FBI Overstated Forensic Hair Matches In Nearly All Trials Before 2000

schwit1 writes The Justice Department and FBI have formally acknowledged that nearly every examiner in an elite FBI forensic unit gave flawed testimony in almost all trials in which they offered evidence against criminal defendants over more than a two-decade period before 2000. Of 28 examiners with the FBI Laboratory's microscopic hair comparison unit, 26 overstated forensic matches in ways that favored prosecutors in more than 95 percent of the 268 trials reviewed so far, according to the National Association of Criminal Defense Lawyers (NACDL) and the Innocence Project, which are assisting the government with the country's largest post-conviction review of questioned forensic evidence. The cases include those of 32 defendants sentenced to death. Of those, 14 have been executed or died in prison, the groups said under an agreement with the government to release results after the review of the first 200 convictions.

Comments

Forensic evidence should not be subjective

[st0nes]

The same thing happened some years back with fingerprint evidence. The people who are responsible for the analysis of forensic evidence should be 'blind', i.e. they should not have access to the context of the case. If they are given two fingerprints to match, they should merely be asked whether or not they are a match, and not told where they come from or even which case they pertain to. Then there would be far less bias. Also, they should not testify in trials, merely issue an affidavit of their results.

Re:Forensic evidence should not be subjective

[SuricouRaven]

That doesn't solve the problem. The FBI can simply fire or transfer elsewhere anyone who doesn't lean towards positive matches. It wouldn't take long for the experts to realise that saying yes a lot is good for their careers, but expressing doubt in court is going to lead to no more court appearances and a demotion to lab tech.

Re:Forensic evidence should not be subjective

[Anonymous Coward]

simple solution. you mix in a control group, and you don't tell the people who determine the match which one is the 'real one'. That way, they can't have a bias, unless the people preparing the material secretly mark it or something, but nothing's perfect.

Re:Forensic evidence should not be subjective

[blankinthefill]

That issue could be solved by an outside auditing body. They would send in samples that are known to be matches or not matches, and search for a statistically significant deviation in outcomes. Of course, having an auditing body truly unassociated with the FBI/CIA/NSA/Local policing force would solve MOST issues with policing these days, and would ruin some of the nice little fiefdoms people have been spending the last few decades building... which is why it will never happen.

Re:Forensic evidence should not be subjective

[binarstu]

Even before that, though, we need high-quality, doubly blinded trials to establish how well any of these comparison-based forensic methods actually work. Evidently, a key problem with hair comparison was that no one actually had any idea how reliable it was for "matching" a sample to a suspect. It is now obvious that the false positive rate is completely unacceptable.

We should have known this long before anyone even thought about using hair comparison evidence at trial, and the sad thing is that the experiments needed to rigorously evaluate this technique aren't even very complicated. For prosecutors, though, it is undoubtedly a lot more fun to impress juries with your scientific-sounding evidence and experts than it is to ask whether the evidence is actually reliable, and you can bet that the hair comparison "experts" were not in any hurry to show that their work was a sham.

Re:Forensic evidence should not be subjective

[CrimsonAvenger]

some of the nice little fiefdoms people have been spending the last few decades building

Last few decades???

IN case you were unaware, the FBI handles kidnapping. Why? Because 80-odd years ago, Herbert Hoover decided the FBI needed some good publicity, and could get it with the Lindbergh kidnapping.

IOW, fiefdom building has been going on forever - all it takes is a chance to get a bigger budget if you do something special....

Re:Forensic evidence should not be subjective

[blippo]

I think that that's actually how it works here in Sweden.

I seem to remember that it's also not always good. Since they only answer questions, more open ended searches are seldom performed.
In one case where an elk killed a woman (unique case, apparently) the police got hung up on her husbands lawn mower (!) which happened to have traces
of blood on the blades (which in fact could have been rust combined with other biological material ) and spent a year or so trying to convict him for murder,
until someone actually saw a YouTube clip of an elk-attack and asked the lab if it could have in fact been an elk. Answer: Yes. Most likely.

Re:Forensic evidence should not be subjective

[Dragon Bait]

You might be able to solve the problem(at the expense of a great deal of additional workload) by larding the caseload with samples specifically constructed to be non-matches; but then blinded and packaged the same as any other sample, to identify people who just lean positive; but that would probably require a lot of additional work to do in enough quantity to counteract the obvious pressure.

Add in that on a random basis you don't have any legitimate suspects in the collection being analyzed. The goal isn't to find best match. The correct answer has to be none of the above on a regular basis as well.

Re:Forensic evidence should not be subjective

[localman]

It's happened with arson experts too. I remember reading a horrible story of a guy convicted of burning his family to death because all the experts described these "pour patterns" in the burnt floor, signifying liquid accelerant. After he was put to death, they figured out it was just carpet glue patterns.

