Killers of the Flower Moon

by David Grann

  Middleton insisted that he didn’t know where she had gone. “I am not a bit afraid,” he said.

  He and his friend didn’t divulge anything. But two witnesses revealed that on the day Lewis disappeared, they had seen, a few miles from her houseboat, a car heading toward a snake-infested swamp. On January 18, 1919, investigators, with their pant legs rolled up, began to comb the thicket of vegetation. A reporter said that one of the lawmen had “scarcely stepped in the water of the bayou when his feet struggled for freedom. When he reached to the bottom to disengage them he brought up a thick growth of woman’s hair.” Leg bones were dredged up next. Then came a human trunk and a skull, which looked as if it had been beaten with a heavy metal object. GREWSOME FIND ENDS QUEST FOR MARY LEWIS, a headline in a local newspaper said.

  Middleton’s companion confessed to beating Lewis over the head with a hammer. The plot was conceived by Middleton: after Lewis was killed, the plan was to use a female associate to impersonate her so that the friends could collect the headright payments. (This strategy was not unique—bogus heirs were a common problem. After Bill Smith died in the house explosion, the government initially feared that a relative claiming to be his heir was an impostor.) In 1919, Middleton was convicted of murder and condemned to die. “There was a point in Mary’s family that they were relieved the ordeal was over,” Jefferson wrote. “However, the feeling of satisfaction would be followed by disbelief and anger.” Middleton’s sentence was commuted to life. Then, after he had served only six and a half years, he was pardoned by the governor of Texas; Middleton had a girlfriend, and Lewis’s family believed that she had bribed authorities. “The murderer had gotten only a slap on the hand,” Jefferson wrote.

  After I finished reading the manuscript documenting Lewis’s murder, I kept returning to one detail: she had been killed for her headright in 1918. According to most historical accounts, the Osage Reign of Terror spanned from the spring of 1921, when Hale had Anna Brown murdered, to January 1926, when Hale was arrested. So Lewis’s murder meant that the killings over headrights had begun at least three years earlier than was widely assumed, and if Red Corn’s grandfather was poisoned in 1931, then the killings also continued long after Hale’s arrest. These cases underscored that the murders of the Osage for their headrights were not the result of a single conspiracy orchestrated by Hale. He might have led the bloodiest and longest killing spree. But there were countless other killings—killings that were not included in official estimates and that, unlike the cases of Lewis or Mollie Burkhart’s family members, were never investigated or even classified as homicides.

  26   BLOOD CRIES OUT

  I returned to the archives in Fort Worth and resumed searching through the endless musty boxes and files. The archivist wheeled the newest batch of boxes on a cart into the small reading room, before rolling out the previous load. I had lost the illusion that I would find some Rosetta stone that would unlock the secrets of the past. Most of the records were dry and clinical—expenses, census reports, oil leases.

  In one of the boxes was a tattered, fabric-covered logbook from the Office of Indian Affairs cataloging the names of guardians during the Reign of Terror. Written out by hand, the logbook included the name of each guardian and, underneath, a list of his Osage wards. If a ward passed away while under guardianship, a single word was usually scrawled by his or her name: “Dead.”

  I searched for the name of H. G. Burt, the suspect in W. W. Vaughan’s killing. The log showed that he was the guardian of George Bigheart’s daughter as well as of four other Osage. Beside the name of one of these wards was the word “dead.” I then looked up Scott Mathis, the owner of the Big Hill Trading Company. According to the log, he had been the guardian of nine Osage, including Anna Brown and her mother, Lizzie. As I went down the list, I noticed that a third Osage Indian had died under Mathis’s guardianship, and so had a fourth, and a fifth, and a sixth. Altogether, of his nine listed wards, seven had died. And at least two of these deaths were known to be murders.

  I began to scour the log for other Osage guardians around this time. One had eleven Osage wards, eight of whom had died. Another guardian had thirteen wards, more than half of whom had been listed as deceased. And one guardian had five wards, all of whom died. And so it went, on and on. The numbers were staggering and clearly defied a natural death rate. Because most of these cases had never been investigated, it was impossible to determine precisely how many of the deaths were suspicious, let alone who might be responsible for any foul play.

