I’m wicked busy today, so have a letter from my past:
Um….huh? What’s the context?
Thanks…I believe you have it.
Funny that the first comment 5 years ago was also one expressing confusion about what this means. Let’s do this again in 2021.
Ah, I forgot that I’d posted that five years ago. Well, the explanation is there.
I’ll bite….what did a professor of Ecology and Evolutionary Biology offer to Simpson’s defense?
Something to do with DNA I would imagine.
If it is not the fittest then it does not survive?
Taking a look at the common American waist makes me think fitness means something different today.
OK. Best teaser post on this site. Ever.
I actually remember reading about it after watching American Crime Story! I was able to find the original article online.
Well well, that explains it!
I was worried this was heading towards another cultural appropriation post.
Who played you in the TV show?
Good question. If they hadn’t OUTRAGEOUSLY and SCANDALOUSLY left Jerry out of the TV show I reckon a good choice to play him would’ve been Liev Schreiber…I’m thinking of him in the role he played in Spotlight. A kind of quiet, furry authority.
Jerry, I had no idea that you played a part in such a huge moment in American societal history. It’s very impressive, not just the restraint and rigour of your contribution but the fact that it’s not announced at the top of the site in huge letters: ‘Jerry Coyne: Star Of OJ Simpson Case – What Have You Ever Done That Compares – I’ll Tell You – Nothing’, as it would be if it were me in your place.
No, but that is what would happen had it been Trump instead of Jerry. Trump might even have named the entire OJ case after himself.
TBH, I think if they had called on Donald Trump for a detailed statistical DNA analysis then I’d have had even graver doubts about the reliability of the result.
A bit late to the game but it has to be said. If Jerry is ever to be portrayed on the screen that job can fall on only one actor, Richard Belzer (think Sgt. Munch of Law & Order: SVU).
Expert witness regarding blood type and DNA matches?
Ha. So it was you driving the White Bronco!
One of my favorite jokes from Arrested Development was when George Sr. was buying an SUV, the salesman said, “They got rid of the Bronco because of the fugitive connotation; they replaced it with the Escape.”
George Sr.: “What a fun name!”
“I pray” — little did he know….
Self-censored nasty comment would have gone here. My error rate would be 100%.
I really hate that prosecutors take a good technique and misrepresent it like this. That sort of thing can only hurt the public’s opinion of science in the long run. Another example: remember when the FBI claimed that their fingerprint conclusions couldn’t be wrong? I wonder how many guilty criminals who were accurately identified through fingerprinting walked free because of that sort of dishonest hype baloney.
DNA is a reasonably reliable way to tell if someone was at a scene. That certainly can be good information to have, depending on the sort of crime you’re investigating. But its not immune to human error and its not a magic criminal-finder.
DNA is a reasonably reliable way to tell if someone was at a scene.
Is it really? I was under the impression that it indicated that a sample of someone’s DNA was at the scene. it’s the SOCO’s job (Scene Of Crime Officer, who is not normally a police officer but a civilian) to document where that sample was found, and what it’s relationship to the rest of the evidence is.
Picking up random fag butts in the street and scattering them around a crime scene is (reputedly) a popular tool for driving up the cost of investigation amongst the more forensically savvy criminals, while reducing the chances of any DNA they inadvertently shed being recognised as significant.
“. . . fag butts . . .”
In this day and age of hypersensitive regressive leftism that phrase is liable to cause an epidemic of aneurysms and hurricanes of spittle. When culturally appropriated into the US it of course means something quite different. Made me laugh out loud.
Excellent reasons for using the phrase. I’m almost tempted to get a gollywog badge because I always liked that jam.
This is a great article but directly relevant at the end where the OJ case comes up.
I didn’t know you testified in court on DNA evidence Jerry. And you did it for free too. You could have made a small fortune off the OJ trial. I’m sure there’s plenty of people in the field of genetics who make a living off of testifying in court. Just another reason to admire you.
Testimony by a professor of statistics at North Carolina University, Bruce Weir, (who is now at the Un. of Washington) was provided as to the likelihood that samples of blood found at the Bundy crime scene could have come from someone other then O. J. Simpson. Professor Ceiling Cat informed the defense team that, in fact, Prof. Weir had made an error in his calculations, an error that exaggerated the likelihood that, in fact, Simpson was the donor.
The problem for the prosecution was that Prof. Weir was the man who computed all the probabilities which were quoted by various witnesses at the trial. As one of the commentators on the trial put it, Weir was the mathematician who couldn’t add.
If the loci don’t match, you must dispatch.
I’m not so sure that I would be proud to have participated in one of the centuries most significant miscarriages of justice.
And, who, exactly is at fault for that? The prosecution? The defense? The judge? The jury? Should they all be ashamed? Perhaps they should have just executed Simpson without a trial.
Jebus, what a stupid thing to say.
