How Reliable Is Ballistic Fingerprinting?

FoxNews.com ^ | 10/18/2002 | Steven Milloy

[JackIV]

How Reliable Is Ballistic Fingerprinting? Friday, October 18, 2002 By Steven Milloy

The sniper spree in the Washington, D.C., area has spawned calls for "ballistic fingerprinting" of firearms.

Sen. Charles Schumer, D-N.Y., announced he would introduce legislation for a national program. The Brady Campaign to Prevent Gun Violence told The Washington Post that ballistic fingerprinting would have "solved this crime after the first shooting."

But an October 2001 report by California state ballistics experts -- hushed up by the California attorney general's office -- concludes that ballistic fingerprinting isn't feasible right now.

Ballistic fingerprinting involves sending a fired bullet and empty cartridge casing from a gun to a government agency before that gun can be sold. The idea is to match -- preferably by automated computer analysis -- pre-sale ballistics data with crime scene data.

Maryland and New York already require ballistic fingerprinting. So far it hasn't helped convict a single criminal in Maryland despite "fingerprinting" 17,000 guns sold since January 2000. New York hasn't had success either.

And there isn't likely to be success any time soon, according to the study.

The report included the test firing of more than 2,000 rounds from 790 pistols.

When cartridges from the same manufacturer were test-fired and compared, computer matching failed 38 percent of the time. With cartridges from different manufacturers, computer matching failed 62 percent of the time.

"Automated computer matching systems do not provide conclusive results" requiring that "potential candidates be manually reviewed," said the experts.

But the experts estimated a California database would grow by about 108,000 entries every year for pistols alone. "This study indicates that this number of candidate cases will be so large as to be impractical and will likely create logistic complications so great that they cannot be effectively addressed," they said.

The test-firing results only scratch the surface of ballistic fingerprinting's problems.

The experts concluded it's unknown whether cartridges fired after typical firearm break-in and wear can at all be matched to the cartridge fired when the gun was new.

"Firearms that generate markings on cartridge casings can change with use and can also be readily altered by the users," said the experts. "They are not permanently defined like fingerprints or DNA."

A file may be used to make scratch marks in a barrel or a breech face, and various parts may be replaced to give a firearm a completely new ballistic identity. Bullets may be treated to alter the machining marks in a barrel.

Not all guns even generate markings on cartridge casings.

Further, "fired cartridge casings are much easier to correlate than fired bullets," noted the experts. Because bullets are severely damaged on impact, they can only be examined manually.

Moreover, Americans already own more than 200 million guns; those won't be included in any ballistics database.

rest of story at: http://www.foxnews.com/story/0,2933,66007,00.html

[Dick Bachert]

It would alter the mechanical profile of the barrel sufficiently to prevent anything close to a reliable projectile match.

Run a wad of steel wool into the chamber a few times and you have also dealt with the "problem" of spent casing matching with -- up to a point -- little or no impact on accuracy.

Then there's always the shotgun with rifled slugs for nearly ballistics-free wet work (at closer ranges, of course).

Gang, this crap is NOT about catching these murderers: It's about DISARMING us so the would-be PRIVATE criminals can have their unfettered way with this nation, something they are currently reluctant to do lest the ARMED natives become dangeroulsy restless.

[SJackson]

There has not been ONE crime solved, or a conviction, in which a firearm was used and "fingerprinting" was used. Yet, another "feel good" law that does NOTHING.

You're wrong, it does something. It provides both states with a databank of recent handgun purchasers, never to be misused, of course.

[backhoe]

Learn how to "unfingerprint" a gun in 20 minutes - with common gun tools!

[wcbtinman]

Send your favorite Congresscritter a copy of "Unintended Consequences".

[sharktrager]

I found this study online yesterday (and now I can't).

Fox misses one point which is vital to the story.

The researchers did find that the more bullets that were fired from a weapon, the less reliable the program became. They found that as few as 7 firings could cause the program to miss the correct match.

They also pointed out that a national database would be so large that investigators would likely get a list of tens of thousands of potential matches. Given that guns change hand fairly regularly, you would be tasked with finding where each of those guns was currently.

The researcher referred to the results of a national database as "looking for a needle in a haystack".

[Puppage]

Of course not. Just like in California where everyone who owned an SKS rifle had to register them. Oh, yeah. Then they made them illegal & they knew who had them. If you didn't turn them in you immediately became a felon. Lovely. But, it WON'T happen here, will it?

[DBrow]

Consider as a talking point: the James Earl Ray case.

A high profile case in the media (with all antigun media, you have short selective memories), with a nationally famous set of ballistics experts.

After 20 or more years of insisting that ballistics testing would exonerate him, Ray got his wish. The aforementioned experts back in 1997 fired 18 rounds from the 30.06 rifle believed to have been the murder weapon. The report came out July 1997. 12 bullets did not match the one removed from King, and according to most of the experts did not much match each other. The remaining 6 had some features matching the King bullet, not many but some.

So now what? Let him go? No, he died in jail from liver cancer.

The real point is, though, if this is a science, it should have been clear. Repeatable data under controlled conditions.

Another case that goes back several years is Sacco and Vanzetti, a Massachusetts capital case in which Sacco was executed based on fraudulent ballistics evidence. In a review held 20 years after the execution, much like the King bullet, the examiners found that the six bullets taken from the victim could all have been fired from different .32 autos, none matching Sacco's auto. There was evidence that the "expert" acted in collusion with the prosecutor to deliberately mislead the jury.

Hardly the makings of a national system, unless arbitrary court ordered disarmament could be a goal of Sarah Brady and her buds. If it was your rifle, would the judge go with the 12 rounds that were exculpatory or the 6 that may, may, be incriminating?

http://www.s-t.com/daily/07-97/07-12-97/a01wn008.htm

http://www.sunspot.net/news/local/bal-te.md.guns15oct15,0,4315047.story?coll=bal-local-headlines


http://ng.netgate.net/~wms/SandV.html