Wrongful convictions are my biggest concern about the judicial system. People make a big deal about the reasonable doubt standard, but in actual fact, juries are willing to convict on shockingly thin evidence. (I realize I have a skewed perspective because the media tends to focus on guilty verdicts, but it really does seem like juries believe just about anything the State asserts. A quick google search tells me that in federal cases, the acquittal rate in bench trials is 38%, but for jury trials it’s only 14%.)
The prototypical dubious case is a drive-by/bike-by/walk-by shooting. A few friends are walking to the corner store, chatting on the porch, etc. A stranger pulls up in a car or bike, or maybe approaches them on foot, and without warning begins firing. There’s no physical evidence linking defendant to the crime. There’s no motive, except maybe vague testimony about an ongoing gang war. There’s only the eyewitness testimony of the surviving victims who identify defendant as the shooter in a photo array and/or lineup.
Problems with this: (1) Most of these crimes happen in the evening or night, so the lighting is poor. (2) The victims may have just a few seconds to see the shooter’s face before he starts firing, after which they’re fleeing and not looking at anyone’s face. (3) Their degree of attention is often questionable. They’re walking to the store, chatting with friends, not really paying attention to the car pulling up to the curb alongside them until they see the muzzle of a gun. After that, they’re paying attention to running for their lives. It’s not conducive to later identifying the shooter. Studies have shown that stress decreases the accuracy of eyewitness identifications.
Lineup procedures are also frequently dubious. For instance, maybe the survivors describe the shooter as a light-skinned guy with dreadlocks. In the ensuing five-man lineup, only two of the participants have dreadlocks, and only one of those is light-skinned. Guess who the witnesses are going to pick! In general, bias in lineup procedure goes only to the weight of the evidence, not to its admissibility; the bias has to be really egregious to get a lineup identification suppressed. And there’s always the possibility of police misconduct. There was a recent high-profile string of convictions that got overturned because the lead detective had a pattern and practice of secretly telling witnesses which lineup participant was the suspect. Not every detective is dishonest, but enough that I mistrust any identification unless I’m convinced the witness would have a strong independent recollection of the criminal.
Yet juries still convict on this kind of evidence. Part of the problem, I think, is that it’s a more comfortable narrative. It’s nicer to think “Yes, we caught the culprit, we are going to put him behind bars” than to think “There’s just not enough evidence, the shooting will probably remain an unsolved crime forever.” Another part of the problem is the witnesses genuinely believe what they’re saying (even if incorrect), and the jury responds to that genuineness. (Studies have shown no correlation between witness confidence in an identification and the accuracy of that identification.) By the time of trial, a witness has been shown defendant’s face repeatedly—in a photo array, in a lineup—so they’ve been primed to stand up at trial and identify the defendant. It’s why I give basically no weight to in-court identifications, no matter how heartfelt they might be. They’re basically the epitome of smoke and mirrors for a gullible jury.
Incidentally, I read about a case where the defense lawyer had defendant sit in the audience, while a random similar-looking person sat at the defense table. Then the State’s witness mistakenly identified the random person as the criminal. Clever! Good advocacy! But the State always gets the last laugh—the lawyer was convicted of contempt of court because the judge didn’t care for his stunt.
===
Here’s an example of an identification that bears strong indicia of reliability: A neighbor heard shots being fired and went to her window to see what was going on. She saw defendant pacing back and forth on the sidewalk for 5-10 minutes. He left her field of vision briefly, then came back without his coat. A getaway car pulled up, he jumped in, and the car zoomed off. Police searched the area and found defendant’s discarded coat, which had his DNA and tested positive for gunshot residue. Also, the neighbor identified defendant in a lineup a few weeks later.
This is a very good identification for multiple reasons: (1) She had a good length of time to observe him. (2) She wasn’t in any personal danger, so she could focus on observation. (3) She had a high degree of attention, since she came to the window specifically to see what was going on. (4) It was the middle of the day. (5) Her observations were corroborated by physical evidence.
===
Now, here's an example of police misconduct: I read about a police officer who witnessed a shooting. He received a tip about a suspect. Then he looked up the suspect's picture on Facebook before viewing a photo array. Of course he “correctly” identified the suspect after that lmao.
===
Not related to eyewitness identifications, but here's another example of a dubious conviction: A man was out for a late-night jog. He jogged past a porch and saw two people standing there, "Lil Lord" (whom he knew from the neighborhood) and another guy whom it was too dark to see. He continued jogging about a house length away and then was shot in the back. The angle of the shot was consistent with having come from the porch. The jogger turned around and didn’t see anyone on the porch or in the vicinity, suggesting that the shooter had fled into the house.
That’s all very nice. But there were two people on the porch. So at best, there’s a 50% chance that Lil Lord was the shooter. We don’t convict on 50% chances. Or at least we’re not supposed to.
What clinched the case was that Lil Lord's "friend" turned State's evidence and testified against him in exchange for a generous plea deal. But accomplice testimony is always suspect. I think there’s certainly a reasonable doubt that he was lying to save his hide.
You must be signed in to post a comment!