New Evidence Suggests Teenager's Conviction in Triple Homicide Arson Case Is Based on Junk Science
By the time a 1995 fire was brought under control in the downtrodden Homewood neighborhood of Pittsburgh, three firefighters had died and a family was left homeless. Within hours, an investigator said he suspected arson.
But it would take agents from the U.S. Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) 14 months to charge Greg Brown Jr., then 17, of conspiring with his mother, Darlene Buckner, to douse gasoline on the basement floor of their Bricelyn Street home and set it on fire with a rolled up newspaper. ATF agents determined Brown and Buckner conspired to ignite the fire so their family could cash in on a $20,000 insurance policy.
A jury ultimately found the prosecution’s arson theory plausible enough to convict Brown of murder in the deaths of Captain Thomas Brooks and firefighters Mark Kolenda and Patty Conroy, the first female firefighter in American history to die in the line of duty.
Brown was sent to prison for life.
Though Buckner was acquitted of the murders, she served a probationary sentence for insurance fraud, which caused her to lose her nursing license.
While Brown has steadfastly maintained his innocence since he was locked down, he has lost every appeal he’s filed. Recently however, two major issues emerged in the high-profile case.
First, a noted fire dynamics expert who conducted an exhaustive origin and cause study of the Bricelyn Street fire, said the blaze was accidental and the fire dynamics theories used to convict Brown amount to nothing more than “junk science.”
Secondly, the testimony of a star witness who prosecuting attorneys relied on to convict Brown has come into question.
At trial, this witness said he overheard Brown admit in conversation on four separate occasions that he set the fire. Investigators from ATF insisted that his testimony was not coerced by the promise of reward money.
In two recent recorded interviews, that witness said he was offered up to $15,000 for his testimony by an ATF investigator, and that he was eventually paid $5,000 in cash, even though prosecutors refused to discuss the payments until they were recently verified through a U.S. Freedom of Information Act request.
In recent interviews, the star witness contradicted portions of his trial testimony and claimed that agents put so much pressure on him that he “had no choice” about whether or not to testify against Brown. His story changed nearly every time anyone asked him about implicating Brown in the fire. In the recorded interviews he said he was coerced by ATF agents.
Meanwhile, the mother of two of the star witness’ babies – who was later so severely battered by him that she had to leave the state – said in a recent interview that the witness repeatedly admitted he lied about Brown’s confession because he was promised money and freedom by ATF agents.
On May 11, some of those findings caused Allegheny County Common Pleas Court Judge Joseph Williams III to reopen Brown’s case, establishing a briefing schedule for an appeal, and setting the stage for a hearing to consider whether new evidence is enough to merit a new trial.
Since then, the Pennsylvania Innocence Project and David Fawcett, of Buchanan, Ingersoll and Rooney, a Pittsburgh-based law firm, have agreed to represent Brown for free. On Sept. 17, 2010, they filed a 38-page petition requesting a reversal of Brown’s conviction and either his release or a new trial. They also asked for an evidentiary hearing on the new evidence that has surfaced.
“These materials demonstrate the untrustworthiness of the only evidence presented at Petitioner’s trial which implicated him directly in the fire that occurred on Feb. 14, 1995, when he was 17 years of age, and they undermine any confidence in the integrity of the verdict,” the petition states.
The Allegheny County District Attorney’s Office has yet to reply to those claims, but in previous court proceedings, it has intensely opposed any relief for Brown, arguing that a jury has already found Brown guilty and appeals courts have upheld the verdict. Specifically, they argue Brown and his previous lawyers already presented expert testimony at trial. By law, they argue Brown has no right to re-litigate expert testimony.
Regarding the intensely litigated issue of payments to witnesses, the Innocence Institute provided all legal stakeholders in this case with:
• A copy of taped conversations with the star witness (1, 2) and his ex-girlfriend (1, 2), in which both claimed that reward money had been promised by ATF agents and subsequently delivered by ATF agents.
• Copies of two cashier’s checks written out by ATF for $5,000 and $10,000 to undisclosed recipients who provided information to prosecutors in Greg Brown’s case.
• A form letter from an undisclosed ATF agent to residents of Bricelyn Street dated Feb. 14, 1996, in which it’s written that ATF “is offering a $15,000 reward for information that could lead to the conclusion of this case.”
Allegheny County District Attorney Stephen A. Zappala Jr. refused interview requests and offered only a brief written response to several inquiries. First Assistant U.S. Attorney Robert S. Cessar dismissed allegations that ATF agents offered rewards in exchange for testimony against Brown prior to the trial.
“Long after the trial was over, ATF decided to pay reward money to two witnesses,” Cessar wrote in a letter to the Innocence Institute. “The agent who paid the reward money to those two witnesses on behalf of ATF stated that the witnesses were of course glad to receive the rewards, but they were also surprised because they had never asked for a reward and had never been told that they would receive a reward.”
That statement is disputed by at least three witnesses contacted by the Innocence Institute to speak about their involvement in the Bricelyn Street investigation. All of them said ATF agents told them they could earn thousands of dollars for testimony implicating Brown or his mother in the case.
Brown hopes the judge will go beyond technicalities to determine whether the fire was an arson, whether the jury heard everything about deals to witnesses, and ultimately whether a 17-year-old was the victim of blind injustice that demanded someone pay for the loss of three fire fighters’ lives.
“I can honestly say I did not do this crime,” said Brown, who is serving a life sentence without the possibility of parole at the State Correctional Institution in Dallas, Pa. (SCI Dallas). “To make this case worse, I don’t even believe there was a crime.”
“I want to clear my name,” he said. “I don’t want these [firefighter’s] kids … thinking I killed their parents.”
His mother, Darlene Buckner, who despite her acquittal has been castigated by the victim’s families for years, sadly laments: “I will never be free until my son is free.”
Dr. Gerald Hurst, 73, an Austin, Texas-based expert who has studied the Brown case, has been involved in several highly publicized cases where arson-related convictions were overturned after he questioned the veracity of what he calls antiquated arson science.
