Reason.com

Free Minds & Free Markets

Forensic Science Is a Mess, and the Justice Department Wants to Keep It That Way

FootprintCreative Commons/Marcy HendricksA White House council has declared junk science all too common in the criminal justice system, but the Department of Justice is refusing to implement recommendations that would increase safeguards against scientifically invalid testimony and require greater disclosure of misconduct and shoddy work.

In September the President's Council of Advisers on Science and Technology (PCAST) released a report finding "a dismaying frequency of instances of use of forensic evidence"—such as analyses of hair, bite marks, and shoe prints—"that do not pass an objective test of scientific validity."

This is not just a theoretical problem. Last year, the FBI admitted that nearly every one of the experts at its microscopic hair analysis lab had given scientifically invalid testimony. The breaches affected almost 270 cases. Of those, 32 defendants were sentenced to death, and 14 were executed or died in prison.

When The Washington Post investigated the probe, it found that "while many prosecutors made swift and full disclosures [to defendants], many others did so incompletely, years late or not at all." One defense attorney only learned his client's case had been affected after he read it in the Post.

Despite these embarrassments, the Justice Department rejected PCAST's recommendations, which include requiring forensic experts to disclose high error rates in their testimony and discontinuing use of methods that haven't been independently verified.

Attorney General Loretta Lynch told The Wall Street Journal that "the department believes that the current legal standards regarding the admissibility of forensic evidence are based on sound science and sound legal reasoning."

One of those "sound" methods she's defending is bitemark analysis. But the PCAST report found that "available scientific evidence strongly suggests that examiners not only cannot identify the source of bitemark with reasonable accuracy, they cannot even consistently agree on whether an injury is a human bitemark."

The vast majority of crime labs are administered by law enforcement, an arrangement that criminal defense lawyers say is akin to a fox guarding the hen house.

Judges, however, are the ultimate "gatekeepers" of what evidence is admissible in court, and some of them are taking notice of these concerns. In a September Wall Street Journal op-ed, Alex Kozinski, the outspoken chief judge of the 9th Circuit Court of Appeals, called on the Justice Department to reject the "voodoo science" plaguing courts.

"Among the more than 2.2 million inmates in U.S. prisons and jails, countless may have been convicted using unreliable or fabricated forensic science," Kozinski wrote. "The U.S. has an abiding and unfulfilled moral obligation to free citizens who were imprisoned by such questionable means."

Photo Credit: Creative Commons/Marcy Hendricks

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  • Lee Genes||

    The FBI goes not give one shit about actual justice. Never has.

  • CatoTheChipper||

    What would you expect from an armed government organization that was formed after the Palmer Raids ... not to correct the government's excesses, but to organize and centralize them more fully?

  • Crusty Juggler||

    This is not just a theoretical problem. Last year, the FBI admitted that nearly every one of the experts at its microscopic hair analysis lab had given scientifically invalid testimony. The breaches affected almost 270 cases. Of those, 32 defendants were sentenced to death, and 14 were executed or died in prison.

    It was only 32 out of 270. Gosh.

  • Citizen X||

    Sometimes you gotta break a few eggs to preserve the myth of government infallibility, na mean?

  • Ron||

    but they are right on when it comes to the climate and everything else they claim

  • Jerryskids||

    Attorney General Loretta Lynch told The Wall Street Journal that "the department believes that the current legal standards regarding the admissibility of forensic evidence are based on sound science and sound legal reasoning."

    Well, the department is free to believe elephants fly by flapping their ears if they want but that doesn't make any less a steaming pile of crap. Too many people think the stuff they see on CSI is plausible science, just as they believe cops can disarm bad guys by shooting the gun out of their hand, speeding buses can jump half-open draw bridges, you can get knocked out and 30 seconds later jump up, shake your head woozily a few times to clear it and chase after your assailant. It's all bullshit, science don't work like that.

    The law, on the other hand, is an ass. So sure, there's sound legal reasoning - we've eaten this shit sandwich before so now it's a permanent fixture on the menu. But it's still a shit sandwich dressed up like a Philly cheesesteak.

  • juris imprudent||

    Just like dog noses are nearly perfectly reliable, objective and disinterested.

