Response to Comments: Natural Selection Doesn’t Help, Gradualism is Out, and so is Evolution

When Charles Darwin presented his theory of evolution one of the main objections was that he had no credible explanation for how biology, with its many designs and intricacies, could have arisen on its own. Darwin’s main argument, for which he presented many powerful evidences, was that biology did not appear to be designed. From its different patterns to its inefficiencies, the design perspective seemed to be badly failing. But this leaves us with evolution in name only. What were the details? How did the world of biology arise on its own? Inefficient or not, biology nonetheless was not trivial. How could it have evolved?

A never ending dialectic?

In the ensuing century and a half evolutionary explanations did not improve very much. The problem in 1859, that evolution did not seem to be capable of producing the biological world, has only grown worse. In fact, while much has been learned since Darwin, the data inevitably are recruited to service the same old arguments.

For skeptics, the new findings simply reinforce biology’s incredible complexity. But for evolutionists, the new data, like the old data, is often irrational, disproving design and so mandating evolution

It’s Cleanthes versus Philo all over again. For every Paley touting complexity there is a Hume ridiculing the backwardness or inefficiency of the design. The design argument is a great challenge, but it is neutralized by the evil in the world. “I needed all my skeptical and metaphysical subtlety to elude your grasp,” admitted Philo, but “Here I triumph.”

Will this dialectic never end? Will all of this thesis versus anti thesis never lead to a synthesis? Like Sisyphus forever pushing the stone up the hill only to have it roll back down, are we doomed to a never ending cycle of complexity versus dysteleology? Cleanthes versus Philo? Paley versus Hume? Perhaps not.

New molecular evidence

One problem with evolution is that since it deals with low probability events taking place over eons of time in the distant past via a long list of candidate mechanisms, it is difficult to test. Darwin tried to put his theory up for a test, but it was nearly impossible. He wrote:

If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find out no such case.

Darwin may sound generous here, allowing that his theory would “absolutely break down,” but his requirement for such a failure was close to impossible. For how could one can show that an organ “could not possibly” have been formed in such a way?

It appeared that Darwin had constructed a theory sufficiently malleable to avoid falsification. This has been highlighted in recent years by the use of the multiverse to explain low probability events. If you have a near infinity of universes from which to draw you can explain just about anything. Astronomically unlikely events suddenly become probable.

But the story does not end here, in futility. In fact an entirely new type of evidence has arisen since Darwin’s day that is much more amenable to detailed, quantitative analysis which cannot be swept under the rug of deep time. It is a result of the twentieth century’s revolution in molecular biology, and it already has provided ample demonstrations of the sort Darwin discussed. It may not be too bold to think that an Hegelian synthesis is nigh.

One good example of a complex “organ” which could not possibly have been formed by numerous, successive, slight modifications is the typical protein. Short of a multiverse, evolution simply does not explain their origin. I have discussed this here, here, here and here. Here are the salient points:

1. DNA and protein sequences provide plenty of violations of evolution’s expected pattern. Importantly, these violations are not “in the noise.” Beyond vague speculation evolutionists cannot explain these as the vagaries of their random process.

2. Although proteins were once thought to be flexible in their design, and so perhaps not so difficult to evolve, we now understand protein design better. Their designs are not very flexible. They cannot tolerate many mutations, and this means evolving a gene from scratch is not feasible, as it was once thought.

3. Given this lack of flexibility, it is not surprising that attempts to evolve real proteins from scratch have failed. Literally millions of millions of attempts are required to stumble upon even the weakest and simplest of partial functions. But even this many evolutionary experiments does not provide natural selection with a stepping stone, for the partial, weak, function provides no useful fitness improvement. Evolutionists who tried to evolve a more realistic, useful protein (but even in that case it was only part of a protein) concluded it is not possible via gradual change. They appealed to non gradualistic mechanisms, such as homologous recombination.

So to summarize, in addition to proteins violating evolution’s expected pattern in important ways, proteins cannot be evolved via gradualistic mechanisms. Do you remember your biology teacher’s lecture on gradualism and how crucial it was to evolutionary theory? Well forget it. It is out. Gradualism doesn’t evolve proteins.

But that is precisely what Darwin set forth as a crucial test of evolution. Remember, “If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down.”

Well forget it. Gradualism is out, and with it Darwin’s falsification criterion has been met. Evolution has been falsified.

But when I explain these things evolutionists cry foul. One professor wrote:

Nonsense. You assume that evolution must search randomly through (half of) 10^100 states to find a specific protein. That is plain wrong. Evolution is not a random search. It is plainly misleading to say so.