He arrived at his very specific conclusions after spending months investigating the scientific basis of the Bricelyn Street tragedy, without pay, at the behest of the Innocence Institute – a journalistic organization unaffiliated with any of this case’s stakeholders. Students of the Innocence Institute investigate and write about claims of wrongful convictions. They work to find the truth in all cases.
After poring over police reports, a video tape of the fire scene, and calculations ATF agents used to convince a jury the fire was arson, Hurst reached his major conclusion: In a rush to judgment, investigators overlooked evidence suggesting the blaze was the result of an accidental natural gas leak.
Hurst contends that calculations made by ATF agents to prove Brown poured gasoline on a basement floor and then ignited the fire that quickly engulfed the home were blatantly wrong and set a faulty scientific basis for the origin of the fire that permeated the entire case.
Hurst also said the lead investigator in the matter, ATF Agent William Petraitis, was incorrect to indicate that burn patterns created during the Bricelyn Street fire were caused by an accelerant, such as gasoline, referring to fire investigation manuals which stress that such burn patterns can be caused by several other circumstances.
While Petraitis had many years of experience as an arson investigator, Hurst suggested he should not have been permitted to testify as an expert in the case because his expertise did not extend to the fire dynamics used to prove the case.
“This is fundamentally a case where they got it damned wrong because they didn’t do what you have to do in every fire scene. You’ve got to eliminate all accidental and naturally occurring causes,” Hurst said, suggesting that was not done on Bricelyn Street.
…and a Reward
Star witness Ibrahim Abdullah’s trial testimony indicates that he spoke directly to Greg Brown on four occasions during which Brown admitted to starting the Bricelyn Street fire. He claimed at trial that he was not paid a reward for his testimony.
Recently, however, he told an Innocence Institute reporter that an ATF agent coerced his testimony with the promise of a $15,000 reward, and that he was eventually paid “only” $5,000 for his testimony. Furthermore, Abdullah said he merely overheard Brown talking about starting the Bricelyn Street fire, and that he thought Brown’s confession was “bullshit.”
Innocence Institute reporters also interviewed Adrian Williams, the mother of two of Abdullah’s children. She took Abdullah’s contradiction to new heights: Williams said Abdullah repeatedly told her that he lied on the stand for money and freedom from legal woes when he testified that Brown told him he set the fire.
According to Williams, Abdullah told her that he was Greg Brown’s roommate while the two served at Vision Quest, a juvenile offender program based in North Franklin, Pa.
Long after they left the boot camp, Abdullah was traced to another jail by ATF agent Jason Wick. Williams said the agent offered Abdullah $15,000 and then – over a long period of time and multiple visits – coached Abdullah through fabricated testimony to implicate Brown.
“[Abdullah] told me specifically that he did not have that information from [Brown],” she said, adding that she didn’t find out that he had lied under oath until years later, when an agent delivered $5,000 in cash to the home they shared.
“I asked him why [Brown] would offer such damaging information to him,” she said. “He stated that he told the ATF officers what they wanted to hear to get the money.”
She continued: “If I would have known that he wasn’t telling the truth, I would have encouraged him to do the right thing.”
Abdullah himself would not go so far to specifically admit that he fabricated his testimony. But in an interview with an Innocence Institute reporter, he said, “Man, that was 15 years ago,” and that he “can’t remember most of it.”
Without giving specific names, Abdullah said that as Brown’s trial date approached, he told ATF agents that “he didn’t want to be part” of the trial – and that he even attempted to escape lockup while the state held him for trial – but that he was given no choice. He claimed that, among other pressures imposed on him to testify, ATF agents placed him alone in a room with the Bricelyn Street fire’s victims families for as long as an hour, while the families urged him to testify.
“I didn’t want to testify,” he said. “My back was against the wall.”
Valentine’s Day Fire in the Alley
It was 15 degrees outside and around midnight on Feb. 14, 1995, when Darlene Buckner said she cajoled her son, Greg Brown Jr., into riding along to get ingredients for a dish she was to prepare for a family wake the next day. Neither of them dallied at the Giant Eagle grocery store at Waterworks Mall as they searched the aisles for macaroni, tuna, dressing, relish, green peppers and onions. Buckner would later produce a receipt for $35.91 in purchases time-stamped 12:37 a.m.
In line to pay for her groceries, Buckner made small talk with the cashier about Valentine’s Day plans while Brown walked alone to the front of the store to flip through magazines. After money and receipts changed hands and the groceries were bagged, the pair exited the store together and began the 10-minute drive home.
As they approached the hillside Bricelyn Street, they noticed smoke billowing into the frigid night sky. By the time they reached a nearby alley, Exley Way, it was clear the smoke and growing number of emergency vehicles were responding to a fire at their house – 8361 Bricelyn Street – where they assumed the rest of their family was asleep.
Buckner would later tell investigators her son jumped out of the still-moving car near Exley Way and ran up the embankment of their yard to find a fireman who told him everyone was out of the house. Meanwhile, a cop on traffic detail witnessed Buckner, two months pregnant at the time, alone in her car when she parked near his patrol car and ran toward her home. The cop did not see Brown, and later testified that he saw an infant’s car seat in the front passenger seat. Prosecutors would later claim that since neither the Giant Eagle checkout clerk nor the cop remembered seeing Brown – and because a child’s car seat was seen in Buckner’s front seat – Buckner’s son was not with her. A Giant Eagle security video was erased before lawmen and defense lawyers tried to obtain it.
Buckner eventually found Brown, her husband, Ron, their infant son, Fred, and Ron’s adult daughter, Catherine, who was with her young daughter. All of them fled the house in their pajamas, leaving them with only blankets supplied by neighbors and firefighters to keep them warm.
As the family watched the fire suppression efforts, they heard whispers from neighbors later confirmed by news reports that Captain Brooks and firefighters Kolenda and Conroy had died in their family room. Coroner and independent reports said their deaths resulted from carbon monoxide poisoning and a lack of oxygen. (Read “The Deadliest Fire Fight in City of Pittsburgh History,” below.)