This isn’t questioning your expert witness contribution. I was surprised that you would tout this letter given it is associated with an appalling miscarriage of justice…the result, without doubt of a completely botched prosecution. The result also, of egregious behavior by the defense.
Are you proud of your involvement?
I have answered you in a post today (Friday). Your letter is insulting, misguided, and, most of all, shows an embarrassing ignorance of how the US judicial system is supposed to work. I suppose you thinkwe should have just given OJ a lethal injection without trial. Jebus.
Since you ask, the jury and the defense were at fault for the miscarriage of justice. And yes they should be ashamed.
I followed it closely, and felt the jury got it just right – they had to acquit because reasonable doubt existed as to the chain of evidence. And that would be the fault of the cops and prosecution.
What a ridiculous comment. Are you suggesting that Jerry should never have got involved with providing expert testimony on the accurate use of statistics in DNA analysis, because in this particular case it might have allowed the prosecutor to deceive the jury on the DNA element of the evidence and get a conviction?
If that’s your approach to due process and justice in general, we may as well just have the audience in Jerry Springer decide who’s guilty based on ten minutes of reading the National Enquirer and a mass shouting vote, who needs DNA at all?
“…based on ten minutes of reading the National Enquirer…”
I doubt the people who read the National Enquirer could last 10 minutes time reading.
Hey, Prof. Ceiling Cat didn’t testify as an expert witness at the trial. His contribution was to correctly inform the defense of an error made by the state’s expert witness in statistics, Bruce Weir, in the latter’s testimony. Are you saying that Prof. Ceiling cat should have kept quiet?
I would be proud. Not for the acquittal, but for doing my best to insure that the evidence presented was correct and accurate.
The prosecution had the opportunity, prior to trial and during the trial, to properly determine the probabilities and to revise their arguments to meet the actual probabilities. The also had the opportunity to not mess up most of the rest of the evidence, not use fabricated testimony, and a host of other issues. They didn’t.
I have no basis to know if OJ did it, though I suspect he was involved, but that doesn’t change that he, as all of us in the US do, theoretically have the right to a fair trial rather than a railroading. Sometimes the guilty get off. I hope this means (though I know that in some cases it doesn’t) that the innocent don’t get convicted.
I would much rather the tendency to type II error with regard to guilt.
No one — no one, ever — should hesitate to say with pride that they honestly participated in our adversarial system of justice. Ever.
You want to blame somebody for OJ walking, blame Mark Fuhrman and Detectives Lange and Van Atter and the whole LA Sheriff’s Department, blame Marcia Clark and Chris Darden and Gil Garcetti and the DA’s Office, blame Medical Examiner Dennis Fung — blame the whole lot of them, since the People of California grossly mishandled the OJ case, start to finish. Although, as far as I can see, all except Fuhrman did so honestly, so have nothing to be ashamed of.
The miscarriage of justice would have been if Simpson had been convicted on evidence that had been wrongly interpreted. If the prosecution fails to persuade the jury, the answer is not to conceal flaws in the case, but to make a better case.
Science is about being honest especially with yourself. I think JAC can be proud that he upheld scientific principles, even though the result turned out to be unpopular.
Tomorrow we’ll see the thanx note from Ken Starr for PCC’s help with the blue dress!
I applaud Jerry’s efforts in this, but I wish that there were wider efforts to improve the use of statistics in the legal system.
In the same trial, Dershowitz argued that among physically abusive husbands, only 1 in a few thousand ultimately kill their wives, therefore Simpson’s prior abusive behavior was not relevant. But this is nonsense, of course. The relevant probability is conditioned on our knowledge that SOMEBODY murdered his wife. When we know that a wife has been murdered, and that the husband has a history of physical abuse, given no other information the probability is around 90% that he is the killer.
It really seems to me that there should be a burden upon any officer of the court who cites statistics in any context to have basic competence. Somebody of Dershowitz’s general intelligence surely knew better, and in my opinion this went beyond vigorous advocacy into deliberate deceit.
It really seems to me that there should be a burden upon any officer of the court who cites statistics in any context to have basic competence
In a perfect world, yes. But not everything can be competent on everything, and Bayesian Statistics is a pretty far reach from Law School.
Besides, as I understand the court system it’s really the judges’ job to catch when a prosecutor (or defense) attorney uses an invalid argument regarding letting data in or out. Its not the lawyer’s job to not-try weak or invalid arguments. So you should be arguing that judges should have an understanding of statistics so that they can effectively rule when lawyers try statistical shenaningans. But that brings me back to my first point…to be a judge already requires a lot of subject matter expertise. Saying they need to be educated in [subject] so that they can’t be fooled by lawyers using [bad argument in subject] is a pretty unworkable requirement, for any subject other than [law].
