14 comments

[ 2.7 ms ] story [ 42.5 ms ] thread
"the purpose of forensics is to solve crimes"

I agree with this article, I think it's a good one. But here's my nit: the purpose of forensics isn't to "solve crimes", it's to get convictions, which isn't the same thing. That's a huge part of the problem.

And the remainder of the problem is that too many of the people inside the system believe that they are the same thing.
Shouldn't the purpose of any forensic investigator be to get to the truth and provide evidence of it? Why is a conviction the measure of success?
It gets used as a measure of success because it's measurable. You can easily count the number of defendants convicted. How do you count the number who are wrongly convicted?

But you get what you measure.

An outsider would think so, but the DA and police force are motivated by convictions and conviction rate. The incentive is in the wrong place.
After reading this who is surprised?
But hey, at least you've got a written consititution. That must avoid a lot of awkward nitpicky arguments about interpretation and individual rights versus governmental powers, right?
Nitpicky arguments are great. It means that both parties are maneuvering closely on each side of a well defined line.
It helps to have some background before assuming that the government will keep an obviously innocent person in jail just because they missed a deadline.

The justice system errs heavily on the side of letting people go. For every person exonerated with DNA evidence, there are tons of people who quite likely are guilty but got off for procedural reasons. E.g. just because a warrant was invalid doesn't mean that the murder weapon recovered from the suspect's car doesn't go to proving guilt.

At the trial stage, a procedural defect is enough to torpedo a conviction. After the trial and appeals are done, however, a conviction is assumed to be valid. Nonetheless, there are outs. Even after a conviction, 28 U.S.C. 2254 (the habeas corpus law), lets you attack a conviction. It puts up hurdles, though, such as deadlines, to force prisoners to bring their claims in a timely manner.

But, and the article doesn't mention this, there's the ultimate out: under McQuiggin v. Perkins, if you can prove "actual innocence" (i.e. that no reasonable juror would've convicted you in light of your new evidence), you can overcome the hurdles in the habeas procedure.

What trips up people at this stage, though, is that the overwhelming majority of habeas petitioners aren't actually innocent. They might have new evidence which at trial could've created reasonable doubt, but their convictions are still supportable on the basis of the other evidence that was presented. It's hard for people to think of it this way, but ultimately the justice system has to be based on probabilities. If someone is pronounced guilty after being indicted by a grand jury, tried, and exhausting his appeals, he's probably guilty. You can draw the line wherever you want, but you do have to draw a line, somewhere, or else give up on the idea of putting people in prison at all.

While reading, I kept wondering what other evidence there was in the case. They make it sound as if it's just the expert witness.
> The new evidence must actually prove innocence, not just be of the nature that, at trial, would've allowed you to lawyer yourself out of a conviction.

That seems to be the problem, no? Proving innocence is close to impossible. How do you prove a negative? From the inside of a prison cell?

There is obviously a concern that discrediting an expert would create a resource problem if every defendant he testified against could get a new trial, but isn't that exactly what is necessary to get prosecutors to ensure their experts are credible?

Sorry, I clarified in my edits. The "actual innocence" standard doesn't require you to prove a negative. But it for require creating more than mere doubt, and has to go to substantive rather than procedural issues.
> If someone is pronounced guilty after being indicted by a grand jury, tried, and exhausting his appeals, he's probably guilty.

And if you're African American, a lie from a 12-year old kid can get you onto death row and in total 39 years jail time, with not even winter clothing or a single dollar in the pocket [1].

Black people tend to be fucked by cops and the justice system, even in Europe.

[1]: source http://www.spiegel.de/panorama/justiz/39-jahre-im-gefaengnis... (german)

You are not required by law to prove your innocence