The Evolution of Cooperation

My friend Tony is one of those amazing sources of things that make my brain go We were chatting about Crowdsourcing – an übersession (which is like an obsession on steroids) of mine, ToingToing’ed here and here.  Tony read a few great books on the subject (notably James Surowiecki’s The Wisdom of Crowds: Why the many are smarter than the few and The Long Tail: How endless choice is creating unlimited demand by Chris Anderson, editor-in-chief of Wired Magazine. In fact, the book is based on an article Anderson wrote for Wired in 2004.

The question of choice in winning strategies reminded me of Robert Axelrod’s book The Evolution of Cooperation and the sequel Complexity of Cooperation; both fascinating investigations of the defection / cooperation axis. Axelrod did some amazing work around an iterative game called the Prisoner’s Dilemma.

The prisoner’s dilemma comes from game theory deliberations, it is based on a hypothetical situation in which two criminals are arrested and put in two isolated police cells. Each of the criminals is then offered the following (identical) deal: if he implicates his partner, he will go free while the partner will get a life sentence. If both avoid implicating each other, they will both get off lightly – say, 5 years. If, however, they implicate each other — both will get a heavy sentence. The game is played several times and players devise a strategy, based on the three possible decisions each ‘criminal’ can make (implicate his partner, avoid implicating his partner, implicate each other.) The best strategy is the one that ensures the player using it minimal punishment or, rather, will gain more then he loses (by spending less time in jail than his partner.)

Axelrod created a computer program that simulated the Iterated Prisoner’s Dilemma and invited others to an IPD tournament in which computer simulations challenged each other (and themselves) to rounds of IPD, and the best strategy eventually emerged. The winner – hands down — was TIT-FOR-TAT – a strategy devised by Canadian mathematician and psychologist Anatole Rapoport. The TIT-FOR-TAT strategy was ingeniously simple: it played IPD game by cooperating on the first move, and then making the same choice as the other player did on the previous move. Amazingly, this cooperation-biased strategy has been a consistent winner against other strategies, including itself.

In his 1984 book “The Evolution of Cooperation”, Axelrod says that the best strategies are “based upon an investigation of individuals who pursue their own self-interest without the aid of a central authority to force them to cooperate with each other.” (p. 6) Axelrod figured out that the most successful IPD strategies had four detectable traits: being “nice” meant the strategies were never the first to defect, being “forgiving” enabled these strategies to cooperate even after the other player defected, being “provocable,” these strategies were, however, capable to repay with defection for the opponent’s defection. Lastly – being “clear” made them transparent and therefore easy to understand by other players.

The key, according to Axelrod, was a strategy’s “willingness” to cooperate. Cooperation, he said, has nothing to do with ethics, friendship or consciousness. Instead, it is based on people’s ability to remember their opponents’ past actions and project these actions on to the future. These traits, according to Axelrod, form the basis to beneficial cooperation. He illustrated the point with examples from trench warfare in the First World War, where soldiers on both sides often ‘rigged’ artillery salvos purposefully to be ineffective – thus ‘cooperating’ (in IPD terms) and indicating to soldiers on the other side that, as long as the Generals can see that the enemy is actually being shelled, they will not pursue any specifically aggressive war tactic, and no one will get hurt. This phenomenon (in which all soldiers involved work together to survive the war) is known as “live and let live“, it illustrates Axelrod’s argument that cooperation is possible also among enemies – and not only among friends or partners. People may cooperate with strangers – and even with adversaries — for some mutually derived benefit.

Axelrod asks: how can cooperation emerge in a world of egoists without central authority? He argues that the IPD computer tournaments show that “cooperation based upon reciprocity can emerge and prove stable provided the shadow of the future is long enough. Applications include politics, economics, and evolutionary biology.” The term ‘political cooperation’ may sound like an oxymoron, but consider power-sharing deal talks between Zimbabwean President Robert Mugabe and opposition leader Morgan Tsvangirai, haggling sessions involving Israeli PM Ehud Olmert and Palestinian President Mahmoud Abbas or even negotiations between the Hollywood Screenwriters’ and the studio representatives, late last year – these examples demonstrate the dangerous power of brinkmanship, where all chances to a peaceful resolution to a conflict rest on two, usually aggressive and belligerent, players who face each other, unrelenting and uncompromising.

45 years ago, Canadian historian Richard Hofstadter wrote “The Paranoid Style in American Politics” – a seminal essay debunking the power of conspiracy theories and what he termed “movements of suspicious discontent” on American politics. He said: “Perhaps the central situation conducive to the diffusion of the paranoid tendency is a confrontation of opposed interests which are (or are felt to be) totally irreconcilable, and thus by nature not susceptible to the normal political processes of bargain and compromise.”

What a pity that most politicians ignore the healthy learnings derived from cooperation strategies like TIT-FOR-TAT and, instead, choose to immerse themselves (and the people they lead) in a political quagmire in which cooperation is seen as weakness, even capitulation.

Contribute to ToingToing!