Between the way police feel free to shoot fleeing non-dangerous subjects these days, planting evidence in full view of other officers, lying on the stand to get convictions, and the labs and experts from every field falsifying results, I'd say our legal system is a disaster.

How many other flaws

[MrKaos]

I'm becoming more convinced that Police are often too lazy to do police work and now reviews of the cases shows evidence procedures stacked in favour of the prosecution. If the Court systems do not have mechanisms to self correct evidence procedures how can there be any trust that policing will lead to outcomes that protect society.

Justice is impossible if the system is not Just.

Re:How many other flaws

[Kiwikwi]

Some facts about the U.S. justice system:

* The Reid technique is widely used for interrogations, a technique notorious for its effectiveness in enticing false confessions.
* Only 5 % of convicted felons had their case tried in court; the rest make a plea bargain (typically under threats of excessively long prison sentences and/or the death penalty).
* Judges are elected, subjecting them to the whims of public opinion and making them more politicians than impartial legal officials.
* At least 4 % of people sentenced to death in the U.S. are innocent.
* The U.S. incarcerates more people than any other country in the world, not just relative to the population size, but in absolute numbers.
* U.S. private prisons sees $3+ billion in annual revenue... Not that that has anything to do with the above issues, I'm sure.

The U.S. justice system is broken in so many ways, I'm certainly forgetting some things.

That's a...polite...way to put it.

[fuzzyfuzzyfungus]

Is there any reason, aside from the reflexive deference to allegedly legitimate authority figures, why they use the phrases 'gave flawed testimony' and 'overstated forensic matches in ways that favored prosecutors' rather than the more honest 'committed a fuckton of perjury'?

Re:That's a...polite...way to put it.

[Eunuchswear]

The perjury, if there is any, is in the expert witnesses claims about the strength of the evidence.

From TFA:

The review confirmed that FBI experts systematically testified to the near-certainty of “matches” of crime-scene hairs to defendants, backing their claims by citing incomplete or misleading statistics drawn from their case work.

"misleading" == perjury.

Re:I'm shocked, I tell you!

[Anonymous Coward]

Truth & justice & the American Way = 0.

As a limey, I always thought they were a list of options. After all, if "the American way" incorporated truth and justice, you would be able to say merely "the American way".

14 already executed....

[Computershack]

And this, people, is why you don't have the death sentence.

Re:14 already executed....

[jcr]

I'm in favor of the death penalty in principle, but I really can't trust the government to only kill people who have it coming. That being the case, I would reserve it for politicians only.

-jcr

American "Justice"

[Required Snark]

Better 100,000 wrongly accused people go to jail rather then one member of law enforcement admit they made a mistake,

Re:Minimum retrial

[tomhath]

The article is very misleading on several points. It gives the impression that there was a problem "in almost all trials"; that's not what happened.

Of the ~21000 requests for an analysis, the lab reported a match about ~2500 times. Of those, the evidence was used in something like 268 trials, and a retrospective analysis of the DNA revealed that the hair did indeed match almost 90% of the time.

The bigger problem (which is where the "almost all") part came from is that when the evidence was actually presented at a trial, the expert witness overstated the reliability of the hair match; if they had stated that the analysis was only 90% accurate there wouldn't have been a problem.

A retrial in the 27 or 28 cases in which DNA revealed a mismatch is certainly in order. Otherwise there is no problem with the conviction; a retrial would simply replace the hair match with a DNA match anyway.

Re:shit

[Lunix Nutcase]

Get compensation for their family for wrongful execution.

Re:I'm shocked, I tell you!

[Archtech]

Lord Acton hit the nail on the head. 128 years ago he wrote that, ""Power tends to corrupt, and absolute power corrupts absolutely". Unfortunately, no one has ever devised a way of running societies without giving power to certain individuals. With a very small number of honourable exceptions (which prove the rule), that power has corrupted them. We see it around us almost every day, and the people who rise to control entire nation states display the corruption due to power in singularly pure and concentrated form.

The technicians working in the FBI labs had a very limited form of power, but within that particular domain their sway was almost unchallenged. What expert would dispute the word of the mighty FBI, what lawyer would challenge it, what judge or jury would not be impressed by it? And the technicians' bosses had more power, which was assuredly focused on the important task of getting convictions. I rather doubt that any lab technician at the FBI ever got much career advancement out of frequently discrediting prosecution evidence. Every bureaucratic organization measures itself according to a limited set of drastically oversimplified metrics, and conviction rate is an important metric for any law-enforcement organization. The higher up the tree you go, the greater the desire for more convictions, and the less the concern for whether they are justified or not.

"One of the many reasons for the bewildering and tragic character of human existence is the fact that social organization is at once necessary and fatal. Men are forever creating such organizations for their own convenience and forever finding themselves the victims of their home-made monsters".
- Aldous Huxley