At Home In the Universe

A review of its view of social and technological systems

By Katarina Ling

SymSys 205: Systems: Theory, Science, and Metaphor

Spring 2005


Stuart Kauffman’s At Home in The Universe suggests the idea of autocatalytic sets as what brings a natural order to original chaos. A natural order means that the order we experience is not an unlikely accident, but one that is bound to occur in some way. Autocatalytic set theory suggests that life arose from a critical mass of diverse molecules, each of which had the ability to catalyze a reaction. Catalysts speed up reactions by lowering the threshold otherwise necessary for a reaction to take place. With enough variety of molecules, the continuous catalyzing of reactions can form a cycle of reactions. This cycle is self-sustaining and can be interpreted as alive.

Although this idea might incite protest by natural selection fans because it reveals a theory that allows life to arise without specifically mentioning DNA or natural selection, this paper will not focus on those topics. Rather, this paper will investigate whether Kauffman’s claims that autocatalytic set theory can also be found in places beyond the molecular level, that of human social systems. Kauffman projects that the concept of an autocatalytic set can be projected to a number of other naturally complex systems such as biological extinction, the development of technological economy, or even politics. Yet, one might ask, can such a simple idea of a large number of catalysts placed together actually be responsible for such macro level phenomena?

Power Law

The power law is the focus of many of these examples. Rather than getting a better idea of the average, or usual size when more individuals are measured, phenomena subject to the power law show that estimates get bigger when more individual cases are measured.  Large individual cases tend to happen rarely whereas small ones happen more frequently. One example is the Bak-Tang-Wiesenfeld sand pile avalanches. Imagine a trickling of sand from above onto a table. The sand pile gets bigger and bigger until it eventually reaches a steady state height. Along the way, there are a number of avalanches. Each grain of sand is roughly the same size yet the sizes of the avalanches they trigger are not.  The size of the avalanche depends more on the particular arrangement of the individual grains and whether the one struck by the new grain was a critical piece of the pile. If a critical piece is removed, it will pull down a number of other grains with it. One thing to note is that there are far more small avalanches than there are large avalanches. The author notes that there are many examples of natural events that follow the power law, including earthquakes and the distribution of extinction events. Additionally, this shows that significant large scaled events can do not necessarily need a large scale cause – a cause the significance of any other will do since the factors that decide how large an avalanche will be is determined by what has actually been laid down before.

To explain how sand piles maintain the steady state size, Kauffman considers an “Invisible Hand” explanation. Adam Smith considered the Invisible Hand to be what maintains an open market economy to operate at an optimal level without the need for government intervention – a natural system of individuals trying to maximize their own welfare at a micro level causes well being at a macro level as well. This is the first of a number of ways Kauffman relates natural events to a broader view of economics – the production and distribution (including exchange) of goods and/or services of limited resources.

For biological extinction, autocatalytic set theory is tied closely with co-evolution and in a way, with economics. Each entity in a particular ecosystem might excrete something. This something can be fatal to some but life-saving for others. Thus, Kauffman calls this a molecular economy. Molecules that get what they need continue subsisting in that area. Those who do not are driven out and only appear in some other place in which their needs are met. Their neighbors are those who also benefit from the environment created by what they excrete. Thus, co-evolution develops. If a new change occurs which violently disrupts a central part of the tightly connected web of coexistence, massive extinction can occur.

Human Systems

Kauffman draws a number of parallels between human systems and the aforementioned biological origin-of-life system. For a technological economy, the general idea is that a new invention creates a new industry, which creates new jobs or new products. From that, other new inventions can be made and the spiral continues, building upon the previous ideas. Kauffman points out that most of now perform jobs that did not exist thousands of years ago. In the hunter gatherer lifestyle, one subsisted by directly gathering food and eating it. Nowadays, it is possible to be a professor to earn food. Lecture giving, diagram drawing, and paper publishing do not result in the magical appearance of food but they are the beginning of and indirect process of earning credits which may be exchanged for food that others produce. Indirect food earning power comes when a society has found enough food so that not everyone has to devote all of their time towards seeking food, rather than the pure number of years of societal evolution. This progress seems natural and bound to happen. Hence, this is another area in which Kauffman’s theory of "We, the Expected" seems to fit.

Furthermore, like molecular synthesis, technology and other new products can be built when people decide to combine current resources into something new – wheat and milk for porridge, for example. Each product can be a catalyst for the creation of either a new product or one that already exists. Kauffman cites research conducted by Canadian economist Jane Jacobs and University of Chicago economist Jose Schenkman that show diversity is positively correlated with economic growth. Innovation in technology is the driving factor in economic growth, according to the Solow-Swan model of macroeconomics. Thus, one might think that the more diversity in the initial pool, the greater the number of possible combinations, which means a greater chance that some will be viable and lead to economic growth.

