Wednesday, July 6, 2005
Death row inmate appeals over flawed evidence
Bullet analysis now discounted
By Andrew Wolfson
awolfson@courier-journal.com
The Courier-Journal
EDDYVILLE, Ky. -- Ronnie Lee Bowling sits on death row at the
Kentucky State Penitentiary for two murders he claims he didn't
commit -- convicted in part through forensic analysis that scientific
experts now say is deeply flawed.
No eyewitnesses or direct physical evidence tied him to the execution-
style slayings of two gas station attendants in Laurel County a month
apart in 1989. There were no fingerprints, no DNA.
But at Bowling's trial, an FBI examiner testified that tests showed
some of the .38-caliber bullets recovered from both bodies -- and
from bullets in an attempted murder at a third station in Rockcastle
County -- came from the same batch of lead.
More crucially, FBI Special Donald Havekost said, they matched some
of the cartridges found in Bowling's mobile home.
Havekost told the jury he'd never seen a bullet match in crimes that
weren't related. Asked about the likelihood of finding another bullet
matching Bowling's, Havekost said, "I'd spend a good part of my life
looking and probably never find it."
But last year, a National Research Council panel found that FBI
examiners repeatedly had failed to tell juries that such bullet
matches might be mere coincidence. The bureau responded by suspending
lead analysis, which it had used in 2,500 cases over 40 years.
Then in March, the first conviction in the nation was reversed based
on concerns about the reliability of bullet-lead analysis. Ordering a
new trial for a man convicted in New Jersey of murder, an appeals
court said "the integrity of the criminal justice system is ill-
served by allowing a conviction based on evidence of this quality …
to stand."
The prosecutor in the Kentucky case said there was plenty of other
evidence to convict Bowling, now 36.
But Bowling is staking his life on the argument that he wouldn't have
been convicted without that forensic evidence -- arguing on appeal
that the new disclosures about such testing demand that he get a new
trial.
"If there is any truth and justice in this country, my wrongful
conviction shouldn't be allowed to stand," he said in a two-hour
interview at the penitentiary, where he has been held since Dec. 24,
1992, for the murders of Ronald L. Smith, 28, and Marvin Hensley, 48.
Overstated significance
After a yearlong study, a research council panel reported in February
2004 that FBI analysts repeatedly overstated the significance of
bullet-lead matches and underplayed the likelihood of a false match.
"FBI examiners should not rely on bullet lead analysis to testify in
criminal cases about the statistical probabilities that a crime-scene
bullet originated with the defendant," concluded the report of the
Washington-based council, the operating arm of the private, nonprofit
National Academy of Sciences and National Academy of Engineering.
Bowling cited that report and other attacks on bullet-lead testing in
motions seeking a new trial that he's filed in Laurel Circuit Court
and U.S. District Court in London. No execution date is pending,
although he is entering the last phase of his appeals.
One of his lawyers, Vince Aprile, likened bullet analysis to hypnosis
and Victorian-era phrenology -- the abandoned belief that
intelligence could be determined by the shape of the head. Aprile
said the national council's report "exposes it as an insidious
endeavor to clothe biased speculation in the impressive robes of
science."
But relatives of the victims say Bowling is just grasping for straws.
And former Commonwealth's Attorney Tom Handy, who won Bowling's
conviction and has since retired, said, "I don't have any doubt that
he was the perpetrator."
Handy concedes that bullet evidence was important to his case -- he
told the jury in his closing statement that the trial came down to
"cold, analytical facts … tested by an FBI agent."
But he said that stronger proof came from Ricky Smith, owner of the
third station, a Sunoco in Rockcastle County. Smith, who is not
related to Ronald Smith, testified that three days after Hensley was
murdered, Bowling came into his station and opened fire, then fled,
leading police on a 32-mile chase that ended with Bowling's capture
outside his home in Clay County.
