New at Reason

Roger Koppl says America's forensics system is rife with errors and offers and eight step plan to fix it.

Comments

Peder | October 24, 2007, 1:05pm | #

Are there enough forensic experts for the reforms that Mr Koppl is recommending?

dbust1 | October 24, 2007, 1:10pm | #

Probably not. So let's create another bloated gov. agency to take care of the problem.

Michael Pack | October 24, 2007, 1:18pm | #

How many wrongful convictions are caused by breath machines?We see police using a magic box for a money maker and yet it wasn't mentioned.

creech | October 24, 2007, 1:24pm | #

There certainly should be a ton of male forensic scientist students in the pipeline...at least if the television depiction of female forensic scientists
(all stacked, with low cut blouses) is true to life.

Randy | October 24, 2007, 1:35pm | #

She blinded me with SCIENCE!

Russ 2000 | October 24, 2007, 1:38pm | #

To establish their independence from police and prosecutors, crime labs should be organized by the courts, not the cops.

Yes, because the courts aren't pro-police.

/sarcasm

Russ 2000 | October 24, 2007, 1:45pm | #

With 238,135 requests for latent fingerprint comparisons in 2002 alone, a false positive error rate of 2 percent implies up to 4,800 false convictions or guilty pleas made in hopes of a lighter sentence each year in the U.S., 1,700 of them in felony cases. (The number of improperly matched fingerprints is not completely clear. A 2005 study of fingerprint analysis suggests that the false positive rate may now be as low as 0.8 percent. But another recent study suggests it could exceed 4 percent.)

This has me curious as to the error rate on DNA comparisons. I know I supposed to believe that DNA comparisons are infallible, which is exactly why I don't beleive they are.

Legate Damar | October 24, 2007, 1:45pm | #

Private labs are subject to civil liability and administrative fines for poor performance.

To me, this is the killer. In the pharmaceutical industry, we have to produce mountains of data and paperwork to get FDA approval. Once we get that approval, it in no way protects us from lawsuit if there was something that we (and the FDA) missed. If an idiot in the police lab determines that the gun "must have" been fired by two people simultaneously, there's nothing the defendant can do once the evidence is exposed as garbage. Not to the lab, not to the forensic scientist who was clearly just making shit up.

Legate Damar | October 24, 2007, 1:50pm | #

I'm to the point now, where if I were on a jury, I would require non-enforcement confirmation of everything before I was ready to convict. There would need to be a third-party confirmation of forensic evidence and a third-party eyewitness confirmation of any account given by the LEO. Otherwise, I'm not gonna be able to overcome "reasonable" doubt.

J sub D | October 24, 2007, 1:58pm | #

There would need to be a third-party confirmation of forensic evidence and a third-party eyewitness confirmation of any account given by the LEO. Otherwise, I'm not gonna be able to overcome "reasonable" doubt.

I can't say I'd blame you.

Dave Woycechowsky | October 24, 2007, 2:25pm | #

I'm to the point now, where if I were on a jury, I would require non-enforcement confirmation of everything before I was ready to convict. There would need to be a third-party confirmation of forensic evidence and a third-party eyewitness confirmation of any account given by the LEO. Otherwise, I'm not gonna be able to overcome "reasonable" doubt.

Which is why the OJ jury did the right thing, even though OJ really did do those murders.

Dave Woycechowsky | October 24, 2007, 2:35pm | #

Oh, yeah, and there should be parity of funding for criminal prosecution and criminal defense. Neither should the defense be allowed to spend more than the prosecutors nor vice versa.

That would allow a more efficient, market driven solution to the problems ably noted by Mr. Koppl than what he proposes. It would solve other bad problems, too.

J sub D | October 24, 2007, 2:45pm | #

Which is why the OJ jury did the right thing, even though OJ really did do those murders.

Dave Woycechowsky,
Your post really pisses me off. Why, because it is true. Shit.

Dave Woycechowsky | October 24, 2007, 3:00pm | #

Your post really pisses me off. Why, because it is true. Shit.

I mean we know he did it because we know that if there was a police conspiracy, then the media would have been motivated to uncover it, or raise suspicions or something.

OJ had friends in the media and other high places, and the media and/or people in other high places can crack a police conspiracy if they are so inclined. They didn't. Ergo, there was no police conspiracy in the OJ case. Deep down, that is the real reason we were all so sure that he did it.

But, if you are the juror, you don't get to consider that stuff. You can consider the general reputation of police, and you get to consider credible and demeanor evidence offered in the trial itself. But, you don't get to consider what the media said and (more importantly) did not say. That is true even though a consideration of the full panoply of media investigations / coverage / commentary would have lead to a more objectively accurate verdict in OJ's case.