  Nevertheless, there were strong hints of widespread murder. In the FBI records, I found a mention of Anna Sanford, one of the names I had seen in the logbook with the word “dead” written next to it. Though her case was never classified as a homicide, agents had clearly suspected poisoning.

  Another Osage ward, Hlu-ah-to-me, had officially died of tuberculosis. But amid the files was a telegram from an informant to the U.S. attorney alleging that Hlu-ah-to-me’s guardian had deliberately denied her treatment and refused to send her to a hospital in the Southwest for care. Her guardian “knew that was the lone place she could live, and if she stayed in Gray Horse she must die,” the informant noted, adding that after her death the guardian made himself the administrator of her valuable estate.

  In yet another case, the 1926 death of an Osage named Eves Tall Chief, the cause was attributed to alcohol. But witnesses testified at the time that he never drank and had been poisoned. “Members of the family of the dead man were frightened,” an article from 1926 said.

  Even when an Osage ward was mentioned as being alive in the log, it did not mean that he or she had not been targeted. The Osage ward Mary Elkins was considered the wealthiest member of the tribe because she had inherited more than seven headrights. On May 3, 1923, when Elkins was twenty-one, she married a second-rate white boxer. According to a report from an official at the Office of Indian Affairs, her new husband proceeded to lock her in their house, whip her, and give her “drugs, opiates, and liquor in an attempt to hasten her death so that he could claim her huge inheritance.” In her case, the government official interceded, and she survived. An investigation uncovered evidence that the boxer had not acted alone but had been part of a conspiracy orchestrated by a band of local citizens. Though the government official pushed for their prosecution, no one was ever charged, and the identities of the citizens were never revealed.

  Then there was the case of Sybil Bolton, an Osage from Pawhuska who was under the guardianship of her white stepfather. On November 7, 1925, Bolton—whom a local reporter described as “one of the most beautiful girls ever reared in the city”—was found with a fatal bullet in her chest. Her death, at twenty-one, was reported by her stepfather to be a suicide, and the case was quickly closed without even an autopsy. In 1992, Bolton’s grandson Dennis McAuliffe Jr., an editor at the Washington Post, had investigated her death after discovering numerous contradictions and lies in the official account. As he detailed in a memoir, The Deaths of Sybil Bolton, published in 1994, much of her headright money was stolen, and the evidence suggested that she had been assassinated outdoors, on her lawn, with her sixteen-month-old baby—McAuliffe’s mother—beside her. According to the log, her guardian had four other Osage wards. They had also died.

  Though the bureau estimated that there were twenty-four Osage murders, the real number was undoubtedly higher. The bureau closed its investigation after catching Hale and his henchmen. But at least some at the bureau knew that there were many more homicides that had been systematically covered up, evading their efforts of detection. An agent described, in a report, just one of the ways the killers did this: “In connection with the mysterious deaths of a large number of Indians, the perpetrators of the crime would get an Indian intoxicated, have a doctor examine him and pronounce him intoxicated, following which a morphine hypodermic would be injected into the Indian, and after the doctor’s departure the [killers] would inject an enormous amount of morphine under the armpit of the drunken In
dian, which would result in his death. The doctor’s certificate would subsequently read ‘death from alcoholic poison.’ ” Other observers in Osage County noted that suspicious deaths were routinely, and falsely, attributed to “consumption,” “wasting illness,” or “causes unknown.” Scholars and investigators who have since looked into the murders believe that the Osage death toll was in the scores, if not the hundreds. To get a better sense of the decimation, McAuliffe looked at the Authentic Osage Indian Roll Book, which cites the deaths of many of the original allotted members of the tribe. He writes, “Over the sixteen-year period from 1907 to 1923, 605 Osages died, averaging about 38 per year, an annual death rate of about 19 per 1,000. The national death rate now is about 8.5 per 1,000; in the 1920s, when counting methods were not so precise and the statistics were segregated into white and black racial categories, it averaged almost 12 per 1,000 for whites. By all rights, their higher standard of living should have brought the Osages a lower death rate than America’s whites. Yet Osages were dying at more than one-and-a-half times the national rate—and those numbers do not include Osages born after 1907 and not listed on the roll.”

