Does Darwin Play Dice?

Some recent scientific papers have spoken to the question of how big a role chance plays in Darwinism.  This issue was one of the key points of contention by early critics of Darwin’s theory.  Modern Darwinists argue whether evolution proceeds strictly by chance, or whether the environment constrains evolution to follow certain paths that lead to adaptation.  Those who argue that there is more involved than chance often point to convergence in nature; organisms with different ancestries seem to have converged on nearly identical solutions to problems.  If there is no one guiding evolution, however, it would seem that chance must lie at the root of all change.  Let’s see what some experts say about these issues.

1. Contingency:  In Current Biology,¹ Doug Erwin (Smithsonian Institute) wrote a quick guide to “Evolutionary Contingency.”  Right off the bat, Erwin answered the first question, “What is contingency?” with the response, “Chance, in a word.” 
   

   For instance, all living sea urchins, sand dollars, heart urchins and other echinoids are descended from one (possibly two) species that survived the great End-Permian mass extinction 252 million years ago…. One can argue that the group with two plates was somehow better adapted, or that they simply survived by chance.  In truth, either possibility is equally likely.  Because we only have a single case, we have no way to choose between the two.

   The answer to the next question had an equally short first sentence.  “How do we know?”  His response: “We don’t!”  He claims that convergence is very common, but it’s hard to test for contingency.  He mentions Stephen Jay Gould as the best-known champion of a contingent view of the history of life on earth.
       Erwin answered the final question, “Are contingency and convergence opposing views of how evolution operates?”

   One hopes not, as both have clearly been important in the history of life.  As is so often the case in evolutionary biology, this is an issue of relative frequency, not absolute possibility.  Chance can limit which groups are around to evolve, where they live, and even the range of future morphological possibilities.  Convergence often reflects limited engineering solutions to particular problems, but does not predict that particular groups are likely to survive over the long-term.  And convergence has little to do with many aspects of evolution where selection, genetic drift and chance are free to come up with the remarkable diversity of butterfly wing patterns, arthropod legs or the colors on seashells.

2. Convergence:  In a companion piece in the same issue of Current Biology,² Simon Conway Morris discussed the flip side of evolutionary theory, convergence.  He didn’t explain how different organisms could evolve to similar forms.  He just asserted that they did.
   

   Consider your eye and that of an octopus.  Both are built based on the camera principle, yet you are closely related to a starfish while the octopus is a near cousin of the oyster.  The common ancestor of you and the octopus lived about 550 million years ago and at most possessed a simple eye-spot.  Regarding the eyes, vertebrates and molluscs have arrived at the same solution, and in doing so have solved equally successfully problems such as how to correct spherical aberration.  Camera-eyes are a brilliant evolutionary invention, and so it is less surprising that they convergently emerged in at least five other groups, including cubozoan jellyfish.

   [See 05/13/2005 on cubozoan jellyfish.]  Morris gives other putative examples of convergent evolution in nature.  He says that while instances are remarkable, convergence as an evolutionary principle is not obvious.  Oddly, he says that evolution should not produce convergence at the molecular level, but “is probably far more common than realized.”  For example, he says carbonic anhydrase evolved five times – but “that is modest when compared to C4 photosynthesis, which has arisen at least 30 times.”
       Simon Conway Morris is such a strong believer in environmental forcing of convergent adaptation, he thinks extraterrestrials will think like us.  But they certainly will not think like creationists.  In answering the last question, “Why does convergence matter?” he said,

   It shows adaptation is real, and not some Darwinian conspiracy.  It insists that organisms are functionally integrated and not a heap of character states.  Paradoxically, the very similarities seen in convergence are some of the best proofs of evolution.  Next time you are cornered by a pair of creationists order them a stiff gin and tonic and then ask him why the position of the retina is opposite in our eye to that of octopus (clue: embryology), and ask her why the bacterial flagellar motor has evolved at least twice.  Then when they are sobering up remind them that the way in which Drosophila reacts to ethanol is remarkably similar in terms of behaviour to the manner in which we get drunk.  Please raise a glass to convergence.

   So rather than seeing common design, Morris sees in the remarkable convergence of parts in different lineages an even stronger proof of evolution.  And it matters because it can give you something to argue over with a drunk creationist.

