Wednesday, August 15, 2018

Book Lessons #3: The Selfish Gene Part 2

This is a continuation of the Book Lessons #2: The Selfish Gene Part 1. Unfortunately, I cannot find any investment lesson in this post, only life lessons!

In the later part of the book, The Selfish Gene, the author Richard Dawkins introduced how altruism sometimes work in nature as it becomes beneficial to both the giver and the taker. The simplest example is what we had learnt in Biology 101 - symbiosis, the direct opposite of the parasitism. It was a good reminder that in struggle between good and evil, there are always examples of good triumph over evil. Though strictly speaking, the main theme of the book is to really move the argument away from good vs evil but rather to think in terms of evolutionary stable strategy or ESS.

Faces of Mother Nature I

In short, there is really no good and evil in the real world of Mother Nature. Is the cuckoo bird evil because of her parasitic way of breeding? What matters is whether evolutionary strategies would achieve stability i.e. the genes would be able to propagate into many future generations. Similarly, in our own lives, rather to think in terms of good vs evil, sometimes it is about having better strategies and think about what's the final endgame that matters for us and for future generations. Having said that, we surely must refrain from deploying too many "evil" strategies! We are here do make the world a better place, not turn it into hell!

One example that brings this out well is the story of the Tit-for-Tat strategy. This strategy became famous after in won over all other strategies in the famous Prisoner's Dilemma game when two prisoners are caught and they can choose to betray the other one or not to. As a reminder, the payoffs are as follows:

1) If A and B each betray the other, each of them serves 2 years in prison.

2) If A betrays B but B remains silent, A will be set free and B will serve 3 years in prison (and vice versa).

3) If A and B both remain silent, both of them will only serve 1 year in prison (on the lesser charge bcos there is not enough evidence).

In just one game, betrayal is always the logical outcome unless you know 100% that the other guy surely won't betray you. However when the game becomes iterative, things get very interesting. We can devise as many strategies as we like. So when competitions for multiple strategies were played via computers, it was discovered that the best strategy is actually the Tit-for-Tat strategy which means that we only betray when we were betrayed previously. In some cases, Tit-for-Two-Tats also work. 

This has found to work in nature as well with birds (again). Birds usually have ticks growing in their feathers and they can pick them off with their beaks but not those ticks on their heads. So birds rely on other birds to help and usually this works on the tit-for-tat basis. If you help me, then I will help you. A bird that keeps betraying would soon realize no one will help it get rid of ticks. Its head will get so itchy that it fails to procreate, or explode, or whatever. Just kidding. But surely they have less energy to procreate when they are always itchy.

Faces of Mother Nature II

Similarly a bird that keeps helping might find disappointment from time to time if the other bird betrays by not returning the favour. The tit-for-tat works because it helps to figure out who the betrayers are and hence avoid them. When it finds a nice guy, it would continue to cooperate as long as the other guy cooperates which works out well and both would live happily ever after, walking into the sunset.

In life, most of the time, it doesn't pay to be the good guy. Hence the phrase, "good guys finish last". Nature has also shown that evil does triumph 58% of the time as shown in the last post (in a population of hawks/aggressive and dove/docile, 7 hawks to 5 doves being evolutionary stable i.e. the ratio of hawks to doves will not change after many iterations) but there is always room for good. In fact, the moral of the story seemed to be that we should be good when it pays to be good and we should fight evil with evil when the situation calls for it.

Interestingly, the Always Betray strategy can beat Tit-for-Tat sometimes when many iterations of the game is played. There seems to be a knife edge whereby Always Betray will win and become evolutionary stable on one side and Tit-for-Tat on the other side. This knife edge depends on how the strategies evolve at the start, if the bad strategies become too populous, then Always Betray will come out winning.

However, once a population succumbs to Always Betray, it will always remain there. If we think back in terms of the jail sentences, everyone will be getting the same sentence. There is no room for "upside" anymore. The Tit-for-Tat strategy is different. Even if there is just a tiny portion of the population that continues with Tit-for-Tat, usually in a local cluster with no Always Betray strategies around, it can grow again and in time crosses the knife edge and converts the whole population to Tit-for-Tat.

The author further explains why Tit-for-Tat is so powerful. First, it is nice, it always begins by cooperating. Second, it is forgiving, it betrays only after being betrayed and it never remembers beyond the last betrayal. Third, it is not envious. It doesn't care if the other guy gets more money or a lesser sentence. Hence it is a strategy that strives to beat the system, not just the other guy. Perhaps that should be the motto how we should live our lives too! 

Faces of Mother Nature III

The truth is, Tit-for-Tat happens a lot in human societies and in nature as vividly illustrated by the author using how the British and German soldiers behaved during WWI and also in vampire bats. Not knowing when the war will end, British and German soldiers started cooperating rather than fight and die in the cold winter. They even celebrated Christmas together! On the other hand, vampire bats are known to donate blood to their kins and sometimes to stranger bats too when these poor bats did not find enough blood to fill their stomachs on unlucky nights.

To end this post, here's quoting the author,

Vampire bats rise above the bonds of kinship, forming their own lasting ties of loyal blood-brotherhood. They could form the vanguard of a comfortable new myth, a myth of sharing, mutualistic cooperation. In fact, they could herald the benignant idea that, even with selfish genes at the helm, nice guys can finish first.