Ballistics in Forensics Bogus?

[parishioner]

Last night before I went to bed, I was watching a segment called "Convicted by a Canine" on Anderson Cooper. It was about a guy who was wrongly convicted of Murder and spent 30 years in prison because a Police dog was not properly trained and made a mistake.

Anyway, what shocked me was what CNN's senior analyst said afterwords. He claimed that there are studies that show that ballistics testing are not valid and therefore many innocent people could be behind bars. He calls it "junk science" and says that these tests have never been subjected to real scientific scrutiny.

Is there any validity to his claims or is he a quack?

Here is the transcript of their conversation.


COOPER: Unbelievable. Florida's Innocence Project is investigating other cases involved in the guy John Preston and his dog. It says dozens more around the country may have been wrongly convicted.

Preston died last year. He was never charged with perjury, never punished.

We're joined now by -- by Jeffrey Toobin

You say this kind of junk science actually happens much more than we think.

TOOBIN: You know, "CSI" has convinced us, especially jurors, that science can answer all sorts of questions and that the kind of tests that go on in courtrooms are very reliable.

But real science has established things like ballistics testing, bite marks, arson investigations, a lot of that is bogus. And there are people in prison because of it, and it's really a scary thought.

COOPER: There's got to be thousands of people in prison based on ballistics or based on a bite mark or based on an arson investigation.

TOOBIN: DNA is the gold standard. DNA is not junk science; it's real science. It's been tested. There are double blind tests. Most of these other tests have never been subjected to real scientific scrutiny.

The National Academy of Sciences just issued a big report saying we need to look a lot more closely at all these technologies, or we shouldn't let them in our courtroom.

COOPER: I think it was the "New Yorker." I read an article about ballistics, where you know, we all have this idea that a bullet can be pinpointed to an exact batch, and every bullet has a unique marking on it...

TOOBIN: Lands (ph) and grooves.

COOPER: ... which is not -- apparently not the case.

TOOBIN: Not the case. I mean, even...

COOPER: Someone actually just kind of made that up, and no one ever -- as I'm now just remembering the article -- no one really had studied it and looked back to prove that it was true. And when they actually did, they found out that actually, it's not true.

TOOBIN: And the whole idea of scientific rigor, of testing of the testing, is something that's relatively new and hasn't been applied to these old technologies like bite marks. Bite marks is a particularly notoriously unreliable one. But it can be very persuasive. And it's just not scientifically valid.

COOPER: It's incredibly scary to think how many people may be incarcerated because of that.

Jeff Toobin, thanks.

[ra2bach]

science, law, drugs, and and even fire are inarguably beneficial to mankind. it's their misuse that makes them bad. as usual, it's the human application of any of that that's the weak link.

[TriumphRat675]

He is right. There are an increasing number of valid challenges to many of the most common forensic techniques used to put people behind bars. Bite marks, ballistics, even fingerprint technologies are based on forensic techniques developed and used by law enforcement, not scientists. Bite marks and the "experts" who testify on them in particular are being discredited by good defense attorneys. These techniques have been used and accepted for years but many of them have no real basis in scientific research.

One exception is DNA analysis, which when properly done is very, very, very accurate. Even then, recent debacles in, among others, the Houston crime lab shows that DNA evidence is not foolproof and can be twisted by unscrupulous law enforcement and evidence techs to achieve the desired results, or simply contaminated through sloppy work. Remember, most of the techs in the crime labs are not really scientists, just technicians - some trained more than others, some more resistant to political pressure to get a conviction than others.

Ballistics in particular are problematic. You're looking for marks made on the bullet to compare to another bullet shot from the same barrel - but bullets can deform or fragment and many are designed to do so. You can look at markings on fired cases, too, but think about what happens if you're comparing different ammo brands, or if the gun has had a couple hundred rounds through it. Barrels and parts wear out and leave different marks on the bullets and the cases, sometimes very quickly.

Forget everything you've seen on CSI. Forensic crime scene analysis is not foolproof and like anything else is occasionally peformed by fools.

One of the more disturbing trends I saw during my stint as a prosecutor was that juries would demand a ridiculous amount of evidence to convict someone of a misdemeanor DWI charge but would convict someone of a rape or murder on very little actual evidence. Add this to a predisposition to unquestioningly buy the testimony of forensic scientists and technicians and you can see why there are so many convictions being overturned, and why there may be a lot more wrongly convicted defendants out there.

[RemMan700]

Go out and read about the poly-graph. Sounds to me like it is just an interrogation prop.

[parishioner]

Go out and read about the poly-graph. Sounds to me like it is just an interrogation prop.

Polygraph tests are admissible in most courts across the country. The reason they are rarely admitted as evidence is both the plaintiff and the defendant have to agree to having the results of the test be admissible prior to the examination being conducted. This agreement is highly unlikely since the test is going to hurt one party and help the other, which is why it is rarely used.

