|
Comments on the September 2006 issue of Awake!
Does it seem logical to you, then, for highly trained researchers who crudely mimic systems in nature to solve difficult engineering problems to attribute the genius of devising the original idea to unintelligent evolution? If the copy requires an intelligent designer, what about the original? ("What Does Nature Teach?" Awake! September 2006, p. 6)
The claim that complex arrangements in nature (whose construction may suggest subtle engineering principles) must have an intelligent designer skirts the issue. Can evolution account for such complexity? That is the question.
You may be surprised to learn that "unintelligent" evolution, guided by natural selection, can produce wondrous arrangements of the most subtle complexity! The principle of natural selection, at the heart of it, is a universal principle that goes far beyond biology. Therefore, it can be tested in realms totally unrelated to biology, such as inside a computer. (Note that such a test is not an actual test of biological evolution itself but a test of the principle of natural selection, the main engine that drives evolution. If the principle works, then biological evolution should work as well.) The universal principle of natural selection requires several ingredients in order to work, and we shall name them after their equivalents in biology.
In demonstrating natural selection on the computer, the first thing we need is a "creature." In the computer, that might be nothing more that a collection of lines joined together in certain ways. Lines might be longer or shorter and they might join at different angles or in different ways. The rules governing the length, angle and other basic properties of the lines, and how they connect, are the "genes" of our "creature." We must also mate two of our "creatures" in order to get the next "generation." That's not too difficult. We use a randomizer to select half of the "genes" from each parent to produce an "offspring." We make several "offspring" that way--the next "generation" as it were. Each "offspring" will look a lot like its "parents," but each one is a little different as they are in real life. Variability is a key requirement for natural selection. Natural selection also needs mutations, new genes being the ultimate source of continuing change, so we will randomly alter one of the "genes" in each off our "offspring." As in real life, most such changes will be harmful to our "creature."
Finally, we need to simulate an "environment." A lot more "offspring" will be produced than will survive, as in real life. A desert environment, for example, will favor some traits (deeper roots, better water storage, less water loss) while working against other traits. So, we have to supply some rules that will in some way assign survival values to our "creatures." The choice is pretty much arbitrary. Hence, the fact that the computer programmer makes them up is perfectly okay. After all, it's the universal principle of natural selection that is being tested, not the choice of survival traits. The universal principle of natural selection simply requires that there be some kind of filter to eliminate the "unfit." One choice is as good as another. We might use some kind of point system; if an "offspring" inherits "genes" that total less than a certain value it "dies" before having "offspring" of its own. We then "mate" the survivors and repeat the whole cycle thousands of times or more. As a practical matter, we could select the top few survivors with the best point system scores as the survivors of their "generation."
Can the universal principle of natural selection do anything with these basic ingredients? Will we eventually "evolve" some elaborate "creatures" or will the whole thing just go nowhere?
Richard Dawkins, in his book "The Blind Watchmaker" talks about a test very similar to the above. He began with a "creature" whose several lines constituted a simple tree. His own words best describe the results he got.
When I wrote the program, I never thought that it would evolve anything more than a variety of tree-like shapes. I had hoped for weeping willows, cedars of Lebanon, Lombardy poplars, seaweeds, perhaps deer antlers. Nothing in my biologist's intuition, nothing in my 20 years' experience of programming computers, and nothing in my wildest dreams, prepared me for what actually emerged on the screen. (Richard Dawkins, "The Blind Watchmaker" p.59)
Dawkins expected tree-like forms for his "creatures," because he started with a simple tree-like "creature" and had supplied a set of "genes" that only made basic, limited modifications such as line length, line angles, etc. Thus, he thought that various types of "trees" would evolve on his computer. Even though Dawkins is an evolutionist, he had totally underestimated the power of the universal principle of natural selection. He got complex "creatures" as well that in no way resembled trees. Some resembled insects, and further experimentation by way of changing the "environment" led to all kinds of amazing "creatures." By playing around with different "environments," Dawkins had demonstrated the vast potential in this simple system of line-creatures governed by a few, basic "genes."
If the universal principle of natural selection could perform such wonders with so crude a system, what might it do in other settings?
New Scientist (15 November 1997) reported a remarkable test in an electronic setting. Adrian Thompson (who worked with Phil Husbands at the Centre for Computational Neuroscience and Robotics at the University of Sussex) had managed to set up an experiment whereby a computer could actually manipulate a special set of electronic parts, combining them into circuits. The whole idea was to see if the universal principle of natural selection could be applied directly to hardware to evolve useful circuits.
