In Stephen Wolfram’s New Science he fully develops a theory that complex outcomes are likely to be based on simple rules. When I started reading this huge book, I was already familiar with cellular automata, and how a very simple program reiterated many times can and often does result in a complex pattern. His aim to apply this concept to the whole universe, while ambitious, is largely very convincing. His idea is that nature, not constrained the way human inventors are by needing goals, is free to try a lot of different things. This sets the stage for us to consider the basic processes we see as simple programs. Are chemical reactions like simple programs? I think so. They have set rules of engagement that are finite and easily characterized. Can we think of particle interactions, say within an atom or within a molecule as a simple program? I think so. Looking at the physical universe this way gives us a new way to filter the huge number of details, a new way to compare behavior across the board.
Wolfram set up a wide range of initial conditions in his models, some were random. Most created orderly behavior of some kind, often exhibiting patterns that are visually familiar to a biologist such as myself. He argues that natural selection could not have found optimal programs for organisms, that there hasn’t been enough time for that. But that the physical world we see around us came about because it represents the easiest growth arrangements resulting from simple interactions. I can certainly see that a complex organism, such as a human, is an arrangement of parts that could be based on simple rules. Where the whole becomes more than the sum of the parts is an interesting story, and that story includes natural selection in a much more dynamic way. Yet selection CAN act on even the simplest level, so it can’t be discounted for the development of life. Wolfram argues that it couldn’t have been the main element in optimizing outcomes, if the underlying simple rules didn’t provide the basic trajectories of growth that we see.
Wolfram has a very interesting analysis of what definition to use for randomness. The patterns he develops using simple rules can look very regular, or look very random, but in both cases we are not capable of discerning the simple rules that gave rise to the pattern. In the case of the more regular patterns, we could make up a rule to describe it, thereby shortening the description of the pattern. For the patterns that appear random, we can’t even make up a simpler way to describe them. Therefore, he argues, for all intents and purposes, those patterns are truly random, even though we know we created them using a simple rule. This concept is a component of complexity theory – what we call random is often non-random but too complex for us to interpret.
So, perhaps this new way of analyzing the universe wont really give us much information. We can say there is a simple pattern or pattern underlying it all, but we have no way of discerning what it is?
I’m not quite finished with this book, perhaps he’ll show the way at the end. I’ll keep you posted.
