Honor and Loyalty

Let’s do a detailed analysis of a simple social concept, towards developing our engineering-grade science of social technology. The concept of loyalty is dear to me, as loyalty has been one of the most important determinants of success or failure in some of my more practical projects.

So what do we mean by it, and how does it work?

Loyalty can mean a lot of things:

The first case is the simplest, and the analysis covers foundational features of the other types of loyalty.

Let’s boil it down to the essential core:

There are two parties; Alice, and Bob. Alice and bob have some kind of ongoing relationship, from which they are supposed to be deriving some value. Alice finds herself facing a tempation: there is something she could do or get that is of value to her, but to do so would destroy her relationship with Bob. Will she stay loyal?

First of all, to stay loyal, first she should percieve the trade-off as such. If she doesn’t realize she’ll lose or compromise the relationship with Bob, she doesn’t even have a chance to be loyal, only a chance to be thoughtless.

Second of all, the option to stay loyal to Bob has to more elegantly achieve her goals than the option to defect. Roughly speaking, there are a number of points of value on the loyalty side which must not be outweighed by the temptation.

For each point of value, it is her belief that matters. But it is generally best to break such things up into the thing being true, and then Alice percieving that correctly, rather than targeting belief directly. Targeting belief brings to mind manipulation and deception, whereas targeting truth + correct perception brings to mind working on true fundamentals, and helping one’s allies. False belief is also an unreliable decision factor; she could at any time realize she was mistaken.

The points of value on the loyalty side are two:

  1. Value of the particular relationship.

  2. Value of being a loyal person (honor).

Each of these in turn has a real instrumental component, and a psychological-moral component. The instrumental components are easy to examine:

The instrumental value of the particular relationship is basically what you expect to gain from that relationship. Maybe it’s a lucrative business relationship. Maybe Alice and Bob are married happily. Maybe its some kind of political scheme where they are allied over the long term on a shared plan. Whatever the case, there is some major value to that relationship which Alice would no longer be able to expect in the case where she defected.

The instrumental value of loyalty is what you expect to gain from being the kind of person who doesn’t defect on relationships like this. Largely, the value comes from being correctly known to be this. More exactly, it is what you will gain from having not defected on this relationship, outside the context of the particular relationship. If you defect, word tends to get around. Maybe you are in a business that requires that others will trust your word. Maybe you are in a social circle that will shun you if you are known to defect. Maybe contract law is involved. Maybe Bob will get mad and retaliate against the disloyalty. Again, there is value to not defecting, or a cost to defecting, beyond the particular relationship, which Alice must consider.

The psychological-moral components are a bit harder to understand, not necessarily following any particular rational logic. But they will tend to follow the instrumental values, or the instrumental values of the culture that shaped Alice’s psychology.

Relying entirely on the moral as opposed to instrumental componenets of loyalty is possible, but not a good idea. In particular, it is not scalable with Alice’s level of self-discipline and philosophical ability. A person with more agency in these areas will be more able to self-modify their psychology if it is leaving real value on the table. That is, the moral/psychological values shrink relative to the instrumental as Alice grows in skill.

This is not to say that highly skilled people are less moral, just that their morality comes from the instrumental fundamentals, rather than from psychological limits and culture. This is an important fact with implications much beyond this thought experiment.

Additionally, especially in highly replicated social technologies, the psychological setups will tend to follow the instrumental setups.

Therefore we can focus our analysis on the instrumental component, and not miss too much.

This gets us to the following decision criteria of loyalty:

  1. The actual value of the relationship with Bob, plus the actual value of the honor tied up in the commitment, exceeds Alice’s temptation.

  2. Alice correctly percieves the value of the relationship and of her own honor.

  3. Alice correctly percieves the temptation as a trade-off with her relationship with Bob.

(“Plus” here is to be understood metaphorically, not literally. The values of plan components do not behave as scalar under addition.)

So if Bob wants to assure the loyalty of Alice, he should be sure of the following facts:

  1. Real Value Case. The temptation never actually exceeds the combined value of the relationship and her honor. The relationship has ongoing value at all points, in a way that will actually be delivered, and that Alice actually values, and/or Alice’s honor is invoked and involved in a real way such that if Alice defects, she will in fact forgo some other value that she actually cares about.

  2. Identification of Trade-Off. Alice is aware that the tempting behavior would lead to a loss of the relationship and her honor.

  3. Subjective Valuation. Alice is able to correctly and rationally percieve the value of the relationship and her honor relative to the temptation.

  4. Psychological Discipline. Alice is psychologically capable of reliably acting according to the rational instrumental incentives in this case. This handles the bumps and kinks introduced by the moral/psychological dimensions.

If these conditions are met, than Alice will be loyal. Otherwise, Bob is on shaky ground.

In my experience, 2 and 3 are technicalities. The key questions are whether the value exists as far as Alice is concerned, and whether Alice is able to act according to that value.

This analysis helps to illustrate and clarify how we might think about social technologies. What assumptions we make, how we structure the problem, the process of thought, and the form of the “answer”.