During one of several interviews over the past year, Darlene Buckner said the magnitude of that tragedy wasn’t lost on her or her family. But she was forced at the time to focus on her own family’s plight. She wondered how long it would take to rebuild after losing nearly everything. At least, she thought, her family recently purchased a renter’s insurance policy after a relative’s house burned down. That policy would cover less than half the cost of their losses, she said, but it was a small comfort. She hoped their family church and fellow parishioners would also offer support.
At that point, she said, being an arson and murder suspect never crossed her mind.
A Pointed Investigation
Now-retired ATF agent Petraitis received an early morning call from Pittsburgh police and fire officials. They requested his team’s assistance in determining the cause and origin of the Bricelyn Street fire.
By that time, Petraitis had probed 300 or more fires since he joined the federal agency in 1971 – a year after he graduated with a Bachelor of Science degree in Environmental Planning and Design from Rutgers University. His formal fire dynamics training was limited to a three-credit course at the University of Maryland, but he had participated in numerous ATF seminars and workshops on those subjects during his career.
Part of that formal training involved studying the U.S. National Fire Protection Association’s 921 Guide to Fire and Explosion Investigations (NFPA921). A bible of sorts that establishes dozens of point-by-point considerations for fire investigators when they attempt to ascertain the cause and origin of a fire, the NFPA921 was first published three years before the Bricelyn Street fire.
The manual states:
• Investigators should take a systematic approach that includes collecting and analyzing data from any fire scene before developing and testing a theory. • Data collection and analysis should be done before an investigator solicits information from outside agencies and witnesses.
• The investigator should identify and collect evidence, which includes fire damage and burn patterns.
• The investigator should check the status of utilities, the role weather conditions could have played in the fire, and how the building’s structure could have played a role in the way the fire started or continued.
• All natural or accidental causes should be eliminated before a cause and origin ruling is made. The manual states: “Any determination of fire cause should be based on evidence rather than on the absence of evidence.”
By the time Petraitis and his team arrived at 8361 Bricelyn Street, the fire had been under control only 45 minutes.
Despite the bitter cold temperatures that night, homicide detectives, paramedics, fire officials, and utility inspectors shared the street with news crews and neighbors bundled in blankets and huddled in clusters to lament one of the worst losses of life in the history of the fire bureau.
After members of the City of Pittsburgh’s arson squad told him what they knew, Petraitis examined the exterior of the house. Then he stepped inside the front door – which opened onto Bricelyn Street – to survey damage to the main floor.
According to his statements and later testimony, Petraitis looked down into the basement and noted the collapsed stairs. Then, Petraitis and his team climbed the still-intact stairs to the top floor. They noted major damage there before descending the steps again to examine the kitchen and family rooms on the main floor. While the Buckners’ kitchen sustained minimal fire damage, part of the floor had collapsed.
“There had been a problem of some sort beneath the floor,” Petraitis would later testify.
Although curious to see the condition of the basement area directly below the sunken kitchen floor, Petraitis next examined the family room, noting that the small, cluttered space where the firefighters had died was damaged by heat, but otherwise unscathed. The team then walked outside and around to the backyard to access the basement through a rear door.
Inside, he scanned the basement and noted a furnace and water heater on one side, along with a double utility sink and a washer and dryer against the wall. He noticed clothes piled high on the floor. There, Petraitis discovered the source of the kitchen floor’s collapse: A portion of the ceiling joists on one side of a steel support beam had been consumed by fire.
That burn pattern caused him to believe that the fire started on the concrete basement floor and shot up a wall more than seven feet six inches to ignite the joists.
“[The fire’s origin was] in the vicinity of the metal cabinet in the center of the floor between the furnace and the metal cabinet, and to the left of the steel pole that bisects the rear basement,” he would later testify.
By the morning of Valentine’s Day, 1995, after the 14-year ATF veteran noticed circular burn patterns around the ceiling joists, he suspected arson.
What proved it, he later testified in court, was a circular burn pattern that developed where the floor joists had been completely consumed by fire. For that to happen, a powerful flame would have had to rage from floor to ceiling and mushroom out in all directions once it got there, igniting a number of joists simultaneously. The only way a flame could get that high would be from an accelerant like gasoline, he testified later.
Petraitis testified that he conferred with electric and gas company representatives, who had examined the appliances. He testified that the furnace was “opened up … taking the protective covers off of it” to confirm the interior wiring was intact and unmelted. The water heater, Petraitis said, “was examined and was in the same condition as the furnace.” A visual examination of the family’s washer and dryer, he said, also showed no signs of malfunction. No further tests were done, even though the Buckner family had reported electrical problems to their landlord just weeks before – issues with fixtures and flickering lights.
Petraitis said these issues were unimportant to his investigation because they were not within the area of the fire’s origin.
“There are two very significant areas of burning and that is the rear basement and the other is the very top floor,” Petraitis said later in court. “This particular fire spread from the basement up through this balloon construction,” he said, describing the house’s framing style, “and broke out on the top floor, where it met with its first horizontal barrier.”
Petraitis said a walk through the area and final examination of the scene yielded more clues. To defense lawyers, these were nothing more than devastating remnants of the loss of everything a family owned:
• A rolled-up newspaper charred at one end was discovered in a litter-strewn vacant lot about 100 yards from 8361 Bricelyn Street.
• A metal, one-gallon, half-full gasoline can was found in the basement where it was stored with the family lawnmower.
• Agents sifting through debris in the family room discovered the Buckners’ newly purchased insurance policy.
• Burn patterns were found, indicating an accelerant, according to Petraitis.
• A police K9 dog sniffed out two sections of the basement where accelerants were suspected near the furnace and a metal cabinet where the Buckners stored their summer clothes.
Within hours of the fire – before these items were back from ATF’s national lab – Petraitis concluded that someone poured gasoline on the basement floor and lit it on fire.
By 5:30 a.m., a member of the Pittsburgh Arson Squad submitted a report making it official for that agency, too.
