Sorted by New

Wiki Contributions


Complexity and the Search for Leverage

Great post, and you have arrived at the beginning. Yes, in most cases, or perhaps all cases, in the developing world, the most effective thing one might do is to effect system change. Even in the First World, one might see a far greater return on efforts and investment if one could find and understand basic principles of human group psychology and how complex systems like those we live in function.

That said, the danger of causing unintended harm is obvious. The last 114 years have seen a number of well meaning efforts to change the system in various nations cause vast amounts of harm and the deaths of perhaps a hundred million people all told, combining the fascist and communist utopian efforts. 

Additionally, for all the talk and intention to be open minded and willing to change our views in EA, it is far less psychologically costly to change our minds on peripheral issues like mosquito nets that we likely had no opinion on before anyway than it is to change our core political beliefs which define who we are to ourselves and our friends. There will be a lot more resistance to system change in EA than there is to anything else, and if system change gains any traction it is basically a given that the movement will fracture because of it. Especially because a real grasp of system change is an entirely new political paradigm and neither left nor right nor centrist.  

Yes, we need to understand complex systems, but we need to be specific here; there are many kinds of complex systems. What we need to understand are complex systems with multiple poorly related selfish agents and nested overlapping subgroups, and to do that we need to understand how selfish agents compete with each other in complex systems, and how and why the system benefits from that competition. 

Saying that selfish are selfish seems to be no insight, but for decades evolutionary theorists wondered why selfish failed to overrun and replace all the altruists in a group of altruists, because they forgot this. Groups of altruists are host bodies for selfish, they are fitness limiting resources for those pursuing the strategy of selfish defection, and selfish, being selfish, do not want to share the group of altruists with other selfish. In fact, too many selfish in the group causes group collapse, as seen in many hundreds of communes, and there actually is no such thing as a group of selfish humans (though there can be sub-groups of selfish): that is a herd or a flock. 

A selfish actor who finds themselves alone in a group of altruists will maximally exploit them. However, when multiple selfish actors are in a group, they compete with each other, attempt to remove each other from the group, attempt to win the group over to their side against the others (for example, one such process is what we refer to as "politics"), and attempt to thwart the ability of the other to selfishly exploit the group in a process that has been labelled "selfish punishment." The result is that the selfish strategy is self-limiting; selfish do not overrun groups of altruists because selfish are selfish and compete with each other. 

Understanding selfish competition, we should expect that more competition between selfish will be beneficial for the group, and less competition between selfish will be harmful for the group. When we look at other complex systems with multiple poorly related selfish agents, this is what we see over and over. Gut microbiota with more diversity and more competition result in healthier humans who live longer on average.  Ecosystems with competition within niches are more stable and fertile and robust. Economies with more diversity and more competition within industries are more stable and grow faster. Political systems with more competition for power are more stable and provide more benefits for the society. This is why Democracy is beneficial, this is why the division of powers is beneficial, this is why anti-trust laws are beneficial. On the other side, a lack of sufficient competition is dependably harmful. Monopolies and cartels are harmful, dictatorships are harmful, systems that combine political and economic power are harmful, monocultures are harmful, invasive species are harmful, economies dominated by one industry or by one resource such as oil are harmful (and usually not Democracies, competition between economic and political forces is extremely important), and this is why. 

A property of complex  systems with multiple poorly related selfish agents then is that more competition is beneficial to the system or group, and too little competition is reliably harmful. Based on this knowledge we can see why Communism and Fascism and Monarchies and dictatorships and Anarchist and Libertarian ideas should be dismissed from the political debate; they will always result in harm because they reduce or eliminate competition in either political or economic spheres, or both. We can also see that efforts to increase economic diversity and competition in developing nations, for example, could result in long term system change in the most positive way, as diverse economic interests seeking to protect themselves from political power is how Democracy arises. 

This then could be the strongest and most beneficial lever to grasp for: it is hard to envision the developing nation that would object to efforts aimed at helping to create a more thriving economy, and long term that thriving economy will tend strongly to create the fertile ground for Democracy. One way we can do this is to help protect developing nations from the exploitation and dominance of large multi-national corporations, and this would be a highly effective project, in my opinion, for EA. Yes, this is the opposite of "free trade," but this is where the facts and theory lead us, in my opinion. 

For much more details and proper citations and support for my ideas, read my paper on competition here: