jim.shamlin.com

5: Let's Get Together

Most logic-puzzles place the subject in a situation in which a decision must be made independently - but in reality, people readily avail themselves of the ability to communicate and negotiate with one another. Unfortunately, communication itself becomes a method of exacerbating rather than mediating conflict, as the parties involved attempt to mislead and deceive one another in order to achieve their goals in a competitive situation.

Communication and Negotiation

In the animal world, communication is done by various means - sound, motion, even scent - but is generally unambiguous. Communal animals such as ants or bees have not been observed to deceive their communities with false communications, and even solitary creatures communicate in earnest (a bear may reconsider and flee rather than attack - but at the moment he growled, he meant it).

Also in the animal world, most species transmit two basic messages: go away (territory) and come here (mating). Species that work cooperatively (hive insects and pack hunters) have a slightly broader range

While humans have a much broader range, the language of negotiation generally comes down to the same two notions: threat and reward. Threats tend to be cheaper, as if your threat is effective (and why tender one if you don't expect it will be), then you need do nothing - whereas a reward must be granted if it is effective in motivating action.

It's generally accepted that civilized negotiation relies more on promises than threats, and that the negotiation process is an exchange of promises between two parties. A simple purchase is effectively the buyer and seller promising to give one thing in exchange for receiving another; and if the initial proposal is not acceptable to both parties, the discussion aims to refine the promises toward an equitable exchange.

An interesting aside (and good advice) is that some differences among cultures - in some countries (particularly those where everyday prices are haggled), it's best to carry small bills. If you haggle for a price of 10 rupees for a sack of rice and hand the merchant a 50 rupee note, he's likely to give you five sacks of rice (rather than one sack and 40 rupees change).

Coalitions

In game theory, two parties who negotiate to a resolution are considered to be forming a coalition. If effect, they were acting competitively until the moment of agreement - but once the agreement has been made, the interaction is cooperative. In effect, taking the agreed-upon action is acting in accordance with terms of an alliance, albeit a temporary one.

In this sense, the resolution to any conflict can be seen as a "win-win" outcome.

The difficulty in forming coalitions is a matter of trust. The author gives the example of two children, inexperienced at trade, who are reluctant to engage in a transaction that both of them want. They don't understand, or trust in the other, and worry that the other party will not reciprocate ("If I give him my item, he might not give me his, and just keep both.")

In this sense, a third-party facilitator is necessary to overcome the lack of trust between parties. In the case of children, a parent can "enforce" the trade agreement; in commerce, consumers and exchanges expect government to be the guarantor if one or the other fails to live up to the terms.

The author mentions the notion of side-payments to facilitators to compensate them for their service. There is some moral hesitation to making such payments, especially when the facilitator is a government agent, they are seen as bribes. However, when side payments become institutionalized as a "sales tax" charged on every transaction, there seems to be fewer moral qualms, though it is functionally the same thing.

When three or more parties are involved in a coalition, the principles are the same, but the alliance becomes more tenuous. In any group of three, there is always the potential that two will gang up on the third. In any large group of people, such as a family, neighborhood, school, corporation, or government, factions and cliques form within the organization as individuals seek to connect with a smaller, more manageable group.

Commitment

The notion of commitment is a long-term coalition that has longevity. This can be difficult to perpetuate and maintain, as situations change, such that the value of the association may be decreased to one of the parties that may seek to break the coalition to pursue its interests independently. Naturally, the scenario that best exemplifies commitment is marriage, and the phenomenon that illustrates its difficulty is divorce.

Commitment entails a coordinated, cooperative outcome. The example given is of two men in a rowboat, who pull the oars in unison. They do not need to discuss or negotiate because it is clear that if either one fails to keep up his end, the boat will not move forward.

In situations where commitment entails disproportionate effort (one party must contribute more than the other), disproportionate reward, or both, the relationship remains productive if there is still a net positive effect for both parties - though the input and outputs are disproportionate, each is gaining more reward for less effort than it would gain on its own.

There is also a notion of "minimally effective cooperation," which is a common outcome of many negotiations in which each side gives the least possible concession to the other in exchange for the lowest acceptable amount of consideration in return. If any less is given or received by either party, the cooperation fails.

Are We Rational?

In an "ultimatum game" experiment, an individual was asked to offer a second party some portion of a sum of money. If the second party accepted, they would both get the money; but if the second party rejected, neither would get anything. It was a once-and-done decision with no negotiation after the fact.

Rationally, the second party should always accept, regardless of the sum offered. Even one dollar is a greater benefit to them than the nothing they would get if they reject the offer - and the amount of benefit the other party gains from the situation should be of no concern.

However, when the second subject was offered less than about 30%, they chose to reject the offer. This was even found to be true when the amount offered was substantial (two weeks' wages). Follow-up interviews suggest that the subjects who rejected the offer are motivated by the desire to punish the other party's greed, even if it harmed themselves to do so.

The interpretation of this action is that people will act irrationally, out of motivation to harm another party, even if it means foregoing a benefit to themselves (though it might be argued that they gain the pleasure of punishing someone else)