Later that month, ATF received word that two of 12 evidentiary samples tested positive for accelerant, which was about the time University of Maryland fire dynamics professor – and Petraitis mentor – Dr. James Quintiere examined the remains of the home. He also concurred the fire was the result of arson. Only an accelerated fire, and not an accidentally occurring one, would produce flames high enough to reach the ceiling.
To eliminate other possible naturally or accidentally occurring fire causes, Petraitis later testified that he relied on oral and written reports prepared by certified electricians and other fire investigators. However, during cross-examination, he couldn’t specifically name any of these electricians or investigators.
To this day, Brown’s initial defense attorney, Al Lindsay, of Butler, Pa., said much of the information on which Petraitis based his expert opinion – referenced as both verbal and written reports – have not been entered into the court record.
As it turned out, the early analysis of the scene would be the only one available; the Buckner’s home was torn down weeks after the fire and months before charges were filed. The removed rubble included the furnace, water heater, washer and dryer and all potential forensic evidence the home contained. Why the home – a major crime scene – was razed later became an issue at trial.
Targets of Murder Probe
The Buckners’ Sworn Statement of Loss was listed as $52,898.87, and Keystone Insurance Company paid out $20,000. The Buckners eventually settled into a Penn Hills home. There, they began to rebuild their lives – even if it was without numerous personal mementos and family heirlooms. Darlene said it was clear that the family had been grossly underinsured, but she hoped through her job as a licensed practical nurse and her husband’s job driving a Pepsi truck that they could work to rebuild their lives.
To make matters worse, they soon learned the fire was ruled an arson. Even though no one in the Buckner family had ever faced criminal charges before, they also realized they were the focus of the ATF’s investigation.
While his mother implored Greg Brown not to do anything to give authorities reason to doubt their innocence, in March 1995 he was charged with drug and gun charges and sentenced to a 90-day stint with Vision Quest that began in May 1995.
That stint would change his life forever.
After he got out of boot camp in November 1995, Brown moved to Billings, Mont., to live with his aunt and uncle, and to escape Homewood – a gang-addled neighborhood with one of Pittsburgh’s highest crime and murder rates at the time.
Meanwhile, the investigation stalled. While Petraitis and his crew had documented cause and origin data, found the rolled up, partially burned newspaper in a vacant lot next door and had the circumstantial fact that a renter’s insurance policy was secured in the months before the fire, they had not found much else.
For many months, investigators could not find an eyewitness or other evidence implicating anyone, even though they had searched nationwide for corroborative evidence.
Then, ATF agent Jason Wick traced Brown back through the Vision Quest program, and began contacting Brown’s former bunkmates. Wick would later write in a report that two of them, Ibrahim Abdullah and Jesse Hughes, said the teenager confessed to setting the fire at his home. Despite that report, Hughes would later write letters to Brown’s lawyers saying he knew nothing about the Bricelyn Street fire or if Brown was involved. That left the ATF with Abdullah as the only former Vision Quest resident to testify. Two others who were in the same bunkhouse have also stated Brown never uttered a confession.
About 11 months after the fire, neighbor Keith Wright agreed to speak with investigators. Wright told agents he saw Brown across the street from his burning home before fire engines arrived. Wright also said Brown was watching the fire burn during the time he and his mother said they were at the Waterworks Mall Giant Eagle. The Innocence Institute sent letters to multiple addresses associated with Wright’s name and his court records, but did not receive a response.
Wright’s mother, Margaret McCord, in an interview with the Innocence Institute, said she was offered a reward – which she said she turned down – and contacted on multiple occasions by ATF’s Wick. She could not confirm or deny that her son – who was living on the third floor of their Bricelyn Street home while she and her daughter lived on the ground floor – saw Brown standing on the street as the fire raged.
“I don’t know what he saw,” she said. “But I know we were offered rewards for what we said.”
McCord said she did not know whether her son received any payments, and indicated she has not been in contact with him for years.
The ATF cause and origin findings, the confessions from jailhouse snitches, and the singular testimony from Wright was enough to file charges against Brown, which would be brought against the teenager and his mother on April 12, 1996, more than a year after the fire.
The pounding on Darlene Buckner’s front door that morning interrupted her as she was enjoying a cup of coffee and preparing breakfast for her two young sons as they watched Sesame Street. The pounding startled her boys, so she was already anxious when she opened the door to find uniformed police officers ready to read her Miranda rights. As an officer clasped handcuffs on her wrists and tossed a hooded sweatshirt over her face, Buckner was ushered out the front door.
“I remember asking [the female officer] if I could grab my purse,” Buckner recalled. “And she said, ‘Where you’re going, you won’t need a purse.’”
In the few seconds it took for officers to walk her from the front porch to the waiting cruiser, she worried about her two wailing young sons: Where would police take them? And would someone notify her husband?
Then, Buckner remembered the hooded sweatshirt that was obscuring her face and shook it off.
“Criminals needed to hide,” she said. “And I didn’t do anything wrong.”
At that time, Buckner didn’t know agents were also taking Brown into custody from the Montana high school where he had studied while living with an uncle after his release from the juvenile offender boot camp.
Darlene Buckner would be confined until after her acquittal.
Her son would never walk the streets again.
Trial and Error
The joint trial for Buckner and Brown began in January 1997 with a court gallery packed with onlookers, many of whom were uniformed Pittsburgh firefighters sitting at attention in a sign of solidarity and mourning.
“It was as if they were there to see that justice was done that somehow justify the deaths of three fine firefighters,” defense lawyer Lindsay said.
The prosecution opened its case by calling two of the firefighters who responded to the Bricelyn blaze.
They testified about the progression of the fire and outlined suppression efforts. Captain Edward J. Wyland, now retired, was the fourth firefighter trapped in the Buckner home during the suppression efforts and the last to speak to Brooks, Kolenda, and Conroy. He testified about the frantic last moments of consciousness he and the other firefighters shared in the house.
He said he encountered Kolenda, who said: “Eddie, I can’t breathe. I can’t breathe,” before Wyland said he “felt Kolenda collapse.”
“Sweet God in heaven, please save our souls because we’re all going to die in here!”
