Saturday, March 05, 2011

Tikkun - DOES EVOLUTION HAVE A DIRECTION? Where Is It Going? - Andrew P. Smith

Andrew P. Smith
has posted an article on the directionality of evolution over at Tikkun - he references integral theory in general and Ken Wilber's work in particular. However, Wilber seems to be a simple jumping off point, after which he is fleshing out the basic idea that Wilber calls "Eros" with a more scientific model and explanation that does not rely on a New Age image or metaphor. This is a very long article, so I am only sharing a small part - read the whole article at the Tikkun site.

Does Evolution Have a Direction?

Reflections on evolution and consciousness from "the Integral World"

Andrew P. Smith, who has a background in molecular biology, neuroscience and pharmacology, is author of e-books Worlds within Worlds and the novel Noosphere II, which are both available online. He has recently self-published "The Dimensions of Experience: A Natural History of Consciousness" (Xlibris, 2008).


Where Is It Going?

Andrew P. Smith

Wilber seeks to use the eros concept in places where evidence strongly suggests it's not needed.

David Lane is the most recent of several authors at this site to point out the flaws in Ken Wilber's view of evolution.* In doing so, he joins an even longer list of scientists, philosophers and other academics who—beginning with the celebration of the 150th anniversary of publication of The Origin of Species two years ago—have spoken out clearly and forcefully in support of modern evolutionary theory.

If there is one single aspect of this theory at the center of controversy, it is surely embodied in the word “random”. If even Einstein resisted the notion that the universe was a game of dice, it's difficult to blame lay people for trying to find design, or at least consistency, in the story of their origins. David Lane takes great pains to point out that randomness is only one aspect of Darwinism; natural selection, what Jacques Monod (1972) called “necessity”, seems to ensure that evolution has, if no purpose, at least some kind of direction. But if it has a direction, where is it going? If survival of the fittest is not a tautology, as some critics have claimed (Wilkins 1997), shouldn't evolution to some extent be predictable? Using evolutionary theory, we can look back into the past and understand how and why certain forms of life evolved. But what can we say about the future?

We need to address this question not simply to reassure people that life is more than just a matter of chance. Predictability goes to the heart of what science is and does. Science can be succinctly defined as the attempt to identify the causes of phenomena. But the process does not stop there. When we think we know what causes a phenomenon, we attempt to use this knowledge to predict other phenomena, and ultimately, by causing them by our own actions. Indeed, scientific theories are generally validated only by the successful predictions they make. Einstein's relativity theory received an immense boost when it was demonstrated that the gravitational pull of celestial bodies did in fact result in the curvature of light. Conversely, modern string theory, for all its mathematical elegance, has failed to displace other so-called theories of everything (TOE) because of the difficulty of testing any predictions it might make (Smolin 2006).

Darwinism, nearly unique among widely accepted scientific theories today, has largely received a pass here. To be fair, the theory has proven, at the very least, to be quite consistent with many phenomena that were unknown when Darwin formulated it. For example, Darwin's ideas, when coupled with Gregor Mendel's experiments demonstrating the segregation and independent combination of hereditary properties, could have been said to predict the existence of genes. While heredity did not have to involve a macromolecule consisting of two helical strands of mutatable nucleotides, this structure turns out to have precisely the necessary properties. Moreover, comparison of DNA molecules of different but related organisms has shown that the further back in time these organisms began evolving separately—as estimated by fossil evidence—the greater their differences in nucleotide sequence (Cooper et al. 2003). This correlation between molecular and fossil dating, along with other fossil evidence of transitional or intermediate forms (Prothero 2007), constitutes powerful evidence for the theory, and should count as successful predictions. Indeed, as Theodosius Dobzhansky (1973) put it, “nothing in biology makes sense except in the light of evolution.”

Nevertheless, a key aspect of evolution is that it is not simply an historical process, but an ongoing one. It would of course be highly desirable that a theory of evolution should be capable of not just accounting for what forms of life appeared in the past, but for providing some insights into what might emerge in the future. This is where randomness becomes a problem. This randomness occurs not only in the gene mutations that are the source of hereditary variation, but perhaps also to some extent even in the selection process. The late Stephen Jay Gould, who referred to this latter kind of chance as contingency (1990), argued that random events in the past sometimes have had a major effect not just on what variants appeared, but on whether they survived. If we could rewind the evolutionary tape, Gould suggested—go back to any arbitrary date in the distant past and have evolution begin anew from that point—the results would be very different.

