Menu Close

Musings on Thermodynamics, Complexity, and Evolution

I’m an evolutionary theorist. Whether I’m a good one or not can be debated, but I am always interested in learning about existing theory on the topic, and hope to improve my own understanding, as well as contribute in some meaningful way to the development of our understanding of evolution.

However, I don’t care to view the topic from a single perspective, and I think that a lot of work can be done by integrating statistical physics and thermodynamics, with complexity theory, in order to not only better understand evolutionary dynamics, but also abiogenesis and its relationship to evolution. And honestly, I think I’m on to something! This article is a commentary on my own thought processes and hopefully it’ll be insightful to others interested in the topic. At the very least, this article will leave a written record of my thoughts, and perhaps someone else can expand on them in the future.


Quite a while ago, I came across a discussion on work conducted by Jeremy England. The work was fascinating. England looked at the second law of thermodynamics, and used some tools from statistical physics to investigate how self-replicating systems might function in open systems.

Generally, the second law of thermodynamics is described in closed or isolated systems. The idea is that components of such a system tend to move towards increasing disorder. We can see how this idea plays out in a room in a house. As we go about our lives, a room gets messier and messier. To bring the system back to an ordered state, we need to do work. However, in an open system, exposed to a constant heat bath — the Earth is something like that kind of system — self-replicating systems are preferred. At least that’s my rough understanding of his work.

I started writing my own musings on the topic, and began reviewing other work that might give me further insight. This investigation led me to work conducted by Taeer Bar-Yam, et al. Their work is based on the concept of requisite variety, developed by cyberneticist W. Ross Ashby. Here variety is the number of possible states in the system. Using a multiscale varient of the law, developed by Yaneer Bar-Yam, the authors formulated an ”inverse second law of thermodynamics” which states that open systems exposed to a structured environment tend to yield to the order of that environment.

Identifying Commonalities

While the two groups of authors took very different approaches, it seems that in many ways the results are the same. The goal of this project is to see if it is possible to generate a unifying mathematical framework that relates the two concepts and a meaningful way, and also learning what can be accomplished with such a framework.

Observer Discrimination

An important feature of Ashby’s original work is that “a set’s variety is not an intrinsic property of the set: the observer and his powers of discrimination.” As stated earlier, Bar-Yam et al. (2018) rely on a multiscale variant.

Question: Is there an analog to the observer discrimination factor within England’s work?

Answer: Yes. The analysis of self-replication requires an understanding that it is “something that happens relative to an observer: only once a classification scheme determines how many copies of some object are in the system for each microstate can we talk in probabilistic terms about the general tendency for that object to affect its own reproduction, and the same system’s microstates can be classified using any number of different schemes.”

Misbehaving Components as Observers

Bar-Yem et al. suggest that when a component of a system begins to misbehave, it can be treated as part of the external environment. When talking about “observer” people often consider the observer to be human, but it shouldn’t necessarily be a conscious observer.

Question: Given that the misbehaving components are part of the external environment, can those components be seen as an observer and if so, how can we utilize it?

Answer: I don’t know!

Evolutionary Direction

Many people claim that evolution has no direction. Part of this gut reaction boils down to a rejection of a conscious director (a god), but a direction in no way requires consciousness. It just requires a gradient.

Question: Does a direction in evolution exist?

Answer: It seems reasonable. Both of the theories addressed in this commentary suggest that there are global pressures acting on evolving systems, and thus an overall gradient along which evolution flows.


I’m a fan of studying and developing theory for the sake of simply bettering our understanding. However, discussion of application is important as well.

Abiogenesis + Evolution: Starting with a more abstract use for these ideas, it seems that the work by England, Bar-Yam, et al. will go a long way to rejecting the claim that abiogenesis and evolution are two separate processes. While it is my conjecture, the results of this work seems to suggest that the two processes are really just different aspects of the same underlying dynamics.

Systems Development and Management: Ashby’s work focused heavily on systems theory and complex systems. Bar-Yam et al.’s work extends his ideas further, and an integration with England’s work in statistical physics could open up an a new understanding of how complex systems work, and how we can develop and maintain them.

Post-Government Future: I don’t know what the views of cited researchers are, when it comes to government, but I am an anarchist. Yes, I am probably a little biased due to this trait, but I think that a better understanding of complex systems will lead us towards a society that is managed voluntarily, rather than through the threat of violence.