Date sent: Thu, 8 Apr 2004 10:44:24 -0700 (PDT)

The parts Roger has written are in italics. The parts Toby has written are in normal text. 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Toby: well, then its a valuable tool, but not predictive since it only can feret out "perceptions" or lead to speculation (even less valuable) after choices were made. because a priori your bang for the buck, unless emotion clouds your judgement or you see the other side as emotionally entangled. as you pointed out regading divorces, this would be true for many contract disputes and politics and lawyer too.

the obvious problem is: what parts of life fit the diliemma model at all vary as the key points: relative benefits, length of game.

its good training for participating in any pol-econ theory, but is a "game" not a philosophy.

> but is optimizing or zero sum? personally i don't think zero sum has much validity, but it has predicative value since people often think they are in zero sum; the same applies to prisoner's dilemma it's a false dicotomy for most situations, although those situations can be created, hence "prisoners' diliemma"

Prisoner's Dilemma is not zero sum. I'm not going to lay out the whole idea, but I will say that each player sees four variables:

• their reward when both players cooperate,
• their reward when both players defect (usually less than when both cooperate),
• their reward when they defect and the other player cooperates (usually higher than when both cooperate),
• and the possibilty of future interactions with the other player.

The last variable -- the potential for more "games" -- is sometimes overlooked in Prisoner Dilemma discussions, but it is really very important. The longer the playing window a player sees, the more beneficial cooperating looks.

As an example of it's predictive value, the Prisoner's Dilemma suggests why divorces and bankruptcies tend to be nasty affairs. The players see the end of their game cycle (relation) looming up, so the benefits of continued cooperation start looking to be of less and less value.

Roger

Date sent: Thu, 8 Apr 2004 20:16:10 -0700 (PDT)

> well, then its a valuable tool, but not predictive since it only can feret out "perceptions" or lead to speculation (even less valuable) after choices were made. because a priori you bang for the buck.

It is a tool -- which means you must have practice and skill to use it well. It has enormous predictive value because it is helping define what the players will percieve as bang for buck, and this valuation can easily happen before choices are made. It can happen as soon as choices can be seen as coming up.

it is predictive only if the person anticipating the choice can (or will let others) quantify the variables... this is of course an arbitrary process itself, more or less like the diecsion process we "consultants" foisted upon or clients so they would make a decision and feel good about it, win or lose. when they were unhappy we said, thoughtfully (and self-servingly), "Hmm. Well, we will have to a better job of quantifying the variables next time. but take heart your disappointment which this outcome gives us, and you, better insight into your tolerance for risk.... yada yada yada -- oh by theway here is our bill ." I grant you these tools have usefulness, but it is certainly not a philosphy, its a tool to assist in decision making, not to predict.

Example: When will a poltitician be hypocritical? He/She will be hypocritical (which can be defined as defecting on their consituency) when they see the benefit of being hypocritical as higher than the risk of loss associated with being discovered to be hypocritical (a condition which will encourage their constituency to defect on them).

This is framing a future action as a Prisoner's Dilemma issue, and in that frame predictions can be made.

> its good training for participating in any pol-econ theory, but is a "game" not a philosophy.

It is not a game, it sprang from game theory, but it's a valid decision making tool in it own right. It's not elaborate, but it does produce answers to the question, "What should I do in this situation?"

so its just a tool, not a philosophy, a saw is a good tool to if you want to cut wood but it's not a philosophy -- it can be used, with exprience to predict how long it will take to cut a piece of wood into pieces.

i mean the game no insult, but it is a tool and only a tool. its prediction is based upon assumptions about yourself and another person's motives and philosophy.

The parts Roger has written are in italics. The parts Toby has written are in normal text. 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36