Eight people were killed in Villisca, Iowa, on the night of June 9, 1912. The investigation that followed was compromised from the first hour: the crime scene was contaminated by curious neighbors before investigators arrived, a confession was coerced from a mentally ill man who was then acquitted twice, and a private detective agency’s politically motivated theory destroyed a career without producing charges. The case was never solved. What happened to it afterward is its own indictment: 1912 newspapers ran sensationalized serial killer narratives that muddied the evidentiary picture, and modern true crime coverage has finished the job by turning a farmhouse where six children were murdered into an overnight ghost hunting attraction.
Josiah Moore, 43; Sarah Moore, 39; their children Herman, 11, Katherine, 10, Boyd, 7, and Paul, 5; and overnight guests Lena Stillinger, 12, and Ina Stillinger, 8. All eight were killed with an ax in the Moore family home in Villisca, Iowa.
No. Reverend George Kelly was tried twice — once with a hung jury eleven-to-one for acquittal, once with a full acquittal. His confession was almost certainly coerced after lengthy interrogation, and he immediately recanted. No physical evidence tied him to the crime.
Iowa State Senator Frank F. Jones, who had a documented business rivalry with Josiah Moore; William Mansfield, a hired-killer candidate; Reverend George Kelly, who was present in Villisca the night of the murders; and Henry Lee Moore, an itinerant convicted of a similar ax murder in Missouri. None were ever convicted of the Villisca killings.
Yellow journalism amplified unverified theories, created a national “Axeman” serial killer narrative linking Villisca to similar Midwest crimes, and generated the kind of public hysteria that made disciplined investigation harder. It is the same structural failure as modern true crime media, with different technology.
There is no documented evidence of paranormal activity. The haunted framing is a commercial product. It is not supported by the evidentiary record and is not relevant to an analysis of the murders or the failures of the investigation.
The mechanisms differ. The effect is the same. Both eras prioritize spectacle over evidence, name suspects without adequate basis, and use the deaths of eight people — six of them children — as raw material for a product aimed at an audience that wants to be entertained by horror.
The town of Villisca, Iowa, had about 2,500 residents in 1912. It was, by all accounts, the kind of small agricultural community where people left their doors unlocked because crime was not something you worried about. Everyone knew everyone. The Moores were a respectable family. Josiah ran a hardware store. Sarah was active in the Presbyterian church. Their children were regular kids in a regular town on the edge of the Iowa plains.
On the evening of Sunday, June 9, Josiah and Sarah took their four children and two of the children’s friends — sisters Lena and Ina Stillinger, who were guests for the night — to Children’s Day services at the Presbyterian Church. They came home. They went to bed. Before dawn, someone killed all eight of them.
That is what is known for certain. Everything after that point is contested, contaminated, or invented — and the inventory of who did the inventing runs from the 1912 coroner’s inquest all the way to a ghost tour company renting out EMF detectors in the gift shop.
What Happened in the House
The bodies were discovered the morning of June 10 by a neighbor, Mary Peckham, who noticed that the Moores had not emerged to do their morning chores. She contacted Josiah’s brother Ross, who used a spare key to enter the house. The coroner and investigators were called, but so was much of the town.
What they found: Josiah and Sarah had been killed in the master bedroom. Josiah received more blows than any other victim — the autopsy documentation describes injuries so severe to his face that his eyes were missing. The killer then moved through the children’s bedrooms on the upper floor, killing Herman, Katherine, Boyd, and Paul. Lena and Ina Stillinger, sleeping in a downstairs guest room, were killed last. All eight victims had been covered — faces obscured with cloth after death. All mirrors in the house had been covered. The ax used in the killings had been left in the downstairs guest room, wiped but not clean.
The medical determination was that the murders occurred between midnight and 5 a.m. No one heard anything. The house sat in ordinary silence for hours before the morning discovery.
The covering of victims’ faces and the covering of mirrors is consistent with behavior seen in other organized homicides — a killer who does not want to look at what he has done, or who is performing a ritual. The ax was wiped. No forced entry. The killer either knew the house well enough to move through it in the dark, or had prepared for the layout. None of this is conclusive. It is what the physical evidence suggests. It is not ghosts.
The critical forensic problem began within hours of discovery. Before investigators could secure the scene, neighbors and townspeople moved through the house, handled objects, sat in chairs, examined the bodies, and by at least some accounts, cooked meals in the kitchen. The crime scene was contaminated by civilian foot traffic before any proper documentation could occur. In an era before forensic photography and evidence protocols were standard practice, this was catastrophic. The first investigators on the scene were working inside a compromised record before the investigation had formally begun.
