Uncommon Descent Serving The Intelligent Design Community

# Can Natural Selection Defy the Odds? Not When They’re This Long.

Share
Flipboard
Print
Email

Alan Prendergast has this story in Westword about Robert Hannum, a professor of “applied probability” at the University of Denver. Recently the management of a casino hired Professor Hannum to investigate a roulette player whom they suspected might be cheating. The house has a huge mathematical advantage in roulette, which is why the casino suspected something other than random chance was involved when the player parlayed a few thousand dollars into over \$1.4 million.

Professor Hannum crunched the numbers, however, and told the casino that while the player’s run was very unlikely (about an 80:1 shot), it was not so unlikely as to suggest cheating. And sure enough, over the next few gaming sessions the player blew his entire \$1.4 million stack.

What was the key assumption underlying the casino management’s request of Professor Hannum? They assumed that some events are just too improbable reasonably to attribute them to the interaction of random chance and the physical laws of nature working on the roulette wheel (i.e., “physical necessity”), and and if an event is not caused by the interaction of chance and necessity, the most likely cause of the event is design by an intelligent agent. In the particular case of gaming “design by an intelligent agent” goes by the name of “cheating.” Finally, the very fact they hired Professor Hannum suggests they understood that design leaves behind indicia that can be sussed out objectively.

Consider an example from poker. Suppose a poker dealer deals himself 13 royal flushes in hearts in a row in a five card game. The odds of this happening are easy to calculate. They are about 2.74^-71. To put that number into perspective, the dealer could deal the same 13 hands to every atom in the universe, and it is less than even money that any atom would receive that same series of hands. Conclusion: It is not, as a matter of strict logic, impossible for random chance to result in 13 royal flushes in a row, but the odds of that happing are so low that the inference to design is overwhelming.

Now the odds of the information content of even the simplest strand of DNA forming though pure random chance are even less than the odds of dealing 13 royal flushes in a row. Yet Neo-Darwinian evolution (NDE) theorists routinely discount the design inference. How can this be?

It seems to me that NDE proponents have two responses to this question. First, they inform us that when it comes to evolution, chance is only half of the equation. Just as with the roulette wheel, the other half of the equation is physical necessity. And, NDE proponents go on to say that NDE has a force of nature working for it that roulette players do not — natural selection. Natural selection, they say, sorts though all the randomness and creates specifications, such as the information content of DNA, that only appear to be designed by an intelligent agent.

Natural selection is, of course, a real force of nature, as demonstrated by the development of drug resistance by the malaria microbe through purely Darwinian processes. But, as Michael Behe has convincingly demonstrated, the power of natural selection is limited. Natural selection can provide a selective advantage by degrading a genome, as it does in the malaria example. But its power to BUILD a complex genome has never been demonstrated in the laboratory. In fact, the laboratory has shown as that over countless trillions of reproductive events, natural selection has NOT created complex new additions to the genome.

When Darwin observed the beaks of Galapagos finches, he was observing small changes in an organism’s phenotype (i.e., the organism’s body plan) that gave the organism a selective advantage and thereby increased its predominance in the population. From this observation Darwin made an inference that has literally changed the world. He inferred that the same process was responsible for creating finches in the first place. Obviously, Darwin did not observe this process create finches. He reasoned, however, that a process that could create one small change in a population of organisms could create other small changes, and over time, those changes would accumulate, and when sufficient changes had accumulated over a long enough time, an entirely new species would emerge. This entirely natural process, Darwin reasoned, was responsible for the creation of all life, from the first single-celled organism on though to human beings themselves.

The important thing always to keep in mind is this: “Darwin inferred . . .” Again, Darwin did not observe one species morphing into another through the process of natural selection. The finches remained finches. They did not change into another kind of bird, much less another kind of species altogether. Nor has anyone since Darwin observed a species morph into another.

