Complex social sciences

Some weeks ago, I read a review of three books in the Week-end Financial Times on Complexity Economics. I bought the three of them, written by Brian Klaas, J.Doyne Farmer and Maja Göpel. They are a good complement of  “How China scaped the poverty trap”, by Yuen Yuen Ang who also uses complexity arguments to explain the historical evolution of China, with references to other societies.

Klaas book, “Fluke,” is perhaps the most interesting and ambitious of the three, in the intersection between philosophy, natural sciences and social sciences. The book reviews the properties of complex systems, such as emergence, self-organization, non-linearity, interconnection, randomness and difficulty to make predictions (the latter, as in Taleb’s books). Complex is different from complicated in that when a part fails in a complex system, the other units change and adapt. Feedback loops, tipping points, and reverse causality are also characteristics of complex systems. These can be analyzed using network analysis and evolutionary dynamics.

In a complex evolving system, small causes can have big consequences. Applied to the human world, we are all part of an interconnected reality, the result of multiple contingencies that result in our existence. When we are born, we do not come into this Planet, but we emerge from this Planet. Klaas urges us to downplay the importance that our Western culture allocates to the individual. Our brains are the result of the interaction of many neurons, and an anthill is the result of many ants. These social insects invented agriculture before humans. We have evolved to develop a sense of self-awareness to survive, not to seek truth, and other beings have developed other skills that we do not have (flying wings, radars). 

We control nothing but we can influence everything. The complexity of human societies suggests to the author that instead of using the expression that something is or is not “rocket science,” we should instead say that something is or is not “social science.” At our state of knowledge, economic systems are more difficult to predict than some physical systems (such as planetary orbits).

To facilitate the connection with social scientists trained in traditional methods and models, books for a general audience discussing complexity should emphasize some elements of continuity with the more advanced existing methodologies. For example, I see a continuity with game theory, just with more players and less (or different) rationality, as in Bowles 2004 book on microeconomics.

For example, in the last part of the book I found useful thoughts that will help in my course on soccer and economics. When the context changes, randomizing strategies may be useful, as a tribe in Borneo does with the selection of the exact location of rice fields. Or as the animal species that follow mathematically perfect random rules in the ocean. It provides a new perspective on the use of mixed strategies, which so far I justified only in terms of being unpredictable in contexts such as penalty kicks or military strategy.

I also found useful the reference to Moneyball (the book and movie about how statistics revolutionized baseball) as a not necessarily desirable trend if brought to the extreme, because it makes the game more predictable. The notions of contingency, complexity and randomness may help explain why Moneyball techniques have been more successful in baseball (although there seems to be a backlash there as well according to Klaas) than in soccer, where the game, less dependent on set pieces, is more fluid and difficult to predict.