When animals have nothing to do they go to sleep. But humans have a drive to know and understand everything. Attentive humans noticed fossils and asked where they came from. Very intelligent humans invented the theory that life evolved over a period of 3.5 billion years. Marshaling the evidence, rational humans judge the theory to be true. Humans then ask: What caused evolution?

Creationists and advocates of intelligent design think that God did it. While there is evidence that God exists (free will, finite beings), there is little evidence supporting this theory. There is no evidence supporting the theory of natural selection. This theory only explains how giraffes got long necks, not how giraffes evolved from bacteria. The reason is that a giraffe is so much more complex than a bacterium and 3.5 billion years is only about a hundred thousand trillion seconds (17 zeros). Not enough is known about the innovations natural selection acts upon for humans to understand how evolution occurred in so short a time. Evolutionary biologists generally speak of "adaptive evolution" in connection with natural selection.

Atheists, creationists, and advocates of intelligent design are responsible for the misinformation that natural selection is intended as an explanation for the complexity of life. The author of the following quote has a Ph.D. in linguistics, not biology. Pinker is Steve Pinker (Ph.D. in linguistics), and Bloom is Paul Bloom (Ph.D. in psychology). Notice that Charles Darwin (Ph.D. in biology) doesn't think natural selection explains the complexity of the human eye:

"They [Pinker and Bloom] particularly emphasized that language is incredibly complex, as Chomsky had been saying for decades. Indeed, it was the enormous complexity of language that made is hard to imagine not merely how it had evolved but that it had evolved at all.

"But, continued Pinker and Bloom, complexity is not a problem for evolution. Consider the eye. The little organ is composed of many specialized parts, each delicately calibrated to perform its role in conjunction with the others. It includes the cornea,...Even Darwin said that it was hard to imagine how the eye could have evolved.

"And yet, he explained, it did evolve, and the only possible way is through natural selection--the inestimable back-and-forth of random genetic mutation with small effects...Over the eons, those small changes accreted and eventually resulted in the eye as we know it." (Christine Kenneally, The First Word: The Search for the Origins of Language pp. 59-60)

McShae and Brandon state the limitations of natural selection explicitly:

"The history of life presents three great sources of wonder. One is adaptation, the marvelous fit between organism and environment. The other two are diversity and complexity, the huge variety of living forms today and the enormous complexity of their internal structure. Natural selection explains adaptation. But what explains diversity and complexity?" (location 78, Kindle)

The second bit of nonsense McShae and Brandon refute is the idea that evolution obeys the second law of thermodynamics. In November, 2008, the American Journal of Physics published an article "Entropy and evolution" that includes an absurd equation combining thermodynamics and probability theory to prove that evolution does NOT violate the second law of thermodynamics. The article was probably written in good faith because there are a lot of books and articles about evolution and the second law of thermodynamics.

According to the second law of thermodynamics, a gas will fill up the entire container it is in because this is the most probable distribution of non-interacting molecules. In other words, nature goes from knowledge to lack of knowledge about the location of molecules, order to disorder, complexity to simplicity, or low entropy to high entropy. When shuffling a deck of playing cards the chance of the deck getting back into its original factory order is about one in 10, 000 vigintillion (67 zeros). I mention cards because physicists label the molecules in a gas No. 1, No. 2, etc. and a deck of playing cards comes automatically labeled. Notice that there are almost four times as many zeros in this number as in the time for evolution.

In trying to understand evolution, biologists use as a model for a protein, not a deck of playing cards, but the English sonnet. A protein has four levels of complexity, but the primary structure has hundreds of amino acids and each amino acid has to be in exactly the right location for the protein to work. As there are 20 amino acids and 26 letters, biologists calculate how long it would take a computer to generate a sonnet by the random selection of letters. (Marc Kirschner and John Gerhart, The The Plausibility of Life: Resolving Darwin's Dilemma, pages 32).