  • CatoTheChipper||

    Here we have what amounts to per se evidence that the US government is dismally failing in the primary purpose of government.

    However, the vast majority of Americans have no first-hand experience with the US legal system, especially the so-called criminal justice system, and must rely on what they learn from the mainstream media. Unfortunately, most of what they learn comes from television shows like Law and Order. So, they believe that the system is just, and since the US government has already achieved justice within its own borders, it should go forth and spread its wonderful system around the world. Furthermore, given its success in establishing justice throughout the land, foolish Americans have come to believe that the US government should expand the definition of justice to include universal healthcare and a $15/hour minimum wage and other tenets of social justice.

  • Zero Sum Game||

    foolish Americans have come to believe that the US government should expand the definition of justice to include universal healthcare

    Every other first world nation in the West has this. Why shouldn't we?

  • Zero Sum Game||

    I observe that nobody felt like taking the bait. I was hoping for something funny in response to that. :(

    "But who will make the roads?" :P

  • Col. Chestbridge||

    We're wise to your zero sum games!

  • sarcasmic||

    The primary purpose of government, or more accurately the only purpose of government, is to maintain and increase the power of the people in it. That's it. Everything else, from universal health care to minimum wage to the war on terrorism or even criminal justice, is merely a means of getting the general public to demand that the people in government be given more power. It's all about keeping and increasing power.

  • Zero Sum Game||

    The primary purpose of government, or more accurately the only purpose of government, is to maintain and increase the power of the people in it.

    I don't agree. Those are certainly goals that many in power seek, but a goal isn't the same thing as a purpose. You can build a tool for a purpose but have no goal in mind with it. For example, that hammer you made has a purpose: hitting nails, but sits on a shelf until someone has a goal for it which could range from building a bird house with your kid all the way up to working on a skyscraper.

    Protection from force and fraud are legitimate purposes. Maintaining property rights, enforcing contracts, detecting and punishing fraudsters, punishing violent people, etc. are all things that many people agree that a government should be doing. It's mostly just libertarians saying that those functions are the only things a government should be doing, and butt out on everything else. ANCAPS can speak for themselves, though I know they disagree with that opinion.

  • sarcasmic||

    What you describe is the abstract purpose of government. Government as an abstraction. What it in fact is is people. People who look out for their self interest. Their self interest will almost always be more power and control. They will seek to cover their asses and the asses of their fellows. That is the reality of government, and of any self regulating agency.

  • Zero Sum Game||

    If we aren't trying to convince people of that ideal, why bother trying at all? If you're saying we should not have government at all because it is impossible to have constrained government due to the behaviors of real people, that is a different argument. It's the difference between a minarchist and an ANCAP. Then you're into the territory of having to argue why we should not have prisons of any type or a military to defend against invasion, as just a couple of examples.

    I don't think that the argument is wrong that a small, heavily constrained government can stay that way if properly maintained. The ideal is to get it small enough that there is no motivation for rent seekers to lobby anything because the government doesn't have the funds or power to give it.

  • sarcasmic||

    I don't think that the argument is wrong that a small, heavily constrained government can stay that way if properly maintained.

    How is it constrained and maintained? Who limits those with the monopoly on the initiation of violence?

  • cavalier973||

    The only way a government can be kept small is to have a place to which people can flee as that goverment tries to increase its power.

  • sarcasmic||

    It is no coincidence that the size and scope of the federal government exploded after Manifest Destiny was achieved and there was no more frontier.

  • Jerryskids||

    The purpose of any system, regardless of its stated purpose, is to perpetuate itself. Goals don't matter, goals are the "end" in the means to an end and self-preservation is not a means to an end, it is an end in and of itself.

  • Zero Sum Game||

    The purpose of any system, regardless of its stated purpose, is to perpetuate itself.

    Borderline tautology. Almost useless to say at all. By your definition, the purpose of your hand is to perpetuate yourself so that you can have children that also have hands. In the absolute most general sense, yes, that is true (with the caveat that hands also get used to fulfill many goals). It ignores the reality that there are other solutions to the problem for which other appendages may also suffice and denies the possibility of meaningful comparison (hands with opposable thumbs versus hands without). All nuance is lost. While interesting to discuss from the thousand-mile-up philosophical view, it isn't of practical value to a discussion of politics.

