Tom White’s Osage murder investigation

Tom White’s Osage murder investigation was a federal effort to impose order on a case clouded by failed inquiries, missing evidence, compromised witnesses, and fear. White took command in 1925 with a mandate from J. Edgar Hoover to direct the investigation. He pursued a disciplined standard of proof and organized undercover work, but the later archival account makes clear that an evidentiary method could expose some deceptions without recovering every lost record or resolving every death.

A case that threatened the Bureau

When Hoover summoned White, the Bureau of Investigation had already spent years failing to solve the murders. In 1923, after the Osage Tribal Council sought federal help, an agent had worked in Osage County for several weeks before concluding that continued investigation was useless. Other agents followed without success. The Osage had been made to finance part of the investigation themselves, an expenditure that eventually reached $20,000. Hoover, newly in command, had then decided to return the case to state authorities rather than inherit responsibility for the Bureau’s failures.

That retreat became harder to sustain after the Bureau’s undercover operation went disastrously wrong. Agents had helped secure the release of Irvin Thompson, a convicted bank robber, so he could gather evidence among oil-field criminals. They lost track of him in the Osage Hills; he robbed a bank and killed a police officer before authorities recaptured him months later. Oklahoma’s attorney general blamed the Bureau for the investigation’s failure, while John Palmer and other critics raised the possibility that corruption had obstructed the pursuit of the killers. Hoover feared that a public scandal, arriving so soon after the Teapot Dome scandal, could endanger his position.

White was summoned because Hoover needed an experienced investigator to take responsibility for the case. He was ordered to Oklahoma City to command the field office. The assignment placed a substantial burden on his family, but White accepted it, aware that previous agents had been dismissed or sent to distant posts and that some people trying to catch the killers had themselves been killed. The mission was therefore both an attempt to solve the murders and a high-stakes test of the Bureau’s credibility.

Reconstructing the record

After taking over the Oklahoma City field office in July 1925, White confronted a mass of files accumulated over two years. The records were not a coherent account but a jumble of reports, leads, and statements, often based on hearsay. More than four years had passed since Anna Brown and Charles Whitehorn were killed. White had to search not only for new information but for overlooked clues buried in the old material.

Earlier investigators had concentrated on six cases they thought most likely to solve: the bombing deaths of Rita Smith, her husband Bill Smith, and their servant Nettie Brookshire, and the shootings of Anna Brown, Henry Roan, and Charles Whitehorn. White saw indications that wealthy Osage people were being targeted and that several victims, including Anna Brown, Rita Smith, and their mother Lizzie, were related. Yet investigators had not even interviewed Lizzie’s surviving daughter, Mollie Burkhart. The case files said little about Mollie beyond noting that she had diabetes and had secluded herself at home, an absence that showed how little the earlier inquiry had recorded about the people at the center of the violence.

The pattern of killings suggested a conspiracy rather than a single, repetitive method. The victims died in different ways, and bodies appeared in different parts of the state and country. White considered that a mastermind might be using henchmen and planning crimes over time. But the abundance of apparent leads was itself a problem: many traced back to private investigators or local lawmen repeating rumor, and even plausible theories could not substitute for evidence admissible in court. White’s working rule was to separate what could be proved from what merely made a persuasive story. He sought what he called an “unbroken chain of evidence.”

An undercover team and a method of elimination

White preferred to work alone, but the number of murders and leads made a team necessary. Witnesses often refused to cooperate because of prejudice, corruption, or fear of being killed. White therefore planned to serve as the investigation’s public face while most of his agents worked undercover. He selected experienced “Cowboys” who could manage dangerous assignments, maintain cover, and operate in outlaw country. His brother Doc White was not included; the brothers had avoided serving on the same cases to reduce the risk that their family might lose both of them.

The operatives entered Osage County under ordinary-seeming identities. A former sheriff posed as an elderly Texas cattleman, and a former Texas Ranger presented himself as a rancher. An experienced undercover agent opened an insurance business in downtown Fairfax, while John Wren arrived as a medicine man looking for relatives. White had told his men to keep their covers simple. The cattlemen gained William K. Hale’s confidence and introductions to prominent townspeople; the insurance agent visited suspects’ homes; and Wren attended tribal gatherings, where Osage people might speak more freely to someone who was not acting as a white lawman. Headquarters sent the agents coded orders directing them to report under cover to White.

The investigation began with serious evidentiary obstacles. Records from the coroner’s inquest into Anna Brown’s death had disappeared after the justice of the peace’s desk was broken into. Little had been preserved from the crime scenes. One surviving object was Anna’s skull, secretly kept by the undertaker. White examined the bullet hole at the back of the skull and noted that there was no exit wound. Since the bullet must have remained inside, its absence from the autopsy record raised the possibility that someone had removed it. The doctors, David Shoun and James Shoun, said they had searched diligently for it; with numerous people present at the autopsy, White could not establish who might have taken it.

