Generalized Evolutionary Theory

Over the years evolutionary theory has itself evolved to encompass new and more disciplines: social Darwinism, genetic algorithms, co-evolution of biology and culture, evolutionary psychology, economics, psychoanalysis, and more. Attempts to formalize evolution typically have focused on several elements or preconditions for natural selection:

  1. a POPULATION of individual agents
  2. a REPRODUCTION mechanism
  3. a MUTATION mechanism that yields differential fitness of agents
  4. a SELECTION mechanism which favors highly fit agents over others for reproduction

For example, at any given time in a culture, there is a large (but finite) population of memes which exist in one or more human minds. These memes reproduce by being transmitted from one mind to the next through standard communication (talking, writing, mass media, etc). Memes clearly mutate over time; for instance it was once thought that the biggest factor in obesity was consumption of fat molecules, but now it is generally thought that carbohydrates are the biggest culprit. Finally, individual human minds select which memes they transmit and which they do not based one personal criteria such as perceived truth, “compellingness” of content, personal self-interest, etc.

The generalized evolutionary argument goes that if all of these preconditions are met within a given system, natural selection (aka Darwinian evolution) occurs ipso facto. Meaning that wherever you can identify these preconditions you will find a system undergoing evolution. Evolution (in the technical sense) is the dynamic process of a population of differentially fit agents bearing offspring which are similar but not identical to themselves. The type of agents is unimportant, it can be biological agents, memes, computer programs, business plans, etc. Natural selection amongst biological agents has no primacy or special place in the pantheon, it is one of a class of dynamics that occur in complex systems.

One aspect of evolutionary systems that was not understood or appreciated until recently is the importance and prevalence of co-evolution. That is, populations evolve in the context of an environment (everything external to the agents themselves). But we cannot forget or ignore that in most cases, the environment consists of many other evolving populations, each of which making up an important part of the respective environment of the others. Thus, when one population, say the antelope population evolves faster and faster offspring, this changes the environment for the cheetah and other predators who need to be able to catch and eat the occasional antelope to survive. The two populations co-evolve such that traits like speed maintain a balance and neither population completely wipes out the other. Co-evolution can take the form of symbiotic relationships, parasitic, competitive “arms races”, and a variety of other forms. As one population evolves it changes the fitness of the other populations in its environment, creating new selective pressures for these populations to evolve as well.

Several factors have lead to the obfuscation of general evolutionary theory, much of it having to do with the language and model of standard biological evolution. The notion of distinct “populations” is an over-simplification, and a human construct to be sure. The problem with relying on the population model is that in naturally occurring systems, there is no boundary line that says this creature belongs to this population and that creature belongs to that one. By drawing lines, we ignore many important interactions between agents, and we also infer interactions that don’t necessarily exist. On the one hand, bacteria and viruses are in direct competition with their hosts for survival, but on the other hand, they are dependent on the host’s survival and health. Conversely, parents are partially competing with their own offspring, and in many species actively exterminate or eat a portion of their own. We may be tempted to say that each species is a distinct population, but any one agent mates with only a small fraction of the entire population. And geographically separated sub-populations never even have a chance to mix with one another. So, is the population of interest a geographically near sub-population, or the much smaller population of those agents that actually mate? The reality is that the term population is a construct that helps us frame the evolutionary dynamic, but we cannot ascribe overly to it lest we miss the (literal and figurative) forest for the trees.

Other limiting constructs include “selection” and “fitness”. The term selection implies an active, intentional action, which implies an outside actor with foresight and a will of its own. In evolutionary systems, no outside actor exists; selection happens by default as some agents die off before reproducing. Imagine the scene where a general is looking for volunteers to fight the enemy. His troops are lined shoulder to shoulder and the general says to them, “if you will serve, take one step forward”. As we know from the comic routines, if enough soldiers take a step backwards, the brave “volunteers” will be easy to spot. They’ve been selected for in the evolutionary sense, but not in the standard sense of the word. Fitness is also problematic because it implies a static, absolute measuring stick by which all are judged. Fitness in an evolutionary system is both dynamic and relative. The fitness of a giraffe’s ultra-long neck is high in the savanna where trees are a certain height and edible leaves are relatively scarce. But in another environment that same asset can easily be a liability. And we know that environments are continuously changing, due to migration, climate, co-evolution and a variety of other factors.

A subtly confusing factor in understanding evolution is the emergence of genotype. Richard Dawkins’ selfish gene argument made an important point: we can consider genes themselves as an evolving population of agents which use their expressed phenotypes (i.e. the organisms in which they reside) to propagate themselves. In other words, from the standpoint of the gene, it is the the phenotype and the organism is the genetic mechanism for the gene’s reproduction. But the selfish gene argument replaces the arbitrary primacy of organism (phenotype) with that that of the gene. Clearly, genes can’t and don’t just exist in the wild naked of their organisms. It is more logical to think of genes and organisms (and other subsystems thereof) as co-evolving populations of agents. The inter-dependency of gene and organism is so tight though — a form of ultra symbiosis if you will — that we fail to recognize this inherent truth.

A final distraction from understanding complex systems evolution is the over-reliance on the model of reproduction, which is not general enough to encompass the variety of evolutionary phenomena observed in the world. As pointed out here, reproduction is one (of many) ways agents preserve their existence and exhibit system stability. The descent chain of parent, child, grandchild, etc is a system that is preserved and renewed through time via a reproductive mechanism. But individual organisms use reproduction and generation at the cellular, sub-cellular and super-cellular levels in much the same way. Lost in the discussion of evolutionary systems are more simple forms of stability such as stasis, movement, self-repair, etc. When viewed from afar over the course of many months, the Namib sand dunes are structures, dynamic systems, which evolve over time. They move in the direction of the prevailing winds, change shapes, converge, decompose, and ultimately co-evolve with one an other and their environment. To focus exclusively on reproduction, misses other important dimensions of evolution.

Darwin’s conception of evolution was “descent with modification”. This view is more general and more accurate than the description of evolution a four piece harmony of population, reproduction, mutation and selection. A truly general evolutionary theory goes something along the following lines. Agents — which is to say systems that preserve themselves through various stabilizing mechanisms — change through time, space and other dimensions, subject to selective pressures (which themselves change).

In other words, evolution is the balancing act that results from a system’s internal pressures to maintain integrity and external pressures that lead to disintegration.