4 ms·

The New Yorker had a really interesting article about arson forensics, a few years ago. It interleaved the case of someone being executed for arson, with a thor

The New Yorker had a really interesting article about arson forensics, a few years ago. It interleaved the case of someone being executed for arson, with a thorough debunking of the science behind his conviction. And then fiber analysis has been debunked too. And the list goes one: hysteria over child abuse combined with well-meaning but incompetent interviewers creating false memories in supposed victims. The problems with coerced confessions. It's all quite terrifying.

lonelappde 7y ago
Forensic "science" is full of pseudoscience that charlatans use to put innocent people in jail for money. https://youtu.be/ScmJvmzDcG0 https://youtu.be/ScmJvmzDcG0
fsh 7y ago
Link to the article (which is definitely worth a read): https://www.newyorker.com/magazine/2009/09/07/trial-by-fire https://www.newyorker.com/magazine/2009/09/07/trial-by-fire
djsumdog 7y ago
There was a Citations Needed podcast about pseudoscience in the criminal justice system: https://citationsneeded.libsyn.com/episode-94-the-goofy-pseudoscience-copaganda-of-tv-forensics https://citationsneeded.libsyn.com/episode-94-the-goofy-pseu... It's an interesting podcast, but their interviewee sucked and she didn't really cite any actual science; just hand waved away things. I've found the studies they cite on fingerprints (your fingerprint could match up with 1 in ~300 people; they're not as unique as once though -- the study that's often cited is from the FBI) but I can't find anything on gun/bullet print identification. Supposedly you can't match a bullet to a gun, especially with more modern firearms with precise milling. Yet all the sourced I've found should photos of bullets with distinct patterns that seem to match a particular firearm.
- manicdee 7y ago
  Then we find that all these photos are stock photos from the depths of FBI labs and none of the photographs show an actual slug collected from a crime scene.

lonelappde 7y ago

Forensic "science" is full of pseudoscience that charlatans use to put innocent people in jail for money. https://youtu.be/ScmJvmzDcG0 https://youtu.be/ScmJvmzDcG0

fsh 7y ago

Link to the article (which is definitely worth a read): https://www.newyorker.com/magazine/2009/09/07/trial-by-fire https://www.newyorker.com/magazine/2009/09/07/trial-by-fire

djsumdog 7y ago

There was a Citations Needed podcast about pseudoscience in the criminal justice system: https://citationsneeded.libsyn.com/episode-94-the-goofy-pseudoscience-copaganda-of-tv-forensics https://citationsneeded.libsyn.com/episode-94-the-goofy-pseu... It's an interesting podcast, but their interviewee sucked and she didn't really cite any actual science; just hand waved away things. I've found the studies they cite on fingerprints (your fingerprint could match up with 1 in ~300 people; they're not as unique as once though -- the study that's often cited is from the FBI) but I can't find anything on gun/bullet print identification. Supposedly you can't match a bullet to a gun, especially with more modern firearms with precise milling. Yet all the sourced I've found should photos of bullets with distinct patterns that seem to match a particular firearm.

manicdee 7y ago
Then we find that all these photos are stock photos from the depths of FBI labs and none of the photographs show an actual slug collected from a crime scene.