The Salem Witch Trials Digital Archive: How and Why

The Salem witch trials spread across twenty-four different communities in the eastern Massachusetts Bay Colony in 1692. At least 155 people were arrested and jailed, 30 were convicted, and 19 were executed by hanging. One man was crushed to death with stones, and 5 more people died in prison. This was, compared to larger European witch-hunt,s a small scale tragedy. It lasted seven months, from March to mid-September 1692 and led to the collapse of the Puritan government in Massachusetts. It was the worst witch-hunt in American history. 

In 1999, I was a faculty member in the department of Religious Studies at the University of Virginia, and I created the web-based Salem Witch Trials Documentary Archive  for the purpose of teaching a seminar for about twenty Religious Studies majors. The seminar, using primary sources, would enable Religious Studies majors to not only learn about a new subject but also how to evaluate different methodologies that scholars use to interpret the primary sources, for example, the methods of sociology, theology, history, psychology, and comparative religions. I chose the Salem episode purposely as the subject of the new seminar without knowing anything about it. I wanted to be fresh to the subject without any preconceived ideas. I was also unaware that I had a personal connection to it. I will write more about that below.  

At this time, I also joined forces with Professor Bernard Rosenthal of the English Department at SUNY Binghamton, N.Y. who was looking for a partner to create a digital archive of Salem court records in order to assist an international team of scholars who were going to create a new transcription of the 1692-93 court records for a new print publication. Rosenthal’s scholarly team, which included Professor Meija Kytö of the English Department at Uppsala University, needed access to the out-of-print book of transcriptions titled the The Salem Witchcraft Papers and to the images of the original manuscripts of the court records.  Later, thanks to several substantial research grants, I was able to develop the archive further by adding various search tools, several out-of-copyright books, and dozens of previously unlocated court records for the purposes of higher-level scholarly research.  My aim for the site was that it should be used by advanced scholars as well as high school and university teachers and students, and by the general public, so the site’s design and functionality should be as obvious and easy to navigate as possible.  

I knew my students had heard of the Salem tragedy, largely because of the popularity of playwright Arthur Miller’s famous play The Crucible, which is widely taught in American high schools. In the course of teaching the seminar, I discovered that four of my ancestors were directly involved as defenders of one person accused of witchcraft. My family name was then spelled Rea, and two married couples signed a substantial document in defense of the victimized 71-year-old Salem Village resident Rebecca Nurse. One frightened 12-year-old named Jemima Rea, whose parents had defended Rebecca Nurse, accused Nurse of bewitching her. Fortunately Jemima’s accusations (reported by her cousins) were not deemed worthy of legal attention, and they were not used by the prosecution. I will return to Rebecca Nurse’s fate at the end of the article. 

After learning of my family’s involvement, my goal was to enable as many voices as possible of the people involved in the trials to be heard: those of the accusers, defendants, their family members, husbands and wives, confessors, magistrates, jury men, ministers, sheriffs, constables, jail keepers and so on.  Altogether there are over 1,400 names of people in the court records. For this reason, the site also serves as an important genealogical resource, as all the names are searchable. 

  The print edition of the court records had contained many errors, dropped words and sentences, and it was also an incomplete collection.  Furthermore, nothing of the material was available online. Indeed, in the 1990s none of the vast historical archives in the Boston area had an online presence. If students were to critically evaluate different scholars’ approaches to the Salem episode, each student would need to consult the primary documents, that is, the original court records. The transcription team would use the existing transcriptions of the court records as a baseline for their ongoing work. 

At about this time, the University of Virginia were encouraging faculty members in the humanities departments to engage in digital scholarship. The simple solution to the problem above was to digitize the three out-of-print volumes of court records, with the publisher’s permission, and to put them online on their own web site at the University’s new Electronic Text Center, directed by David Seaman. I received a small grant for this purpose, and the three volumes of the print edition were keyboarded in India. The website retained the architectural format of the print edition. Besides creating the digital archive, my job in this collaborative effort, with the help of Margo Burns, Rosenthal’s project manager, was to go into the archives, which were mainly in the Boston area, and locate all the original manuscripts of the court records, scan them and put them online together with the transcriptions of The Salem Witchcraft Papers. In the process Burns and I located over 70 previously unidentified documents.  https ://

The transcripts in the print edition had been organized alphabetically according to the names of the accused individuals, each person having their own case file of court records. Following English criminal law (the trials were under the authority of the state, not the church), the documents reflected all the stages of the legal process: complaints, arrest warrants, summonses, preliminary hearings, grand jury hearings, indictments, trials, sentences. The records of the trials of the 30 individuals who were found guilty in 1692 are missing, – Some believe that the reason for this is that the authorities wanted to protect the reputations of the various judges.  Thankfully about 90 percent of the other records which were used in the trials survive.  Altogether, there were 140 case files of the individuals involved. There were also files of other related legal proceedings from the 1692-93 proceedings extending through 1750. All the names in the files had also been cross referenced. 

