Skip to main content



Dr. Strangelove: A critique of Game Theory as it applies to Nuclear Chicken

DrStrangeloveFINAL

For those who are not familiar with Stanley Kubrick’s 1964 film Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb, the premise is as follows (spoilers).  An American Air Force general, acting based on his own delusions of a soviet conspiracy to poison the american water supply, orders a nuclear strike on the soviet union without the knowledge or approval of his superiors.  The planes, en route to eastern europe, can only be called off using a special radio code that only the (deranged) general knows.  At the same time, the soviet union has developed a nuclear dead man’s switch that, in the event of an American attack, will automatically and instantaneously trigger several nuclear devices rendering the earth uninhabitable for 100 years. This doomsday device has already been implemented with no way to disarm or remove it, but its existence has not yet been revealed to the world.  The fairly predictable nuclear apocalypse ensues.

The outcome of the film can be seen as a criticism of using game theory when the risks are too large.  In her article, Marli Wang provides the payoff table for nuclear chicken, a hawk-dove game where the negative payoffs are astronomical (in this case, the end of the world).  Like all iterations of hawk-dove, the states of equilibrium occur when one person attacks (or is aggressive) and the other is not.  Because of this, a player often can use the strategy of brinkmanship, forcing the opponent to back down by showing them that the player’s only option is to be aggressive.  In the perfectly rational world of game theory, this works just fine as the opponent, with perfect knowledge of the game, backs down each time, avoiding the extreme negative payoffs of retaliating.  However, Dr. Strangelove claims that humanity cannot be depended on to be rational.  Indeed the soviet’s brinkmanship move of constructing the dead man’s switch would have worked, had they told the world about it before it was activated.  Since they did not, a US general ordered an attack in a brinkmanship play of his own, causing the most catastrophic outcome (attack, attack) of the nuclear hawk-dove game and ending all life on earth.

Dove Hawk
Dove Tie,Tie Lose, Win
Hawk Win, Lose Armageddon

While this critique relies on an incredible sequence of events,  it does bring up a very good point about the dangerous nature of hawk-dove games where the negative payoffs are extreme.  When looking at the problem using game theory, nuclear deterrence only works when other players don’t have nuclear weapons themselves.  As soon as two have nuclear capabilities, the situation becomes a dove hawk game in a state of nonequilibrium (dove,dove).  At that point, both countries would have a higher payoff attacking the other if they could be sure that the other wouldn’t retaliate (hawk,dove).  While boiling global nuclear politics down to a simple hawk-dove game is certainly an oversimplification, it still poses some interesting questions.  Could modern conflicts involving the possession of nuclear weapons all come down to the fact that the global hawk-dove game is not currently at equilibrium?  What does that say about our future if we maintain the status quo? Are sequences of events like those in Dr. Strangelove really that unrealistic?

 ridingbomb

Marli’s Post: https://nuclearchickencollusion.wordpress.com/2011/06/19/how-we-learned-to-stop-worrying-and-love-the-game/

Comments

Leave a Reply

Blogging Calendar

September 2015
M T W T F S S
 123456
78910111213
14151617181920
21222324252627
282930  

Archives