In 1991, Cameron Todd Willingham was awakened by his children’s cries and found his house engulfed in fire and smoke. He survived the fire, but his three daughters died. Thirteen years later, the Texas man was executed for murdering his kids by arson. But later, fire experts scientifically proved each and every one of the indicators of arson had been invalid.

On the morning of 23 December 1991, neighbors found Cameron Todd Willingham’s house engulfed in fire while he stood outside pleading for help. Firemen brought the fire under control, but unfortunately, none of his three daughters survived.

In 1991, two days before Christmas, everyone in the Willingham household was asleep except for Stacy. When Cameron Todd Willingham woke up and saw his wife Stacy going out of the house to pay the bills, he heard his one-year-old twins, Karmon and Kameron, crying and gave them bottles. His two-year-old daughter Amber was asleep in the same room. After closing the childproof gate to their bedroom, Willingham went back to his bedroom and fell asleep. About an hour later, he heard Amber crying “Daddy! Daddy!” The house was full of smoke, and he couldn’t see the doorway. So, he crouched low, crept into the hall, and then made his way to the girl’s bedroom. But by then, debris started falling from the ceiling burning his shoulder. He fled out of the front door screaming for the neighbors to help.

His neighbor, Mary Barbee, saw Willingham in the front yard and ran to her phone to call for help. When Barbee returned, she saw the windows had blown out, and fire and smoke were billowing out of the house. Soon, the firefighters arrived on the scene. Willingham repeatedly tried to enter the house, and the firefighters had to restrain him. After some time, a fireman brought Amber from the house, but she was unconscious. Willingham was taken to a hospital to receive treatment for his burns. There, he was told that Amber had died of smoke inhalation. The twins were discovered in the children’s bedroom, their bodies severely burned.

Fire investigators, Douglas Fogg and Manuel Vasquez, investigated the home and gave their verdict that the fire was set intentionally using a liquid accelerant. The fire was considered a triple homicide, and Todd Willingham became the prime suspect. On December 31st, the authorities brought Willingham in for questioning and later charged him with murder. The trial ended after just two days when the jury returned a unanimous verdict: guilty.

Four days after the fire, two arson investigators went inside the Willingham house to determine the cause of the blaze. One of them was Douglas Fogg, the assistant fire chief in Corsicana, and Manuel Vasquez, a deputy fire marshal. They toured the perimeter and concluded that the fire was deliberately started using some kind of liquid accelerant poured all over the children’s room. The only person present in the house beside the victims was Todd Willingham, so he became the prime suspect. When police interviewed the witnesses, they initially claimed that Willingham was devastated by fire. But later some of them changed their claims and said that Willingham seemed calm and controlled.

Willingham was interrogated, but the police couldn’t find any clear motive. He was arrested on January 8, 1992, and charged with murder. Willingham couldn’t afford to hire a lawyer so was assigned two by the state: David Martin, a former state trooper, and Robert Dunn, a local defense attorney. Sometime after Willingham’s arrest, a prison inmate in the same jail, Johnny Webb, alleged that Willingham had confessed to him. Webb said Willingham had said that he took “some kind of lighter fluid, squirting (it) around the walls and the floor, and set a fire.” After this allegation by Dunn, Willingham was offered a life sentence in exchange for his guilty plea.But he declined the offer in spite of his own counsel’s advice. In August 1992, the trial commenced and quickly ended with the jury unanimously returning a guilty verdict.

After spending eight years in jail on death row, Willingham was visited by Elizabeth Gilbert who had volunteered to be a pen pal for an inmate on death row. Elizabeth’s interaction with Willingham and the witnesses and her investigation of the case, made her realize that Willingham might be telling the truth.

In 1999, Elizabeth Gibert visited Willingham. She was a 47-year-old French teacher and playwright who volunteered as a pen pal for Willingham. She met and wrote letters to him first with the conviction that he was indeed the culprit. But after exchanging numerous letters, she became curious about the case. So, she went to the courthouse to review the trial records. She also met the people in the community who were present as witnesses. In the records, she noticed that the testimony of witnesses grew damning after January 1992 when authorities concluded that Willingham was likely guilty of murder.

She even interviewed the jailhouse informant Johnny Webb. As Gilbert chatted with him, she thought that he seemed paranoid. During Willingham’s trial, Webb disclosed that he had been given a diagnosis of “post-traumatic stress disorder” after he was sexually assaulted in prison in 1988 and that he often suffered from “mental impairment.” After months of investigating the case, Gilbert found that her faith in the prosecution was shaken.

Elizabeth sent the file describing the evidence of arson to an acclaimed scientist and fire investigator,  Dr. Gerald Hurst. After examining the evidence, Hurst concluded that there was no evidence of arson.

After her own investigation, Gilbert began believing that Willingham may have been telling the truth from the beginning. So, she sent the files describing the evidence of arson of Willingham’s case to Gerald Hurst. Hurst was an acclaimed scientist and fire investigator who had earlier saved prisoners from death row in arson cases by proving the evidence used to convict them was scientifically invalid. In Willingham’s case, Hurst individually discredited each piece of arson evidence. He used scientifically supported experiments backed by his re-creation of the elements in question.

Hurst rebutted all twenty of the indications listed by Vasquez of an accelerant being used. He concluded that there was “no evidence of arson.” Since there were only very few days left before Willingham’s scheduled execution, Hurst wrote the report in such a hurry that he didn’t even have time to fix the typos.

Hurst later said, “The whole case was based on the purest form of junk science … There was no item of evidence that indicated arson.”

Four days before Willingham’s scheduled execution, Hurst’s report was sent to the Board of Pardons and Paroles, but they denied the petition. Willingham was executed on February 17, 2004, via lethal injection. On July 23, 2010, Texas Forensic Science Commission acknowledged that that state and local arson investigators used “flawed science” in determining that the blaze had been deliberately set.

Hurst’s report was sent to Governor Rick Perry’s office as well as the Board of Pardons and Paroles along with Willingham’s appeal for clemency. There was no response from the governor and the board denied the petition. On February 17, 2004, Willingham was executed at the age of 36. On his death certificate, the cause was listed as “Homicide.”

In December 2004, questions about the scientific evidence in the Willingham case began to surface. Nearly two years later, the Innocence Project commissioned four top fire investigators to conduct an independent review of the arson evidence in the Willingham case. The panel concluded that “each and every one” of the indicators of arson had been “scientifically proven to be invalid.” In 2005, Texas established a government commission to investigate allegations of error and misconduct by the forensic scientists involved. Noted fire scientist, Craig Beyler, who was hired by the commission, investigated and reported that the investigators in the Willingham case had no scientific basis for claiming that the fire was arson. They ignored evidence that contradicted their theory, had no comprehension of flashover and fire dynamics, relied on discredited folklore, and failed to eliminate potential accidental or alternative causes of the fire.
[source: 1, 2, 3, 4, 5]

Also see: June and Jennifer Gibbons, the twins who only talked with each other; later one of them died so that the other could lead a normal life