State police who pursued him later found a gun along the road and
determined, through traditional ballistics testing, that it was the
same weapon used in the murders.
Bowling claimed that he hadn't shot at Ricky Smith and that the gun
wasn't his. Bullets found at the Sunoco station were too mangled for
conventional ballistics tests, in which marks on bullets from the
crime scene are compared with marks on test bullets fired from the
same gun.
That is where Havekost came in.
He testified that one of the bullets from the Sunoco station had the
same chemical makeup as five of the bullets found in a partially
filled box of ammo at Bowling's home, meaning, he said, that they
originated from the same manufacturer's batch of bullet lead.
He also said that one bullet from each of the two murders matched and
that three of the bullets that killed the second victim, three of the
bullets found at the scene of the third shooting, and 16 of the 24
bullets found in Bowling's home all matched.
How testing works
The FBI did bullet-lead analysis in cases where no weapon was found
or where ballistics analysis was impossible. Since 1986, five
examiners testified 521 times about bullet-lead tests, the bureau has
said, although it's also said it doesn't know where or against whom.
The Washington-based Forensic Justice Project, part of the nonprofit
National Whistleblower Center, has sued the FBI to get that
information, contending the government won't release it for fear it
might trigger hundreds of new trials.
No one knows how many times bullet-lead analysis has been presented
in Kentucky courts, but it figured in the case against Shane Ragland,
whose conviction for the murder of a former football player was
reversed on other grounds last year by the Kentucky Supreme Court.
Another defendant, Tyron Anton Wilcox, was acquitted of murder in a
2002 trial in Jefferson County in which an FBI examiner presented
bullet-matching evidence.
Here's how it works:
Bullets from the crime scene are tested for trace amounts of seven
elements, including arsenic and tin. The same test is conducted on
bullets linked to the suspect.
If bullet lead from both sets is found to contain the same amounts of
those elements, FBI examiners have historically concluded that the
samples came from the same batch of lead. Some examiners went so far
as to testify that the bullets came from the same box.
The findings have rested on twin assumptions -- that every batch of
lead was consistent throughout and that no two batches were exactly
the same. The FBI's own collection of about 13,000 bullets seemed to
confirm the theory: no two bullets in it seemed to match unless they
came from related crimes.
But a statistician hired by the FBI found the bureau's database was
too small to draw valid conclusions. And other researchers, including
the FBI's former chief metallurgist, William Tobin, began challenging
the findings.
Tobin and colleagues went to lead smelters and found that different
samples from the same batch sometimes had different compositions, and
that samples from different batches sometimes had the same
composition. In other words, a bullet match might just be a coincidence.
"It's like saying that because two people have the same chemicals in
the blood, they have the same parents," said Tobin, now an
independent consultant and expert witness whose pro bono research
helped win the New Jersey defendant a new trial and Wilcox his
acquittal.
His research also found that in rural areas, where there are fewer
stores that sell ammunition, the chances are great that different
customers will buy bullets that match.
In Juneau, Alaska, for example, the odds that an "innocent" purchaser
would buy bullets likely to match the "suspect's" bullet ranged from
87 percent to 100 percent, depending on the brand, he found.
Tobin, who also has advised Bowling's lawyers, said the bullet
testimony in that trial was scientifically invalid and unsupported.
"If somebody is guilty of capital murder," he said, "send them to the
electric chair. But don't do it with comparative bullet-lead analysis."
The National Research Council panel, for its part, found that, while
the actual testing of bullet lead is sound, the testimony of FBI
examiners often was not -- that juries weren't told about the
possibility of coincidental matches or that a single batch of lead
can produce between 12,000 and 33 million bullets.
"It would be like a Nike shoe print, size 10, found at a crime
scene," said Case Western Reserve University law professor Paul
Giannelli, who served on the committee. If a suspect were found with
the same brand and size sneakers, "it would be admissible in court,
but there would be a large number of people with that type of shoe."
