5 ms·
> By definition of non-cooperative and cooperative[0] game, nation-states are playing cooperative game. I pondered using either of those concepts. The reason b
by r0naa 11y ago
> By definition of non-cooperative and cooperative[0] game, nation-states are playing cooperative game.
I pondered using either of those concepts. The reason behind my choice is that while there is indeed a multi-level struggle for power where we have coalitions playing against coalitions, there is also some inter-nation rivalries that highlight the fundamentally non-cooperative nature of the "game".
The US weakening the influence of the French leader Charles de Gaulle in the 1960s for example is an example of such tension between members of a same "coalition". The US will first and foremost (legitimately) defend its own interests over those of its allies.
> Also, nation-states are not atomic entities who can be considered "players".
I disagree. The decision-making apparatus whether democratic or dictatorial of nation-states is similar to the brain of a player. See my third point.
> Also, world is too complex to be described by rule-based model.
I agree and it was not my intent to describe the world using a single rule-based model. That being said "non-cooperative games" fully capture, what is perhaps the most important property of our world's geopolitics.c
- vdaniuk 11y ago>The decision-making apparatus whether democratic or dictatorial of nation-states is similar to the brain of a player. See my third point. Nation A is able to bribe government(decision-making apparatus) of Nation B to implement such measures that will reduce competitive advantage of Nation B. This scenario is possible, happened multiple times throughout the history and undermines the "brain" model. Also this scenario violates assumed rationality of players in game theory. Do you agree?