The main point is that the power of natural selection to create large, as opposed to small, changes in the genotype and the phenotype of organisms remains, to this day, an inference from the data, not the data itself. If any NDE proponent commenting on this post disputes this assertion, I invite him or her to cite a single example of one species being observed changing into another since Origin of Species was published in 1859.

This gets me back to our discussion of probability. As I said, NDE proponents assume that natural selection has the power to beat the odds and create, for example, highly complex and specified strands of DNA, the creation of which is beyond the power of mere chance. But since no one has ever observed natural selection create complex changes in a strand of DNA (much less create the strand of DNA from scratch in the first place), how can NDE proponents be so dead certain of the staggering, almost God-like powers of creation they attribute to natural selection? One would think they would be more modest in their claims for a process that has never actually been observed. Instead, they bombastically assert that their theory has the same epistemological standing as the theory of gravity.

And what happens when someone fails to bow at the altar of NDE? Well, this brings me to the NDE proponents’ second response that I mentioned above, which I call the “shut up and sit down” response. As I noted above, NDE proponents have inferred (as opposed to observed) from the data the vast creative powers of natural selection. Sadly, the more intellectually strident among them suggest that this is the ONLY allowable inference that can be made from the data. Why is this the only inference allowed?

Because, for those who worship at the altar of materialism, the creative power of natural selection is true as a matter of simple logic before we even get to the data. Indeed, there is no need to even look at the data, because the data are irrelevant. The metaphysics is what matters, and the metaphysics leads inexorably to the inference that NDE through natural selection or something very like it is true.

An unbiased review of the data might suggest there are substantial observational difficulties with NDE, not the least of which is the fossil record, which is characterized NOT by the gradual accumulation of small changes Darwin’s theory predicts, but by sudden emergence and stasis, exactly the opposite of what Darwin’s theory suggests should be the case. But, as I said, the data are irrelevant. This is why when the data do not fit the theory NDE proponents blame the data, not the theory. In other words, if the data do not fit the theory, so much the worse for the data. In the case of the fossil record, for example, no NDE proponent denies the data, on the whole, run counter to the theory. Instead, they say that even after 150 years of development, the record is far too incomplete to make any firm conclusions. In other words, even after 150 years, the problem is with the fossil record, not the theory.

And if an ID proponent stands up and says, “Wait a minute. There is another inference that can be drawn from the existence of specified complexity in DNA and the irreducibly complex structures we observe in all living things. In fact, the most plausible inference is that both the specified complexity and the irreducible complexity are designed.” In discussions of this sort between the two camps the NDE proponents invariable get around to saying something on the order of “IDiot! Shut up and sit down. Don’t you know that under our materialist assumptions, the truth of natural selection is actually derived as a matter of logic, and therefore any assertion that denies that truth is quite literally absurd.”

Never mind that, as I and others have demonstrated, ID theory works perfectly well even given materialist assumptions. NDE proponents insist that it cannot. Never mind that an argument (such as the creative power of natural selection) that rests on an unspoken metaphysical assumption that assumes the conclusion of the argument is not very satisfying. Sit down and shut up IDiots. We have rigged the game so that you can never win.

The history of ideas has shown, however, that strategies like this, while than can win in the short run, can never win in the long run.

