Friday, April 8, 2011

Anarchist's Dilemma game

The core idealist anarchist principle is that of non-aggression.  It is in fact a core tenet of any utopian philosophy: A pledge to not initiate aggression.  The Anarchist's dilemma game is a simplification of an anarchist world of 2 people/groups, and the payoffs each can expect if they either choose peace, or lie about choosing peace (choose war).  Note that whether or not one side is a war monger, it still wants to pretend to want peace so as to convince the other side to be unprepared and unaggressive. Thus , the non-aggression principle is susceptible to hypocrisy: "Non-aggression is something you should believe, so that I may take advantage of your gullibility."

                                 Peace    War
Y    Peace                   1            1
O    War                    -1000      0
 Above "chart" is payoff matrix for Me, in an anarchist's dilemma game between me and you.

  • If we both choose peace, we both create and keep 1 unit of wealth
  • If we both choose war, there is stalemate, but neither of us produced any wealth
  • If I choose war, and you choose peace, I steal your unit of wealth
  • If I choose peace, and you choose war, I die and can no longer keep playing.
Many of you will recognize this as a variation of the Prisoner's dilemma game. (See first comment below for fuller explanation of game). Probably the most important philosophical and game theory concept.  The Anarchist Dilemma variant is an important reformulation of the payoff and precept with historical insights into the formation of religion, currency and civilization, and insights into the future of civilization.

Note that the win-win solution occurs if both sides choose peace.  2 units of wealth are created, and split evenly between the two.  We all call that the "good" encouraged outcome, and so would god, and every religion.  The need to convince everyone else that peace is the good choice motivated religious worship of that good force, and quite possibly, fictionalized personified exaggeration of that force for good.  "Peace is good because we have faith that peace is good" is a valid argument even if it is circular.  Similarly, reinforcing faith in currency or faith that your neighbour will/should not murder and pillage you naturally enhances value.

Promotion of the good outcome has also justified the state's role of administering police forces to punish its citizens who choose violence over peace.  Note (in the anarchist dilemma game)  that despite the mutual profit of choosing peace, each person individually gains greatest expected utility (by a wide margin)  by preparing for war, and thus not producing any surplus value/wealth.  By adding punishment for crime, socialized police forces can add sufficiently negative consequences to compel citizens to choose peace. Socialized costs should be less than the sum of individual protection costs, and if it is successful in compelling people to be peaceful, it helps achieve that good and productive outcome that happens to be otherwise against individual participants' interest.

The obviousness of the police solution for guiding citizen behaviour would imply an equal obviousness in sovereign nations associating to pool military funds into a punishment organization.  The less we all spend on military, the more we can enhance wealth and happiness in our own countries, and less military spending brings about even less defensive military spending requirements.  The 2 main arguments against a global punishment force happen to be the same as those against a state punishment (police) force: 
  1. Punishment force efforts can be politically biased instead of principled.
  2. Individual protection efforts can be more effective in protecting the individual's interests.
Addressing the 2nd point first, consider driving regulations.  We can all individually get to where we need to faster by being allowed to drive 200mph+ while multitasking and enjoying alcoholic beverages.  We're collectively safer by forbidding/punishing those behaviours.  Similarly, preventing individuals from amassing the most powerful weapons (including nuclear) and militias, and not having to rely on their judgement for application of that force, makes us collectively safer and richer.  Under the anarchist absence of a punishment force, the principle of vengence and cost-ineffective deterrence leads to more death and destruction, and poverty.

Political bias and errant use of police resources are strong arguments equally against states and a global association of states.  The primary rationale for natural governance is addressing this issue head on.  Any socialized police/enforcement force cannot have any influence on any other socialized association.  And no influence on creating laws.  The police chief should have a mandate to be effective and fair, and direct electoral accountability for only that mandate.  This is the best we can do to prevent political bias in enforcement, and political bias for creeping legislation after an association is formed.  In the case of global punishment force, the simplest mandate is to enforce the territorial integrity of all sovereign members, and in order to "eliminate" civil strife, support a procedure for all secessionist movements.  On the latter point, the option for divorce is an option for peaceful resolution of any disagreement.  It is both distracting, and destructive, to entertain reasons for either side to be more deserving of murder or punishment.