When presented with their own claims evolutionists consistently reject them. Evolutionists are their own judge. Yes, evolution must indeed search randomly. There is no such thing as selection “pressure.” That is a euphemism evolutionists use to refer to scenarios where selection could potentially work. But the design change needed must arise without selection’s guidance. No hints allowed.

When critics are not around evolutionists are busy telling the world that evolution is unguided. There is no logic, no design, no teleology in evolution. It does not have a goal in mind as it meanders mindlessly.

Evolution must indeed “search randomly” before selection can work. This is the problem that was intuitively understood in 1859, and today the quantitative example of proteins illustrates why. The professor continues:

It has been shown experimentally [1] that fitness can be increased substantially by local moves alone. We have discussed that here. In that particular instance, a randomly scrambled system reached 52% of its original fitness through single substitutions. Nonlocal moves are required to move through the rugged landscape at the top, but evolutional variations do include nonlocal moves. Frame shifts and duplications to name a few.

[1] Y. Hayashi et al., "Experimental Rugged Fitness Landscape in Protein Sequence Space," PLoS ONE 1, e96 (2006); doi:10.1371/journal.pone.0000096.

Once again the evolutionary version of the evidence is riddled with misrepresentations. In this experiment, to which the professor refers, only a fraction of a protein machine was randomized and then mutations in the randomized segment were evaluated for their improvement to the overall machine performance. Not surprisingly some mutations helped. But the improvements were minor, not “substantial.” The “improved” machine was still far less functional than the native version. The difference in productivity in the experiments was enormous.

But to make the results appear more hopeful evolutionists divide the logarithms of the productiveness of the native and evolved machines. Imagine a salesman who sells only 100 cars a year while his co-worker sells 1000 cars. Would his boss buy his argument that the difference is really just 50%? In fact the difference is 900.

In fact even the evolutionists who did the research admitted that gradualistic evolution cannot do the job. They estimated that 10^70 [a one followed by 70 zeros] evolutionary experiments would be required to break the logjam, and even then the machine nonetheless would not be fully function.

And of course this was when only a fraction of the machine was randomized. The remainder of the machine was assumed to be fully evolved and assembled. And a functional biological environment was assumed to be in place in which the protein machine could be tested and evolved.

Even given all these advantages gradualistic evolution failed. 10^70 evolutionary experiments is completely unrealistic, so the evolutionists appealed to non gradualistic mechanisms, such as homologous recombination. But homologous recombination isn’t available. It comes after proteins, not before. And furthermore, even if some magical non gradualistic mechanism just happened conveniently to be available, it wouldn’t help anyway. All that does is start the experiment over again. Remember, evolution has no foresight.

Imagine a toddler trying to climb Mt. Everest. He walks a few hundred feet and realizes he will never make it to the top. His solution? Start over again. This is what the evolutionists are telling themselves.

As if sensing a problem, another professor came to the rescue. He wrote:

Probability calculations such as this depend on the model being proposed. Dr Hunter’s calculation is based on the arbitrary assumption that the entirety of protein sequence space must be randomly sampled to yield functional proteins.

No, I’m assuming no such thing. As I have explained several times, evolution need not search the entire protein sequence space, not even close. But it nonetheless must explore an astronomically large space, well beyond its means.

The professor continues:

But the incorporation of some limiting factors into the model yields a different outcome:

“We suggest that the vastness of protein sequence space is actually completely explorable during the populating of the Earth by life by considering upper and lower limits for the number of organisms, genome size, mutation rate and the number of functionally distinct classes of amino acids. We conclude that rather than life having explored only an infinitesimally small part of sequence space in the last 4 Gyr, it is instead quite plausible for all of functional protein sequence space to have been explored and that furthermore, at the molecular level, there is no role for contingency.”

How much of protein sequence space has been explored by life on Earth?
David T.F Dryden, Andrew R Thomson and John H White

http://rsif.royalsocietypublishing.org/content/5/25/953.long

This paper the professor refers to is, unfortunately, another example of how evolution has damaged science. It presents several obviously flawed arguments that have no place in a scientific journal. But the reviewers of the paper are evolutionists and approved of the pseudo science.

The paper attempts to make two general points. First that evolution can succeed with a much smaller protein sequence space and second, that evolution can easily search the entire protein sequence space. Both conclusions are scientifically ridiculous and are inconsistent with what we do understand about proteins.

But why should that matter, for the paper is written from an evolutionary perspective. In other words, evolution is assumed to be true, and so proteins must have evolved somehow. Let’s have a look at the arguments.