Origenes I think you're on to something very good and very important. This point has not been emphasized or analyzed enough, in my opinion. We all know that the term "natural selection" is a misnomer because nothing is selected. But more importantly, as you point out, selection just means a reduction of some part of a population.
Sure, ‘selection’ is a real destructive force, but, within the context of NDE, mutation—blind chance—causes drug resistance, not ‘selection’.
Selection is dependent upon death. If there is no threat of death by competition, then there is no selection. That part also does not get explained correctly. We hear of 'fitness benefit' and often it assumes this can occur without death to all who don't have the fitness benefit. But that's not the case. If the threat doesn't kill off all organisms that don't have the beneficial mutation, then there is no "selection". How many times does that actually happen in nature? Nobody has ever even tried to explain it.
Elizabeth Liddle @28: No, of course natural selection doesn’t produce (how could it?) the very variance that results in differential reproduction. That has to have some other cause …
She makes it sound like "of course, everybody knows that" - but no, they don't. In many research papers we read that "natural selection produced" whatever result. So, in a three word phrase a combination of two deceptive terms is used: First, "natural selection" where nothing is "selected". The second "selection produced", when selection produces nothing. This all points to the key claim: "Evolution is non-random". But that is false. It is claimed that "selection is non-random" because it "selects" the fittest. There again, the deceptive terminology. Let's just paraphrase E. Liddle - "of course natural selection does not select". Evolution is entirely chance. Mutations = random chance. Selection = reduction of the population so only the fittest are left. The Fittest = dependent on the environment. Environment = random chance. Selection chooses the fittest. Organisms are the fittest because of random mutations and random environments. So, selection reduces the population until a random output is left (whatever is fit is based on mutation and environment, both random). So, evolution is entirely random. Sounds like a tornado in a junkyard to me. Silver Asiatic
Going through some older posts in order to find out how ‘natural selection’ is understood.
Barry Arrington: Now the odds of the information content of even the simplest strand of DNA forming though pure random chance are even less than the odds of dealing 13 royal flushes in a row. Yet Neo-Darwinian evolution (NDE) theorists routinely discount the design inference. How can this be? It seems to me that NDE proponents have two responses to this question. First, they inform us that when it comes to evolution, chance is only half of the equation. Just as with the roulette wheel, the other half of the equation is physical necessity. And, NDE proponents go on to say that NDE has a force of nature working for it that roulette players do not — natural selection. Natural selection, they say, sorts through all the randomness and creates specifications, such as the information content of DNA, that only appear to be designed by an intelligent agent.
My position is that ‘selection’ (read: elimination) creates nothing.
BA: Natural selection is, of course, a real force of nature, as demonstrated by the development of drug resistance by the malaria microbe through purely Darwinian processes.
Sure, ‘selection’ is a real destructive force, but, within the context of NDE, mutation—blind chance—causes drug resistance, not ‘selection’.
BA: But, as Michael Behe has convincingly demonstrated, the power of natural selection is limited. Natural selection can provide a selective advantage by degrading a genome, as it does in the malaria example.
'Selection' does not degrade the genome. Mutation degrades the genome. Behe has demonstrated that the power of mutation is limited, given the size of the search space and given that organisms must retain viability.
BA: But its power to BUILD a complex genome has never been demonstrated in the laboratory. In fact, the laboratory has shown as that over countless trillions of reproductive events, natural selection has NOT created complex new additions to the genome.
‘Selection’ (read: elimination) has never created anything — neither complex nor simple. - - - In this very thread there are several people who make the same simple straightforward point as I do, unfortunately the matter gets little attention. Why is that?
Meleagar @6 : I don’t understand why natural selection is a help when trying to acquire targets of high specific function. All natural selection does is end pathways; it doesn’t create them ...
Matteo @ 17: You have nailed it precisely. The materialists would have us believe that natural selection is a probability enhancer, when, in fact (since all it is is death), it is a probability reducer.
Elizabeth Liddle @28: No, of course natural selection doesn’t produce (how could it?) the very variance that results in differential reproduction. That has to have some other cause …
Origenes
No response from Lizzie. The question was simple: What is the cause, and what is the effect? Mung