  Louis F. Burns, the eminent historian of the Osage, observed, “I don’t know of a single Osage family which didn’t lose at least one family member because of the head rights.” And at least one bureau agent who had left the case prior to White’s arrival had realized that there was a culture of killing. According to a transcript of an interview with an informant, the agent said, “There are so many of these murder cases. There are hundreds and hundreds.”

  Even cases known to the bureau had hidden dimensions. During one of my last visits to the reservation, in June 2015, I went to the Osage Nation Court, where, in many criminal cases, the Osage now mete out their own justice. An Osage lawyer had told me that the Reign of Terror was “not the end of our history,” adding, “Our families were victims of this conspiracy, but we’re not victims.”

  In one of the courtrooms, I met Marvin Stepson. An Osage man in his seventies with expressive gray eyebrows and a deliberate manner, he served as the chief trial court judge. He was the grandson of William Stepson, the steer-roping champion who had died, of suspected poisoning, in 1922. Authorities never prosecuted anyone for Stepson’s murder, but they came to believe that Kelsie Morrison—the man who had killed Anna Brown—was responsible. By 1922, Morrison had divorced his Osage wife, and after Stepson’s death he married Stepson’s widow, Tillie, making himself the guardian of her two children. One of Morrison’s associates told the bureau that Morrison had admitted to him that he had killed Stepson so that he could marry Tillie and get control of her invaluable estate.

  Stepson’s death was usually included in the official tally of murders during the Reign of Terror. But as I sat with Marvin on one of the wooden courtroom benches, he revealed that the targeting of his family did not end with his grandfather. After marrying Morrison, Tillie grew suspicious of him, especially after Morrison was overheard talking about the effects of the poison strychnine. Tillie confided to her lawyer that she wanted to prevent Morrison from inheriting her estate and to rescind his guardianship of her children. But in July 1923, before she had enacted these changes, she, too, died of suspected poisoning. Morrison stole much of her fortune. According to letters that Morrison wrote, he planned to sell a portion of the estate he had swindled to none other than H. G. Burt, the banker who appeared to be involved in the killing of Vaughan. Tillie’s death was never investigated, though Morrison admitted to an associate that he had killed her and asked him why he didn’t get an Indian squaw and do the same. Marvin Stepson, who had spent years researching what had happened to his grandparents, told me, “Kelsie murdered them both, and left my father an orphan.”

   Marvin Stepson is the grandson of William Stepson, who was a victim of the Reign of Terror. Credit 72

  And that was not the end of the plot. After William Stepson and Tillie died, Marvin’s father, who was three years old at the time, became the next target, along with his nine-year-old half sister. In 1926, Morrison, while serving time in prison for killing Anna Brown, sent a note to Hale, which was intercepted by guards. The note, filled with grammatical errors, said, “Bill, you know Tillies kids are going to have 2 or 3 hundred thousand dollars in a few years, and I have those kids adopted. How can I get possession or control of that money when I get out. You know I belive I can take these kids out of the State and they cant do a dam thing…they Could not get me for Kidnapping.” It was feared that Morrison planned to kill both children. An Osage scholar once observed, “Walking through an Osage cemetery and seeing the gravestones that show the inordinate numbers of young people who died in the period is chilling.”

  Marvin Stepson had the judicious air of someone who had spent his whole career serving the law. But he told me that when he first learned what Morrison had done to his family, he feared what he might be capable of doing. “If Morrison walked in this room right now, I’d…” he said, his voice trailing off.

  In cases where perpetrators of crimes against humanity elude justice in their time, history can often provide at least some final accounting, forensically documenting the murders and exposing the transgressors. Yet so many of the murders of the Osage were so well concealed that such an outcome is no longer possible. In most cases, the families of the victims have no sense of resolution. Many descendants carry out their own private investigations, which have no end. They live with doubts, suspecting dead relatives or old family friends or guardians—some of whom might be guilty and some of whom might be innocent.

