Join 3,564 readers in helping fund MetaFilter (Hide)


"High-tech forensic perfection is a television fantasy, not a courtroom reality"
March 25, 2009 2:04 PM   Subscribe

The American National Academy of Sciences recently released a report that punched a few holes in the credibility of the forensic sciences: often seen (and portrayed) as infallible, in practice they're non-standardized, subjective (warning: pdf with gory image), accepted without rigorous testing (pdf), and lousy with dilettantes. A Canadian inquiry into the work of a pathologist whose testimony wrongly convicted a man of anally raping his four-year-niece to death says that forensic science is useful, but that we're doing it wrong. It's beginning to dawn that what we used to think of as a few bad apples may actually be symptoms of a deep rot in the field itself.

As for the wrongfully convicted Canadian, after 12 years in prison for a crime that didn't exist, he forgave (.mov).
posted by hayvac (29 comments total) 24 users marked this as a favorite

 
related new yorker article the csi effect
posted by andywolf at 2:26 PM on March 25, 2009 [2 favorites]


NPR on the topic.
posted by cjorgensen at 2:32 PM on March 25, 2009


They said the cause of death was homicidal asphyxia, not anal rape.
posted by merelyglib at 2:36 PM on March 25, 2009


Wow. The effect that lead to the false determination of physical abuse in the wrongful-convcition case is the same effect that plays a crucial role in Season 5 of the Wire.
posted by molecicco at 2:39 PM on March 25, 2009


They said the cause of death was homicidal asphyxia, not anal rape.

Read a little further.
posted by hayvac at 2:39 PM on March 25, 2009


Maybe it is considered an unimportant thing, but I wish people would avoid electing phrasing such as "anally raping his four-year-niece to death," when in fact the charge was that death was caused by strangulation.
posted by nanojath at 2:41 PM on March 25, 2009


Metafilter: lousy with dilettantes.
posted by penduluum at 2:41 PM on March 25, 2009 [2 favorites]


I love how on CSI they have a question and then the next scene shows them in the lab getting the answer. If you don't know why this is funny, you haven't ever worked in a lab.
posted by telstar at 2:41 PM on March 25, 2009 [14 favorites]


And they were wrong about that, too.
posted by mr_roboto at 2:42 PM on March 25, 2009


Maybe it is considered an unimportant thing, but I wish people would avoid electing phrasing such as "anally raping his four-year-niece to death," when in fact the charge was that death was caused by strangulation.

I wasn't going to split hairs above, but to be clear the charge was that he strangled her to death while raping her. The reason it's important is that it was a big factor in getting him convicted: the forensic expert says, "Yes, I can tell from X and Y that she was anally penetrated during the time that only Mr. Mullins-Johnson was with her," when in fact no one ever laid a hand on her.
posted by hayvac at 2:47 PM on March 25, 2009


It just goes to show you can't be too careful.
posted by Schmucko at 3:08 PM on March 25, 2009 [5 favorites]


I am clear on the the details, I read the entirety of the linked report on the alleged crime. The way you stated it was inaccurate (in that the alleged rape and murder were separate and that the murder had a distinct cause not related to the rape) and phrased in such a manner as to amplify the gruesome and horrific nature of the charge and I personally believe that we should avoid such tricks with language which are constantly being used to emotionally manipulate us, and instead opt for the more accurate presentation of facts. I don't think you phrased it thus out of some bad motive, hayvac, but I don't think it's accurate either. Whether it matters is I guess a personal opiniont.

I honestly hope it can be left aside as a minor derail because the central discussion is more relevant and important.

As a somewhat accidental professional in quality assurance (you'd think the only one on Metafilter judging by the non-response to my last AskMe question) I think the question of whether criminal forensics needs its own set of quality assurance standards is a fascinating one. It's ironic that from the most basic principles of science the frequent premise of shows like CSI is utterly wrong: nothing could be worse than for individuals examining evidence to be intimately involved with and knowledgeable of the crimes under investigation.

One thing I am left wondering is whether the family of the child who died accepts the outcome and whether the apparently estranged family is reconciling. It seems like there is no bottom to the harm of this kind of miscarriage of justice:
posted by nanojath at 3:08 PM on March 25, 2009 [2 favorites]


(opiniont is not a word and it looks like I had more to say but I don't - the : should have been a . - poor quality control).
posted by nanojath at 3:10 PM on March 25, 2009


Reason magazine and in particular Radley Balko has been on about this for years.

Here is a November 2008 article on the firing of Stephen Hayne by Mississippi and there are more in Balko's article list.

Experts need to be on tap, not on top.

In a little remarked upon part of the President Eisenhower's farewell address that talked about the military industrial complex he said:

Yet, in holding scientific research and discovery in respect, as we should, we must also be alert to the equal and opposite danger that public policy could itself become the captive of a scientific-technological elite.

A lawyer who I know who worked in liability claims for the government described how the different sides had lists of doctors who they called on, there were some who were guaranteed to say that in a say, a back injury case, that the plaintiff was faking while others would be guaranteed to say that work had exacerbated the injury. It is unsurprising that criminal cases should be subject to similar manipulation.

The problem of trusting experts for the legal system and governments is very deep. The current financial crisis involves people having to trust economists, i.e. the very people who got us into the mess. Military matters must involve consultation with generals.

Experts often have very strong opinions and there are often majorities in various fields. These problems can be addressed by making sure expert advice is checked by other experts and that dissenting voices are sought.
posted by sien at 3:22 PM on March 25, 2009 [1 favorite]


Yet, in holding scientific research and discovery in respect, as we should, we must also be alert to the equal and opposite danger that public policy could itself become the captive of a scientific-technological elite.