These were Conroy’s last words, Wyland said.
In an exclusive interview, Wyland said he knew his purpose at trial.
“[The prosecution] wanted me for the sympathy vote,” the retired fire official recalled recently. “They wanted me to make a victim impact statement at the end of the trial, but I declined.”
It was also during Wyland’s testimony that the question of water in the basement first arose. While the firemen described falling into the basement after the stairs collapsed under them, Wyland indicated that it was completely engulfed in flames. Then on cross-examination, defense attorneys quoted from an incident statement that indicated Wyland told a first responder that when he fell, there was water all over the floor from suppression efforts.
Other accounts of the fire response indicate that a hose leading down into the basement ruptured during the response, causing an inundation of water that one fireman claimed was hip-high. But the judge ruled that statement inadmissible because it was unclear who had prepared the report from which the statement was extracted.
As a result, Petraitis was never asked if basement water levels came into play during his investigation. Brown and Buckner’s defense attorneys, Lindsay, James Ecker and Sidney Sokolsky, argued that if they were allowed to pursue that line of questioning, they would have asked Petraitis what effect two feet of water would have had on his examination of the forensic evidence in the basement.
Also paramount to the prosecution’s case was the testimony about what Brown said to whom while he was in boot camp. ATF agent Jason Wick’s reports suggest he possessed written statements from Abdullah and Jesse Hughes, which detailed four occasions when Brown admitted to conspiring with his family to burn down their home for insurance money. While the handwritten statements were nearly identical, only Abdullah was called to testify. Before trial, Hughes, then confined to the State Correctional Institution at Greensburg on unrelated charges, wrote Lindsay twice, stating both times: “I don’t know nothing about [Brown’s] case.” Hughes was not a witness for the prosecution.
Abdullah – who had been in and out of the court system since he was 13 for fire-related and other crimes – testified that Brown repeatedly bragged about setting the fire while they lay in the bunks at night. Some later reports said the two only had four conversations ever. Pressed for details on the witness stand, Abdullah could recall little other than Brown told him that his family had promised him a cut of the insurance money for his part in setting the fire. Abdullah said Brown planned to use the money to purchase a car. Brown says no such exchanges ever occurred, but at trial, Abdullah testified he was not to receive money or freedom or anything else for his testimony. He said he came forward because “it seemed like the right thing to do.”
Abdullah’s testimony was damaging, even after Brown’s friend, Raoul Gibson, who was living in Charleston, S.C., testified that Wick met him at his home 650 miles from Pittsburgh with an offer. If he could get his friend to confess to the crime during a recorded telephone call, he’d reap thousands of dollars, Gibson testified.
Gibson testified he refused the offer, explaining he neither heard Brown take responsibility for the fire, nor did he believe his friend was capable of carrying out the arson. Gibson said placement in the federal witness protection program was also offered.
After Gibson’s account, he was challenged by Assistant U.S. Attorney Shawn Sweeney, who Lindsay argued in a later appeal “indignantly scoffed at” Gibson: “You think there is a fund somewhere with $7,000 that agents can just go grab and pay you?” the prosecutor asked. “Are we to understand that Jason Wick says, ‘Raoul, let’s take a walk, and I can give you $7,000 to call Greg Brown’?”
Gibson said those weren’t his exact words, but added, “He said, ‘Start at $7,000. There is a lot of money in it for you, Raoul.’”
Wick was never called to testify.
In November 2009, the Innocence Institute filed a federal Freedom of Information Act request over payments to witnesses. In August 2010, the ATF released a letter written to Bricelyn Street residents promising a $15,000 reward for information related to the Bricelyn Street fire, and copies of two cashier’s checks written out to undisclosed recipients for $5,000 and $10,000. The Innocence Institute’s appeal to determine the identities of the recipients was denied because ATF said identification of the payees was private information.
When Lindsay learned of the verified payments, he took umbrage with the government’s responses to repeated demands that they disclose payments to witnesses. “[It] wasn’t so much that they denied [payments],” he said recently, “but rather [that] they implied it was an unthinkable thing that they would pay money to witnesses.”
Now Abdullah and the mother of Abdullah’s children say he was the recipient of the $5,000 payment long after the trial.
“I wish I could say I was shocked,” Lindsay said. “I’m shocked in the sense that it confirms what we’ve always suspected: that there was money paid to witnesses. Give them money to testify in a certain way and don’t disclose that … there could be no more significant impeachment evidence than a witness getting paid to say this.”
Then there’s the question over who got the second payment. Margaret McCord, Keith Wright’s mother, says Wick pursued both of them during multiple visits and offers of reward money. While the federal agent wanted McCord to testify that she saw Buckner but not Brown in her car outside the fire, her son’s testimony would become a linchpin of the case – even if he did not come forward for nearly a year and his testimony was vague.
At that late date, Wright told ATF agents that just after midnight on Valentine’s Day in 1995, he saw smoke billowing from the house three doors away. Fire equipment, he said, had not arrived. When he looked closer, he told them he saw Greg Brown across the street from the burning house watching the fire burn. Wright said Brown was wearing a hat, coat and jeans, but could not specifically describe the clothing.
That eyewitness account contradicted the defense’s assertion that both Buckner and Brown were at Giant Eagle when the blaze was first discovered. This information became critical when neither the grocery store clerk nor the cop at the fire scene that evening recalled seeing Brown with his mother. While the Giant Eagle had a surveillance camera, the tapes could not be found when requested by their attorneys because so much time had passed between the night of the fire and the arrests.
On cross-examination, defense lawyers asked Wright why he never called 911, and why it had taken so many months for him to come forward with the eyewitness account.
He never answered the question.
“On that particular street there isn’t too much that happens that people don’t see,” Wright said of the relatively remote, hillside location above Frankstown Road in Homewood.
Defense lawyers later called McCord, who testified that she called her son to tell him about the blaze after firemen arrived on scene. She assumed at the time that she had awoken him when the phone rang more than twice. This testimony challenged the notion that Wright saw anything prior to the arrival of firefighters.
