Game theory emerged to solve a concrete problem: how people make decisions when their outcomes depend on the actions of others. Von Neumann, Morgenstern, and Nash laid the groundwork eighty years ago, building a solid framework for analyzing cooperation, betrayal, reputation, and institutional collapse. When designing governments, organizations, or incentive systems, we tend to ignore those lessons.

Giacomo Bonanno, in his game theory text, organizes key ideas that decades of study have validated. Revisiting them helps clarify what actually matters in large-scale coordination. What's revealing is that many failures we blame on human nature are, in fact, design errors.

Nash equilibrium starts from a simple premise: a structure is stable when no one gains by unilaterally changing their action, assuming everyone else stays put. For institutions, this means that asking for cooperation isn't enough; you have to make cooperation the rational choice. The Experience Points (XP) and Social Reputation Points (SP) in the Ludist Manifesto follow this approach: aligning the collective good with individual benefit, without forcing anything.

In a single game, Nash equilibrium can lead to poor outcomes. The prisoner's dilemma illustrates this: two rational people choose the worst option for both due to a lack of coordination. Repetition changes everything. In repeated interactions, cooperation is sustained through reputation, without external authorities. A track record of actions creates a memory that guides the future. This isn't just theory; it's precise math that backs up cumulative SP, an indelible record of past behavior.

The folk theorem extends this: in indefinitely repeated games, almost any cooperative outcome can be sustained if players value the future enough. Long-term models generate dynamics that short-term thinking can't achieve. That's why the Ludist Manifesto fits better as a generational initiative. Institutional impatience isn't just a bad habit; it undermines the mathematical foundations of lasting cooperation.

Mechanism design tackles head-on how to align personal interests with the collective. The XP, SP, and levels in the Manifesto aren't playful decorations; they're technical structures that define information, actions, and consequences. I've seen organizations where noble values fail because their internal processes reward the opposite. The failure isn't in the people—it's in the design.

An underrated factor is incomplete information. Without full knowledge of others' intentions or actions, equilibria get distorted. Bonanno details how the availability of data drastically alters outcomes. Radical transparency isn't a luxury; it's an operating condition. Without it, decisions rest on faulty assumptions, and the resulting failures are unfixable because no one has the data to diagnose them.

This generates adverse selection, where the worst actors dominate due to informational asymmetries, and moral hazard, when someone dodges the costs of their own mistakes. Both erode public institutions. The structural fix is visibility: the theoretical basis for Transparent Surveillance and Auditable AI in the Manifesto. Not to punish, but to enable. Cases like Openplanter or Meta's Ray-Ban glasses show the flip side: when opacity benefits elites, the model inevitably corrupts itself.

Signaling games solve another puzzle: how to verify real capabilities or commitments. Credible signals are costly to fake, like a genuine diploma. Many, however, have degraded as they became easy to obtain. XP in the Manifesto function as auditable signals of real contributions, not self-reported ones. This addresses the fragility of reputation models where, without verification, everything collapses into propaganda.

What explains our decades-long neglect of these ideas? Applying them demands transparency, and that's uncomfortable for those who thrive in the shadows. What's more, good designs disperse power, which discourages those who concentrate it. The folk theorem also explains stable extractive equilibria, like ancient Rome or modern monopolies: changing them requires a coordination that the model itself blocks.

Even so, theory offers concrete ways out: extended interactions, aligned processes, reliable signals, and balanced information. Historical records of long-lived communities confirm it: transparency, lasting reputation, and long time horizons sustain cooperation. This isn't a novelty; it's knowledge we've chosen to shelve.

Open questions remain: how to scale these processes without them being captured, how to deal with impatient actors, how to preserve signals against intense fraud. I'm still working through these complexities. But the direction is clear: for viable institutions, game theory can't be intellectual decoration; it has to be built into the design from the start.

Stones don't lie, but historians sometimes do.

Sources:

1. Bonanno, Giacomo. Game Theory. Open-access textbook, University of California Davis.

2. Nash, John F. "Equilibrium Points in N-Person Games." Proceedings of the National Academy of Sciences, 1950.

3. Fudenberg, Drew & Tirole, Jean. Game Theory. MIT Press, 1991.

4. Hurwicz, Leonid. "The Design of Mechanisms for Resource Allocation." American Economic Review, 1973.

5. Akerlof, George. "The Market for Lemons: Quality Uncertainty and the Market Mechanism." Quarterly Journal of Economics, 1970.