1st Degree Murder Conviction Reversed Because of Failure of DA to disclose Police Dog Sniffing Dog's Prior Bad Acts!

Nice reversal of a first-degree murder conviction out of Los Angeles County. I heart Judge Pregerson, although it is a Fletcher opinion.

What had happened was:

LA DA Steve Cooley's deputy (I spent some time trying to find out who the constitution violator was because JJC ALWAYS NAMES NAMES but I couldn't :( if you know, email it to me) violated Brady in failing to disclose evidence in their possession that the dog sniffing dog who identified the defendant's scent on the passenger seat of a car where the shooter reportedly sat had, on more than one prior occasion, been wrong with his scent sniffin.

Side note, is there a confrontation clause problem with the fact that you cannot cross-X the dog and show his biases, motivation, and financial interest to lie??? (j/k. Or am I?) This doggie eventually got fired because he was so bad!!

Anyway, in this case the DA's office had stipulated to the dog's unreliability in a different trial several months before Aguilar's which led to the exclusion of the evidence in that trial.

The LA Public Defender--and the bad ass constitution enforces that they are--had actually written a letter to the DA after that trial saying that the DA had a Brady obligation to disclose this doggy's unreliable nose to every defense attorney going forward if they planned to use that evidence against that person.

Cooley apparently ignored the letter or didn't ensure his deputies complied.  It is unclear why, if the Public Defender's Office knew about the dog's unreliability in the prior trial, the particular public defender in this case did not.  I believe there are over 700 public defenders in that County. It is understandable that one deputy doesn't know about something in another deputies case.  But, this is an example of why PDs offices should have a master database of police officers and Brady material on any of them (or their dogs) and deputies should always search that database when they get a new case.  Obviously, the database could not include information learned through pitchess motions (we need to change this).