The Investigation: A Sequential Account of What Went Wrong
The coroner held an inquest on June 11 — two days after the murders — interviewing 13 witnesses. The limitation of that process is documented: witnesses had already been talking to each other, to journalists, and to anyone who asked. Their accounts were already shaping around a shared narrative rather than independent recollection. The evidentiary contamination was not just physical. It was testimonial.
The early focus of local suspicion settled almost immediately on Iowa State Senator Frank F. Jones. The logic was accessible: Josiah Moore had worked as a clerk at Jones’s hardware and implement dealership. About five years before the murders, Moore left Jones’s employ and opened a competing business. Jones was a hard, politically connected man, and local accounts suggested he did not take the defection well. When the bodies were found, the town split roughly along religious lines — the Methodists insisting Jones was innocent, the Moores’ Presbyterian congregation inclined toward his guilt. No formal evidence placed Jones anywhere near the crime.
The national private detective firm known as the Burns Detective Agency dispatched an investigator named James Newton Wilkerson, who arrived with a theory he was prepared to prove regardless of what the evidence supported. Wilkerson’s theory was that Jones had hired a professional killer named William Mansfield to murder the Moore family. The motive he alleged went beyond business rivalry: Wilkerson claimed that Jones’s daughter-in-law had been involved in an affair with Josiah Moore, and that the combined humiliation of both professional and personal betrayal drove Jones to order the killings.
Wilkerson held public meetings in Villisca. He named Jones and his son as orchestrators and Mansfield as the weapon. He successfully derailed Jones’s re-election campaign. He eventually got a grand jury convened. Mansfield was arrested in 1916 and extradited to Iowa. The problem: Mansfield had an alibi backed by payroll records placing him several hundred miles away in Illinois on the night of the murders. The grand jury declined to indict. Jones, who never faced charges, sued Wilkerson for slander — and lost the suit, which tells you something about what a jury thought of Jones’s conduct even if it couldn’t put him at a murder scene.
The Burns Detective Agency was retained privately. Wilkerson was not a disinterested investigator — he was a hired advocate for a theory that required Jones’s guilt. He went public with accusations before he had evidence to support them, destroyed a career, contributed to a decade of legal chaos, and produced nothing provable. Private detective investigations with a predetermined conclusion are not forensic analysis. They are advocacy with the appearance of investigation. The Villisca record shows how much damage that can do.
The Kelly Problem: A Confession That Never Should Have Been Used
Reverend Lyn George Jacklin Kelly was, by any honest description, deeply mentally unwell. He was a traveling English-born preacher, slight of build at five feet two inches and about 119 pounds. He had a documented history of peeping in windows, confessed arson crimes, and what investigators described as erratic, grandiose behavior. He had been in Villisca on June 9 — he attended the same Children’s Day services as the Moores and left town before dawn on June 10, on a train that departed at 5:19 a.m., reportedly telling fellow passengers that eight people had been killed in town before the bodies had been publicly discovered.
That last detail is the most legitimately suspicious thing about Kelly. It is also the kind of detail that gets laundered into certainty through repeated telling. Whether he actually said it, whether the witnesses accurately reported what he said, and whether there were other ways he could have known — from rumor on the train platform, from an unconfirmed early report, from genuine involvement — cannot be established from the record that survives.
Kelly was not arrested until 1917 — five years after the murders. By then, witness memories had faded and shifted. Under interrogation described in court as lengthy and coercive, Kelly signed a confession. His account claimed that God’s voice had instructed him to “slay utterly” and “suffer the little children to come unto me,” and that a dark shadow led him to the Moore house where he found an ax and obeyed divine instruction. He recanted almost immediately. The confession was internally inconsistent across the multiple versions Kelly gave during interrogation.
At his first trial, charged only with the murder of Lena Stillinger, the jury deadlocked — eleven votes for acquittal, one for not guilty by reason of insanity. Not one juror voted to convict. The second trial returned a full acquittal. There was no physical evidence tying Kelly to the crime. His slight physical build was repeatedly noted as a reason investigators had not originally pursued him — the coroner and private detectives had concluded he likely lacked the physical capacity to inflict the injuries documented. The confession, coerced and immediately recanted, was the entirety of the prosecution’s case. The prosecution lost it twice.
“Aside from the confession, a dislike of children, and the coincidence of his hand preference, there was no physical evidence that actually tied Kelly to the murders.”Crime Library, summarizing the Kelly prosecution’s evidentiary basis
Kelly was tried for one murder. The acquittal on that charge did not legally bar prosecution on the others. No further charges were ever brought. He eventually left the Midwest and settled in New England. He was a disturbed man who almost certainly did not commit eight murders. He also might have had some knowledge of events that night that was never fully accounted for. The record does not resolve that question, and the record is what matters.
The Suspects: What the Evidence Actually Supports
Four figures remain the primary subjects of serious historical analysis.
Frank F. Jones had motive — business rivalry, the alleged affair theory — and the means to hire someone if the hiring theory was true. He had no alibi problem because he was not a suspect who personally committed the act. The evidentiary record against him is: community suspicion, Wilkerson’s privately funded accusation campaign, and the fact that he lost a slander suit. None of that is criminal evidence. It is the residue of a decade-long public campaign that substituted pressure for proof.
William Mansfield had the right general biography for the hired-killer theory — he was a suspect in a similar ax murder of his own family in Illinois in 1914 — and Wilkerson worked hard to tie him to the Villisca killings. The problem is documented: his alibi was verified by payroll records. He was not indicted. The coincidence of his later Illinois involvement is legitimately notable. It does not substitute for evidence of his presence in Villisca.
Henry Lee Moore — no relation to the victims — is the suspect most discussed by researchers who believe Villisca was part of a pattern of Midwest ax murders targeting sleeping families. Moore was convicted of murdering his mother and grandmother in Columbia, Missouri, in 1905, using an ax. He was released and was at large during the period of several Midwest ax murders that share methodological similarities with Villisca. He was convicted again in Missouri for the Columbia murders and imprisoned. He was never charged with Villisca. Connecting him to multiple crimes requires accepting a serial killer hypothesis for which the primary evidence is methodological similarity — which is suggestive but not conclusive.
George Kelly remains the figure most media coverage defaults to, largely because he is the only person who was actually tried. The evidentiary analysis above stands.
The documented record supports this: eight people were killed by someone who had access to the house, moved through it without disturbing anyone, and covered the victims after death. The killer was almost certainly familiar with the property. The business rivalry between Moore and Jones provides a documented motive for the Jones-hired-Mansfield theory. Kelly’s presence and early departure are suspicious. Henry Moore’s pattern of ax violence is notable. None of these observations constitute proof. The case is unsolved because the evidence required to establish proof was contaminated, lost, or never collected in the first place.
The 1912 Press: The First Layer of Sensationalism
The American press in 1912 was operating in the final years of what historians of journalism call the yellow press era — competitive, circulation-driven, and structurally incentivized to make crimes more dramatic than the evidence warranted. The Villisca murders arrived at the exact right moment to become a national story, and the national press made them into something the evidence did not support.
The specific distortion that did the most damage was the serial killer narrative. Newspapers began linking Villisca to a series of similar ax murders that had occurred across the Midwest in the preceding years — in Colorado Springs, Monmouth, Illinois, and other communities. The crimes did share methodological similarities: sleeping victims, ax wounds, families killed in their beds, organized scenes. Whether they were connected by a single perpetrator, committed by copy-cats, or methodologically similar by coincidence is not established by the record.
The press invented “Billy the Axeman” — a name for a putative serial killer responsible for all of them. This was not a conclusion drawn from evidence. It was a narrative constructed to sell papers. It created the impression of a connected series without doing the investigative work required to establish whether the connection was real. It also had a direct operational effect on the Villisca investigation: it encouraged investigators and the public to look for a wandering stranger rather than examining the local relationships that the evidence actually pointed toward.
When newspapers run theories as established fact, the public absorbs those theories as the frame for evaluating evidence. When an investigation is being conducted in a community where the local newspaper has already decided the killer was a traveling stranger, witness accounts tend to conform to that frame. Evidence that doesn’t fit the narrative gets discarded or ignored. The 1912 press coverage of Villisca did not just sensationalize — it actively shaped the investigative environment in ways that made disciplined analysis harder. That is a specific harm, not a general complaint about irresponsible journalism.
The press also fueled the Jones accusation campaign. When Wilkerson went public with his hired-killer theory, newspapers ran it. They ran it before the grand jury, before any official finding, and before Jones had been charged with anything. Jones’s public career was destroyed in the press before the legal system had determined he had done anything. The slander suit he brought against Wilkerson — which he lost — is a document of how those two things intersected. A private investigator pushed a theory to the press, the press amplified it into the community, and the community’s response made the theory feel like a verdict even though it never became one.
This is not a problem unique to 1912. The mechanism is the same one that appears in any high-profile case where media coverage outpaces the evidentiary record. The technology changes. The structural dynamic does not.
Modern True Crime Coverage: Same Problem, Different Format
The Villisca murder house is now owned by a commercial ghost tour company. You can book a 45-minute tour for $15. You can rent an EMF detector for an additional $7. You can book an overnight stay. The property is listed on the National Registry of Historic Places, which was meant to preserve its historical significance, and which the current ownership uses as a marketing credential for haunted tourism.
