Comment by 4409

Entered on: 3/21/2009 3:28:51 PM

5 stars!

Comment by Powell Gammill

Entered on: 3/22/2009 10:43:51 AM

The problem is game theory requires trust that others will do the same. Further, anyone trying to organize such a response could be charged with conspiracy to _____________. Thwart justice, interfere with .... And lastly, maximum sentence overt examples would be selected and made to pressure those into switching back to copping a plea. Laws would change alright. Just not in the manner so as to diminish the ever flowing slaves into the maw of the prison industrial complex.

Comment by Brock

Entered on: 3/22/2009 11:44:16 AM

Starting from the end, Powell, you are absolutely right. Even if a new law were created, it would merely create a new set of unintended consequences and do nothing to address the root of the violence problem - the prohibition apparatus. That's exactly why I don't have even the slightest time for a legislative solution.

You are also absolutely right that, since the payoff matrix can be manipulated by the court, it will be manipulated to incentivize the outcomes they favor.

However, if you look at the marginal case, where somebody has every reason in the world to plead "not guilty", to change the payoff matrix enough to get a guilty plea would necessarily involve zero punishment for a guilty plea. Working back from there, we can theoretically find a point at which manipulation of the payoff matrix is just less satisfactory than the status quo.

Skipping conspiracy, for a second, to your first statement on game theory, it is true that outcomes based on trust do not establish stable equilibriums. What you are probably thinking of is the Prisoner's Dilemma, which has long been considered to have a Nash Equilibrium of defect/defect. While not widely known, a second stable equilibrium has been discovered which cannot be discounted without adding additional rules to the game.

Basically, if the two prisoners have any knowledge of each other greater than zero, the stable equilibrium is cooperate/cooperate and, in fact, defect/defect becomes unstable.

So, if my goal were "conspiracy to _____", which it is, the key tasks for me are:

a) effectively communicate the unseen payoffs in the matrix - attempting to approach that point where manipulation of the payoff matrix is just worse than the status quo for prohibition-favored outcomes.

and, b) establish a common link between defendants, such that their knowledge of other, present and future, defendants is greater than zero.

This seems like a monumental task, however, consider that it is not unprecedented. There is a very well understood, zero-financed, uncoordinated campaign with astounding results called "stop snitching". The effect of the stop snitching (maybe culture is a better word than campaign) is a lot of prohibition apparatus resources devoted to gaining, keeping, and protecting the few snitches they can, exactly as I predict for a "not guilty" campaign/culture.