I am not in favor of my colleagues ideas about the usefulness of law enforcement having cameras nor individuals paying for their own (legal difficulties in using a privately owned camera on the job?). I would rather have the enormous costs of cameras go towards 1) their salaries and 2) massive overhaul of "use of force" policies 3) nationwide training in deescalation. The problem isn't evidence, the problem is police and citizens fearing one and other.
We've had enough filmed deaths and altercations to prove that the presence of cameras doesn't stop cops from whipping guns out in excess. In states that have started using either body or dash cams there have been instances of police turning off or readjusting cams before they go forward with questionably scandalous interactions with citizens. Tapes can also be detained even while going through investigations. In the video of Sandra Bland's arrest, the video had been edited to loop despite the officers audio continuing. Even while being filmed police can be caught doing legally questionable acts and it takes public outrage for it to be called into question. Think about the officer who became a viral meme during the Occupy Wallstreet protests. After he point blank sprayed pepper spray into the eyes of a row of peaceful protesters, not an uncommon occurrence, he was held accountable by the outrage of the nation. Initially he was being considered for a paid leave of absence until the storm passed, but as the storm raged on he was sacked while being awarded with $38k in workers compensation for the psychological damage of becoming an internet joke. This was a rarely exhibited transparent case of how the judicial system gives law enforcement the benefit of the doubt even in cases of excessive force.
While I don't want to discount the real danger that officers face, it is imperative to include the fear people also face when dealing with an entity that has a robust history of excessive force especially against non-white peoples. Consider the findings of this UC Davis study,
“evidence of a significant bias in the killing of unarmed black Americans relative to unarmed white Americans, in that the probability of being black, unarmed, and shot by police is about 3.49 times the probability of being white, unarmed, and shot by police on average.”**
and from Washington Post data,
“when factoring in threat level, black Americans who are fatally shot by police are, in fact, less likely to be posing an imminent lethal threat to the officers at the moment they are killed than white Americans fatally shot by police.”**
These statistics point to deeper roots that cause distrust between law enforcement and the public, and it isn't a lack of communication that can be solved with greater transparency.
( **http://www.vanityfair.com/news/2016/07/data-police-racial-bias )