tl
Member
Not ballistic fingerprinting, but thought it might be interesting to some (from the Des Moines Register):
http://desmoinesregister.com/news/stories/c4788993/23493859.html
Washington, D.C. - A National Research Council report to be issued today questions the FBI's use of bullets as evidence in thousands of criminal cases, upholding the findings of Iowa State University researchers, according to congressional aides familiar with the report and its written summary.
ISU researchers in 2000 completed an FBI-financed study that raised similar challenges to the bullet analysis technique, but, the researchers said, it was kept quiet because the findings were not to the agency's liking.
The ISU study, which cost the FBI $100,000, surfaced after a reporter obtained it through a Freedom of Information request. Defense lawyers have used the study in at least three murder trials since. Facing criticism for its continued use of the bullet analysis, the FBI requested the National Research Council report.
Aides to Republican Sen. Charles Grassley of Iowa were briefed on the new report Monday. They said the report's findings probably will not mean the reversal of thousands of cases in which the bullet technique was used, although there's no guarantee. Often the technique was just one piece of evidence, aides said.
However, the report suggests that the FBI use a more rigorous statistical method in its analysis. It also recommends that FBI expert witnesses be more precise and uniform when describing their findings to juries. Grassley aides said the FBI will not resist those suggestions.
"Attorneys, judges, juries and even expert witnesses can easily and inadvertently misunderstand and misrepresent the analysis of the evidence and its importance," said the report summary.
FBI experts who testify should not say that two bullets that cannot be distinguished from each other came from the same production run, or box, or even were made the same day, the report said. "None of these assertions . . . can be justified by the available data," it said.
Grassley, a member of the Senate Judiciary Committee, for years has been a harsh critic of the FBI crime lab and an advocate of whistle-blowers who brought public attention to what they said was sloppy work at the lab not backed up by good science.
"It is absolutely crucial that law enforcement use science that is based on valid, proven techniques," Grassley said in a letter to the National Research Council.
"Law enforcement must follow the facts, and let the truth convict."
The criticism swirls around the use of bullet fragments.
For 40 years, FBI investigators testifying in state cases around the country have analyzed bullet fragments - often badly mangled - and compared them with unused cartridges found in possession of suspects.
The fragments were found at crime scenes or from the victims.
This analysis has been used in 2,500 cases since 1980, Grassley aides said.
The FBI turned to ISU in the late 1990s to see whether statistics researchers could predict bullet matches as accurately as DNA and other biological samples.
Alicia Carriquiry, professor of statistics and associate provost, and other researchers analyzed a sample of 800 .22-caliber bullets, a tiny sample considering that 9 billion bullets are fired in the United States each year.
Four major bullet manufacturers in the United States produce bullets in similar ways but use different concentrations of elements in the lead mixtures that make up the bullet tip.
Bullets are manufactured in groups of up to 700,000.
FBI experts had concluded that the composition of the bullets would stay the same, so that bullets with a similar makeup were thought to come from the same group.
Carriquiry's research didn't back up that logic.
Bullets are manufactured from vats of molten lead.
Bullets from one end of the manufacturing cycle have been shown to have a different composition from bullets at the end of the batch, Carriquiry said.
"The results were not exactly what they were hoping for," she said.
Carriquiry had not read the report Monday, but "it's kind of nice to see that the academy recognized the importance of statistical analysis," she said.
"It sort of confirms what we said a few years ago. "
Another problem with bullet samples is that they require expensive equipment to analyze, change composition over time and are difficult to track.
ISU research used at trials
Defense attorneys have used Iowa State University research on bullet analysis in at least three murder trials. For Jose Mateu, the research was partly responsible for two hung juries.
The Ketchikan, Alaska, man was charged with first-degree murder and evidence tampering in the death of his father in January 2000. His father was found dead in his home, shot in the head.
Prosecutors linked the .22-caliber bullet found in the father's head to a bullet of the same caliber found in the family home. They concluded that the bullets came from the same manufacturer. Police found no blood on Mateu, whose alibi included spending the night with his girlfriend.
"That was their case, essentially," said Louis Menendez, a Juneau attorney who defended Mateu in the first two trials. "They were jumping to these conclusions based upon readings from a computer."
A ballistics expert pointed Mateu's attorney to scientific research, including the ISU research, that threw doubt on comparing the two bullets. Attorneys say that led to two hung juries.
"I think they presented a dangerously bad case of scientific reasoning that could have resulted in this boy going to jail for the rest of his life," Menendez said.
