The Solitaire Principle: Game Theory for One

Do I contradict myself?
Very well then I contradict myself;
(I am large, I contain multitudes.)

This post is an exercise in taking Whitman seriously. If the self is properly understood as a loose coalition of many agents with possibly distinct values, beliefs, and incentives, what does game theory have to say about self-improvement?

The Solitaire Principle is the principle that human beings can be usefully thought about as loose coalitions of many agents. Classes of interpersonal problems often translate into classes of intrapersonal problems, and the tools to solve them are broadly similar. The Solitaire Principle is a corollary of the paradigm that the universe is self-similar at every level of organization: the organizational principles and faults of a civilization are not wildly different from those of a single human mind.

Self-improvement is often framed in terms of optimization of a monolithic whole. Instead, the Solitaire Principle suggests that self-improvement can also be achieved by alignment of pieces within the whole to cooperate more efficiently.

First, I fractionate the self across the time dimension and investigate self-improvement as an iterated game for one. This is partially inspired by this essay on becoming more legible to other agents.

Second, I fractionate the self into multiple sub-personalities and investigate self-improvement as a single sub-personality taking unilateral action to improve the whole.

Read the rest of this entry »