Further development 

In 2000 I was appointed a Research Fellow of the University’s new Institute for Advanced Technology in the Humanities at the University, then headed by John Unsworth, enabling the project to be developed further.  

One of the problems for students and researchers was that about half the case files were undated and hence not retrievable by date, which is critical for historical study.  Nor were the names of the individuals searchable across all the case files. This being the late 17th century, many of the 1,400 personal names were not spelled consistently.  The solution was to create a standardized index of searchable name tags and date tags. Another necessity was the creation of a word search tool, a key feature, linked to the date tags, which would enable the tracing of certain concepts and procedures across time and make it possible to follow key concepts through the course of court activities, and to determine which themes had become the focus of the court’s attention. Finding patterns in this large number of documents would be one of its primary functions, patterns that are chronological and reveal the role certain people played over time, or patterns in the legal process. For example, all those who falsely confessed to acts of witchcraft, to save themselves, were required to estimate the size of the threat to the village ministry since it was believed that the Salem Village church was the focus of Satan’s attack. According to one informant, “Satan’s design was to set up his own worship, abolish all the churches in the land, to fall next upon Salem and so to go through the country.” Satan’s aim was to defeat Christian worship and replace it with Satanism.  Indeed, a search of the records turned up the fact that confessors gradually increased the number of spectral witches they “saw” holding Satanic rituals opposite the minister’s house in order to reinforce their confessions, which would save them from trial. Key word searches turned up dozens of reports of these Satanic rituals next to the minister’s house in Salem Village, which drove the court’s witch-hunt efforts relentlessly forward. For the first time, the allegedly growing size of the threat and the lengthy legal process could be documented over time and could identify the people who were responsible for it. 

It was also necessary to digitize several extremely rare 17th century books that dealt with the witch trials which were held in the University library’s Special Collections. We digitized well-known 19th century works, and a most valuable early 20th century abridged edition of these books that was out of copyright.  This (latter) was keyboarded and put online. There were also 9 volumes of court records from Essex County Court from 1636-1686. Among the most valuable sources is the “Church Book Belonging to Salem Village” of 1692 written by the aggressive witch-hunter pastor of Salem Village, the Reverend Samuel Parris.  This was digitized and transcribed in electronic text and reveals his crucial role in supporting the young Salem Village accusers. <

Other important resources were geographic maps of the town of Salem in 1700, a separate map of Salem Village in 1692, and a map of the house locations of 17th century Andover – three communities with a large number of accusers and accused victims. Images of these maps were digitized as part of the Archive Most important in all this was the creation of a dynamic map showing the spread of the witchcraft accusations from their epicenter in Salem Village across 23 other towns, starting February 29, 1692, and running through early November 1692. This map would give the students a sense of the chronological and geographic process, showing the locations where the 151 accused individuals lived and giving a sense of the whole episode in geographic space. It is the only map of its kind based on the dates and locations of the accusers in the court records. 

“Key word searches turned up dozens of reports of these Satanic rituals next to the minister’s house in Salem Village, which drove the court’s witch-hunt efforts relentlessly forward. For the first time, the allegedly growing size of the threat and the lengthy legal process could be documented over time and could identify the people who were responsible for it.” 

One of my main priorities throughout the work with the archive was to make the web interface as transparent as possible so it could be navigated by any user, with all the contents listed on the entry page under major group headings:  records of various courts, church record books, historical maps, contemporary books, literary works, selected short essays by undergraduates in the years 2000-2002, and press reviews of the web site.  

What are the historic fingerprints of Rebecca Nurse’s case? Unfortunately, like other defense efforts, the petition about Nurse’s exemplary Christian character had no effect on the court’s decision and Nurse was despite the defense convicted and executed.  Her two sisters were also accused and one of them was executed. The accusations against the sisters were based on a conflict over a property boundary with the large and powerful Putnam family of Salem Village, some of whom also signed the petition for her innocence. The petition can be found at . Notice the signatures of 39 of her neighbors. All are individual signatures, suggesting that the document was taken from house-to-house. But such petitions had no standing in court because they were not sworn to, lest they be untrue and would then endanger the signer’s soul. In Rebecca’s case, the jury first returned a verdict of innocent, but were told to continue their deliberations because Nurse’s responses to the judges were not always clear. After further deliberations the jury returned a guilty verdict and Nurse, as stated above, was later executed.  Her Death Warrant still survives, In the aftermath her family members took her unburied body from the high ledge at the base of Gallows Hill during the cover of night back to her home for burial. 

Salem’s long suppression of the memory of its brutal witch-hunt resulted in the loss of the knowledge of the location of the infamous execution site.  GIS played a role in its rediscovery using view shed analysis, and this evidence contributed to the site’s recent memorialization in 2017. and <>. It is rare that historians, using digital technology, can change how and where events can be studied and remembered, and indeed be suitably memorialized. The Salem Digital Archive is now fully integrated in the University of Virginia Library as a permanent resource. 

About the author

Professor Benjamin Ray, Distinguished Teaching Professor, Emeritus
Place of work: Dept. of Religious Studies University of Virginia