- Jane Norman
http://desmoinesregister.com/news/stories/c4788993/23493859.html
Washington, D.C. - A National Research Council report to be issued today questions the FBI's use of bullets as evidence in thousands of criminal cases, upholding the findings of Iowa State University researchers, according to congressional aides familiar with the report and its written summary.
ISU researchers in 2000 completed an FBI-financed study that raised similar challenges to the bullet analysis technique, but, the researchers said, it was kept quiet because the findings were not to the agency's liking.
The ISU study, which cost the FBI $100,000, surfaced after a reporter obtained it through a Freedom of Information request. Defense lawyers have used the study in at least three murder trials since. Facing criticism for its continued use of the bullet analysis, the FBI requested the National Research Council report.
Aides to Republican Sen. Charles Grassley of Iowa were briefed on the new report Monday. They said the report's findings probably will not mean the reversal of thousands of cases in which the bullet technique was used, although there's no guarantee. Often the technique was just one piece of evidence, aides said.
However, the report suggests that the FBI use a more rigorous statistical method in its analysis. It also recommends that FBI expert witnesses be more precise and uniform when describing their findings to juries. Grassley aides said the FBI will not resist those suggestions.
"Attorneys, judges, juries and even expert witnesses can easily and inadvertently misunderstand and misrepresent the analysis of the evidence and its importance," said the report summary.
FBI experts who testify should not say that two bullets that cannot be distinguished from each other came from the same production run, or box, or even were made the same day, the report said. "None of these assertions . . . can be justified by the available data," it said.
Grassley, a member of the Senate Judiciary Committee, for years has been a harsh critic of the FBI crime lab and an advocate of whistle-blowers who brought public attention to what they said was sloppy work at the lab not backed up by good science.
"It is absolutely crucial that law enforcement use science that is based on valid, proven techniques," Grassley said in a letter to the National Research Council.
"Law enforcement must follow the facts, and let the truth convict."
The criticism swirls around the use of bullet fragments.
For 40 years, FBI investigators testifying in state cases around the country have analyzed bullet fragments - often badly mangled - and compared them with unused cartridges found in possession of suspects.
The fragments were found at crime scenes or from the victims.
This analysis has been used in 2,500 cases since 1980, Grassley aides said.
The FBI turned to ISU in the late 1990s to see whether statistics researchers could predict bullet matches as accurately as DNA and other biological samples.
Alicia Carriquiry, professor of statistics and associate provost, and other researchers analyzed a sample of 800 .22-caliber bullets, a tiny sample considering that 9 billion bullets are fired in the United States each year.
Four major bullet manufacturers in the United States produce bullets in similar ways but use different concentrations of elements in the lead mixtures that make up the bullet tip.
Bullets are manufactured in groups of up to 700,000.
FBI experts had concluded that the composition of the bullets would stay the same, so that bullets with a similar makeup were thought to come from the same group.
Carriquiry's research didn't back up that logic.
Bullets are manufactured from vats of molten lead.
Bullets from one end of the manufacturing cycle have been shown to have a different composition from bullets at the end of the batch, Carriquiry said.
"The results were not exactly what they were hoping for," she said.
Carriquiry had not read the report Monday, but "it's kind of nice to see that the academy recognized the importance of statistical analysis," she said.
"It sort of confirms what we said a few years ago. "
Another problem with bullet samples is that they require expensive equipment to analyze, change composition over time and are difficult to track.
ISU research used at trials
Defense attorneys have used Iowa State University research on bullet analysis in at least three murder trials. For Jose Mateu, the research was partly responsible for two hung juries.
The Ketchikan, Alaska, man was charged with first-degree murder and evidence tampering in the death of his father in January 2000. His father was found dead in his home, shot in the head.
Prosecutors linked the .22-caliber bullet found in the father's head to a bullet of the same caliber found in the family home. They concluded that the bullets came from the same manufacturer. Police found no blood on Mateu, whose alibi included spending the night with his girlfriend.
"That was their case, essentially," said Louis Menendez, a Juneau attorney who defended Mateu in the first two trials. "They were jumping to these conclusions based upon readings from a computer."
A ballistics expert pointed Mateu's attorney to scientific research, including the ISU research, that threw doubt on comparing the two bullets. Attorneys say that led to two hung juries.
"I think they presented a dangerously bad case of scientific reasoning that could have resulted in this boy going to jail for the rest of his life," Menendez said.
- Jane Norman