Im mainly concerned with the validity of ballistics not polygraphs.

[RemMan700]

I saw something on tv back awhile ago.. maybe something John Stassle did a story on. They were talking about how the FBI would recover a bullet from the victim and then recover a box of unfired bullets from the suspects house. They would then "test" the lead to see if the bullet matched the ammo from the suspects house. They would use this as evidence in court to convict people. The developer of the techniques who was then retired from the FBI came out later after doing more research on his own and said that the test was not accurate. He found that the compound or molecular makeup of bullets made from the same batch of lead could vary. The story said that the FBI was still using the faulty test to this day. It was a pretty neat program.

[Dave Berryhill]

I would question the opinion of anyone claiming to be an expert who refers to forensic firearms examination as "ballistics." Ballistics is a completely different discipline. While it is sometimes used by Firearms Examiners, it is not the science of microscopic comparisons of tool marks that is the primary focus of a Firearms Examiner.

There have been many validations of the microscopic comparisons of fired bullets and cartridge cases. One test in particular has been repeated numerous times over the years which is to take firearms made with 10 consecutively rifled barrels from a manufacturer, have someone test fire them and then give the firearms and fired bullets to Firearms Examiner and let him determine which bullets came from which firearms. When you consider how similar 10 consecutive barrels are (and how minute the differences are) this is quite a feat.

There was one test that was recently proven to be false and that was a lead analysis test sometimes used for .22 rimfire bullets. The problem was that it was developed and used by only one agency and I don't know if it was ever subjected to scientific peer review.

The biggest problem with any forensic science is not the science itself but the person doing the testing. There are some extremely competent Firearms Examiners out there just as there lazy ones and so called "experts" that are nothing more than a liar-for-hire.

[ballistic]

I would question the opinion of anyone claiming to be an expert who refers to forensic firearms examination as "ballistics." Ballistics is a completely different discipline. While it is sometimes used by Firearms Examiners, it is not the science of microscopic comparisons of tool marks that is the primary focus of a Firearms Examiner.

There have been many validations of the microscopic comparisons of fired bullets and cartridge cases. One test in particular has been repeated numerous times over the years which is to take firearms made with 10 consecutively rifled barrels from a manufacturer, have someone test fire them and then give the firearms and fired bullets to Firearms Examiner and let him determine which bullets came from which firearms. When you consider how similar 10 consecutive barrels are (and how minute the differences are) this is quite a feat.

There was one test that was recently proven to be false and that was a lead analysis test sometimes used for .22 rimfire bullets. The problem was that it was developed and used by only one agency and I don't know if it was ever subjected to scientific peer review.

The biggest problem with any forensic science is not the science itself but the person doing the testing. There are some extremely competent Firearms Examiners out there just as there lazy ones and so called "experts" that are nothing more than a liar-for-hire.

Well said. The science isn't "bogus" but rather comes down to the training, experience, and competence of the individual firearms examiner. Same could be said for any discipline. Validations like the ones mentioned will help quatify known error rates and statistical certainty.

[KYPD]

The two pieces of information can reliably be obtained from bullet forensics. The first is the number of lands and grooves and the twist of the barrel. And even then, that information is reliable ONLY if the bullet has NOT distorted on impact, something nearly all bullets involved in a crime do.

The second piece of information is the brand of bullet. With luck, a skilled analyst may be able to identify high-end, name-brand bullets, such as the Federal Hydrashok or a Speer Gold Dot or a Winchester Ranger SXT hollowpoint, for example. But only IF the bullet is not too deformed. Again, a rare thing. If the bullet is a 9mm or 45 ACP FMJ cheapo brand from Eastern Europe, for example, there is no way to precisely identify brand.

So if the bullet is not distorted, a skilled, and lucky analyst (a very rare individual) might be able to tell you about the general nature of barrel and the bullet. BUT, no matter how skilled or lucky they are, they cannot reliably identify a specific gun, nor any particular lot of bullet simply from analyzing a recovered bullet. Anything beyond this is smoke and mirrors and lies. Welcome to the American justice system.

[TriumphRat675]

There have been many validations of the microscopic comparisons of fired bullets and cartridge cases. One test in particular has been repeated numerous times over the years which is to take firearms made with 10 consecutively rifled barrels from a manufacturer, have someone test fire them and then give the firearms and fired bullets to Firearms Examiner and let him determine which bullets came from which firearms. When you consider how similar 10 consecutive barrels are (and how minute the differences are) this is quite a feat.

Agreed; it is amazing what a skilled technician can discover with a microscope in a controlled environment, but the world is not a controlled environment, especially those parts of it that are crime scenes.

[boltcatch]

Out in the real world, different types of rifling can bugger up the results, as well.