It is comforting to know that OJ's case is the exception in this regard. In many cases, if you make your decision based on media accounts, then you will make a worse decision than you would by following the trial.

sv | October 24, 2007, 3:47pm | #

This all sounds like it would go a long way towards solving the problem put forth. It also sounds really, really expensive, if it's feasible (enough practitioners, etc.), and besides, the public probably doesn't care about those accused of crimes, and even if it did, they probably wouldn't want to shell out for it, and even if they might be persuaded to do THAT, the government would surely interpret this as being a soft-on-crime strike against them (us vs. them) and find a way to scuttle it, probably budgetary fudging.

Avoid government, keep your nose clean, avoid police, type rant-y run-on sentences on blogs to vent, carry on.

sv | October 24, 2007, 3:52pm | #

re: the idea of the costs being offput by the savings in not incarcerating the innocent: like the govt. is interested in that (drug war anyone?)

Jessica | October 24, 2007, 9:28pm | #

While I in no way think San Antonio represents the state of forensics country-wide, based on my experience as a juror in a murder case....Some of Koppl's suggestions are starting to be used. There is room for improvement though.

Specifically, the suggestions of Privatization and the Redundancy were followed for DNA testing but not fingerprinting. For my jury case, the DNA testing was done by an outside lab who then verified their results by sending out to another lab (who had no info on what the test was for). Both reports were submitted to the jury for evidence.

But I do see your point about independence in fingerprinting analysis. In SA, the people ID-ing the fingerprints are in the Evidence division of the police force. If I remember correctly the main fingerprint guy and the person who double-checked his analysis didn't visit the crime scenes. BUT they were both aware of what case the print ID was for so your point about masking is valid.

It seems to me the newer tests for DNA have better control procedures possibly because there's less of a history of poor practice to overcome. But the evidence development should be more consistent between methods (DNA vs. fingerprinting vs. hair&fiber). Just my 2cents.

Iain McKie | October 25, 2007, 6:10am | #

Roger Koppl does us all a service by highlighting the problem of forensic science error rates.

Separating the scientists from the police is an essential first step. The former are not involved in a ‘war against crime’ nor should they become caught up in the culture that such a battle creates. The scientist’s role is the objective analysis of data free from emotional and cultural prejudice.

In addition the ‘adversarial’ approach, where one expert is set against another and where clarity and understanding are the biggest losers, should be dropped and the court should have the power to call its own experts to ascertain the truth affording all defendants and prosecutors a level playing field.

For more on the problem worldwide see www.shirleymckie.com

Crime Lab Report | October 25, 2007, 8:58am | #

We have been following professor Koppl's comments over the last several months, and will continue to monitor his characterizations of forensic science.

It would be inappropriate for our editorial board to describe all of Professor Koppl's recent recommendations as being unworthy of consideration. However, we are interested in learning what fiduciary advantage he creates for himself and his university by inaccurately portraying the forensic sciences as being inadquate and "rife with errors." We challenge Professor Koppl to present the statistical evidence for this claim.

We also wonder if his ultimate goal is to secure grant monies to fund the research he would need to test the validity of his recommendations. If this is the case, which we believe it is, Professor Koppl and many of his colleagues in academia should have the professionalism to acknowledge their conflict of interest and refrain from making the ad hominem attacks that they have come to be known for.

If you would like more information about forensic science and how it is portrayed in the media, please visit us as http://www.crimelabreport.com.

Joe M. | October 25, 2007, 5:16pm | #

My brother used to laugh when anyone mentioned forensic "science" to him. He told me that as a research scientist he would be lucky get 2-5% of his results to be repeatable and verifiable to the standard required for publication. That's under the strictly controlled environment of a university chemistry lab. Yet forensic "scientists" are able to get 100% results from a pair of old underpants buried in a swamp for ten years or find foetal blood sprays under dashboards where none exist (google Australia, lindy chamberlain), and send someone to prison for the rest of their life based on their "evidence".

Most of the disasters which result from forensic evidence are inevitable because they are based not on verifiable scientific techniques but on subjective estimates and educated guesses, which is mostly what forensic labs produce. Even the magic of DNA evidence will wear off once the mistaken convictions resulting from mishandling and misinterpretation (no, it's not just matching the dots despite what's on TV) start filtering through.

Forensic technology has an important role to play in helping police identify or eliminate suspects and to try to reconstruct the timing and scenario of crimes - especially violent crimes. But it should never be used in a court as evidence. As many people are beginning to realise, it never is!

Atabrat | October 25, 2007, 6:59pm | #

Where I work, we have been preparing for three years to be an accreditated lab so we can test materials for the states of Washington and Oregon. We have to have a complete quality control system in place, including procedures for calibration of equipment, rules for reporting information, and so on. We've been receiving samples almost every month for the last three years to test our proficiency, we've spent thousands and thousands of dollars on equipment and training, and we have two massive inspections by outside agencies before we can finally be approved. And this is all just so we are able to test soil and rock for highway jobs. Maybe if we were able to send people to prison the requirements wouldn't be as strict.