Like most evolutionary biologists, though, Gould was interested in the details of evolution—the specific types of structural adaptations developed by organisms. While it may be beyond our power to predict these in advance, what about more general trends? Perhaps there was nothing predictable about the emergence of a four-limbed creature that learned to walk upright and manipulate objects, but was the evolution of a brain, or some similar organ capable of processing information about the environment, inevitable? What about consciousness?

Here I will discuss the evidence for several major evolutionary trends, then briefly speculate on how they might extend into our future. As an important sidelight, I will make the case for the existence of several evolutionary processes other than Darwinism. The never-ending war between evolutionists and creationists has long tended to obscure the important point that one can fully believe in evolution without accepting that Darwinism accounts for all of it.


Complexity is difficult to define, and several different definitions are in use. I will not attempt a precise definition here, but use an approximation that I believe is close enough to be useful. I define it in terms of the number of different states, or possibilities, that a system (a living thing or a machine) can exist in: the more states, the greater the complexity. This definition of complexity is, I believe, reasonably close to more precise definitions based, for example, on the number of computational steps required to create a system (Chaitin 1973; Bennett 1988; Lloyd 2007). In any case, it doesn't really matter if my definition fits exactly with these other, more precise definitions, because I am mainly interested in identifying a certain evolutionary trend, regardless of what name we give it.

We can easily appreciate my definition of complexity by considering what is often taken to be the most complex form of existence on earth: the human brain. With its billions of neurons and trillions of synaptic connections among these neurons, the brain clearly is capable of existing in an enormously large number of different states. Each state is distinguished from every other by the particular set of neurons that are active at that moment, and in the view of modern neuroscience, any particular such neuronal state—or at any rate, many of them—represents some specific form of information. Thus when we think or feel certain thoughts or emotions, a certain pattern of activity occurs in our brains, and this pattern is constantly changing as our thoughts and feelings change. Though the relationship of brain activity to thinking, feeling and other cognitive properties is not entirely understood, it's apparent that the enormous variety of behavior we are all capable of is closely related to the enormous number of different states our brains can exist in.[1]

Defined in this way, it seems obvious that there has been a major increase in complexity during evolution. If we consider the progression: small molecules, macromolecules, cells, invertebrates, vertebrates, humans—it's clear that over time forms of life have appeared that, by this definition, or indeed, anyone's reasonable definition, are more complex than any that preceded them. What makes this view controversial, though, is the implication that there is an inherent drive or purpose to evolution—shades of Wilber's eros—that produces this complexity. Some authors, like Robert Wright (2001), have embraced this idea, and find evidence for it throughout the history of both natural and human social evolution. Others, such as Gould (1997), have argued strenuously against it, conceding that while there has been some increase in complexity over evolutionary history, it results from chance, not any trend. In Gould's metaphor, evolution began with its back against a wall, representing zero or minimum complexity. Any change at all had to result in some increase, a step away from this wall.

I think the most reasonable position here—and the one probably accepted by the majority of scientists—lies between these two extremes. While there may not be an intelligent force guiding evolution to create life forms of increasing complexity, several scientific studies have provided evidence for an increase that does not appear to be a result of just the statistical fluctuations that Gould implied (McShea 1996; Adami 2002). David Lane, referencing Hermann Muller, points out that a process of gradually accumulating variations has the potential to create life forms of greater complexity. I would go further and say that an increase in complexity can often result in greater fitness, so that complexity is not simply possible, but often favored. Richard Dawkins (1997), one of today's strongest proponents of Darwinism, essentially made this point in a debate with Gould. Furthermore, other evolutionary processes that do not involve random variation and natural selection have also been described that could result in an evolutionary increase of complexity (Cavalli-Sforza and Feldman 1981; Kauffman 1993; Barabasi 2002; McShea 2005; Jablonka and Lamb 2005). I will be discussing some of these other processes later.

Those who argue against this idea, that natural selection very often results in increasing complexity, frequently cite the fact that some of the simplest life forms appear to be among the most well adapted. Bacteria, for example, have survived on earth for billions of years, and might well continue to survive under conditions that could result in the extinction of most or all multicellular organisms, including our own species. But those who make this argument generally overlook the fact that though bacteria are far simpler than any multicellular organism, they represent an enormous increase in complexity over what came before them—the primordial abiotic soup of organic molecules. In fact, most of the organic molecules found within bacteria and other kinds of cells today no longer exist on earth outside of these cells. They have been able to survive only by virtue of being part of a lifeform that can control, for example, the degradative processes that quickly destroy these molecules when they are found outside of living cells. So the complexity increase manifested in the evolution of bacteria, and in the still more complex eukaryotic cells that evolved later, represented, from the viewpoint of the cell's macromolecular components, a clear case of greater adaptive fitness.