    Regardless, self-perpetuation is still a goal, not a purpose. Your hand may help you to survive to procreate, but it does not cease being useful to you once you've fulfilled that "obligation" to the species. There are still plenty of goals you can apply your hand to. ;)

    Again, I ask: do you guys wish to argue that there is no purpose for a government whatsoever and that all government should be abolished? The implication seems evident if that there is no point in defining purpose for any given system is that it ought not exist, that it is a novelty, a Rube Goldberg machine that is only fit for temporary entertainment and not to be used to solve problems.

  • Zero Sum Game||

    I ought also note that it veers close to proggy reasoning. For example, a gun is a tool. A gun can be used to murder people. Ergo, a gun ought not exist (at least in private hands).

    The nuance lost in that reasoning is that, of course, guns are used for other purposes and that murder is a comparatively rare use. Too much generalization leads to weaseling around without nuance.

    One of the nicer things about libertarians is that we usually focus on where the rubber meets the road as much as we focus on the ideals themselves. That often gets lost in other political ideologies because most of them would much rather ignore the unintended consequences of their policy proposals.

  • sarcasmic||

    These systems are comprised of men. They are not inanimate objects. To compare a system of men to a gun just doesn't work. Those men have self interest.

  • bacchys||

    A considerable amount of how our government was constructed was based on an understanding of the base nature of Man and set the ambitions of those in power against any of them succeeding in expanding their own beyond its place.

    That design ultimately failed, slain by factionalism, but the alternative to government isn't a solution to ambitious and unscrupulous men.

  • ||

    There's almost zero forensic "science" that is valid. No double blind, inadequate controls (need positive and negative controls, hidden replication, all double blind, NEVER happens). Any case based on the shoddy practices that are judicially accepted should result in jury nullification- and of course, rarely does, since anyone who actually approaches science with rigor will get tossed at or before voir dire.

  • GSL in E||

    Yep. And don't get me started on DNA testing. Where we all just pretend that tests don't have false positives, and that positive predictive values are terrible when the event rate is low.

  • ||

    It's easy to avoid false positives, but that requires a well-structured experiment, which slows throughput.

  • Zero Sum Game||

    This is not a good example. DNA testing is one of the only things they actually do get right fairly often. The science of it is not wrong and its predictive power is not weak.

    There are still problems with it, but they're procedural. Cross-contamination of samples is a big one. Another is the problem of DAs hiding exculpatory DNA evidence from the defense.

    If a good lab has determined a match and they followed the protocols correctly, then it is a match and within the mathematical probabilities quoted at trials.

  • Zero Sum Game||

    Instead of "predictive power" I should have said "evidentiary value." Been awake too many hours. :P

  • ||

    The science is right, the experimental design is wrong at forensic labs. Again, no double blind, no positive and negative controls, no hidden replications.

  • Zero Sum Game||

    Yep, protocol can be poor. As far as replication is concerned, usually both the defense and prosecution have their own tests done. Sometimes that is a problem because small samples may be all you have and get destroyed in the process, which is totally understandable if unfortunate. Fortunately, PCR techniques have come a very long way and it should not happen much anymore.

    As for solving the problem of not having controls, obviously that brings the cost up substantially. What do you propose to counter that? One might think more like a businessperson than a scientist on solving that problem. Send the occasional control sample through the process to test each of your workers at random, just like quality control checks of random parts in big batches of machined parts. It helps keep the costs down, and it really does result in few fuck-ups getting through. Of course, as with any state employee, would they actually get fired for failing? Probably not. They have a culture of forgiving even egregious mistakes in government work.

  • ||

    Well, that's the thing- if you want to do science right, there's a cost associated with it. If you're not going to do it right, an intelligent jury ought to toss the case. Of course, the government does its best to prevent an intelligent jury, and judges consider precedent as valid science.

    Basically, we're fucked and this is just a pretty veneer so we can pretend that we're doing justice.