Rather than accept every suspect’s story or discard it by intuition, White methodically tested alibis. Agent Burger’s earlier work helped him eliminate suspects, including Anna’s former husband, Oda Brown. The accusation against Brown had come from a forger who, White learned, had fabricated his story in hopes of bargaining for better prison conditions. Other suspects identified by the former sheriff Harve M. Freas also fell away under scrutiny.

A false confession attributed to Rose Osage offered a particularly clear example of the method. A private investigator had obtained a statement claiming that Rose killed Anna after a romantic rivalry involving Joe Allen. A woman identified as a witness said Rose had confessed; the Fairfax marshal also recalled a dark stain on the backseat of Rose’s car. But Rose and Joe had an alibi confirmed by the rooming-house owner. The account of the supposed confession also conflicted with the medical and crime-scene evidence: doctors had not reported signs that Anna’s body had been moved, and investigators found no trail of blood from the car to Three Mile Creek. When agents questioned the witness, she admitted that Rose had never told her the story. A strange white man had written it down and forced her to sign. The episode showed that the investigation had to identify not only evidence that had been erased, but evidence that had been manufactured.

The limits revealed by later records

The book’s later archival investigation revisits what White’s inquiry could and could not establish. In a 1924 ceremonial photograph displayed at the Osage Nation Museum, a section had been removed. Kathryn Red Corn showed the researcher a small print of the missing panel: William K. Hale stared out from it. The Osage had removed his image, the account emphasizes, not because they had forgotten the murders, but because they could not forget. The image’s removal makes visible a contrast between the case’s official record and the memory carried by the community.

Other records raised questions about H. G. Burt and the deaths of George Bigheart and W. W. Vaughan. Secret grand-jury testimony from 1926 showed that an insurance agent who had issued Henry Roan’s policy, naming Hale as beneficiary, said Burt had recommended another Indigenous person as a target for an insurance scheme. Bureau reports described Burt and Hale as close associates and said they had divided money obtained from Bigheart. Hale later claimed $6,000 from Bigheart’s estate using a forged creditor’s note. Burt had also become guardian of Bigheart’s daughter, giving him access to her finances. Records further indicated that Burt had been with Bigheart around the time he became ill and that Burt had taken money Bigheart intended for Vaughan.

The Vaughan evidence remained suggestive rather than conclusive. A newspaper account placed Burt on the train with Vaughan and said Burt reported Vaughan missing from his berth. An informant close to Hale told the Bureau that he thought Burt had killed Vaughan; the Fairfax marshal’s later account also implicated Burt and said a local mayor had helped throw Vaughan from the train. Burt moved to Kansas in 1925, when he was reportedly frightened by the investigation. These details formed a circumstantial case, not a proven account of who killed Vaughan or precisely how he died. The later researcher explicitly recognized the unfairness of accusing a dead man who could no longer answer the evidence.

A handwritten Office of Indian Affairs guardian log widened the question beyond the cases White’s team had prioritized. It listed guardians and their Osage wards, marking a ward who died with the word “Dead.” Burt was listed as guardian to Bigheart’s daughter and four other Osage, one of whom had died. Scott Mathis had nine listed wards, including Anna Brown and Lizzie; seven had died, and at least two of those deaths were known murders. Other guardians’ lists contained similarly striking death counts: one had eleven wards, eight deceased; another had thirteen, more than half deceased; and one had five, all dead. Since many of these deaths had never been investigated or classified as homicides, the records could not establish who was responsible, or even precisely how many deaths were suspicious. They did, however, expose how much the official case count left outside its frame.

An investigation still open to history

The later account presents White’s casework as a serious attempt to replace rumor with corroboration, not as a complete explanation of the Osage deaths. White’s undercover team could penetrate social spaces closed to openly identified investigators, test alibis, and uncover fabricated testimony. Yet missing inquest records, vanished or altered evidence, frightened witnesses, and deaths never classified as homicides placed limits on what the Bureau could prove.

The book also questions the triumph attached to the convictions of Hale and his co-conspirators. Once the principal defendants received life sentences, authorities treated the cases as closed, although the Bureau had not connected Hale to all twenty-four murders. The later archival work asks who abducted McBride in Washington, D.C., and who threw Vaughan from the train; it finds no evidence that Hale’s usual associates followed McBride to the capital or were with Vaughan on the train. The investigation’s significance therefore lies both in the evidence it gathered and in the unanswered questions it left behind. Its central discipline—distinguishing proof from an attractive theory—also makes the remaining gaps impossible to mistake for certainty.