What about a comparison of different multicellular organisms? There is a good argument to be made that, here, too, more complex life forms are frequently more adaptive. If our species goes extinct, it is most likely to be the result of a cataclysmic event that will also eliminate all other vertebrates and a great many invertebrates. Global warming, which along with other forms of pollution and habitat destruction is considered by many today to have the greatest potential to result in such a cataclysmic event, has already been shown to be a greater threat to many species considerably less complex than we are, including coral reefs and their associated marine invertebrate species and tropical rainforests, which are composed of an enormous diversity of plant and invertebrate as well as vertebrate species (Flannery 2005; Kolbert 2006; Wilson 2007). The World Conservation Union's Red List of endangered species as of 2007 listed 5742 vertebrate species, but also 2108 invertebrate species and 8447 species of plants. These numbers do not begin to tell the full story, as the status of the great majority of species, and particularly of invertebrate species, has not yet been evaluated; but it is telling that the number of threatened species as a percentage of species evaluated is far higher for both plants (70%) and invertebrates (51%) than for vertebrates (23%).[2]

To summarize, while there is no strict correlation between evolution and greater complexity, the fact remains that over time the process has created a significant number of different species which are more complex than any that preceded them. This has occurred not just once or twice or a few times, but literally dozens of times. While there are numerous examples of evolution going down paths that do not lead to greater complexity, it does seem that there is a bias in the selection process that ensures that greater complexity will be created more often than would be predicted on the basis of purely statistical fluctuations. From such a bias, can a trend be born?


If natural selection seems capable of at least occasionally creating life forms of greater complexity, how exactly does it do this? It turns out that there is often a very simple answer: through combination of pre-existing life forms—whether they be molecules, cells or organisms—into societies. Such combination can potentially result in an enormous increase in the number of possibilities or states.

How societies become more complex

This can be demonstrated very easily mathematically.. Suppose we begin with a life form—what Wilber, and others including myself would call a holon—that can exist in just two different states. For example, a nerve cell might either be active, sending an electrical impulse down its axon, or inactive. If we combine this neuron with a second neuron, the combined system can exist in four different states—both neurons active, both inactive, or one or the other active. More generally, if a system consists of n components or holons, each of which can exist in x different states, then the total number of states that system may theoretically exist in is x raised to the nth power: xn. In other words, as more components are added, the complexity of the system increases exponentially.

However, the complexity increase in real-life social systems is potentially much greater than even this. Suppose that a holon adopts a different state every time it interacts with another holon. Then as new holons are added to the system, the number of states that both they, and the existing member holons, can exist in increases. This is shown in Fig. 1. For two holons, the system can exist in only two states, depending on whether or not the holons are interacting. For three holons it can exist in eight different states, while a system of four holons can exist potentially in sixty-four different states. The complexity or number of states increase does not follow the simple xn rule I mentioned previously, because the states of any one holon are not completely independent of the states of all other holons. But this may be more than compensated for by the increase in potential new states as holons are added to the system.

Fig. 1. Relationship between individual and social complexity. (A) In a system consisting of just two components, each component can exist in just two possible states, interacting with the other component, or not interacting with it. These are also the only two possible states of the system as a whole. (B) In a three-component system, each component can exist in four possible states. These are shown in the top row, for the component indicated by the filled circle. This three-component system as a whole, however, can exist in eight possible states, including the four shown in the top row, and an additional four shown in the bottom row. (C) In a four-component system, each component can exist in eight possible states. This is shown for the component indicated by the filled circle. The four-component system as a whole can exist in 64 possible states, which are not shown here. As discussed in the text and endnotes, this scheme has some simplifying assumptions.

So the first lesson here is that complexity is created by social organization. Societies—and I use this term very generally, to mean not simply human or even animal societies, but combinations of molecules and cells as well—are generally more complex, and frequently, far more so, than any of their individual members. This is a point that has been lost not simply on Wilber—who clings to the erroneous and easily disproven idea that societies always have the same degree of complexity (and level of hierarchical development, an issue I will address shortly) as their individual members—but to some extent, I believe, on the entire evolution community.

The really important changes in evolution almost always involve combinations of holons into societies.
Read the rest.

No comments: