Sand Piles, Avalanches, and the Next Catastrophe
(the third of four stories about entropy and homeland security)
This is one in a series of posts dealing with the Second Law of Thermodynamics, or Entropy Law, and the origin of homeland security threats. In previous posts we have explained that, due to the Entropy Law, closed systems tend towards maximum disorder, or entropy, while in open systems, the importation of matter, energy, and information can cause localized self-organization to occur. Due to the Entropy Law, total entropy must still increase, so this means that the part of the system from which the inputs that facilitate the self-organization come from becomes even more disordered as inputs are extracted from it.The system that arises from this interplay of self-organization and entropy is called a dissipative system. As the system dissipates increasing amounts of energy, it becomes more complex while exporting more disorder. It turns out that the manifestation of the disorder from dissipative systems does not always occur smoothly over time, but can occur in a “lumpy” fashion, in which long periods of calm are shattered by sudden, or even violent or catastrophic events.
Swedish physicist Per Bak and his associates studied this tendency of nature in sand piles, and then noticed it’s applicability to many different kinds of systems. Bak and his team named this property of nature “self-organized criticality”, which they noted applies to “dissipative dynamical systems” –exactly the systems that this series of posts is dealing with. By self-organized, they meant that the system naturally evolves to the state independent of its initial conditions, or in other words, the critical state is an attractor of the dynamics. Energy is dissipated at all length scales of the system. Once the critical point is reached, the system stays there.
Bak and his associates noticed that systems that exhibit self-organizing criticality follow a power law distribution, which has a “long tail” to rather than a symmetrical bell curve. The long tail means that while there are many events that have small significance, as you move out towards the end of the tail on a graph the events are very infrequent, but very significant. For example, a geological fault may have thousands of small tremors that are hardly noticeable, but every once in a while, a really big earthquake occurs that has major significance. A financier named Nassim Taleb, who incidentally made a fortune by betting on “long tail”, major, financial events, named these events “black swans” due to their rarity. While many believe that in hind sight that many major events in history were predictable, Mr. Taleb believes that many times they are black swans, and that these unpredictable catastrophic events, and the adaptations that occur in response to them, are main drivers of world history.
Ted Lewis applied the concept of self-organizing criticality to many different kinds of systems of homeland security relevance. Lewis noted that complex systems tend to become more structured or organized as they age, and that old or highly evolved systems will develop more self-organized criticality over time. Lewis applies Bak’s insight regarding the self-organization of “networks of minimally stable clusters” to actual networks related to homeland security concerns and describes the progression to increasing complexity and criticality: Networks evolve from random to clustered and eventually scale-free networks, which are totally integrated and are near their critical points — their point of collapse. Lewis sees self-organization is a property of all ‘intelligent systems’ that evolve into increasingly complex structure, most typically as the result of some form of optimization, efficiency, defense, survivability, or plain ordinary ‘growth’. Whether these systems are intelligent could be debated, but the increasing self-organization and complexity are consistent with the open dissipative systems that we have been dealing with in this series of stories. But, as noted above, in accordance with the Entropy Law for open systems, with increasing energy use, self-organization, and complexity also comes increasing entropy, and therefore we should expect to see more disasters, and we do, with declared natural disasters in the US going from less than 10 per year in the 1950’s to over 50 in the 2010’s.
Catastrophes observed in systems with “long tails” include financial meltdowns, earthquakes, forest fires, disease pandemics, breakdowns of various types of critical infrastructure, economic collapses , species extinctions and long cycles in economic, political, social, and cultural processes. Considering the observation that as systems become more self-organized and complex that they tend toward greater self-organized criticality, we should expect increasing self-organizing criticality as global development continues, and we do. Here is how globalization is contributing to self-organizing criticality in some systems:
Threat of Pandemic As means of transportation have improved and proliferated, and transportation networks have become more integrated, the ability of communicable diseases to spread and cause pandemics has increased.
Spread of economic contagion Due to economic globalization, with the internationalization and integration of financial markets, and the increasing diversity, complexity, and inter-relatedness of financial instruments made possible by financial deregulation and the decrease in the cost of computing power, financial and economic crashes can now propagate through the world economy as never before.
Threats to various infrastructure: telecoms, the power grid, the internet — A couple of trends are increasing self-organized criticality of these infrastructure systems. One is the growing integration of these systems which allows the propagation of faults over long distances. The other is that, due to the impetus to maximize profit, these systems are being optimized to the point of self-organizing criticality.
Mitigating the natural tendency of dissipative systems to develop self-organized criticality requires planned intervention, and that takes scarce resources. Who should pay for this?
By the way, in spite all the rumors and motivational speeches you might have heard about the bell curves, the middle class and the American Dream, wealth and income distributions also follow a power law, and for the US, the curve has been getting steeper lately. Does that mean we can expect some kind of catastrophe in near future? For an answer to this question, you will have to read the 4th installment of this series of posts on entropy and homeland security — “Globalization and the Coming Social Crisis.”