  • Zero Sum Game||

    an intelligent jury ought to toss the case

    I don't quite agree. A jury should be looking for reasonable doubt. The standard isn't "beyond all doubt."

    That a jury should be told what the downsides of the testing protocol are is unquestionable. It should be left up to them if that is unreasonable, given all the other evidence in a case.

    Being imperfectly scientific doesn't necessarily imply that it is wrong or not of evidentiary value. A competent defense attorney should be having independent tests done and hammering the government's case when they don't do a good job. There should be an expectation of follow-through when a government lab is shown to be incompetent. Heads need to roll, and that they usually do not is a serious problem.

  • ||

    Shit protocol is absolutely reasonable doubt. For me, that defines reasonable doubt.

  • Zero Sum Game||

    I think that depends on the nature of the mistake. And, again, it is also the defense's job to make sure that the prosecution isn't cheating.

    There is "evidence" that is so clearly flawed that no jury should hear it. There is evidence that may be flawed in its collection or processing that might be able to be heard, with the admonition that it alone should not weigh heavily in the decision. A jury is supposed to look at the evidence in totality. We know that eyewitness testimony can be complete garbage, and a jury should be told that. That doesn't imply that juries should never hear eyewitness testimony, because some of it will be perfectly valid.

    More often than not, the defense has to acknowledge that their own labs say the same thing as the government labs. In such cases, I think that it's fit evidence for a jury to hear.

  • Jerryskids||

    I posted this link a few days ago but it bears repeating.

    The Austin PD crime lab was so screwed up the state had to step in and attempt to clean it up. After working with the lab techs on proper training, the state threw up their hands and said most of the techs were too retarded to even train.

  • DenverJ||

    Feature not a bug.

  • sarcasmic||

    Everyone is guilty. Their attorney may convince a jury that they are innocent, but they are still guilty. So even if the evidence is bunk, that doesn't matter. People in government don't make mistakes. The perp is still guilty because some cop or DA said so, regardless of the evidence.

  • ||

    It should be noted that "forensic science" fits the "science" pattern: anything that has to call itself "science" in its name is not science. E.g., political science, social science, climate science, environmental science... I don't tell people my bachelor degree is in "physics science" or that my doctorate is in "chemistry science."

  • sarcasmic||

    My degree is in Computer Science, which isn't really science. It's more of a black art.

  • ||

    There ya go!

  • Jerryskids||

    I have an ex-brother-in-law that refers to himself as a medical scientist. He's really more of an electrical engineer except his area of expertise in electricity happens to be the human brain and he holds (jointly) several patents on probes they use in research on how exactly the brain works. Smart guy, but annoying as hell to talk to because of his mannerisms. Much as I'm interested in the work he does, I can't stand talking to the guy for more than about two minutes.

    Now if you were to put him on the witness stand as an expert witness, with somebody like Bill Nye as his opposite number, I know which one I'd trust to know what the hell they're talking about but I also know who the jury would find persuasive. And that's the problem with the "dueling expert witnesses" method of arguing complicated scientific issues before a jury of laymen, the prosecution has unlimited resources to find credible expert witnesses who spend far more time working on their credibility rather than their expertise while the defense struggles to find decent experts they can afford. Prosecutors firmly hold to the idea that dazzling them with brilliance is far less a winning proposition than baffling them with bullshit so they hire experts who aren't really experts, they just play one on TV. A good expert witness isn't necessarily one who knows what he's talking about, it's one who knows how to convince the jury he knows what he's talking about.

  • Uncle Jay||

    RE: Forensic Science Is a Mess, and the Justice Department Wants to Keep It That Way

    The good ol' USA is number one in incarcerating people in the world.
    We must keep these numbers up if we are to maintain our status of a gulag nation.
    If tens of thousands of innocent people are wrongly put into prison to keep our ranking high, then so be it.
    That's the price we all must pay to be number one.

  • Ron||

    A relative who is a biologist told me that they can rarely identify finger print with any accuracy

  • bacchys||

    Fat chance that America-hating piece of shit Sessions does anything except push for more junk science.

GET REASON MAGAZINE

Get Reason's print or digital edition before it’s posted online