While each technological change may be about the same size, depending on how critical the piece that the new technology knocks out is, the avalanche that accompanies it (loss of jobs, significant change in lifestyle) varies. For example, the invention and popularization of the car caused the loss of the horse and buggy as the primary mode of transportation. That brought down with it all the manufacturers of buggies, buggy drivers, and even horse shoe makers. (If buggies, and thus horses as the powering of the buggies, were no longer in high demand, then one may suppose that the need for their re-shoeing was less frequent). Yet, the transition between walking and horseback riding was not that great of a technological loss. The avalanche was smaller. Perhaps couriers by foot were no longer as valued as before, but the car as a disruptive technology was probably greater than the horse and buggy idea.

Without going into too much detail, Kaufman also ponders the idea of extending autocatalytic set theory to politics. Namely, one way to find a balance the interests of a large and diverse society is to have a democracy composed of a number of smaller distinct jurisdictions. This is like a large number of molecules all operating independently but all contributing to a greater whole. He also points out that autocatalysis might apply to cultures as well. Cultures meet and sometimes react, causing either one to dwindle and die, to be assimilated, or to create a new culture from the originals. For example, immigrants to the United States bring their home culture. Their children grow up in a mix of their ethnic culture and the American culture, often creating a new one from the blend of the two. A combination of existing cultures might serve to facilitate the creation of a new one, something like a catalyst. Kauffman’s perspective of how a large number of an otherwise chaotic organization of individuals come to survive together and maintain a seemingly homeostatic existence fits with the history lessons that high school students learn.


One way to support the theory is through simulation. This idea of the origin of life can be modeled by a computer program. Thus, this model is a simulation of how life arose and different parameters can be adjusted to see how they each affect the final results. Recently, the Virgo Consortium put together the Millenium Simulation, the largest simulated universe ever created. While its primary focus has been in seeing how the universe evolved very early on in the life of the universe (that is, far before life on Earth was created), one can see that a similar experiment with equal complexity could probably be constructed to test Kauffman’s theory. Kauffman also cites a few examples of (binary) symbol string experiments in which strings can run into one another and evolve according to the rules of the experiment or create a new string from its parents. Other scientists have used simulation to check on mass extinctions.  Simulation provides another means of proof beyond thought experiments, but saves the time and resources of having to actually create the universe or the primordial soup once more. However, this book does not mention actual simulations on human systems. One reason may be the amount of protest about the generality of the simulation, and not being complex enough given the many different perspectives and personalities of people. Almost all simulations must sacrifice exact accuracy in order to make the code more manageable.

However, biological evidence has shown the autocatalytic theory to work at least some of the time at the molecular level. From a 1998 email with Kauffman, Gert Korthof found that

Gunter von Kiedrowski, then at U. Freiburg in Germany, several years ago published work on a collectively autocatalytic set of two DNA hexamers that mutually ligated the two pairs of DNA trimers composing the two hexamers. Meanwhile, Reza Ghadiri at the Scripps Institute in La Jolla, California has made an autocatalytic peptide, Nature August [1996], and nearly collectively autocatalytic sets more recently [1998].

With society currently accepting experimental science (biological, physical, and chemical) as more “real” than simulation  or thought experiments, the added evidence are a boon for Kaufman’s theory, at least at the molecular level. A link and a few references to these experiments are listed at the bottom of this page.


The idea of a large number of catalysts pulling each technological or biological development forward is appealing. Yet, Kauffman does not consider inhibitors, or what might counteract catalysts as a problem in his models. It is possible that an inhibitor to an inhibitor will effectively cancel out the inhibitor effect, as if neither were present in the first place. Another reason that this may not be an issue with respect to the origins of life or the technological development is that any inhibitor is likely to not be permanent. The process may just be slowed, but will not be permanently destroyed. It would be interesting for Kaufman to explain his perspective about inhibitors.


Kauffman argues that his theory is scalable and what might first appear to be very different systems actually have much in common. Furthermore, these systems would also be expected to occur and did not evolve by pure chance in our world. From the evidence that Kauffman gives, his theory is plausible. His points about the similarities between molecular and human systems are astute. He also has the support of a number of computer simulations, in addition to the more traditional biological and chemical approach of experiment.  Hence, while the claims are not deductively proven to be the new definitive theory of the origins of life and self-organization, there seems a high likelihood that the systems Kauffman has illustrated might be related by a natural universal occurrence, like autocatalytic set theory. It may be another natural law.


Kauffman, S. (1995). At Home in the Universe: The Search for Laws of Self-Organization and Complexity. New York: Oxford University Press.

Kauffman, S. (1996) Self-replication: Even peptides do it. Nature 382.

Korthof, G. (1998).  Kauffman at home in the Universe.

Lee, D. H. et al., (1996) A self-replicating peptide. Nature 382, 525 - 528.

Lee, D. H. et al., (1997) Emergence of symbiosis in peptide self-replication through a hypercyclic network. Nature 390, 591-594.

Lee, D. H. et al., (1998) Correction: Emergence of symbiosis in peptide self-replication through a hypercyclic network. Nature 394, 101.

Maddock, J. (2005). Millenium Simulation – the largest ever model of the Universe.