Natural Selection – Evolution Magic

Natural Selection is described as “the differential survival and reproduction of individuals due to differences in phenotype”. To this, some add “blind, mindless, and purposeless environmental process” that nonetheless is imagined turning random genetic mutations into superior new features enhancing descendants’ survivability (fitness). Accumulation of these features supposedly turns one lifeform into another over time. Natural Selection seeks to explain the appearance of design in nature without appealing to a designer.

This definition however fails the simplest test as different phenotypes survive different environments thus delinking phenotype from survivability. In a small farm, only organisms closely related to their wild cousins survive, but agribusinesses select for chickens with oversize breasts and research labs select for populations with specific genetic mutations requiring tight environments to survive. Although all these have different phenotypes, they do not possess an intrinsic phenotype “fitness” independent of the environment. In addition, who decides what is natural and what is not? Darwin considered domestication natural enough to include it as supporting argument. And as far as “blind, mindless, and purposeless”, all these are impossible to prove in addition to being utterly incompatible with the anthropic concepts of “better adapted” and “better fit”, both of which cannot be evaluated independent of survivability anyway.

Natural Selection is supposed to tie both ways survivability with phenotype, but this leaves out the environment which not only affects survivability directly, but also phenotype, itself a sum of genotype plus the environment, and even genotype that is a recurrent function of previous genotypes and the environment again. So in the end, survivability is a recurrent function of genotype, an infinite continuum of environments, and other unknown factors. While survivability can be measured as can be the individual genotype, measuring a population’s genotype is daunting at best, and the impact of the ever changing environment is simply impossible to evaluate. Phenotypes are impossible to define and measure in entirety even for one individual and, in addition, phenotype changes constantly from birth to adult to old age. We do see genetic mutations (unknowable if random) and we do know that, given a similar environment, extreme genotypes reduce survivability, yet we also know that a large variety of genotypes survive just fine in any population.

Fitness is never defined independently of survivability – this renders the fitness concept redundant especially since survivability can be measured while fitness cannot. Evolutionary Fitness is defined as the quantitative representation of natural and sexual selection (reproductive success) of a genotype or phenotype in a given environment. “Survival of the fittest” is interpreted as: “Survival of the form (phenotypic or genotypic) that will leave the most copies of itself in successive generations.” Not only is survivability the only measure, but survivability also changes with the environment.

Natural Selection is Intelligent Selection which is always done by an Intelligent Selector such as Darwin’s breeder which is an intelligent and willful player that takes intentional actions to reach preset goals. Predators, plants, birds, insects or bacteria, all show intelligence and the willful pursuit of predetermined goals. When interacting with the inert environment, organisms self-select rather than being selected by this environment. As soon as the organism dies and becomes part of the lifeless universe, all selection of that entity ceases.

Selection is limited to a narrow set of possible adaptations – what is not there, cannot be selected. Among the most common adaptations are body color/size/shape, hair type, antibiotic/chemical resistance, and behavior, and even these are limited in scope. Farmers would like to grow walking chicken breasts the size of hogs that grow much faster and come in various flavors, but this is not happening despite their best efforts. Antibiotic resistant bacteria still cannot survive extreme temperatures and chemical concentrations and their resistance decreases when the stimulus is removed. Rabbits cannot turn green when hopping over grass and white just over winter, despite the clear advantage such camouflage would bring. Size of tails, horns, beaks, trees, etc. are all stable over time as tradeoffs limit their growth. Human intelligence, flying, swimming, venom, and all other desirable capabilities remain restricted to specific organisms. Domestication has greatly helped mankind’s progress, but it has not changed the nature of the target animals and plants despite intensive efforts to accelerate their evolution. Instead, humans only enhanced the built in characteristics of domestic organisms and simply did without – a huge civilization disadvantage – when those plants and animals were unavailable. Hence, selection does not “design”, is limited in scope to a few available characteristics, and is reversed as soon as the selection pressure ends.

Extinct organism were not flawed and their features were not “selected away”. Most characteristics of the extinct survive in current organisms of which some changed so little over time they are called living fossils. Sure, the mammal eye might provide superior vision to insect eye, but nothing comes for free and tradeoffs ensure both survive. Organisms that have completely vanished cannot be characterized as flawed and it would not take much imagination to see them thriving in a current landscape. The environment may have changed dramatically over time, however on a macro scale, the environment affected all organisms making the “natural selection” explanation highly doubtful regarding why some organisms survived in their old form, why some went extinct and why others would survive in a changed form. Humans and apes shared the same environment in Africa so common genotype would not have caused our dramatic differences just as lions are not that different than leopard, the cheetah and the others.

What if anything should replace Natural Selection? Humans have applied the most intensive and targeted selective pressure on us and others with great results for our existence. Yet we have not transformed even one organism into another – not even the lowly eColi after decades of laboratory work (Lenski). Our dogs are still basic canines and our cats are still basic felines, not much different than their wild cousins. If anything, we had to adapt to them rather than them to us. The finch, the moth, the antibiotic resistant bacteria are still the original organisms, their hailed changes having reverted or proven simple adaptations. We are no smarter, more powerful or longer living (in absolute) than out primitive ancestors. Selection is not transformative, much less creative.

Humans would apply the Natural Selection method if feasible. But we don’t because it isn’t. A Natural Selection software would use a random generator and a selection criteria to maximize survivability in an available niche. For instance, a family vehicle should optimize the transport function (survivability) given a set of environmental constraints (regulations) and an existing design as starting point. Random minute changes could be tested and retained if the transport function is improved. However, this method can only remove minor oversights but will never create any new designs. Any significant departure such as a new fuel, material or environment either results in a suboptimal design, or requires a cascade of changes to improve the survivability function. That is why the auto industry introduces minor redesign annually and major revamps every few years. And while even the minor improvements must come in harmonized packages rather than one off (to reduce negative ramifications), in the absence of those major redesigns a firm would shortly go extinct.

Pro-Con Notes

Con: What about organic design? Isn’t that natural selection at work?

Pro: No. This is just iterative optimization of a given design. In this case, the wing shape, the material, the environmental forces and the optimization target are all given. The algorithm will not generate a new wing shape or material and it will stop converging as soon as the environment is less than perfectly defined. In addition, this design is radically different from the previous one, and the next iteration will certainly be radically different than this one (no gradualism).

Con: You just don’t understand natural selection.

Pro: If “natural selection” were hard to understand it would not be taught to young children. Instead, “natural selection” is more like very bad street magic where the bus is covered with the cloth and we then are asked to imagine it disappeared without even removing the cloth and showing us the empty space.

Posted in Views and tagged , , , .