A spokeswoman for the FBI laboratory, Special Agent Ann Todd, said
the bureau is in the final stages of reviewing the National Research
Council's report.
In an interview, Havekost, who retired from the FBI soon after
Bowling's trial and now lives in Nebraska, said that if he'd been
asked, he would have testified that many people could have bought
matching bullets in a rural area like Laurel County.
But nobody asked, he said.
Other evidence
Ricky Smith had just opened his Quality Sunoco station at 6 a.m. on
Feb. 25, 1989, when Bowling came in asking about a job, then pulled
out a revolver and started shooting, according to court records.
Smith ducked for cover, escaping injury, and returned fire with his
own gun, he testified, grazing Bowling's head with one bullet.
Bowling didn't testify at his Laurel County trial but has since
claimed that he wasn't armed and that Smith started shooting at him
because he was nervous after the recent murders in a neighboring county.
Asked why he would run from police if he hadn't done anything wrong
-- state troopers testified they chased him at speeds up to 100 mph
-- Bowling said an interview that he was rushing home so his wife
could take him to the hospital.
To win a new trial based on newly discovered evidence, Bowling's
lawyers -- Aprile and assistant public advocates David Harshaw and
Susan Martin -- will have to show there is a "reasonable certainty"
that his trial would produce a different result with the recent
disclosures about bullet-lead analysis.
In an interview, the foreman on Bowling's jury said the bullet
evidence wasn't much of a factor for the jury. "I don't remember
anybody really bringing that out in deliberations," said Lisa Trett
of Corbin.
Jackie Steele, the Laurel assistant commonwealth's attorney who is
fighting Bowling's motions, noted that Bowling was convicted of
attempted murder at a separate trial in the Rockcastle shooting
without bullet-lead analysis. He was given an additional 20-year
sentence on that charge.
Steele said there was plenty of other evidence to support the Laurel
County conviction.
Two witnesses, for example, said they saw Bowling at Hensley's
station a couple of weeks before he was murdered, and that he seemed
to be casing the store. Another witness, who had shared a jail cell
with Bowling, testified that he'd confessed to shooting a service-
station owner five times in the back.
And while Bowling had told police he didn't own a .38-caliber gun,
the prosecution at trial presented a family picture showing Bowling
with that caliber weapon in the background.
Bowling and his lawyers dispute that evidence.
They say the men who saw Bowling at the first station were friends of
Hensley whose court testimony went beyond what they initially told
police.
In a two-hour interview in the penitentiary's visiting room, Bowling
said the gun belonged to his father, as his father testified at
trial, and that the jailhouse snitch -- Timothy Chappell -- "lied and
ratted his way" to win lenient treatment on pending federal charges.
Court records show that Chappell, who faced 20 years in prison on
four counts of mail fraud, pleaded guilty to one count and got a two-
year sentence. He was later convicted as a child molester, according
to Georgia's sex offenders Web site, which says he has failed to
register his most recent address. He couldn't be found for comment.
'Horrific crimes'
Bowling has been incarcerated for more than 16 years. He said he
spends his time reading, studying Zen Buddhism and working as a legal
aide for the 34 other men on Kentucky's death row. A prison
spokeswoman said he has had only one disciplinary write-up in 13 years.
He acknowledged in the recent interview that the Laurel County
murders were "horrific crimes."
One of the victims, Hensley, was known for frequently giving gas,
food and shelter to people in need. Besides operating his station, he
was pastor at Lighthouse Assembly of God Church near Mount Vernon.
Ronald Smith, the other victim, was divorced and had one daughter. He
lived with his mother, Oda Proffitt, who said in an interview that
her son was just starting to get his life together. "I think about
him every day," she said.
But Bowling said there is nothing that connects him to the Laurel
County crimes other than bullet-lead analysis. "The link doesn't
exist," he said. "They misled the jury in my trial by presenting
evidence that was flawed, unreliable junk science."
http://www.courier-journal