3. Evolvability:  The evolution of the ability to evolve (see 08/04/2004) is the topic of another article in Current Biology by Sniegowski and Murphy.³  This concept is recent and controversial, they begin:
   

   Increasing numbers of biologists are invoking ‘evolvability’ to explain the general significance of genomic and developmental phenomena affecting genetic variation.  What exactly is evolvability, and how important is it likely to be for our understanding of evolution?  Definitions of evolvability are almost as numerous as the papers and books that have been written on the subject.  All definitions agree that evolvability has to do with the capacity of populations to evolve – no surprise there.  In actual use, however, evolvability can be a rather slippery concept with a variety of meanings and implications.  The goals of this primer are to try to pin down some of the meanings of evolvability and to explain why evolvability is a controversial subject.

   Whatever it means, though, it does not mean purpose or goal.  That is made clear in a paragraph entitled, “The problem of teleology” –

   The idea that variability has been fine-tuned in order to maximize the evolutionary potential of populations is certainly controversial, although it is not new.  The obvious reason to be suspicious of this idea is that it suggests a teleological view of evolution.  Natural selection cannot adapt a population for future contingencies any more than an effect can precede its cause, so any future utility of the capacity to generate variation can have no influence on the maintenance of that capacity in the present.  As Sydney Brenner supposedly remarked many years ago, it would make no sense for a population in an early geological period to retain a feature that was useless merely because it might “come in handy in the Cretaceous!”

   Clearly, then, evolvability must be an unguided (i.e., chance) process.  The authors explain their ideas on how this capacity might have come about in a contingent world.  Evolvability is a by-product of the random walk of evolution, they say.  Seeming to realize they have wandered into the bypass meadow of speculation, they repent and ask how the evolvability-as-byproduct hypothesis might be testable.  Surprise: it isn’t—

   In fact, it is rather easy to pile up examples of genomic and developmental features that may affect evolvability, and this is a bit troubling: How do we know when it is necessary – rather than just appealing – to invoke evolvability differences in order to explain evolutionary histories?  The problem here is that the evolvability-as-byproduct hypothesis is probably correct in a very broad sense that tells us little we did not already know: because newly arising variation modifies existing organismal blueprints, large differences between taxa imply differences in the kinds and amounts of new variation that can arise….
       Invoking variability as a retrospective explanation for why one clade has diversified or changed more than another does not rule out the possibility that the clades evolved differently for reasons unrelated to variability.  And finding related to distinctive variability mechanisms � for example, mutations of major phenotypic effect caused by transposable elements � provides only anecdotal evidence for the importance of such variability mechanisms in evolution.  As other commentators on evolvability have noted, there is a need for quantitative, testable predictions concerning evolvability rather than retrospective and anecdotal arguments.  Approaches such as computer simulation and long-term experimental evolution may yield some progress in this direction because they allow direct manipulation and assessment of the effects of variability differences on evolution, but even these kinds of approaches may not provide dependable insights into whether and how variability differences have actually affected the evolution of natural populations.

   In fact, their last sentence says that it’s only an interesting – but largely untested – hypothesis.”  This may be a disconcerting conclusion to readers looking for something a little more substantive in modern evolutionary thought.  But that’s not all: they said earlier that in spite of major advances in molecular biology in the last few decades, “our fundamental genetic understanding of natural selection developed before 1950 and has not changed in major ways since then.”  Is evolutionary explanatory power stuck in a rut?  Maybe so, but when working with hypotheses based on chance, don’t expect nice and neat answers.  This, however, is no excuse for storytelling:

   To some, this historical disjunction suggests that evolutionary theory cannot account for the origin and maintenance of mechanisms affecting variability and is overdue for major revision.  It is indeed attractive to suppose that the most important evolutionary feature of organisms – their very capacity to evolve and adapt – is itself an adaptation, but this is probably only true in highly restricted circumstances.  Instead, variability is probably most often a byproduct of the messy and intricate ways in which genomes have evolved.  And the possibility that incidental differences in variability between populations have caused differences in evolvability with profound consequences for evolutionary history remains an interesting � but largely untested � hypothesis.

