This is a guest post by Sonia Kahn, detailed Reference Librarian in the Geography and Map Division.

Today on Halloween children often dress up as witches and wizards with brimmed hats and broom sticks. Many find joy in the celebration of all things spooky by donning a costume and an alternate identity for a single night. Of course 300 plus years ago this innocuous enough tradition was non-existent. No, in fact, quite the opposite. To be viewed as a witch or magic user three centuries ago was a dangerous path to tread. The hysteria and fear around witchcraft led to one of the most infamous and dangerous cases of defamation in American history, the Salem Witch Trials.

The witch no. 1. Lithograph by Joseph E. Baker, 1892. Prints and Photograph Division

Paranoia and sensationalism around witches had existed in Europe since the latter half of the Middle Ages. The fad was actually dying down by the time of the Salem Witch Trials in 1692. But Salem in the late 17^th century had the perfect conditions to stoke fear and distrust amongst neighbors. Many may not know that in 1692 Salem was actually an amalgamation of two separate populations – Salem Town which is modern day Salem, and Salem Village which is modern day Danvers, Massachusetts. Salem Village formally separated from the original Salem Town in 1672 after amassing a sizeable populace, but a rivalry between the two areas remained, and limited resources created resentment and distrust between the villages.

In January 1692 several young girls from Salem Village started acting strangely and a medical examination declared them bewitched. These girls accused three other women from the village of being witches, thus igniting a firestorm. When one of the accused confessed to being a witch and professed that there were more witches hiding across Salem, mass hysteria broke out. Accusations abounded, quickly spreading to Salem Town. When the smoke had finally cleared over 200 people, both men and women, had been accused of witchcraft, and 19 were sent to the gallows.

Over time, though the notorious reputation of the trials remained, more precise information about the events was lost to history. One of the pieces of the puzzle that was slowly ensconced by time was the specific location at which the accused witches lost their lives. Until recently, speculation had dictated that the witches were hung atop Gallows Hill, a large hill in the southwestern portion of modern day Salem. This was in part due to the findings of a 19^th century historian by the name of Charles Wentworth Upham who determined that Gallows Hill was the likely site of the executions.

This map published in Charles W. Upham’s 1867 work, Salem Witchcraft, depicts the town as Upham believed it to be at the time of the trials. A “W” marks the location Upham believed to be the execution sight atop Gallows Hill. Map of Salem Village, 1692. Made by William Phineas Upham, 1866. Geography and Map Division.

Upham’s findings were eventually called into question. Certain characteristics of Gallows Hill simply didn’t line up with the information left in the primary resources from the events as well as oral histories which had been passed down through the generations. In the early 20^th century a new historian named Sidney Perley posited that the likely hanging site was not Gallows Hill, but rather a significantly smaller knoll called Proctor’s Ledge at the base of Gallows Hill.

This tourist map of Salem from 1915 created by Sidney Perley’s son, Richard Hood Perley, appears to label the execution site as Proctor’s Ledge, where his father theorized the killings took place. A New Map of Salem. Made by Richard Hood Perley, 1915. Geography and Map Division.

Perley put forward several reasons as to why Gallows Hill was unlikely to be the true execution site. The first was the fact that Gallows Hill is quite steep which would make it immensely difficult to push the carts carrying the condemned witches to its peak. Proctor’s Ledge by contrast, at the base of Gallows Hill, would be more reasonable to summit with the additional human cargo in tow.

This 1851 map shows some of the topography of Salem. You can see the relative elevation of Gallows Hill, along with a smaller nearby ridge which appears to be Proctor’s Ledge. Map of the City of Salem Mass. Made by Henry McIntyre, 1851. Geography and Map Division.

Aside from the problematic elevation, other factors contradicted Gallows Hill as the ostensible location of the killings. Stories passed down through generations of Salem residents recounted how the son of Rebecca Nurse, one of the accused witches, rowed a small boat back to the execution site in the cover of night to recover his mother’s body. This tale would have been implausible if the witches were executed atop Gallows Hill, as the mount is not bordered by any water ways, nor has it ever been adjacent to water. Proctor’s Ledge, however, was located next to a stream in the late 17^th century, which has since been filled in. Moreover, information from the primary documents left behind by the witnesses of the murders just did not add up to Gallows Hill being the culpable site. The testimony and descriptions given by some of those who witnessed the murders helped provide context and locate the site of the executions. For example, some original documentation places witnesses in areas from which Gallows Hill would have been obscured from view. Proctor’s Hill, on the other hand, would have been visible from the locations of these eyewitnesses.

Despite the fact that Sidney Perley hypothesized that Proctor’s Ledge was the true execution site of the condemned witches, this map from 1930 continued to identify Gallows Hill as the location of the killings. Salem Massachusetts. Made by Helen Washburn, 1930. Geography and Map Division.

Despite Perley’s reasoning, he simply did not have the technology necessary in the early 20^th century to make the indisputable claim of where the accused witches met their end. However, a team of scholars nearly one hundred years later would have the capability to do just that. A group of five researchers spearheaded the Gallows Hill Project in order to formally identify the site of the executions. The team, which included a Geographic Information Systems specialist, examined hundreds of pages of primary documents consisting of maps and court testimony. Information gathered from these original sources were then input into modern day GIS technology to map Salem as it would have looked during the trials, and to spatially document the location of witnesses. With this 21^st century rendering of Salem in 1692, the scholars were able to narrow and ultimately pinpoint the location of the executions by determining which sites would have been visible from the vantage point of each unique witness. The GIS analysis was consistent with Sidney Perley’s claims nearly 100 years earlier that Proctor’s Ledge would have been the most readily visible location for the killings. The group announced their findings to the world in January 2016, and a little over a year later, a small memorial was erected at the site.

In 2022, three hundred and thirty years after the trials, Proctor’s Ledge finds itself in a quaint and quiet suburban area, nestled between trees and private residences. As the young witches of today’s Salem head door-to-door tonight seeking treats, they are likely none the wiser to what happened to the witches of the past, just down the street.