The Selfish Climber!

Let us suppose that there are two competing climbers (Merrill and Melvin) who are tied by all possible measures going into the final round of a bouldering competition.  The winner of this competition will be rewarded a small purse, as well as the potential for a life changing sponsorship from major vendors including PETZL, Arc’teryx, and theDIHEDRAL.  There are four possibilities for the outcome of the final climb.

1. Merrill sends while Melvin falls.
2. Melvin sends while Merrill falls.
3. Both Merrill and Melvin send.
4. Both Merrill and Melvin fall.

The sponsorships that are up for grabs will only be awarded to the winner of the competition, unless both climbers send the final route.  If both climbers send the final route, the vendors will believe the route was too easy, and thus not a worthy metric by which to award such generous sponsorships.  Now, if neither climber sends the final route, the competition will end in a draw.  In this case the vendors have decided to split the sponsorship evenly between the two climbers as they have both proven to be equally strong on a difficult route.  Splitting the sponsorship will be boon to both climbers, but not the lifechanging opportunity offered by sole attainment of 1^st place.  There are three possible payoffs depending on the outcome of the final climb.

1. Sponsorship = $
2. Partial Sponsorship = P
3. No Sponsorship = N

Merrill and Melvin sit in isolation unable to communicate with one another prior to climbing the final route, they also return to isolation after their climb, thus preventing them from knowing how their opponent fared.  The sponsors explained the situation prior to the final climb, and now Merrill and Melvin must decide what to do. 

The scenario:

While Prisoner’s Dilemmas can be found throughout the history of the universe, it was originally framed as a thought experiment in game theory by theorists Merrill Flood and Melvin Dresher in 1950.  Essentially, a prisoner’s dilemma models circumstances involving strategic behavior in which entities stand to gain benefits from cooperation or suffer consequences from failing to cooperate.

The typical contemporary prisoner’s dilemma involves two prisoner’s each of whom are made an offer in which their decision may help or hurt their state of being, depending on how their accomplice responds to the same offer (much like the example stated above).

There are an array of alternative examples and uses of the prisoner’s dilemma.  The Tragedy of the Commons is a type of prisoner’s dilemma in which the benefit of the whole can be disrupted decisions of the parts.

Currently the world is faced with a global prisoner’s dilemma regarding AI research and progress.  Most developers understand the potential existential risks involved with advanced AI but scaling back on research and development runs the alternative risk of falling behind competing developers who refuse to scale back despite the potential existential risks involved.  If one cannot rely on their competitors to slow progress, then why fall behind?  If you can’t join them, beat them!

In philosophy, prisoner’s dilemmas can be used to raise moral questions, specifically regarding the rationality of self-interest.  Prisoner’s dilemmas can be used to suggest that acting in our own self-interest isn’t always in our own self-interest.  When we all act by purely self-interested reasons, then we all run the risk of suffering unnecessarily.

So, now that we have a little more information about The Prisoner’s Dilemma, let’s take a look at what Merrill and Melvin should do.

Regardless of what the other decides, each climber reasons that it is best to look out for themselves.

Merrill reasons: If I send and Melvin falls, then I get the sponsorship.  If I decided to fall in order to split the sponsorship, but Melvin sends, then I get nothing.  Given these options it is better if I send.

Melvin reasons: If I send and Merrill falls, then I get the sponsorship.  If I decided to fall in order to split the sponsorship, but Merrill sends, then I get nothing.  Given these options it is better if I send.

Merrill and Melvin then, reason in the same way and both send the route.  Ultimately making life worse off than if neither sent the route.

Now, even if we allowed Merrill and Melvin to communicate and both agree to fall off the route, provided that self-interest was the driving force to their decision, each would reason in the same way noted above, and realize that it would be better to double-cross their accomplice, and send the route.  Again, resulting in the worst-case scenario in which both athletes leave with nothing.

Self-interest has been the driving force for all life on earth, and while there’s nothing wrong with looking out for oneself, there are times when cooperation can raise us all beyond what we might achieve alone.