There is a lot of interest in the nature and origin of 'complexity' these days. The word can be a buzz-word or stand-in for careful thought, or a label to help make one's work sound more original, deep, novel or important than it actually is. We're all vulnerable to such usages.
Still, many phenomena in life really do seem to be complex in the sense sometimes referred to as 'emergence': this blog post is an emergent trait relative to the ink-spots and letters of which it is composed. One can never understand it from a catalog of those components. If one can call our blog 'intelligent', it is because the inkspots were organized by an outside cause -- us -- with its organization known in advance.
Many things in life may be intelligently emergent in that sense, probably more that we tend to think; at least many animals, and even plants, seem to do organized planning rather than being simple automotons. Nonetheless, many emergent traits in both the living and non-living world seem to manifest what is called self-organizing complexity. The idea is that the emergent trait arises on its own, without an external 'designer', by the individual, local activity of its components.
You and we develop in this self-organizing way as embryos, for example, a major point in our book The Mermaid's Tale. Each cell responds to its local conditions, and step by step, the interactions of our countless cells ends up as a highly organized human. We are emergent traits relative to our cells, which in turn are emergent traits relative to the genes and other components of which they're composed.
The complexity of slime mold
In this context there is an interesting recent report of self-organizing complexity in the humble slime mold. The slime mold Physarum polycephalum exists as independent, autonomous cells much of the time. They just motor along minding their own business. But like cells anywhere, they monitor their environment, and are capable of responding to it. Under some circumstances, the individual cells interact in a striking example of self-organizing complexity.
The report shows how slime mold mimics the Tokyo rail system in efficiency, tolerance of faults, and cost. That is, as it grows and responds to environmental stresses, it forms networks connecting the individual cells, that are similar, when modeled mathematically, to such a 'real-world infrastructure network'. This is reported in Science on Jan 22, in a paper called "Rules for Biologically Inspired Adaptive Network Design," by Atsusho Tero et al.
The idea is not that we can use rail networks to understand biological systems, but the other way around.
As the authors say,
Some organisms grow in the form of an interconnected network as part of their normal foraging strategy to discover and exploit new resources. Such systems continuously adapt to their environment and must balance the cost of producing an efficient network with the consequences of even limited failure in a competitive world. Unlike anthropogenic infrastructure systems, these biological networks have been subjected to successive rounds of evolutionary selection and are likely to have reached a point at which cost, efficiency, and resilience are appropriately balanced. Drawing inspiration from biology has led to useful approaches to problem-solving such as neural networks, genetic algorithms, and efficient search routines developed from ant colony optimization algorithms.The slime-iness of railway systems
And slime mold self-organizes without any centralized organizer, or global information: no central nervous system, no consciousness -- at least of the kind we associate with ourselves, as far as anyone knows. So, if the rules of Physarum network formation can be captured mathematically, they should be useful for planning and optimizing the functioning of non-biological networks.
Our biologically inspired mathematical model can capture the basic dynamics of network adaptability through iteration of local rules and produces solutions with properties comparable to or better than those of real-world infrastructure networks. Furthermore, the model has a number of tunable parameters that allow adjustment of the benefit/cost ratio to increase specific features, such as fault tolerance or transport efficiency, while keeping costs low. Such a model may provide a useful starting point to improve routing protocols and topology control for self-organized networks such as remote sensor arrays, mobile ad hoc networks, or wireless mesh networks.
This idea is based on the assumption that biological systems are optimized, in terms of cost, efficiency and resilience, the same outcomes that should be optimized in man-made networks, and that there are rules that determine this, that can be discerned and readily translated from the humble slime mold to the sophisticated man-made world of teleological -- purpose-designed -- networks. But slime mold cells have no blueprints, organizing committees, or overall goals. All each cell can do is react to its immediate surroundings. Any overall efficiency or resilience must have evolved indirectly by selection working on the net result of the aggregated activity of cells each working only locally.
That is very different from a committee designing a subway system or economic network. How selection's effects on the net result of slime mold aggregate patterns is manifest at the underlying genetic level is one of the major challenges in evolutionary genetics.
Selection is so indirect at the level of such complex phenomena that there is no reason to think it will be simple -- much less unitary -- at the gene level. It will have been assembled non-purposively, and with no need to optimize any of its parameters, and will likely be of no use to human urban planning committees. Evidence from attempts to map complex traits in all sorts of organisms supports the idea that it will not be simple or unitary at the gene level.
As we say in Mermaid's Tale, if there are biological rules, they can be broken. There are certainly no universal biological rules for efficiency or cost. The peacock's tail is proof of this, in that by usual considerations it's a risky, costly thing to have -- yet it purportedly has such a net advantage as to have offset its costs. Instead of some econometrician's idea of efficiency, whatever works is what evolution allows. Resilience comes closer to being a principle of life, but uncoupled from efficiency or cost. Slime mold's ancestors presumably ran into situations often enough that genes that produced sensors, signals, receptors, or whatever that led them under some circumstances to connect with neighboring cells, provided them a fitness advantage.
And there's another important way in which biological and man-made networks like rail or mobile phone systems differ, and that is that man-made networks are designed to maximize profit, and that's not an attribute relevant to biological systems, that presumably can only maximize what leads to better reproductivity. Unless survival itself is equivalent to profit, but if so, then these systems are even less equivalent than supposed because, as we say in M. T., the idea that only the fittest survive is patently false. Instead, selection weeds out the weakest organisms, while most others stand roughly comparable chances. 'Failure of the frail' is, we think, a more accurate descriptor of an organisms's chances in life than 'survival of the fittest'. In life, good enough is good enough. Period.
Somehow higher emergent properties have evolved, even in what appear to be simple organisms. It's not a surprise, because even a single cell is hard for contemporary science to explain in genetic evolutionary terms. But we still are far from understanding it adequately, which means that applying the 'lessons' of biological systems to man-made networks is largely based on assumptions that may or may not be valid, and may or may not be relevant.