Natural selection refers to no more and no less than the phenomenon of heritable traits having phenotypic effects that affect the probability of successful reproduction in a given environment.
What is the cause, and what is the effect? Mung
Well natural selection can explain the appearance of design, but that doesn't mean that it isn't the same thing as differential reproduction! The two terms mean the same thing. The only caveat I would make (but it applies to both terms) is that both normally applied to differential reproduction as conferred by heritable traits. But natural selection/differential reproduction can also be applied to non-heritable traits (or traits not heritable by genetic means). I wish myself that differential reproduction was used more widely, as it is less likely to mislead people into thinking of it as an external agent, and using phrases like "natural selection selects". Natural selection refers to no more and no less than the phenomenon of heritable traits having phenotypic effects that affect the probability of successful reproduction in a given environment. Elizabeth Liddle
Mung, are you seriously querying that natural selection and differential reproduction are the same thing?
Absolutely. Natural selection and differential reproduction are the same thing when the evolutionist needs them to be. When the evolutionist needs for natural selection to mean something else then it means something else.
In what sense are they not?
When natural selection needs to be used to explain the appearance of design, or become explanatory rather than tautological, or be applied to a specific case in nature, or be testable, etc., etc. Even the Wikipedia page on natural selection admits to different meanings of the term. Mung
Mung, are you seriously querying that natural selection and differential reproduction are the same thing? In what sense are they not? Of course one doesn't "cause" the other. That really would be a tautology. Elizabeth Liddle
Formulations of natural selection fall into four groups: tautologies, special definitions, metaphysics, and lame formulations (T, SD, M, L). - Tautologies are not testable scientific explanations. They are definitions masquerading as explanations. - Special definitions are a multitude of conflicting explanations masquerading as a single unified theory. - Metaphysical explanations are not testable, therefore they are not scientific. - Lame formulations do not even address the problem of adaptation, therefore they cannot solve it. None of these formulations scientifically solves the problem of adaptation and design. The illusion that "natural selection is science" was created by shifting back and forth between formulations. The shifting was concealed by various factors: - Vague and ambiguous keywords (like fitness - Rapid shifting between formulations - Over-emphasis of peripheral issues, like reproduction and probability Walter James ReMine, The Biotic Message
Mung
Inventive natural selection is the essential evolutionary mechanism for the origin of life's adaptations. The mechanism includes survival of the fittest (acting with reproduction and mutation) plus many other mechanisms and assumptions.
Mung
Naive natural selection assumes that survival of the fittest (acting with reproduction and mutation) is sufficient to account for the adaptations of life.
And boy have we seen plenty of examples of naive natural selection here of late. Mung
natural selection IS differential reproduction
Well, I would certainly hope that it isn't something that it's not!
natural selection IS differential reproduction
Except when it's something else, as needed by the evolutionist. But at times, I swear, it sounds like you're claiming that the cause of differential reproduction is natural selection. That is to say, the cause of natural selection IS natural selection. No wonder people are commenting on whether you are confused about the cause/effect relationship between natural selection and differential reproduction. Cheers. Mung
heh. Thanks ba77 :) Yes, there's a lot of research into that kind of thing (showing that kind of result). We aren't as rational as we think we are :) Elizabeth Liddle
Elizabeth, I think you may this recent article from ENV interesting: Psychologists Discover Their Own Biases http://www.evolutionnews.org/2011/06/psychologists_discover_their_o047081.html bornagain77
Mung: natural selection IS differential reproduction. Elizabeth Liddle
Science tries to explain effects in terms of their causes. As cause and effect are different, the two sides of a causal explanation cannot be the same. In a tautology the cause and effect are the same. Therefore there is no explanation. A tautology gives the appearance of being explanatory when, in fact, it is not. - Walter James ReMine, The Biotic Message
Mung
Chuckie:
so different from each other, and dependent upon each other in so complex a manner, have all been produced by laws acting around us.
In so complex a manner as to be untestable. Great. Having been produced by laws, which we're still waiting to discover.
a Ratio of Increase so high as to lead to a Struggle for Life, and as a consequence to Natural Selection