  When McAuliffe tried to find the killer of his grandmother, he initially suspected his grandfather Harry, who was white. By then, Harry had died, but his second wife was still alive and told McAuliffe, “You should be ashamed of yourself, Denny, digging up things about the Boltons. I can’t understand why you’d want to do such a thing.” And she kept repeating, “Harry didn’t do it. He had nothing to do with it.”

  Later, McAuliffe realized that she was probably right. He came to believe, instead, that Sybil’s stepfather was responsible. But there is no way to know with certainty. “I did not prove who killed my grandmother,” McAuliffe wrote. “My failure was not just because of me, though. It was because they ripped out too many pages of our history….There were just too many lies, too many documents destroyed, too little done at the time to document how my grandmother died.” He added, “A murdered Indian’s survivors don’t have the right to the satisfaction of justice for past crimes, or of even knowing who killed their children, their mothers or fathers, brothers or sisters, their grandparents. They can only guess—like I was forced to.”

  Before I left Osage County to return home, I stopped to see Mary Jo Webb, a retired teacher who had spent decades investigating the suspicious death of her grandfather during the Reign of Terror. Webb, who was in her eighties, lived in a single-story wooden house in Fairfax, not far from where the Smiths’ home had exploded. A frail woman with a quavering voice, she invited me in and we sat in her living room. I had called earlier to arrange the visit, and in expectation of my arrival she had brought out several boxes of documents—including guardian expense reports, probate records, and court testimony—that she’d gathered about the case of her grandfather Paul Peace. “He was one of those victims who didn’t show up in the FBI files and whose killers didn’t go to prison,” Webb said.

   Mary Jo Webb Credit 73

   The open prairie north of Pawhuska Credit 74

  In December 1926, Peace suspected that his wife, who was white, was poisoning him. As the documents confirmed, he went to see the attorney Comstock, whom Webb described as one of the few decent white attorneys at the time. Peace wanted to get a divorce and change his will to disinherit his wife. A witness later testified that Peace had claimed his wife was feeding him “some kind of poison, that she was killing him.”

  When I asked Webb how her grandfather might have been poisoned, she said, “There were these doctors. They
were brothers. My mother said that everyone knew that’s where people would get the dope to poison the Osage.”

  “What was their name?” I asked.

  “The Shouns.”

  I remembered the Shouns. They were the doctors who had claimed that the bullet that had killed Anna Brown had disappeared. The doctors who had initially concealed that Bill Smith had given a last statement incriminating Hale and who had arranged it so that one of them became the administrator of Rita Smith’s invaluable estate. The doctors whom investigators suspected of giving Mollie Burkhart poison instead of insulin. Many of the cases seemed bound by a web of silent conspirators. Mathis, the Big Hill Trading Company owner and the guardian of Anna Brown and her mother, was a member of the inquest into Brown’s murder that failed to turn up the bullet. He also managed, on behalf of Mollie’s family, the team of private eyes that conspicuously never cracked any of the cases. A witness had told the bureau that after Henry Roan’s murder, Hale was eager to get the corpse away from one undertaker and delivered to the funeral home at the Big Hill Trading Company. The murder plots depended upon doctors who falsified death certificates and upon undertakers who quickly and quietly buried bodies. The guardian whom McAuliffe suspected of killing his grandmother was a prominent attorney working for the tribe who never interfered with the criminal networks operating under his nose. Nor did bankers, including the apparent murderer Burt, who were profiting from the criminal “Indian business.” Nor did the venal mayor of Fairfax—an ally of Hale’s who also served as a guardian. Nor did countless lawmen and prosecutors and judges who had a hand in the blood money. In 1926, the Osage leader Bacon Rind remarked, “There are men amongst the whites, honest men, but they are mighty scarce.” Garrick Bailey, a leading anthropologist on Osage culture, said to me, “If Hale had told what he knew, a high percentage of the county’s leading citizens would have been in prison.” Indeed, virtually every element of society was complicit in the murderous system. Which is why just about any member of this society might have been responsible for the murder of McBride, in Washington: he threatened to bring down not only Hale but a vast criminal operation that was reaping millions and millions of dollars.