Evolution can succeed with a much smaller protein sequence space

For the first claim, the evolutionists argue for a smaller protein sequence space because:

A. “the actual identity of most of the amino acids in a protein is irrelevant” and so we can assume there were only a few amino acids in the evolution of proteins, rather than today’s 20.

B. Only the surface residues of a protein are important.

C. Proteins need not be very long. Instead of hundreds of residues, evolution could have used about 50 for most proteins.

For Point A, the evolutionists use as support a series of simplistic studies that replaced the actual protein three-dimensional structure and amino acid chemistries with cartoon, two-dimensional lattice versions.

The evolutionists next used the fact that a type of protein can be found in different species with very different amino acid sequences. So doesn’t that mean the amino acid identities don’t matter very much? No, this is yet more evolutionary blowback. Evolutionists conclude this because they believe the proteins all evolved from a common ancestor. But science tells us that proteins, generally, do not tolerate amino acid substitutions very well.

As I have explained before, the amino acid sequence is not merely “along for the ride.” Because the evolutionists believe the proteins evolved, they mistakenly confuse sequence variation (which is observed) with sequence irrelevance (which is not generally observed).

Likewise Point B is at odds with science, and again is an unwarranted extrapolation on a simplistic lattice study.

For Point C, the evolutionists note that many proteins are modular and consist of self-contained domains “of as few as approximately 50 amino acids.” But the vast majority of protein domains are far longer than 50 residues. Single domain proteins, and domains in multiple-domain proteins are typically in the hundreds of residues.

While it is true that one can envision a smaller amino acid alphabet, and that shorter proteins can sometimes work, the evolutionists stretch the science far beyond its reasonable extrapolation point.

Evolution can easily search the entire protein sequence space

To defend their second claim, that evolution can easily search the entire protein sequence space, the evolutionists present upper and lower bound estimates of the number of different sequences evolution can explore.

Their upper bound estimate of 10^43 (a one followed by 43 zeros) is ridiculous. It assumes a four billion year time frame with 10^30 bacteria constantly testing out new proteins. First, even for an upper bound estimate their time frame is about two to three orders of magnitude too large. And furthermore, from where did these bacteria come? Bacteria need thousands of, yes, proteins. You can’t use bacteria to explain how proteins first evolved when the bacteria themselves require an army of proteins.

The lower bound of 10^21 is hardly any more realistic. The evolutionists continue to use the four billion year time frame. And they also continue to rely on the pre existence of an earth filled with a billion species of bacteria (with their many thousands of pre existing proteins).

So with these two claims in hand, the evolutionists conclude that the evolution of new proteins is no big deal. “We hope,” they explain, “that our calculation will also rule out any possible use of this big numbers ‘game’ to provide justification for postulating divine intervention.”

This shifting of attention to “divine intervention” does not remedy their several scientific errors. The scientific fact is that the numbers are big. This isn’t a “game.”

For instance, consider an example protein of 300 residues (many proteins are much longer than this). With 20 different amino acids to choose from, there are a total of 10^390 different amino acid sequences possible. Now let’s simplify and assume only four different amino acids are needed. This reduces the problem to 10^180 different sequences.

Next let’s assume that only 50% of the residues are important. At the other 50%, any amino acid will do. That is, fully half of the amino acid sequence is inconsequential. These are extremely aggressive and unrealistic assumptions, yet nonetheless we are left with a total of 10^90 sequences. 90 may not appear to be a big number, but a one followed by 90 zeros is. It is completely impractical for evolution.

And if you don’t agree with my example, then we have the evolutionary experiments, described above, which concluded that 10^70 tries would be required. And even that was only for a fraction of the protein machine, and it assumed a pre existing biological world with its many proteins already in place.

So let’s take the evolutionist’s own numbers at face value, giving them every advantage. The number of experiments required is 10^70 and the number of experiment possible is 10^43. Even here, giving the evolutionists every advantage, evolution falls short by 27 orders of magnitude.

The theory, even by the evolutionist’s own reckoning, is unworkable. Gradualistic evolution—the test that Darwin himself set forth—or non gradualistic evolution, it does not matter. Evolution fails by a degree that is incomparable in science. Scientific theories often go wrong, but not by 27 orders of magnitude. And that is conservative.

The ghost of Hume

But it all comes back to the Humean rebuttal. Those big numbers are impressive, but “Here I triumph.” Evolutionists will never adhere to the science, because this never was about science in the first place. From the Epicureans to the evolutionists, it is about the failure of design. We must not have divine intervention for this gritty world. As Lucretius put it:

That in no wise the nature of all things
For us was fashioned by a power divine-
So great the faults it stands encumbered with.

Religion drives science, and it matters.