Thursday, May 12, 2011

Bargaining and Promise Keeping

You are an intentional agent with desires that motivate you to act so as to realize states of affairs in which the propositions that are the objects of those desires are true.

You are surrounded by other intentional agents.

However, desires only motivate the agents that have them. Therefore, the fact that you have these desires does not automatically give anybody a motivating reason to realize states of affairs that would objectively satisfy your desires. Going up to somebody and pleading, "I have a desire that P" can well lead to a shrug of indifference or even motivate the agent to realize not-P, if (for example) he hates you.

So, what options do you have in interacting with these other agents?

One option is the option to bargain or trade.

You find somebody with a desire that Q, where, that person can realize Q more efficiently with your help, and you say, "If you perform these acts for realizing P, then I will perform these other actions for realizing Q."

With this, you give that person a reason to realize P - by turning it into a means for realizing Q where he already had a motivating reason to realize Q.

Of course, this assumes that P and Q are jointly realizable. If P implies not-Q or Q implies not-P then there is a problem. But there are many cases in which “P and Q” is possible.

Unfortunately, under the assumptions we are making here, a substantial percentage of potential bargains are doomed to failure before they are even made. These are bargains where one agent fulfills his terms before the other one does. As soon as you complete your side of the bargain, and Q has been realized (or, at least, you have completed your steps for realizing Q), then his acting to realize P ceases to become a means for realizing Q. Consequently, his motivation for acting to as to realize P disappears.

You would be quite foolish to trust that he will realize P after you have completed your side of the bargain, unless he has some other motivation supporting the realization of P even after it is no longer an effective means for realizing Q.

One potential motivator could concern reputation. If he wants to enter into future bargains with others, he has reason to avoid being known as somebody who does not complete his side of a bargain. That would reduce his ability to enter into future bargains and reduce his ability to realize states that could best be realized through bargains.

However, this only applies if (1) the other agent has a reason to enter into future bargains, (2) you have the ability to threaten his reputation, and (3) you have the will to use that ability (which might be hindered by threats of violence or just a general aversion to causing trouble). Remove any of these elements, and your bargaining partner no longer faces the motivation of reputation.

Another potential motivator is an aversion to breaking promises. A person with a strong aversion to breaking promises will be strongly disposed to choose actions that will keep the proposition, "I have broken a promise" false. This is true in the same way that a person with a strong aversion to pain will be strongly disposed to keep the proposition, "I am in pain" false.

The person concerned solely with reputation will break a promise when he can get away with it. The person with an aversion to breaking promises will not break a promise even when he is the only person who will ever know about it. This is true in the same way that a person with a strong aversion to pain will avoid situations in which he is in great pain even when he will be the only person to know about the pain.

For that person, the instrumental value of the bargain - that realizing P becomes a means for realizing Q - is his motivation for making the promise. The aversion to breaking promises become his motivation for completing his side of the agreement even if you should finish your part first, and realizing P ceases to become useful for realizing Q.

To the degree that you can reliably detect this aversion to breaking promises in others, where you will complete your part of the bargain before the other party completes theirs, it makes the most sense to bargain with somebody who has this aversion to breaking promises. Furthermore, it would be in your interest to become a reliable detector of the aversion to breaking promises in others.

Now, you should also realize that, what is true of you in this case is true of others as well. Those other intentional agents that exist in the world around you also have reasons to enter into bargains. They have reason to prefer to bargain with people who have an aversion to breaking promises. And they have reason to work on improving their capacity to reliably detect who has this aversion – just as you do.

This means that, if you can acquire this aversion to breaking promises, then you will probably have a comparative advantage over others that those people would seek to bargain with.

At this point, I will not get into the question of whether it is possible to cultivate a desire or aversion. I will leave this discussion at the point that says that if it were possible to cultivate such an aversion, and others are reliable detectors of those who actually have such an aversion, then you almost certainly have a motivating reason to do so.

Of course, it is also true that if you can discover a way to exploit their detection methods and fool them into thinking you have this quality, you have reason to do that as well. Though it will also pay you to teach others how to reliably detect when others are using this technique, so that they can help in identifying and flagging those others.

This leads to a further implication of this system. Not only do you have reason to bargain with those who have an aversion to breaking promises, and to reliably detect those who have this aversion, you have reason to join with others in a campaign to identify and remove or, at least, flag those who lack this aversion to breaking promises. Of course, in doing so, you risk that they will flag you as such a person if you have not cultivated this aversion to breaking promises yourself.

Such is the nature of bargaining and promise-keeping under the assumptions we are working with here. People have no automatic reason to consider the desires of others, but they have reasons to enter into bargains. Bargains will fail where one person completes his terms of the bargain before the other one does – unless the other person faces some other motivation. Reputation is a good but flawed motivator in that people motivated by reputation will break promises they can get away with breaking. On the other hand, people with a strong aversion to breaking promises will keep promises they can get away with breaking. We have reason to prefer bargaining with those agents, with improving our ability to detect those agents, with cooperating with others in detecting those agents, and in being one of those agents.

3 comments:

Anonymous said...

I'm not sure… this is pretty complicated… can you break it down and simplify it some more? ;-) Just kidding… I don't think this could be any more clear.

Jesse said...

Matters like "aversion to breaking promises" and "aversion to losing reputation" can be generalized as precommitment.

In addition to making this post much more compact, precommitment has many more specifications than just promise-keeping and reputation. For example, breaking a legal contract often has financial or criminal penalties, so the legal system acts as a reliable context for making precommitments.

Alonzo Fyfe said...

Jesse

The "many more specifications" mean that "recommitment" will not work for what I have in mind.

The "more specifications" that you mention are not mental states. Many of them involve threats (which are the subject of the post following this one, and which have some important differences from bargains).