Kirschner and Gerhart only report a calculation simulating natural selection for a the phrase "to be or not to be." My guess is that nobody did the calculation for a sonnet because nobody cares. The primary structure of a protein doesn't even begin to describe the complexity of life. The dozens of proteins that make the flagellum of a bacterium rotate is a static kind of complexity, like the primary structure of a protein. By contrast, genetic engineering is the ability of cells to detect changes in the environment and create new proteins in response to the change. The ability of a fertilized animal egg to develop into a multicellular animal is another example of complexity. The instinctual behavior of animals and the knowledge of grammar that human infants are born with are other "sources of wonder," as the authors express it.

The only connection between evolution and the second law of thermodynamics is that evolutionary biologists and physicists both perform probability calculations. The AJP article singles out for criticism the creationist Henry Morris for saying evolution violates the second law of thermodynamics. The article argues that the second law only applies to closed systems without mentioning the criterion that it only applies to non-interacting particles.

The second law of thermodynamics does not apply to very large numbers of hydrogen atoms in outer space because the force of gravity causes stars to be formed from these atoms. A living organism is anything but a system of non-interacting particles. This is the real reason evolution does not violate the second law. With the goal of understanding evolution, the earth is a closed system. It is true that the sun bathes the earth with its energy. But energy from the sun generally heats things up and makes them more disordered, not less disordered.

This is the way the McShae and Brandon explain that the complexity of life has nothing to do with the second law of thermodynamics:

"Based on what we have said so far, some will be poised and ready to make a leap, from the notion of accumulation of accidents to the second law of thermodynamics.... We advise readers against this, for their own safety. We are concerned that on the other side of that leap there may be no firm footing. Indeed, there may be an abyss. First, we think the foundation of the ZFEL [zero-force evolutionary law] lies in probability theory, not in the second law or any other law of physics. And second, our notions of diversity and complexity differ fundamentally from entropy, in that entropy, unlike diversity and complexity is not a level-related concept. " (location 220 on Kindle)

I found this book to present an interesting argument, but I was continually bothered by one of their basic conclusions related to the implications of random variation among independent groups.

The authors assert that among two different groups whose components vary randomly and independently from each other, the two groups will be become more and more distinct from each other, on average, over time. I don't believe statistical theory supports this assertion.

For example, two piles of leaves (a favorite analogy of the authors) are 100 ft. apart from each other. Random wind patterns may distribute the leaves in each pile randomly and the variation in the leaf movements are independent between the two piles. Over time, the both piles will be obliterated but if the wind patterns are truly random than the leaves in the first pile will, on average, still be 100 ft. from the leaves in the second pile, on average.

If the authors do not consider average distinctness of one group of leaves or cells from another to be measured in this way, they do not explicitly describe how they would propose to measure it differently. I believe the issue would remain nonetheless.

I. Brief Thoughts

This book posits the "Zero-Force Evolutionary Law", or ZFEL (which the authors suggest pronouncing like "zeffle"). The book is not long, but it is dense, and the authors are what an old professor of mine would call "good philosophers" -- they tell you what they are doing as they do it. The lines of reasoning are easy to follow, and examples, objections, and distinctions are offered in due course to make sure readers keep the discussion well sorted in their minds. The ZFEL, briefly put, is a hierachical and probabilistic explanation for the widespread observation that organisms tend to diversify and become more complex over time. The authors argue the ZFEL is the best null hypothesis in biological situations, and they point to a change in how we formulate biological explanation. They discuss both relevant empirical and philosophical points in a wide-ranging but deceptively simple book.

- - - - - - -

II. Further Comment

The ZFEL law is argued to be the fundamental background condition in any biological situation. Considering, as Dobzhansky put it, that nothing in biology makes sense except in the light of evolution, it will come as no surprise that "biology's first law" is an evolutionary law. It may come as a somewhat greater surprise that the ZFEL is distinct from natural selection. The ZFEL, though, is posited in a strongly neo-Darwinian context, and the authors argue strenuously that the ZFEL complements the principle of natural selection.

