What once was considered proof of arson, experts say, may actually not be proof of anything.
Now, Superior Court has ordered a new trial for Dougherty, 54, in the deaths of 3-year-old John and 4-year-old Daniel Jr. His defense lawyer's failure to challenge the state's scientific conclusions so skewed the trial that "no reliable adjudication of guilt or innocence took place," the court wrote.
If the jury had heard and believed an authority on the advanced fire science of the time, it "would have had reasonable doubt about [Dougherty's] guilt and would have been compelled to acquit him," the court wrote.
Philadelphia prosecutors oppose a retrial. The District Attorney's Office has asked for reconsideration by the full Superior Court, beyond the three-judge panel that in this case rendered a unanimous decision.
"We stand by the conviction. We stand by the case," said Tasha Jamerson, spokeswoman for District Attorney Seth Williams.
Dougherty's appellate lawyer, David Fryman, declined to comment.
Marissa Boyers Bluestine, legal director of the Pennsylvania Innocence Project, which works to exonerate people convicted of crimes they did not commit, said she was encouraged by the ruling.
"I'm hopeful the D.A. will review this and say, 'What's the right thing to do?' " she said. "I would say, the right thing is to see what a new jury says."
Dougherty is serving a life sentence at the Greene prison in Waynesburg, in southwestern Pennsylvania. His death sentence was vacated in 2012 by an agreement between prosecutors and defense counsel over the ineffectiveness of his trial lawyer, Thomas Ciccone, now deceased.
In a letter to The Inquirer last month, Dougherty said he hoped "the truth will be known and this innocent man will go home to his loved ones."
Prosecutors said a vengeful Dougherty set the fire to hurt two women: his girlfriend, Kathleen Schuler, with whom he lived on Carver Street, and the mother of the boys, Kathleen Dippel, from whom he was separated. He sought to destroy the house of one and the children of the other, prosecutors said.
During the last 20 years, a stream of scientific studies has deflated what once were considered solid indicators of arson. In Pennsylvania and around the country, inmates have challenged convictions that they say are based on old, disproved science.
For instance, burn patterns on a floor were once viewed as evidence of an accelerant such as gasoline. Investigators now know those marks are common when a room is engulfed, even in an accidental fire.
Where finely cracked glass once was a sign of rapid heating caused by an accelerant, scientists now know the opposite: It is caused by rapid cooling, when water from fire hoses hits hot glass.
Dougherty's case has drawn comparisons to that of Cameron Todd Willingham, executed in Texas in 2004 for killing his three young daughters by setting the family home ablaze. Five years later, an expert hired by the Texas Forensic Science Commission called the arson finding doubtful, saying better understanding of fire could have freed Willingham.
In August 1985, Dougherty was a 25-year-old mechanic with a 10th-grade education and an alcohol problem. From the start, police suspected him of having set the fire that destroyed the Oxford Circle home where he lived with his sons, his girlfriend, and her 8-year-old son.
And from the start, he cooperated, waiving his right to remain silent.
"I ain't got nothing to remain silent about," he told homicide detectives in an interview an hour after the fire.
Five days later he took a polygraph test - and failed. Schuler took a lie-detector test the same day. She failed, too.
With insufficient evidence to file charges, police released Dougherty - into a life of drinking, combative relationships with women, and torment over his lost sons.
He wasn't arrested until April 1999, after his estranged second wife, Adrienne Sussman, then battling Dougherty for custody of their child, called police and told them he had confessed to her six times.
She never testified at trial. The only new evidence came from two jailhouse informants, who said Dougherty confessed to them while in custody.
Dougherty always told the same story: He was sleeping on the living-room couch, his children in their second-floor bedroom, when he awoke to see the curtains on fire and ran outside. He tried to reach the boys but was forced back by heavy smoke and flames.
Police Officer Jesse Simonson said he had to physically block Dougherty from barreling inside the burning building. When the officer asked for his name, Dougherty replied: "My name is mud and I should die for what I did."
To police, the statement was not the cry of a distraught father but an admission of guilt.
The trial revealed another discrepancy. On the stand, Dougherty described his attempts to reach the children, saying he'd twice forced his way inside. The day of the fire, the jury learned, he told police he couldn't get inside at all.
Superior Court based its new-trial ruling on a claim of ineffective defense counsel, but fire evidence lay at the heart of that finding.
At the 2000 trial, the key prosecution testimony came from John Quinn, who 15 years earlier was an assistant city fire marshal and the first investigator to enter the burned rowhouse. He said the fire started in three places, a classic indicator of arson: a love seat, a sofa, and a spot beneath a dining-room table.
He ruled out accidental causes such as a gas leak, electrical fire, or careless smoking. And he said Dougherty's descriptions of his efforts to save the children were not believable, because his body showed no exposure to flames or smoke.
Dougherty's lawyer, Ciccone, never rebutted that testimony by consulting or calling to the stand an expert to describe the advances in fire study between 1985 and 2000. Ciccone "could have mounted a convincing challenge to the substance of the charges," the court wrote.
In fact, at Dougherty's initial appeal, two authorities did just that, testifying that Quinn's arson finding could not be supported.
Angelo Pisani, a St. John's University criminal justice associate professor and former coordinator of the New York City Arson Strike Force, testified that forensic fire investigation in 1985 was based on myths and misunderstandings.
"According to the science today, we can't determine whether the fire was accidental or intentional," Pisani said in an interview. "I don't think anybody should be incarcerated or executed because you don't know the answer to whether a crime was committed."
Nationally known investigator John Lentini of Scientific Fire Analysis L.L.C. in Florida testified about two key phenomenas: flashover and full-room involvement.
In flashover, rising heat collects at the ceiling, growing hotter until it radiates downward, causing parts of the room to spontaneously combust. Full-room involvement occurs when a fire consumes the available oxygen, and the heat and flames burn floors, walls, tables, and chairs in search of air.
Those conditions make it nearly impossible to know the starting point or points of a fire - and none could be determined at Dougherty's house, Lentini said. The cause should have been listed as "undetermined," he said.
Efforts to locate and seek comment from Quinn were unsuccessful.
Having an expert refute the commonwealth's scientific findings should have been the top priority of any competent defense lawyer, the court wrote.
"Such testimony," the court said, "would tend to show that Lt. Quinn incorrectly identified multiple points of origin for the fire, that his conclusions lacked scientific underpinning, and that his opinions conflicted with principles of forensic fire investigation that were widely accepted at the time."