It seems to me, however, that the main problem is that the forensic experts are decidedly not the scientific-technological elite. Most scientists would blanch at the certainty with which they state their conclusions. There's little or no characterization of the reproducibility or error of the methodology, let alone peer review. Many of the experts are medical doctors with no formal scientific training (sorry, guys, med school doesn't count).

If the scientific elite were in charge of forensic analysis, you wouldn't see half the nonsense described in this post. Much of what these people do is, from a scientific standpoint, laughable.
posted by mr_roboto at 3:42 PM on March 25, 2009 [12 favorites]


They said the cause of death was homicidal asphyxia, not anal rape.

Read a little further.
posted by hayvacPoste



"It is also now clear that there is no evidence to support a finding of homicidal asphyxia, the cause of death proffered at trial."

--From the 'wrongfully convicted' link.
posted by merelyglib at 3:44 PM on March 25, 2009


Which doesn't change that cause of death at the time of the trial was attributed to "homicidal asphyxia", not "anal rape".
posted by Decimask at 3:59 PM on March 25, 2009


Oh, for the gmail 5 second rule...
posted by Decimask at 4:00 PM on March 25, 2009


Apologies, merelyglib. I'm going to log out until my brain can parse English correctly again.
posted by Decimask at 4:02 PM on March 25, 2009


I wish I could favorite mr_roboto more than once. It is the truth and needs to be heard.
posted by ltracey at 4:05 PM on March 25, 2009


Sigh. Look, I've got quite a few friends who work in this field. They take what they do very seriously, and I honestly think they are contributing quite a lot to making this a more just world to live in.

That aside, there are a lot of dirty secretes in forensic anthropology that simply are not widely known to the general public. Many of the methodologies and tools used are frames of wire held together with wishing and lots of good luck. But that doesn't change the fact that a great many of them do work.

What it does mean is that even well trained human beings make mistakes. Especially under tight time lines, stress, overwork, and budget cutbacks - in what is already what I consider one of the most thankless, horrific jobs on the planet. And if you don't believe me, go spend a weekend cleaning up liquid decomp out of a south Florida attic in the high of summer, while the landlord yells at you and the police are demanding answers.
posted by strixus at 4:44 PM on March 25, 2009


In other news - stuff on TV isn't real. Get over it.
posted by strixus at 4:46 PM on March 25, 2009


But that doesn't change the fact that a great many of them do work.

Which ones? How do you know?
posted by mr_roboto at 4:54 PM on March 25, 2009 [2 favorites]


In other news - stuff on TV isn't real. Get over it.

Except that people are influenced by what they see on television, and one of the elements of this discussion is there being a perception of the validity of evidence that is out of line with its actuality. I have no trouble believing that television shows may be leading to a lot of average jurors having irrationally skewed perception of what may constitute reasonable doubts in a case based on forensic evidence.

It is entirely possible that both strixus and mr_roboto are correct: that criminal forensics is plagued by essentially unscientific methods AND that it is still leading to a net improvement of justice (given the notorious unreliability of things like eyewitness accounts). But since it is a demonstrated fact that failures in this field are putting innocent people in jail and letting guilty people go free you have to ask just how much this happens and the only way to do this is to impose more of the rigor of the true scientific method and apply the methods and metrics of quality assurance and control systems.

I really wonder if we're in any sense prepared to financially accommodate what it would take to get there, though - segregating the examination of evidence from knowledge of its origin and context almost automatically means more staff, for example, things like adding more control tests or statistically analyzing results is by definition extra work. It's in no way an easy situation.
posted by nanojath at 7:37 PM on March 25, 2009


I've watched CSI. Those guys can't even use a pipetter properly. It's no wonder they get shoddy results.
posted by kisch mokusch at 11:30 PM on March 25, 2009


nanojath: One thing I am left wondering is whether the family of the child who died accepts the outcome and whether the apparently estranged family is reconciling.

There was a program on CBC about this. Some of the family accepts that the man didn't do it but still want little to do with him. The dead girl's parents split up and that seems to provide a certain animus toward Mullins, especially from his ex-sister-in-law who adamantly refused to accompany her mother-in-law on prison visits. The dead girl's father is an alcoholic whose reconciliation with his brother seemed forced at best, and didn't lead to any lasting connection between the two. Mullins himself had come to think that his brother might have been the abuser/killer so that connection, between two very close brothers, seems severed forever. The dead girl's sister is convinced he should still be in prison. She said, "He was convicted, wasn't he?" or something like that. Mullins' mother came to visit him in prison and probably (some have their doubts) believed in his innocence, but she's dead now. The worst part is that Mullins himself seems to feel, if not guilty, somehow responsible for everything. All are damaged people but Mullins is a mess.
posted by CCBC at 12:01 AM on March 26, 2009


On a related note, the Phantom of Heilbronn, one of Europe's most notorious criminals apparently is a figment of DNA sample contamination (via).
posted by ghost of a past number at 3:19 AM on March 26, 2009


you don't need forensic evidence anyway.

the west memphis three have been in prison for fifteen years now.
posted by Hammond Rye at 10:17 AM on March 26, 2009


For example, the dilation of the child’s anus that was thought to be so indicative of anal penetration and chronic sexual assault (recall Dr. Zehr and her comment that this was one of the worst cases of child sexual abuse she had seen) is a normal finding in children after death.

So did Dr. Zehr ever actually see a dead body before then? Dubba Tee Eff, man.

Speaking of the West Memphis Three, the police in that ordeal made the same exact mistake, but not the freaking medical examiner.
posted by dgaicun at 2:25 PM on March 26, 2009


« Older Innsmouth: The Musical. Carol of the Old Ones. Sho...  |  Sick of worrying about the glo... Newer »


This thread has been archived and is closed to new comments