I do understand the status quo, and why the system is broken. I don’t accept that we just have to accept that and that it can’t be fixed. The entire notion of “evidence” is bound up with concepts of probability and statistics. The entire purpose of a trial is to determine whether the aggregate evidence passes a probability hurdle.
I’m not talking about technical issues like understanding DNA probabilities – of course that’s something where an expert witness is appropriate. But there’s no excuse for such blatant misrepresentations of elementary principles as the one that Dershowitz got away with.
It seems entirely unacceptable to me that we should accept officers of the court who are totally ignorant of the most elementary concepts that allows us to interpret evidence quantitatively. It should be an integral part of legal training, at least for anyone who’s involved in a trial.
I agree that neither judges or attorneys should or reasonably could be held responsible for being competent in such a broad range of knowledge.
But are you implying, I certainly could be misinterpreting you, that lawyers should be allowed to lie and it should be the responsibility of judges to catch their lies? I suppose it can be a fine line between using an argument you know is bad and lying. I think lawyers should absolutely be held accountable by the law for lying.
I believe this is partially why some jurisdictions allow the presiding judge to call their own experts, etc.
Anyone from (I believe) parts of the Spanish-speaking countries who can confirm this? I seem to remember this being the case in some of those countries.
Dersh did not testify before the jury, this was an off-hand opinion he offered in an interview. Had no effect on the outcome of the trial.
Are you sure a similar argument wasn’t offered by one of the defense attorneys in the actual trial? I seem to remember otherwise, though of course I could be mistaken. The trial was loooooong. Seemingly never-ending.
—-In the same trial, Dershowitz argued that among physically abusive husbands, only 1 in a few thousand ultimately kill their wives, therefore Simpson’s prior abusive behavior was not relevant. But this is nonsense, of course. —–
I remember this argument being made at the time, and knew then that there was something dreadfully wrong and fallacious about it. Though I couldn’t put my finger on it. I just could tell Dersh was trying to pull a fast one.
The correct way to look at the circumstances was not “How often do abusive people murder their abuse victims?”, but rather, “When an abused person is found murdered, how often is the murderer someone other than the abuser?”
I’ve just finished a tutorial for Forensics MSc students where we discuss just that example.
Yes, the important probability is
P(killed wife GIVEN known wife-beater AND wife murdered by someone)
Its surprisingly easy to get wrong though, its so obvious that she is dead (otherwise why the trial) that you don’t notice that you need to condition on it.
Of known wife-beaters whose wife is murdered by somebody, how many of them did it themselves?
Previous post: “The state often got the calculations—the “match probabilities”— completely wrong, biasing the data against the defendant.”
It’s far worse than just that:
Polygraph Testing Too Flawed for Security Screening
FBI admits flaws in hair analysis over decades
ACLU Urges Mass. Court to Vacate 24,000 Drug Cases Based on Tainted Lab Evidence
Bad FBI Science – A forensic whistleblower on the latest scandal.
I’ve long suspected that firearms matching is a deeply flawed procedure (i.e., when a bullet is matched to a gun). To see the reliability of that, you really need to do blind testing: fire a bullet from, say, one of a hundred guns, and then have a technician say which gun fired the bullet based on the scratch patterns.
NAS paper about the reliability of forensic evidence in general – fingerprints, tire tracks, fire evidence…
“Although research has been done in some disciplines, there is a notable dearth of peer-reviewed, published studies establishing the scientific bases and validity of many forensic methods.
As recently as September 2008, the Detroit Police crime laboratory was shut down following a Michigan State Police audit that found a 10 percent error rate in ballistic evidence.”
Absolutely, blind testing is essential.
I really should have explicitly mentioned this example of blinding in the lecture on experimental design I gave our Forensics students this week. I’ll just revise the notes and upload to our system.
Thanks for the reminder!
It should be noted that an FBI hair and fibers expert, Douglas Dietrich, was not allowed to tell the jury that there was a match between hairs found at the crime scene and exemplars taken from Simpson, only that the exemplars came from an African American.
What I would expect is that, in a criminal trial, when an expert in their field is called on to give an expert opinion, that is what they do. And, although I may disagree with the overall outcome of the trial based on reasonable doubt criteria, I can’t fault an honest expert for giving an honest opinion based on the data available. Having said that, I think that there are way too many “paid experts” that causes the process to come into question (And don’t even go there; I am absolutely not discussing the author of this website!)
That was truly the most fascinatingly unexpected thing to have read here. The original post must have been just before I started visiting this site. And you actually said to once that your life isn’t interesting enough for an autobiography…!
Oh, so it’s all YOUR fault.
The wording of the letter is a fascinating example of lawyer hyperbole verging on parody but doesn’t mention what the Professor’s contribution actually was, so I am thinking that every defence “witness” received the same missive.
Notify me of new comments via email.
Notify me of new posts via email.
Enter your email address to subscribe to this blog and receive notifications of new posts by email.
Join 43,358 other followers