Here is how the authors articulate the ZFEL:

"In any evolutionary system in which there is variation and heredity, there is a tendency for diversity and complexity to increase, one that is always present but may be opposed or augmented by natural selection, other forces, or constraints acting on diversity and complexity."

The simplest example the authors use is the accumulation of variation on the pickets of a fence. As paint flakes off, lichen appears, etc., the variation is hereditary (retained through time), and so as time progresses the pickets become increasingly different. Importantly, We can say either that the population of pickets becomes more *diverse* or that the fence becomes more *complex*. Typically, we speak of complexity when examining "down" to a smaller scale, such as cellular machinery, whereas we speak of diversity when looking "up" to a wider scale, such as an ecosystem. Both, however, refer to the number of constituents of some entity -- both are hierarchical. And anyone who's ever seen (or made) a phylogeny should recognize that evolution is manifestly hierarchical.

Thus the authors point to evidence for the ZFEL, arguing "[t]he ZFEL is supported by an enormous amount of evidence, at every temporal and physical scale, at every level of organization, across biology." They specify that "[t]wo of the most compelling pieces of evidence" are "the increase in phenotypic diversity over the history of life and the rise in genomic complexity marked by the divergence of pseudogenes." Thus the ZFEL is a formalization of widespread patterns we are all familiar with. The authors argue that the pattern of increasing diversity/complexity is so pervasive that it constitutes the background condition, the null hypothesis of biology.

It is helpful to contrast the zero-force expectation of the ZFEL against the Newtonian zero-force expectation. The nature of inertia dictates that an unperturbed object maintains its trajectory -- that, absent some additional force, an entity at rest remains the same. The ZFEL says that we should expect an entity to be different (or differentiated) at some later time, as a null expectation, in the absence of any intervening forces.

It is also helpful to contrast the ZFEL against a pair of well regarded biological mechanisms: diversifying selection and drift. Both are already recognized as agents that increase diversity/complexity in the biosphere. The ZFEL subsumes these mechanisms with an important hierachical and probabilistic notion of respective randomness. Here is how the authors explain their position:

"Consider this. Take a snapshot of all of the people on a crowded city street corner at some moment in the middle of the day. Then find these same people 10 minutes later. Find them again 20 minutes later, and then 30 minutes later. With the passage of time, they will become increasingly dispersed, or in other words, the variance in their locations will increase. And this is true even if the trajectory of each person is the deterministic outcome of his or her plans for that afternoon. One is on her way to her office. Another is walking his dog. A third is going grocery shopping. And so on. To the extent those motivations and plans are different from and independent of each other, the individuals movements are random with respect to each other. And dispersion at the higher level - the greater variance in location of the group - is the expected outcome of randomness in the with-respect-to sense at the lower level. This is the principle underlying the ZFEL."

The ZFEL turns on this kind of hierarchial, probabilistic thinking. Thus although the ZFEL is a formalization of conventional data, the novelty of the ZFEL is in interpreting biological causation as essentially probabilistic and hierachical. Now, to be sure, biology has long grappled with both hierarchy and probability. But, the traditional norm has been to treat both notions rather differently than the ZFEL suggests. Discerning the novelty in the ZFEL is tricky, because, as the authors note:

"What we have offered here is a standard sort of scientific argument for a theoretical point of view. We have invoked the huge body of data represented by virtually everything we know about macroevolutionary divergence and argued that it best explained by the ZFEL. Further, we have argued that the standard explanations in the field are, implicitly, ZFEL explanations. So even though we are explicitly stating the ZFEL here for the first time, the data pre-exist and strongly support the ZFEL."

So, what is the value of dwelling on the obvious like this? The novelty of the ZFEL is not that it predicts diversification/complexification -- this is already the intuition of everyone from oncologists to ecologists. The novelty is in the roles of probability and hierarchy in explanation.