“Did you hear your mother testify under oath in this courtroom a few moments ago that she in fact called you on the telephone to tell you that there was a fire a few doors down the street and woke you up? Did you hear her say that, sir?” James Ecker, Buckner’s lawyer asked him.
“If I’m not mistaken, my mother called,” Wright said. “She didn’t know if I was asleep or not.” In fact, he said he was kept awake from traffic throughout the evening on the little-traveled Exley Way.
Also curious was Wright’s description of the scene during the early stages of the fire. Along with the timing of his sightings – which his own mother questioned – there were other inconsistencies: While Wright described the backyard as well lit with thick smoke, a Pittsburgh fireman described a scene with minimal smoke. The first responder also said poor lighting made it difficult to see clearly.
Petraitis was then called to defend his arson determination. On the stand, Petraitis testified that in the 300 cases he probed, he determined the cause and origin in 10 of those. He concluded all were arson. At the time of trial, he was not certified by the International Association of Arson Investigators.
Petraitis explained his use of a fire dynamics equation taught by Quintiere in his book, Principles of Fire Behavior, which he said helped him determine that between a half-gallon and two-thirds of a gallon of gasoline was used to create flames tall enough to ignite ceiling joists 7-feet-6-inches up. Petraitis’ calculations indicated his hypothetical gasoline spill would have created more than 13-foot-high flames, but he was not asked how long a fire started with that amount of accelerant could sustain itself.
Hurst later said there was not enough accelerant to legitimize Petraitis’ calculations.
The defense contested everything about Petraitis’ testimony. They unsuccessfully challenged his claim that he was a fire dynamics expert, arguing that his experience as an investigator did not qualify him to testify as an expert regarding scientific forensic analysis.
He was challenged on his use of burn patterns, which he said clearly indicated an accelerant had been used to start the fire.
“The way [the joists] were burnt away in a circular pattern indicated that the flame impinged upon the ceiling and mushroomed out in a radical fashion,” he said. “In order to create the circular pattern in the joists and in the tongue and groove of the basement, you would have had to have a flame that would impinge on the ceiling and proceed out in each direction” – another declaration Hurst would take to task some 14 years later.
At trial, the defense’s expert witness, Boyd Cochrane, an arson expert from Ohio who had investigated more than 100 fire and explosion cases, noted for the first time the hip-high levels of water in the basement from suppression efforts. Cochrane suggested that the half-filled gasoline can sitting next to the family’s lawn mower could have bobbed around on the water, releasing vapors and spillage that could have contaminated items throughout the basement. Cochrane also refuted Petraitis’ use of burn patterns as an arson indicator and said the investigator failed to eliminate all other possible causes.
Cochrane’s opinion did little to sway the jury.
Buckner said she knew things were going awry when he held a rendering of the Buckner family home upside down while explaining the scene to the jury. Both Brown and Buckner took the witness stand to claim their innocence, but defense attorney Lindsay said Brown’s use of street slang made him look less credible than his more articulate mother, which his lawyer said “was an affect that would not be appealing to a white suburban Allegheny County jury.”
As the defense wrapped up its case, Buckner said she and Brown hoped inconsistent eyewitness accounts coupled with character witnesses and the time-stamped Giant Eagle receipt – even if the store clerk said she never saw Brown and the store’s surveillance video tape of that night was lost – would be enough to sway the jury toward not-guilty verdicts.
Then, hours before its closing arguments, defense lawyers learned about a letter – the same one the Innocence Institute would receive in reply to a Freedom of Information Act request – that had been sent to the residents of Bricelyn Street offering a reward for information about the fatal fire. Defense attorneys lobbied for a hearing to reopen the evidentiary process, arguing that if the prosecution had turned over the letter in discovery, it could have been used to impeach Abdullah, Wright and McCord’s testimony and bolster Gibson’s accusations that money for testimony was being offered.
The trial judge refused to reopen testimony. In a subsequent appeal, a federal judge ruled that defense lawyers did not produce enough evidence to prove the money offer violated Brown’s right to a fair trial. The judge also ruled that defense attorneys provided no evidence that money or anything else was promised or given to witnesses.
“Nothing exists here … to suggest that any of the Commonwealth witnesses were paid or provided money to testify, or that they were the recipient of any reward money by virtue of their testimony,” wrote U.S. District Magistrate Judge Robert C. Mitchell, who was later upheld by another federal jurist. Both agreed with a prosecutor’s argument that the defense was on a fishing expedition that “failed to develop a factual basis.”
That, according to Brown’s lawyer, was because prosecutors hid information from him that prevented him from asking questions of the witnesses about payments.
After lengthy deliberations, the jury found Brown guilty of one count of arson and three counts of second-degree murder. He was acquitted of conspiracy to commit insurance fraud.
The jury found Darlene Buckner guilty of insurance fraud, but acquitted her on all other charges.
Both were sentenced on April 21, 1997.
Brown received three consecutive life prison terms without the possibility of parole for the homicide charges, and an additional five to 10 years for the arson charges.
Buckner received three years probation, 500 hours of community service and was also fined $5,000 for insurance fraud.
“When they read the verdict, my heart was beating out my chest,” Darlene Buckner said recently. “I thought, ‘This stuff just happens in the movies.’”
“I’m in here for one of the biggest frame-ups in American history,” he said.
Vetting the Investigation
Hurst is one of a handful of noted independent specialists who have worked for free to help discredit arson cases fraught with what he calls “folk lore” and “junk science.” In most of the cases he’s attacked, he’s said charges were dropped and sentences were overturned after so-called arson indicators were scientifically debunked.
At this point in his career, Hurst can afford to work pro bono. He holds a Ph.D. from Cambridge University and broke into the science field during the moon race, when he worked on rocket propellants for NASA before moving into the area of covert warfare during the Vietnam War. He later founded an explosives company, was responsible for developing and inventing most of the mixtures used in binary explosives today, and developed patents for the first Mylar balloon. He also developed the newest formula for Liquid Paper (which his wife, Gay, a former secretary, teasingly refers to as his greatest accomplishment).
