centrality of spectral evidence from the trials held after the Court of Oyer and Terminer ended. This centrality of spectral evidence to the trial cases of 1692 is implicitly made clear by the House of Representatives on July 20, 1703, where a Bill was ordered drawn up banning the use of spectral evidence for taking away a person’s life or even a person’s good name.57
After the trials ended, reactions to them soon developed, leading eventually to a broad consensus that something had gone terribly wrong. In the years that followed, specific legal steps were taken to address the injustice, including compensation for families of those executed or condemned but not executed, although no compensation came to those who had been imprisoned but not condemned or executed. Indeed, doubts about the correctness of what had been done were so great that even condemned confessors received compensation, as for example Abigail Hobbs. In September 1710, William Hobbs, her father, asked for 20 pounds compensation so “that our names may be Repayered.”58 In December 1711 she was awarded 10 pounds compensation.59 Other condemned confessors also received compensation then. By 1711, if the government actions reflect the broader view in Massachusetts Bay, as they probably do, the confessions of 1692 had simply been discredited, as had most of the people whose testimony in part led to the death of others. It all had been a grand delusion, and nobody was legally guilty of anything – neither accusers nor accused, nor sheriffs, nor judges, nor anybody.60