Probability is traditionally understood as a tool for verification (the "prob" in the word comes from the same root that gave us words like "prove" and "probity" and "proof"). In a complicated world, we can't expect even valid results to be perfectly self-evident, so probabiilty is used to reveal when re are credible. McShea and Brandon, however, suggest that probability can become the logical basis of biological CAUSATION, like calculus did for classical mechanics. They are not precisely the first to suggest this, and point (for example) to another author's argument that probability is a sufficient explanatory tool for radio-decay.*

Hierarchical scales of biological activity delineate the traditioanl subdisciplines in biology; for example, biochemistry, cell biology, and physiology. Because biologists recognize that different scales are truly different -- there are different types of data to collect and different processess to explain -- the practice is to confine investigation to a single level. Rather less emphasis, however, has been put on the relationship between levels. The ZFEL explicitly states that biological explanation can be INTER-scale, not just INTRA-scale. At a time when interdisciplinary work is coming into its own, the ZFEL might be a helpful guiding tool in what inter-scale relationships investigators could work toward.

Somewhat more abstractly, the ZFEL challenges a person to try a kind of hierarchical thinking that is foreign and yet deeply gratifying to the intuition of anyone familiar with nature. One of the curious things about biota is that we can more fully distinguish two organisms that are more similar. That is, the more similar things are to each other, the more specific we can be about how they are different. If we want to compare toads to (other) frogs, we walk away feeling like we have said something directly relevant to what makes toads toads and frogs frogs. If we want to compare toads to porpoises, we will end up specifying a set of differences and similarities that differs substantially from the toad-frog set (e.g. how many heart chambers, rather than how water-permeable the skin). Comparing toads to sycamores will result in a still-different set of comparisons (e.g. the source of mechanical rigidity). McShea and Brandon recognize this situation in their explanation of how to paramaterize complexity/diversity -- by doing it ad hoc, with whatever characters could be useful for a given comparison -- and they make this situation explanatorily relevant in their formulation of the ZFEL. The slippery nature of inter-taxon comparison is manifestly a hierarchical phenomenon, and it is what makes it impossible to develop a single set of parameters that would describe all organisms fully without redundancy.**

This brings us to the crucial difference between the ZFEL and the 2nd Law of Thermodynamics: entropy can be compared in identical units between any two entities. Colloquially speaking, entropy is everything getting messier, while the ZFEL is everything getting more complicated. As the ZFEL pushes things to become more different, the nature of their difference itself becomes more complicated. This is not the case for entropy. To be sure, the ZFEL complements, rather than contradicts, the law of entropy. McShea and Brandon point to a literature on the relationship between entropy and evolution, citing it as an important influence on their own thinking. For some time, physicists have distinguished dis/organization from entropy. Consider, for example, that before the universe segregated into galaxy clusters, with galaxies of varied structure, with stars having distinct layers, surrounded by distinctly composed planets arrayed in ordered orbits, the rather formless early universe had *lower entropy* than the *highly organized* universe we see now. For me, at least, this situation highlights the weakness of my intuition about what entropy really is. Entropy and organization actually increase together.

If the basic expectation of the ZFEL -- increasing diversity/complexity through time -- is so overwhelmingly confirmed by experience as to be just the status quo recapitulated, the ZFEL's challenge to intuition and the reorientation of causal explanation are so subversive that they might discomfit even ZFEL enthusiasts. I suspect this is what Michael Foote had in mind saying the "ZFEL will be obvious to some, heretical to others" (see editorial review). But I would differ slightly from Foote, as I think it is precisely those who find the ZFEL self-evident and congruous with their former education who will grapple with its devious truth.

* A bias toward sufficiency over necessity is heresy to many. I suspect Monod would smile.

** Whether nucleotide sequence accomplishes this task is an interesting issue, but space limits.