In 1972, Hurst retired into full-time consulting, working for the prosecution in product-liability cases to determine if a particular product was responsible for a fire. He said in those cases, high-powered attorneys fought in court to determine who was at fault and who was responsible for paying the ensuing damages.
“These were intellectual battles,” Hurst recalled. “And so I was miles away from the arson scene.”
But then came Sonia Cacy.* Her uncle, who had raised her from childhood, had died in a 1997 house fire. Investigators suspected arson. Cacy was convicted of murder. Her alleged motive was to claim the cash in his bank account – which amounted to a few hundred dollars. After she was tried and convicted based largely on a medical examiner’s claim that traces of gasoline were found on his clothing, her lawyers contacted Hurst. He read her file and saw the same kinds of what he considers unsound, dubious fire science that he sees in Greg Brown’s case. Hurst testified in Cacy’s case, debunking the prosecution’s arson indicators. That testimony, coupled with updated lab testing that showed there was no accelerant on her uncle’s clothing, helped exonerate Cacy.
“I thought Sonia Cacy was probably the exception, and I found out that bad fire investigation was the rule,” Hurst said. “I was getting bogus cases left and right and that has continued unabated.”
But his most damning – and highly publicized – case was that of Cameron Todd Willingham, a Texas man convicted in a 1991 fire that killed his three young daughters. Hurst was one of many scientists who concluded that the fire Willingham allegedly started was accidental – not arson. Despite those findings, Willingham was executed in February 2004 after Texas governor Rick Perry refused to concede that the State of Texas was executing an innocent man. Since then, extensive investigations by the Innocence Project of New York, as well as the Chicago Tribune, the New Yorker, and ABC’s Nightline have suggested that the case was fatally flawed.
“I am an innocent man – convicted of a crime I did not commit,” Willingham said before he was put to death by lethal injection.
Not only have scientists like Hurst debunked evidence in arson cases across the nation, but in a sweeping indictment of forensic science evidence, the National Academy of Sciences issued a scathing report in 2009. It states long-used methods such as firearms identification and arson analysis have not “been rigorously shown able to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source.” It said police and prosecutors routinely bolster cases using faulty collection and documentation methods, and that poorly trained technicians often exaggerate their accuracy in front of juries who don’t know better.
Hurst said that report, plus the publicity stemming from the Willingham case, helped prompt awareness and government action nationwide. In 2005, Texas created a commission to investigate allegations of forensic errors and questionable forensic scientific conduct, but Hurst said his work on behalf of the wrongfully convicted is far from over.
“Now is the time to attack these old cases,” Hurst said. “You have all this bad fire investigation done for the past 30 to 40 years and you have people sitting in jail all over the United States based on this crap.”
After pouring over evidence the Innocence Institute provided that included video tapes of the crime scenes and hundreds of pages of reports and transcripts documenting other evidence, Hurst came to believe Greg Brown is one of them.
Revisiting the Investigation
After Hurst’s review, he said the investigation in Greg Brown’s case didn’t come close to a comprehensive cause-and-origin analysis.
He said the ATFs failure to eliminate all plausible accidental and naturally occurring causes before making a determination doesn’t adhere to NFPA921 standards – which are set to ensure reliable fire investigations.
“Even if you think you are positive, you still have to eliminate all the reasonably possible natural and accidental causes,” he said.
Hurst said the investigators’ reliance on burn patterns as an indicator of arson goes against the NFPA921, which states that in most cases no single item is sufficient in itself.
While ATF interpreted the plume burn patterns as proof of the presence of accelerant, the manual states the same markings can occur when hot gases from a working fire meet with a cool surface.
“Careful interpretation of these patterns should be exercised, since they may be mistaken for patterns originating from ignitable liquids,” the NFPA921 dictates.
Such conditions, Hurst said, were present on Bricelyn Street that night.
“Most of the patterns that resulted in people being prosecuted and convicted were actually not pour patterns at all,” he said.
Then there is Hurst’s allegation that the ATF agent incorrectly used the fire dynamics equations taught by Quintiere.
After recalculating Petraitis’ numbers, Hurst said the flames would have reached 10 feet, not 13. While this still would have been high enough to reach the joists, he said it would not burn long enough to light them on fire. Petraitis also confused a pour of gasoline with a pool.
“Here’s his first error: working with the false assumption that a gasoline spill one meter in diameter will generate the same [energy] as that of a laboratory pool,” he said, citing government research that debuted in the late ’90s.
“It turns out that a spill of gasoline in a given area does not produce the same heat that a pool of gasoline uses. The measurements that were relied on by Petraitis was a pool, which is an open tank with a finite depth; a depth measured in inches.”
Without a deep pool, Hurst said, the flames would not have been as high or burned as long as would be needed to start the house on fire. When he used the correct input data, Hurst said he discovered that much more gasoline would have been needed to achieve the 13-foot flames that would burn long enough to spread throughout a house like the one at 8361 Bricelyn Street.
Despite his misgivings about Petraitis’ calculations, Dr. Hurst conceded “… it sways juries. Juries listen to it and they think they know what he is talking about.”
Then there are the rational problems with Petraitis’ theorized gas-spill scenario.
“If I pour gas on a concrete floor it is true I can get a very tall flame. But I could only get it for 20 seconds or less at full output and maybe a total burn time of a minute, which is touch-and-go trying to ignite something as heavy as joists,” Hurst said. “Fire investigators always think gasoline has magical properties. We are fed this in movies, and it simply isn’t true.”
Hurst also wondered why the court allowed Petraitis to testify as an expert, but said the Buckner-Brown trial was held at a time when the qualifications for such scientific forensic testimony was ill-defined.
The technical issues aside, Hurst said two eyewitness reports submitted by firemen on the scene – and given short shrift at trial – solidified his personal theory of what caused the Bricelyn Street blaze.
“What you’ve got I’m almost sure is a big natural gas leak,” he said.
He refers to several pieces of evidence, including an incident report from a first responding fireman. In that report, Jerome Jones told homicide detectives he saw a jet of blue flame in front of the window near the furnace of the Bricelyn Street home. Jones said blue flames jetted towards the left center of the basement. The firefighter said it seemed to be splitting and spreading up and out, appearing to be fed by something.