Well, no. No such increase is needed we now here. All that's needed is "differential reproduction." Chuckie's not doing too well. Mung
This debate on NS is hilarious. I got a new copy of ReMine's The Biotic Message and re-read his chapters on Natural Seleciton and I get to see it all in action.
Summary Inventive natural seelction is the distinctive evolutionary mechanism - essential to Darwinian theory. Evolutionists presume it creates new adaptations by somehow traversing the hills and valleys of the fitness terrain. But they do not attempt to defend it as testable science. Rather, for the defense they shift back to the naive version - survival of the fittest. Then they might offer some tautology to help expunge all doubt. When challenged, they shift between various formulations They use naive natural selection to convince the public that evolution is simple, testable, and virtually inevitable. When opponents point out that such continually uphill evolution is refuted by the data, evolutionists effortlessly shift away from naive natural selection. Then they charge that the opponent has a poor understanding of evolutionary theory. In short, evolutionists merely shifted away from criticism, then focused their arguments (and your attention) in a direction that seemed to overcome the criticism. This phenomenon occurs at several levels. Biological adaptation by natural selection is not inevitable, nor is the theory scientific. It had merely lent support to the philosohpy of naturalism.
Mung
Well, either you are listening to people who don't understand the theory, or misunderstanding them. No, of course natural selection doesn't produce (how could it?) the very variance that results in differential reproduction. That has to have some other cause (and there are many sources of variance). Nor does natural selection account for the first self-replicating-with-variance critter. To quote the Great Chuckie himself:
It is interesting to contemplate a tangled bank, clothed with many plants of many kinds, with birds singing on the bushes, with various insects flitting about, and with worms crawling through the damp earth, and to reflect that these elaborately constructed forms, so different from each other, and dependent upon each other in so complex a manner, have all been produced by laws acting around us. These laws, taken in the largest sense, being Growth with reproduction; Inheritance which is almost implied by reproduction; Variability from the indirect and direct action of the conditions of life, and from use and disuse; a Ratio of Increase so high as to lead to a Struggle for Life, and as a consequence to Natural Selection, entailing Divergence of Character and the Extinction of less improved forms. Thus, from the war of nature, from famine and death, the most exalted object which we are capable of conceiving, namely, the production of the higher animals, directly follows. There is grandeur in this view of life, with its several powers, having been originally breathed by the Creator into a few forms or into one; and that, whilst this planet has gone circling on according to the fixed law of gravity, from so simple a beginning endless forms most beautiful and most wonderful have been, and are being evolved.
The first bolded part refers to Variation - where do the fitter traits come from? - which Darwin didn't know (he even backed Lamarckian mechanisms in the end), and the second bolded part refers to the Arrival part, which again he did not know, and indeed, in this passage, attributes to a Creator. Natural selection won't even operate without some mechanism that ensures variability in the progeny of the self-replicators, because without difference there can be no differential reproduction! Nor can it account for the first self-replicator. So while there may be silly people around, I don't know of anyone who knows anything about evolution make the claim you say they make. Keep better company maybe :) Cheers PS: Oh, and I don't demonize anyone. Don't believe in demons :) Elizabeth Liddle
Ilíon: "As was pointed out well over a century ago, the “survival of the fittest” does not explain the arrival” of “the fittest." Silly DarwinDefender: "I don’t think that has eluded anyone, least of all Darwin." And yet, you silly people keep up the tradition started by Saint Chuckie Himself to pretend and assert that it does ... and to, at best, ingore the people who point out that you haven't explained anything; and, more generally, you all don't merely ignore, but rather actively demonize those who point out that there is no there there in Darwinist "explanations." Ilion
And I could cite no less an authority than Saint Chuckie Himself that "natural selection" is equal to "survival of the fittest." "Natural selection" is like a (blind) breeder of fancy pigeons who "selects" by (mindlessly/goallessly) killing off as many individuals as he can catch. Ilion
ilion:
As I said, cause and effect, and differentiation of them, seems to elude you. As was pointed out well over a century ago, the “survival of the fittest” does not explain the arrival” of “the fittest.”
No, it doesn't. I don't think that has eluded anyone, least of all Darwin. Elizabeth Liddle
