December 14, 2010 | David F. Coppedge

Plenty of Time for Evolution?

A mathematician and a biologist have calculated that there’s “plenty of time for evolution.”  Why?  Evolution has a spy: natural selection.  Darwin’s spy vastly decreases the amount of time that “doubters” have argued it would take to create a complex organism like a butterfly.  Will their argument fly?
    The short paper by Herbert Wilf and Warren Ewens was published in PNAS with the simple title, “There’s Plenty of Time for Evolution.1  The gist of their argument is that evolution, contrary to the ideas of critics who envision everything having to evolve sequentially, works like a parallel processor: “Evolution is an ‘in parallel’ process, with beneficial mutations at one gene locus being retained after they become fixed in a population while beneficial mutations at other loci become fixed,” they said.  “In fact this statement is essentially the principle of natural selection.”  They used some equations to show that their model drastically reduces the time required for the evolution of complex systems,2 without referring to any biological cases in particular.  Their model, however, assumes that every beneficial mutation will become fixed during the random mutational search.
    PhysOrg accepted their claims uncritically, even saying in the headline, “New mathematics research proves there’s plenty of time for evolution,” and that “the idea that evolution would require ‘too much time’ to be true is proved false.”  Proof is a strong word.  For support, PhysOrg quoted Wilf comparing evolution to a hacker.  If a hacker tries to guess a complex password, it would take a very long time to do a random search.  But if a spy tells the hacker every time a correct character is found, the time plummets dramatically.  The article tries to clinch the argument:

But what does hacking have to do with the evolution of species?
    Simple, Wilf said.  In the case of evolution, the hacker is evolution itself.  The password is the string of codons that describes, for example, a butterfly.  And the spy is natural selection.

So far we seem to have nothing but a restatement of Darwin’s thesis, or worse, that a butterfly evolves because it evolves.  It also sounds reminiscent of Richard Dawkins’ computer algorithm that converges on “Methinks it is like a weasel,” given that each correct letter mutated in a random string sticks when it is correct.  But who verifies the correctness of each letter?  Who is the spy assisting the hacker?
    Buried in the model is the assumption that each mutation has survival value.  Wilf explained in the PhysOrg article:

“If, when we guess the full string of letters [for a new species], one of the letters is correct – for instance, one that describes correctly the eyes of a butterfly – then that letter has survival value,” he said.
    “It will not be discarded as future mutations take place because the intermediate creatures are seeing very well, and they will live and reproduce.  So although it seems at first glance that the process of random mutations will take a very long time to produce a higher organism, thanks to the spying of natural selection, the process can go very rapidly.
    “In the paper, these ideas are precisely quantified, according to this model, and the extent of the speedup is found.  It is enormous, and shows that there is indeed plenty of time for evolution.

But before Wilf and Ewens can write Q.E.D., a number of questions arise.  Who is watching, and who watches the watchers?  Darwin was trying to rid biology of guiding intelligences that could steer any process toward a goal by design.  The main problem with Dawkins’ analogy was that a target sequence had been preselected by an intelligent agent, and some invisible watcher was verifying the correct letters (in his computer, the algorithm comparing the string with the target sequence).  Evolution has no target sequence; nothing in evolution moves toward the arrival of the butterfly.  Evolution has no comparison algorithm.  Survival cannot be the algorithm; there’s nothing inherent in survival requiring a complex organ to emerge.  Fitness cannot be the algorithm; there is no independent measure of fitness in evolutionary theory outside of survival, which reduces “survival of the fittest” to a tautology – the fitness of the fittest, the survival of the survivors (see “Fitness for dummies,” 10/29/2002).  Additionally, measures of fitness are often based on bad statistics (03/30/2009, 09/05/2008).
    In addition, no single mutation entails any foresight to a distant result.  Wilf and Ewens envisioned beneficial mutations, even if as rare as one in 10,000, having independent, additive benefits to an organism.  This ignores cases where no benefit is conferred till multiple genetic modifications are in place (irreducible complexity), and also fails to account for the damaging effects of pleiotropy – benefits in one gene causing disasters in another (e.g., 06/30/2009) – and epigenetics (07/27/2009, 01/27/2009).  Another challenge is posed by indirect genetic effects that produce “slippage on the treadmill” when fitter offspring facing fitter brethren and enemies (see 03/17/2003).  Arguments from analogy are also fraught with pitfalls (cf. 05/04/2010).  Genomes are “infinitely more complex” than passwords (04/05/2010), containing codes upon codes (04/18/2010, 05/06/2010), and may be fractal in their overall architecture (10/13/2009).  Even a single cell has “staggering complexity” (02/02/2010, 12/29/2009).
    Wilf and Ewens’ paper only included two references and did not answer the criticisms of earlier efforts to overcome the time problem.  See also the 11/14/2010, 09/30/2010, 09/28/2010, 03/08/2010, and 04/02/2008 entries.

1.  Herbert S. Wilf and Warren J. Ewens, “There�s plenty of time for evolution,” Proceedings of the National Academy of Sciences, published online before print December 13, 2010, doi: 10.1073/pnas.1016207107.
2.  For an unknown word of L letters, drawing from an alphabet of K letters, the sequential model would require KL trials to guess the word.  They argue that parallel processing reduces the trials to approximately KlogL.

This paper might have attracted attention in 1860, when Darwin’s X-Men were on a roll (03/04/2004), but not after a century and a half of thought about natural selection and multiple revolutions in genetics.  These guys seem oblivious to all the counterarguments made against similar notions in the past.  Where have they been?  What is most instructive is that such an easily-refuted idea was published by the prestigious National Academy of Sciences, and was offered as “proof” of Charlie Daddy’s story by the popular press.  If this is the best they can offer after so much time, spilled ink and spilled blood, take heart.  Darwin’s totalitarian regime is nearing intellectual collapse.

(Visited 39 times, 1 visits today)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.