Secession also solves terrorism.  Political violence by the relatively powerless stems entirely from the inability to affect the political process.  Policies which frequently intentionally disenfranchise those thought too weak to fight back.   Beyond terrorism, all political violence within a territory is based on the prize of  political control of that territory.  The UN's attempt to create strong(er) states in Africa has directly created violence among groups for the prize of political control.  Beyond civil violence, political acrimony, for instance in the US, should lead to divorce.  Republicans and Democrats appear to have irreconcilable differences, rather than an acceptance of one ruling over the other.  Beyond political acrimony, In Canada, objection to political process where the majority who oppose a clearly corrupt authoritarian ruler resolved to spend $1500 of each Canadian's money on prisons and poorly designed fighter jets, shouldn't force Canadians to accept the political process, or submit to its results.  The option of secession is more important than democracy in resolving political differences or abuses, and guarding against corrupt, or simply biased, governance.

Democracy may be a peaceful means of obtaining political control, but that control is substantially equivalent to any non-democratic ruler.  Revolutionary ideals and principles such as the obviousness of self-determination, or Montesquieu's "Government should be set up so that no man need be afraid of another.", have yet to ever survive democracy.  There are no laws protecting the people from the government, and only a small individual voice  in choosing vague and unaccountable election platforms.  Democracy is not enough to have effective, accountable government.

A final tangent to leave you with.  An alternative to a global punishment force borrows from natural finance's 3 party contract system (comptrollership function) where 2 parties contract with a 3rd in order to resolve differences (or act as punishment decider/enforcer) between the 2 parties.  Unlike typical arbitration processes, care must be taken to guard against systemic bias between the two parties (company-consumer arbitration tends to favour companies because they are more likely to need future arbitration.  The same bias applies towards state prosecutors).  Funding of the arbiter has to be formulaic and independent of the outcome of a complaint, and the appointment of the arbiter, made prior to any complaint.  The purpose of this mechanism is to accommodate many small seceded communities in a regionally efficient manner.


  1. A slower explanation of the game:

    For simplicity, only 2 players (Me and You) are displayed. Each player makes a choice to either make something useful (Peace) or Attack the other player (war). The numbers in the table are the payoff or score that "Me" wins based on the 4 combinations of play decisions You and Me make. (The score for You is mirrored. You be Me, to see what your score would be).

    Collectively, the best total score happens if we both choose peace (we both get 1 point). But individually, the "winning" move is to play war because the expected payoff of choosing war is 0.5, while the expected payoff of choosing peace is -499.5. The outcome to me, if I choose peace and you choose war is scored as -1000, but could just as easily be -infinity. It is simply very bad for me, as I am attacked unprepared, am murdered, and to top it off, the 1-point-trinket I made is stolen.

    The rest of the article explains efforts to help make both players choose peace. Most simplistically a police force that punishes people for attacking each other.

    While there are substantial criticisms of the state included, the game is named Anarchist's Dilemmma because cooperating on security is more efficient than doing so individually (and suggests organization with a budget and resource scheduling). Natural Governance is a minarchist solution to security-- the weakest state function necessary.


    There are opportunity costs involved in spending on security or ninjas or military weapons. Dedicating a portion of your budget and manpower to force means not producing surplus wealth. At the same time, defense is necessary to at least protect your wealth.

    The article focuses on mechanisms to modify the payoff matrix so as to encourage mutual peace.

    One proposal not discussed is basic income:
    Basic income is a non-paternalistic (no means test/compliance) wealth redistribution method. Not only does it reduce motivation for crime by eliminating desperation and oppression, it also can offer something to lose if individuals meet with social disapproval.