tbh, while I know why Darwin chose the term (as an analog of the "selection" practiced by farmers), "differential reproduction" would have been a better term. It amounts to no more or less than that. Differential reproduction means, automatically, that traits that result in more efficient/probable reproduction are going to be better represented in the subsequent generation. It's precisely that mechanism that Darwin dubbed "natural selection". As you imply, natural selection isn't a cause, it's an effect. However it also has an effect, which is the repeated concentration (distillation if you like) of the most fecund traits over generations. And if new traits are also being spontaneously generated, those few that endow their bearers with increased fecundity will feature in the new distillate. Elizabeth Liddle
E.Liddle: "So the probability of a complex structure that confers even greater fecundity is constantly increased." As I said, cause and effect, and differentiation of them, seems to elude you. As was pointed out well over a century ago, the "survival of the fittest" does not explain the arrival" of "the fittest." Ilion
Not at all, ilion. Natural selection IS differential reproduction. I could cite no less an authority than John Sanford on that, although simple logic should suffice :) Elizabeth Liddle
E.Liddle: "Firstly, “natural selection isn’t “all death” – it’s differential reproduction." You people have a really hard time understanding and differentiating cause and effect, don't you? Ilion
Hi Matteo :) You write:
Differential reproduction equates to differential destruction. Destruction is not construction.
. But differential reproduction doesn't "equate to differential destruction". I have one child; my friend has two. No child has been destroyed; yet two people carry her genes into the next generation, only one carries mine.
The point is this: if one were to calculate probabilities of reaching a particular function via random variation based on no constraints of competition and no natural selection, where organisms are free to breed and reproduce their heart’s content, with no culling of the herd, those probabilities are going to be better, not worse, than if natural selection (i.e. death, i.e. differential destruction) were operating. Again: obvious.
Well, no - not only is it not obvious, it's wrong! It's wrong several times. First of all, the "probabilities of reaching a particular function" are irrelevant. What we are talking about are the probabilities of reaching any of an unknown number of functions that confer greater reproductive success. So if all your organisms are always equally fecund, than no such function can exist. Secondly, even if we have infinite resources, and no death, but do have differential reproduction (some traits confer greater reproductive success than others) then the chances of one of that unknown number of reproductively advantageous structures evolving is considerably reduced than it would be if eventually some of the population died of old age and/or stopped reproducing, because the probability of gene exchange between individuals who already possess a fecund trait is less than if there is some "culling" (this is particularly true of course of sexually reproducing species). Thirdly, nonetheless, even in your scenario, as time goes by, a greater and greater proportion of the ever-expanding immortal population will be bearers of the fecund trait, constantly multiplying the number of opportunities for a further variant to build on that trait. So the probability of a complex structure that confers even greater fecundity is constantly increased. At least that seems "obvious" to me :) Elizabeth Liddle
Differential reproduction equates to differential destruction. Destruction is not construction. The point is this: if one were to calculate probabilities of reaching a particular function via random variation based on no constraints of competition and no natural selection, where organisms are free to breed and reproduce their heart's content, with no culling of the herd, those probabilities are going to be better, not worse, than if natural selection (i.e. death, i.e. differential destruction) were operating. Again: obvious. Matteo
Well, it isn't "blindingly obvious" because it's a straw man :) Firstly, "natural selection isn't "all death" - it's differential reproduction. Secondly, the result of differential reproduction is that lineages that reproduce more end up with more individuals bearing the pro-reproductive traits and therefore more opportunities for a new trait that enhances that trait to appear. Cheers Lizzie Elizabeth Liddle
Meleagar (at comment 6), You have nailed it precisely. The materialists would have us believe that natural selection is a probability enhancer, when, in fact (since all it is is death), it is a probability reducer. After all, an ensemble of monkeys isn't going to reach Hamlet faster if you continually machine-gun some substantial fraction of them. It should be blindingly obvious, but there are many who do not want to see it. Matteo
I hope Professor Hannum was a better statistician than that. All sequences of the roulette wheel are equally improbable. I expect the reasoning was Bayesian.
Not necessarily, he might have calculated 1-Pr(nobody getting a big sequence) = 1- Pr(1 person not getting a big sequence)^n (to a good approximation). Heinrich