What I found weird while reading this article was that they all said the results would be "nominal" and nothing will realistically change regardless of if the DNA tests were actually erroneous.
Uhhhh what? If people were sentenced for a crime based on the fact that their DNA matched with the DNA found at the scene of the crime they supposedly committed, then surely there's a massive problem here? Just one case of this would mean an innocent person has been rotting away in prison for the last decade... and it won't just be one case!
I'm confused as to why they're not thinking this is a big deal.
You're completely correct when you say the public was made to believe DNA testing was completely foolproof. If you're in court, being prosecuted and they prosecution says that the DNA matches yours everyone instantly thinks you are in fact guilty. If you've got a jury they'd all think the same too.
I thnk this is a really big deal, but I'm not in the US so I may be missing something (again). :-)
> If people were sentenced for a crime based on the fact that their DNA matched with the DNA found at the scene of the crime they supposedly committed, then surely there's a massive problem here?
I get the impression - though I don't have the data/numbers obviously - that in almost all of those cases the DNA evidence was being used as the cherry on top of the evidence sundae they had already compiled.
That's why it won't realistically change much; either very few or none at all are convicted _only_ because their dna happened to be found at a crime scene.
"... in almost all of those cases the DNA evidence was being used as the cherry on top of the evidence sundae they had already compiled..."
Not to put too fine a point on it...
but that's the whole problem DNA was supposed to solve. Cases like rape X, where all of the evidence pointed to guy A. Then we test a known serial rapist, guy B, 30 years later to find out that he actually committed rape X. The Innocence Project sails in, and in due course guy A is exonerated and sent on his way after having spent 31 years in prison. We're stuck with a sizable financial liability to the ex-prisoner. And for some reason, everyone calls this a happy ending ???
DNA was supposed to protect us from situations like that. But now you say that the evidence against guy A in rape X really would have been enough in any case... so we shouldn't read too much into the fact that the DNA samples don't match ???
I'm just confused as to what evidence would EVER prove guy A's innocence in such a system ???
I was responding to a specific interpretation of OP's post which, upon reflection, might not have been the best reading. In short: are people convicted based solely on the presence of their DNA at a scene.
The way I read it was if it was being used in a frivolous manner, e.g. "There was a murder in Times Square last night. We found a hair of yours in Times Square. Case closed." Obviously there are more serious uses, and clearly a spectrum exists of how important the evidence is in the case.
If, in the cases spoken of in the article, DNA evidence was used as the strongest piece of evidence to get a conviction, then obviously that matters. I didn't mean to imply it didn't.
But if, in the cases spoken of in the article, the DNA evidence was incidental to a lot of other evidence - note this was the scenario I was responding to - and not necessarily enough to convict on its own, and more of the "cherry on top", then I agree with them that it might not change much in the way of these cases.
We won't know what percentage fit either description until they've gone through them.
> The way I read it was if it was being used in a frivolous manner, e.g. "There was a murder in Times Square last night. We found a hair of yours in Times Square. Case closed."
There will always be more evidence than that. You can find circumstantial "evidence" of anything. If you were friends with the victim then "the defendant knew the victim." If you were enemies or competitors then you had motive. If you had never met the victim but the victim was robbed then you had motive again, and the same if you ever expressed opposition to any class of activity the victim had ever engaged in. If you were caught on surveillance near where the crime was committed then it will be used against you, but if you were at home sleeping then you have no alibi. And so it goes.
A prosecutor can whittle the world of facts about your life down to only the ones that imply you might have committed the crime. None of that stuff actually proves anything -- but it sure makes a good show for the jury.
Throw fraudulent DNA evidence on top of it and the outcome is predictable.
The problem lies where the "cherry on top" is that which causes the jury to eliminate the shadow of a doubt. In that way, maybe it's not so much a "cherry on top," as a "linchpin," or "keystone."
This article is about errors in a set of population priors used in software. That could affect the outcome of tests, but it also might not. However, given that the FBI was recently caught having faked an entire field of forensic science (hair analysis), there is little reason to extend them the benefit of doubt.
The important question is: are these crime labs routinely processing control samples, and if so, are they misclassifying any of the control samples?
Polygraphs? They know full well polygraphs are not "lie detectors". And a lot of the general population knows that too. They're still useful when recruiting candidates. Obviously their results are not admissible in criminal cases.
I know in New Zealand, you've got the right to redo tests with whoever you want (i.e. private labs).
So, if the prosecution has said that your DNA was found at the scene or something, you can then get that test done privately (yes it'll cost money) to see if it matches up. You can do this as many times, with as many labs as you need.
Is this a normal thing in the US too? I'm assuming it is. If my assumption is correct, then wouldn't another lab doing the same test come to a different conclusion?
It's not that expensive. About $500 NZD for a basic test. Yes, expensive if you don't have $500 but there's many ways to go about it: ask for invoice, get legal aid to pay, do it through lawyer (who adds to their bill), hope you still have family who loves you or friends.
Alternatively, you can make the court do it to rule out any problems. Most of the time the labs come up with the same result but the courts are more than happy to make sure it's about water tight.
It kind of makes me think/wonder that in criminal cases perhaps forensic biometrics work should be done by two or more labs (maybe using two different methods) to help cross-validate the results. My understanding is that most of this work happens in a single lab under one test, so contamination or procedural problems go unchecked.
Maybe the same lab shouldn't even be testing both samples. Just send the sample from the victim to lab #A and same from the accused to lab #B and compare the digital results.
The two samples shouldn't be sent in together, there is absolutely no benefit in that. Plus the lab shouldn't know what they're testing so there can be no accusations of bias/manipulation (e.g. "this person is definitely a child rapist, if you guys find this DNA matches this child rapist goes to jail, if it doesn't' (wink wink) then they go free").
The anonymous digital results can then be shipped to the defence so that they can double-check the degree of similarity with an expert of their own.
I see where you're coming from, but I'd be concerned if one of the labs has a broken process somewhere (or contaminates the sample) that it would produce a false negative and a possible match would not be found -- thereby leaving a bad person out in the wild.
Just goes to show that humans can even mess up "perfect" evidence. How many convictions will get overturned by this, I wonder. A factor of 10 is pretty large, if they're talking about "1 in 180" levels of probability.
I wonder if innocent men are rotting away in jail that have given up themselves and been given up on. They know they are innocent but I doubt 'the system' will swoop down and fight for them.
Interesting! I suppose there's a perverse incentive for the system not to track the falsely imprisoned down. They have a right to sue for false imprisonment.
The public was made to believe DNA testing was absolutely bomb proof, there are even entire TV shows centered around this "fact" (CSI whatever ...)