4. Creatures of Accident:  In Nature,⁴ Matthew Wills [U of Bath] reviewed a book by Wallace Arthur whose title clearly indicates that chance is king in Darwin’s theory: Creatures of Accident: The Rise of the Animal Kingdom (Hill & Wang, 2006).  Wills began with a rather dogmatic statement of evolution-as-fact:
   

   All life on Earth, no matter how complex, shares a common, very simple, ancestor.  You and the bacteria in your gut have been evolving away from this starting point for precisely the same length of time.  So how is it that you are capable of perusing a magazine, whereas your gut bacteria look little different from the most ancient prokaryote fossils we know?  The answer, in a nutshell, is development.

   With answers that firm, students may not want to ask further questions, unless they want to think outside the nut shell.  Wills proceeded to speculate about the controversy of whether evolution is progressive, egalitarian, or deterministic.  He leans toward Simon Conway Morris:

   The human brain enables behaviours vastly more elaborate than those of even our closest relatives.  But is this an inevitable corollary of evolution?  Arthur –like Simon Conway Morris in his excellent book Life’s Solution (Cambridge University Press, 2003) – thinks it probably is.

   Presumably, the “development” spoken of in Wills’ nutshell answer is a random process.  The question whether complexity is inevitable from unguided developmental evolution led to an aside about theology.  Wills made a timid appeal for less excoriation of theism by Arthur, (as long as it is humanistic and liberal, like that of prominent evolutionary thinkers like “George Gaylord Simpson, Theodosius Dobzhansky, Stephen Jay Gould and even Charles Darwin from time to time”) but the appeasement was more than drowned out by his vitriol aimed at any and all Darwin doubters: “The subtext and closing chapters [of Arthur’s book] are an attack on fundamentalism of every stripe.  Pleasingly, there is little explicit tilting at the straw men of creationism and intelligent design: atheism and theism are equally irrational in Arthur’s view,” he said, lumping together a wide range of views into the narrowest pigeonhole for easy snuffing.

Despite these controversies and deep problems in evolutionary theory, the public gets a very sanitized view.  The Darwin Exhibit from the American Museum of Natural History, visited by half a million Americans during its New York debut, is now traveling throughout major cities, starting with Philadelphia and then Tokyo.  It fails to present the problems in evolutionary theory, and these controversies in particular (see Evolution News on how the exhibit also whitewashes its legacy of eugenics and Social Darwinism).
    Not all are taking Darwin’s explanations without question, however.  A Japanese writer for the World Peace Herald reported that an increasing number of scholars are questioning Darwinism’s ability to explain the complexity of life.  And Carol Iannone writing for Phi Beta Cons said that more students are questioning a theory based on randomness and contingency.  She recommended, “first tell students that Darwin hypothesized that it all comes from chance, and then give them some of the models showing the probability of certain life substances evolving by chance.”  (See online book.)

¹Douglas H. Erwin, “Quick Guide: Evolutionary Contingency,” Current Biology, Volume 16, Issue 19, 10 October 2006, pages R825-R826, doi:10.1016/j.cub.2006.08.076.
²Simon Conway Morris, “Quick Guide: Evolutionary Convergence,” Current Biology, Volume 16, Issue 19, 10 October 2006, pages R826-R827, doi:10.1016/j.cub.2006.08.077.
³Paul D. Sniegowski and Helen A. Murphy, “Primer: Evolvability,” Current Biology, Volume 16, Issue 19, 10 October 2006, Pages R831-R834, doi:10.1016/j.cub.2006.08.080.
⁴Matthew A. Wills, “Evolution’s highest branches,” Nature 443, 633(12 October 2006) | doi:10.1038/443633a.

These Darwinists, who by their own admission are wandering aimlessly in speculation space, nevertheless get free rein in the science rags to trash creationists without consequence.  But what proof do they have for their beloved dogma?  Only 30 proof – the stiff gin and tonic that Morris offered.  No thanks, we don’t drink (10/07/2006 commentary).  When scientific standards return, and the lazy goofballs in the Darwin Party lounges are kicked out (12/22/2003 commentary), there will be a second scientific revolution.  Led by intelligent design instead of chance, it will make the first revolution seem elementary.