Hurst then points to observations made by fireman Joseph Somma, who reported recognizing the presence of natural gas.
“We started moving in,” Somma wrote. “The fire would not go out because of natural gas in [the] kitchen.”
“A natural gas fire would be able to do what a gasoline pour couldn’t – provide a sustained burn that ignited the massive ceiling joists given the sparse ventilation because it can use what air is available and concentrate it,” Hurst said. “The fire fits [the natural gas scenario] like a glove, which fits two separate eyewitness accounts from two firemen.”
At this point, Hurst said there is no doubt in his mind that Brown is innocent.
“There is no way to counter this evidence. They did check the furnace and water heater but they didn’t check out anything else,” Hurst said.
While neither ATF officials nor Petraitis responded to written requests for interviews, Quintiere did, calling the Texas scientist and his findings “off the wall.”
“I know this science and scientists, and I don’t know Dr. Hurst,” the University of Maryland professor, who works extensively with the ATF, said. “Yet, Bill Petraitis teaches this stuff and is certified. I bet Mr. Hurst has never dug out a fire scene.”
He said allegations that Petraitis did not conduct a complete investigation are unfounded, as are accusations that the ATF agent was unqualified to testify at trial as a cause-and-origin expert.
Quintiere confirmed he had visited the scene, but he said he did not testify as an expert because Petraitis is “perfectly capable” and had more intimate details from witnesses.
“You can’t impugn the credentials of Bill Petraitis,” said Quintiere, who offered Petraitis an acknowledgement in his book.
While Quintiere said that gasoline pours and pools – which are the subject of a chapter in his well-regarded book – do have different energy output values, he said Hurst grossly overestimated the margin of error.
“There is only a small difference between a pool and a pour – about 20 to 30 percent,” he said. But even that error would cause flames high enough to impinge the joists, Quintiere said. He said the oxygen-rich void spaces typical of the construction of the Buckner home would have carried them to the upper floors.
And Quintiere said he does not assign much significance to the blue flames reportedly seen in the basement area.
“Mr. Hurst’s science is fouled. Blue flames may also be associated with ventilation-limited fires, which was the case at Bricelyn Street,” he said, explaining that blue flames are also noticed when ventilation-limited fires are close to extinction. He said he still believes Petraitis’ theory of a gasoline accelerated fire is “dead on.”
“If you take a slow-burning fire of clothing on the floor, it would use all the available oxygen and quickly go out,” he said. “If you spill an accelerant, it will give a much taller flame with a larger diameter.”
“What you need for the fire to carry on is the floor joists getting involved. You’re now going to pull air into that space, and the balloon construction pulls those flames up to the attic,” he said.
Quintiere said it was “not likely” that a normally occurring fire would have had the strength to reach and involve the floor joists.
“This has to be a pretty big fire, and what is capable of doing that is an accelerant,” he said.
He said issues surrounding the amount of gasoline used and the misuse of the equation are “secondary.”
But Hurst disagreed: Those ancillary details are all important to a scientifically sound investigation, he said. And he revisited what he said is his central complaint: The lack of elimination.
“The cardinal rule in arson investigation is not whether one can construct a theory which is consistent with an incendiary cause but whether one can eliminate other reasonably possible accidental causes,” Hurst said. “If an investigator fails to eliminate accidental causes, he must declare the cause of the fire to be ‘undetermined,’ despite the presence of what he may believe to be evidence of an incendiary cause.”
Hurst reiterated that the flame was described as more of a churning blue jet, not a wavering flame, once again arguing a natural gas leak fueled the fire.
In Brown’s case, he said the first firefighter on the scene noted a jet of flame moving from right to left across the small basement window and directly toward the heavily damaged area of the ceiling. Given that the basement had no exterior ventilation other than the small window through which the flame was seen, Hurst said the “most obvious source of such a continuing momentum jet” would have been a natural gas leak.
“The fire investigators checked only the gas-fed furnace and claimed to have eliminated that unit as the source of the fire. They did not inspect the gas feed system or check to see if there may have been a problem with underground gas leaks in the neighborhood,” Hurst said.
“Could you have had some kind of other fire? Possibly,” Quintiere said, but he then added that the circumstantial evidence ATF had in the Bricelyn Street fire was compelling. “If someone had that alone it would be enough,” he said. “But if it’s a confession – even a confession by proxy – that’s a slam dunk.”
Quintiere did not respond to a September 2010 query in which he was told about payments to witnesses and suggestions some of them lied under oath.
After learning about the new information in the Brown-Buckner case in August, juror Nancy Fitzgerald said the cause-and-origin theory proposed by Hurst would have weighed heavily on her opinion of the pair’s guilt or innocence. With a background in chemistry, she remembers finding it difficult to believe that gasoline spilled on a concrete floor would have been able to create flames severe enough to produce the damage shown in the court exhibits.
Fitzgerald, of Edgewood, said that if the natural gas theory had been presented at trial, it would have struck her more than news that two witnesses associated with the case had received a total of $15,000 in cash payments from ATF.
While letters detailing the new information in the case were sent to all 12 jurors and both alternates, those individuals either indicated they did not want to be interviewed or did not respond to requests for interviews.
During repeated interviews in the summer of 2010, Ibrahim Abdullah said he merely overheard bunkmates Thomas Blanton and Greg Brown talk about the fire – a recounting of events that contradicts Abdullah’s trial testimony in court. In two interviews, Blanton claimed that he did not have any discussions at all with Brown while Abdullah and Brown bunked together.
In addition, the mother of two of Abdullah’s children who said she lived with Abdullah when he received $5,000 in cash from ATF, says Abdullah lied for money.
Adrian Williams says her stormy relationship with Abdullah, who was listed in court documents as 6-feet-two-inches tall, 380 pounds, was in full swing when he secured thousands of dollars from ATF in exchange for his testimony.
“He made that information up,” she said. “He told them what they wanted to hear.&