A lot of people have been put behind bars using "scientific evidence" that really only has a place as exclusionary tools.

[ballistic]

The two pieces of information can reliably be obtained from bullet forensics. The first is the number of lands and grooves and the twist of the barrel. And even then, that information is reliable ONLY if the bullet has NOT distorted on impact, something nearly all bullets involved in a crime do.

Wrong.

What are your qualifications to make such statements? First, I'm not sure what your use of the term "bullet forensics" means. Firearms examination or metallurgical analysis of components of ammunition? You seem to be misusing the term to apply to firearms examination which leads me to believe that you lack any valid qualifications in the field to make such a statement.

For starters you fail to mention caliber class, land and groove impression widths, which along with number of lands and grooves and direction of twist can provide a list of suspect weapon makes/models using the GRC Database.

Secondly, it is possible to obtain this information from damaged projectiles if there are areas such as the base of the projectile or other undamaged areas where accurate measurements can be taken. It is not as rare as you allege. How many "bullets involved in a crime" have you examined? I've personally examined hundreds and more often than not, there were sufficient areas to measure these characteristics.



The second piece of information is the brand of bullet. With luck, a skilled analyst may be able to identify high-end, name-brand bullets, such as the Federal Hydrashok or a Speer Gold Dot or a Winchester Ranger SXT hollowpoint, for example. But only IF the bullet is not too deformed. Again, a rare thing. If the bullet is a 9mm or 45 ACP FMJ cheapo brand from Eastern Europe, for example, there is no way to precisely identify brand.

A qualified firearms examiner could be able to determine whether the characteristics of a spent projectile are consistent with a known brand. Don't confuse "bullet forensics" or metallurgical analysis with firearms examination. Ogive, cannelures, base type (concave, convex, boat-tail, etc), jacket type (full jacket vs exposed base, etc) hollow point configuration, etc in addition to caliber class and weight are characteristics that can be used to identify possible brands.


So if the bullet is not distorted, a skilled, and lucky analyst (a very rare individual) might be able to tell you about the general nature of barrel and the bullet. BUT, no matter how skilled or lucky they are, they cannot reliably identify a specific gun, nor any particular lot of bullet simply from analyzing a recovered bullet. Anything beyond this is smoke and mirrors and lies. Welcome to the American justice system.

Your opinion is contrary to over 70 years of history in the field of firearms examination and numerous validations. Spent projectiles can be reliably identified as being fired from a specific barrel, even when they are not pristine specimens.

[Dave Berryhill]

Out in the real world, different types of rifling can bugger up the results, as well....

Care to explain that?

[Dave Berryhill]

Agreed; it is amazing what a skilled technician can discover with a microscope in a controlled environment, but the world is not a controlled environment, especially those parts of it that are crime scenes.

We usually brought the firearms evidence back to the lab where it could be examined in a controlled environment rather than at the crime scene ;)

[JimmyB62]

I saw something on tv back awhile ago.. maybe something John Stassle did a story on. They were talking about how the FBI would recover a bullet from the victim and then recover a box of unfired bullets from the suspects house. They would then "test" the lead to see if the bullet matched the ammo from the suspects house. They would use this as evidence in court to convict people. The developer of the techniques who was then retired from the FBI came out later after doing more research on his own and said that the test was not accurate. He found that the compound or molecular makeup of bullets made from the same batch of lead could vary. The story said that the FBI was still using the faulty test to this day. It was a pretty neat program.

I believe that analysis technique is no longer used by anyone, at least for court purposes.

[ballistic]

I believe that analysis technique is no longer used by anyone, at least for court purposes.

Numerous state, local agencies/crime labs stopped submitting evidence to the FBI for that analysis several years ago due to questions about its validity.

[Preferred User]

This must be quite the hot topic. The current issue of Popular Mechanics is devoted to forensics: CSI Myths: The Shaky Science Behind Forensics (http://www.popularmechanics.com/technology/military_law/4325774.html).

As someone that works in forensics I can say one of the biggest challenges is living up to the TV labs. If only we had that kind of budget and could come up with answers and results in 42 minutes. And the guys from the lab going out in the field to confront the bad guys - > INSERT GUT WRENCHING BELLY LAUGH HERE <- is pure fantasy.

[pleaforwar]

This must be quite the hot topic. The current issue of Popular Mechanics is devoted to forensics: CSI Myths: The Shaky Science Behind Forensics (http://www.popularmechanics.com/technology/military_law/4325774.html).

As someone that works in forensics I can say one of the biggest challenges is living up to the TV labs. If only we had that kind of budget and could come up with answers and results in 42 minutes. And the guys from the lab going out in the field to confront the bad guys - > INSERT GUT WRENCHING BELLY LAUGH HERE <- is pure fantasy.

Dang, beat me to it. :p