The house has appeared on Ghost Adventures, Kindred Spirits, Paranormal Lockdown, Scariest Places on Earth, BuzzFeed Unsolved, and what the Roadtrippers travel site describes as “basically every ghost hunting show.” The BuzzFeed Unsolved episode, which reaches an audience of millions, featured hosts walking through the property speculating about paranormal activity. The property is regularly described in coverage as “one of the most haunted houses in America.”
Six children were killed in that house. They have names. They have documented lives — they went to church the evening they were murdered, they had friends who visited overnight, they had a father who ran a hardware store and a mother who was active in the community. None of that is what the coverage is about. The coverage is about whether a temperature fluctuation in the upstairs bedroom might indicate a ghost.
When the primary frame for covering a murder is “this house might be haunted,” the victims become props. Their deaths become the backstory for an entertainment experience. The investigative questions — who did this, what did the record show, how did institutions fail — are background noise. The haunted framing is commercially convenient because it requires no accountability analysis, no evidentiary discipline, and no engagement with the documented ways the investigation failed. It is sensation dressed as history, and it has been running for 30 years in this case.
The comparison to 1912 press coverage is structural, not hyperbolic. In 1912, newspapers invented a serial killer narrative because it sold papers and required no evidentiary rigor. In 2026, ghost hunting shows and true crime content farms produce paranormal content about the same property because it generates views and requires no evidentiary rigor. The beneficiaries changed. The victims lost in both cases.
There is a subtler version of the same problem in the documentary and podcast space that considers itself more serious. The pattern is: extensive time on gruesome details, competing suspect theories presented as roughly equally plausible without evidentiary weighting, a closing note acknowledging the case is “still unsolved,” and no systematic analysis of what the institutional failures were or what they mean. Villisca is treated as a mystery to be speculated about rather than a record to be analyzed.
Who contaminated the crime scene and why was access not controlled? What was the standard of evidence required before Wilkerson could name Jones publicly? How was Kelly’s confession obtained, what was the interrogation methodology, and who authorized the interrogation conditions? Why was no one charged with the deaths of the other seven victims if Kelly was genuinely believed to be guilty of Lena Stillinger’s murder? These are the questions the record raises. None of them are answered by asking whether an EMF detector spiked near the attic stairs.
The Haunted House Business and What It Costs the Record
The commercial ghost tourism model is worth addressing directly because it has a specific effect on the evidentiary record of cases like Villisca. When a crime site becomes a haunted attraction, several things happen. The property is restored to its appearance at the time of the crime — which at the Moore house means recreating the crime scene aesthetic as an immersive experience. The narrative around the property becomes dominated by paranormal claims, which crowd out analytical discussion in media coverage. People who visit develop strong emotional attachments to the “haunted” framing, which creates a constituency resistant to coverage that treats the case analytically rather than supernaturally.
This is not a complaint about the Linn family, who purchased the property in the 1990s and decided to preserve it rather than demolish it. Preservation has value. The historical record of the Moore house matters. The question is what frame is applied to that record. A frame that asks “what does the evidence show about who did this and how institutions failed” is analytically useful. A frame that asks “what entities might be haunting the bedroom where the children were killed” is not.
The former mayor of Villisca, Susie Enarson, said it plainly in a 2014 AP interview: “I would like it to be over. I would like the people to rest in peace and not have all this ghost discussion.” The current ownership’s response, also on record: “I don’t have any qualms about that. It was 100 years ago.” That exchange is a document of a community that lost something and a commercial enterprise that found something. They are not talking about the same thing.
What the Case Is Actually About
Villisca is not primarily about who swung the ax. It is about what happens when every system that should produce accountability fails in sequence. The crime scene was contaminated before investigation began. The detective agency hired to investigate had a predetermined theory and the financial incentive to pursue it regardless of evidence. The press amplified accusations before they were tested. The prosecution used a coerced confession it couldn’t defend at trial. The legal proceedings destroyed at least one person’s reputation and career without establishing guilt. No one was ever convicted. And then, decades later, the commercial entertainment industry converted the site of all that failure into a product.
The eight people who were killed in that house on June 9, 1912 — Josiah, Sarah, Herman, Katherine, Boyd, Paul, Lena, and Ina — are not the background of a ghost story. They are the entire point. Their names are not props for content. Their deaths deserved an investigation that was disciplined, an evidentiary record that was preserved, a press that reported what was documented rather than what sold, and a subsequent century of coverage that engaged with the institutional failures rather than replacing them with paranormal speculation.
They did not get any of that. What they got was a gift shop.
Sources
Continue Your Investigation
If this reporting raised more questions, use the Clutch Justice ecosystem to keep going.
