Since its publication in 2002 Stephen Wolfram’s enormous theory of science, A New Kind of Science, (you can read it for free) has been enormously controversial with different scientific communities who have accused its author of various forms of plagiarism, hype, and commercialism. One obvious reason for this is simple jealousy. Stephen Wolfram is a multi-millionaire software entrepreneur whose flagship product, Mathematica, powers the research that these scientists rely on to make a living. As such, Wolfram can self-finance all of his scientific work and avoid all the protocols of academia, making him something like a genuine scientific pioneer of the enlightenment. Reading the reviews of his treatise on Amazon reveals as much on the scientific community as it does on the content of the book, as it is probably one of the most hated works ever sold by Amazon, by people who have obviously not understood the main point.
The reason this book is so important and so difficult for mainstream scientists to understand is that Wolfram’s “new kind” of science is nothing less revolutionary than a generalization of a priori science, the scientific method economics is founded upon, to every field of science including, most controversially, physics. The arguments that Wolfram employs are similar to those that Mises and other genuine economists have always held against the drift towards positivism in economics. This is crucially important for us in that, should Wolfram’s scientific method lead to major breakthroughs and become mainstream, there would no longer be any reason to deny that Austrian a priorism is the true basis for economics, as it would have in common the same scientific foundation as physics or any other field.
Here is how I’ve understood Wolfram’s thesis. As a mathematician and physicist, Wolfram observed that mathematical modeling had been very effective at solving certain classes of physical problems, but for other phenomenons it had failed or had only succeeded by becoming increasingly arcane and innacurate (a Kuhnian paradigm limit). As a computer scientist he had studied very simple, one-dimensional programs called cellular automata. These cellular automata he observed had some very distinct classes of behavior. For the first classes it would be very simple to model mathematically what the program was going to do as the behavior is essentially repetitive. For the last class however the behavior was random, discrete and deterministic, and no amount of mathematical modeling could predict it.
Because he saw that it was so simple to produce complex behavior on a computer, he assumed that there was no reason why the natural world could not also be producing highly complex behavior all the time. But if one were practicing the traditional scientific method of mathematical modeling, going out into nature and observing this behavior would not inspire any theory. Instead one would go looking for phenomena that could be solved with mathematical modeling, and ignore complex phenomena.
For various reasons Wolfram explores, you cannot truly deduce the rules that generate complex phenomena by observing their behavior. Observing the natural world would be of absolutely no use towards producing a theory. However what is possible is to trace the behavior of programs in the “computational universe” and observe if we find anything interesting. If we do find an interesting program and it does match a phenomenon in nature, then we already have the entire rule to generate it. We know with 100% certainty that this rule generates this kind of behavior.
And so Wolfram has rebuilt Mathematica to make it possible to explore the “computational universe”, finding discrete processes that can be used to explain previously inexplicable physical behavior. His new science is to run searches through every possible program in a set to find complex behavior, and thus know the rule that generates this complexity with full precision and certainty.
In other words, exactly the same thing as praxeology but applied to physics.
Wolfram then arrives at some conclusions based on the nature of this science. One is the principle of computational equivalence, which says that there is a limit on computational power that is almost always reached, and that it is not possible for a process of equivalent complexity to predict another one. So for example if our brains are maximally complex, and clouds are maximally complex, we won’t be able to predict how clouds will change their shape even if we have the rule that explains cloud formation, as that would require all of the specific data involved in the process being computed with full discreteness. (Echos of the economic calculation debate.)
Another concerns the theory of life evolution. For Wolfram, lifeforms consist of the universe sampling all possible programs and finding which ones succeed and which don’t. Complexity in life is therefore intrinsic to the physics of life and is not driven by natural selection, which tends to reduce complexity instead. As an example he shows seashells that reproduce all the patterns in all classes of his elementary cellular automata, a pattern that is hidden by skin while the mollusk is alive and therefore provides no evolutionary benefit whatsoever.
It is a fascinating book and forces one to completely re-evaluate their perspective on science. People who have status to protect will hate it, as all people with status have always hated revolutionary ideas. But for an Austrian economist it provides the arsenal that will demolish the pretenses of econo-physics, neoclassical economics and all other inferior methods.