The goal was to build--by evolution--a circuit that would distinguish between the words "Go" and "Stop," which a human would speak into a microphone connected to the circuit. A limited collection of electronic parts would be available to the computer along with elementary rules describing how they can be connected. Thus, a simple, practical test for the universal principle of natural selection was set up in a unique medium.
Dawkins' "creature" now corresponded to a particular electronic circuit. The "environment" now had the effect of screening out those circuits that did more poorly in distinguishing between "Go" and "Stop." A way was worked out to "mate" the electronic circuit "creatures" to get future generations of them, and random mutations were introduced. In short, the necessary ingredients for natural selection were set up. How would natural selection perform?
The results were astounding and earned Thompson a Ph.D. The result of all this was a circuit whose engineering even Thompson doesn't understand! And, he can't ask the engineer, because the engineer is evolution. That circuit did the required job using less than 1/10 of the components that an electrical engineer would have used! Moreover, the design is so subtle that no one really knows how it works! Apparently, it exploits properties of the individual components that would be overlooked in traditional engineering. The universal principle of natural selection clearly worked exceedingly well, and it did the job in only a few thousand generations!
At Napier University in Edinburgh, Peter Thomson and Julian Miller picked up on Thompson's idea to evolve digital circuits. Their hope was that evolution would teach them new design tricks. Thus, we have the amazing answer to Awake!'s question above. Trained researchers can learn subtle tricks by studying the design of something that has no designer.
The universal principle of natural selection does work, and a whole field of study called "genetic algorithms" has developed to tap its potential. Way back in 1998, Dave Thomas (New Mexicans for Science and Reason) noted some of the accomplishments of genetic algorithms.
The July 27, 1998 issue of U.S. News & World Report has an article by C. W. Pettit entitled "Touched by nature: Putting evolution to work on the assembly line." The article discusses several very successful applications of "genetic algorithms," including new turbine and wing designs (Boeing), trusses for space platforms (Matra Marconi Space), production scheduling (Deer & Co.), development of new antibiotics (Maxygen Inc.), and others. (Dave Thomas, Internet)
Boeing was attempting to design an airplane to carry 600 passengers, but with the wingspan of current jumbo jets. That difficult engineering problem was turned over to genetic algorithms which yielded an unexpected and elegant solution, thus being another success for natural selection!
There is no question that the universal principle of natural selection works extremely well in all kinds of environments (abstract lines, electronics, engineering, medicine, complex scheduling, etc.), so why shouldn't it work just as well for evolutionary biology? With billions of years to work with, and environments that are often stable for millions of years, is it any wonder that marvelous design has arisen in nature?
Quite apart from the fact that natural selection could hardly fail in a biological setting, since all its requirements are met, we have the fossil record, comparative anatomy, the comparative study of macro-molecules and other lines of overwhelming evidence that biological evolution did, indeed, occur.
PROFESSOR BEHE: Many scientists disagree with my conclusions because they see that the idea of intelligent design has extra scientific implications--that it seems to point strongly beyond nature. That conclusion makes many people nervous. However, I was always taught that science is supposed to follow the evidence wherever it leads. (Awake! September 2006, pp.11-12)
What evidence might that be? Behe wrote a popular book (Darwin's Black Box) but he has, to my knowledge, never submitted his ideas to a scientific, peer-reviewed journal. That's where serious scientific ideas are presented to the scientific community to be tested. Since his book was written 10 years ago, intelligent design has been fruitless in suggesting new ideas or areas of research. Is there any science in it at all? Behe is also an odd person to quote, since he apparently thinks that evolution above the cellular level is okay!
Behe presents us with his idea of an "irreducibly complex" system. Take away any part and it fails. How, then, could evolution get there unless it evolved all the parts at the same time? And, that, indeed, would be highly improbable. How would evolution even select for these parts as each part fitted into a system that wasn't even there.
Behe's idea of irreducible complexity has at least two fatal flaws in it. One, components to an "irreducibly complex" system often evolve independently for totally different reasons. Hence, the evolution of the various components in an "irreducibly complex" system does not have to occur at the same time. It's infinitely easier if evolution merely has to cobble together parts that are already in existence.
A second fatal problem with "irreducibly complexity" is that parts often evolve first in a system that does less than the system in question. An eye, for example, doesn't have to have a lens to be useful under some conditions. The nautilus does just fine without one. Thus, some part such as a lens may evolve at a later stage yielding a more capable system. In that manner a complex system may be arrived at in slow, reasonable steps.
For the above reasons, evolutionists don't lose any sleep over Behe's claims, at least in a scientific sense. They do object to shoddy reasoning being passed off as disproof of evolution, even if it only applies at the cellular level.
Is the evidence for macroevolution so strong that it should be considered a fact? (Awake! September 2006, p.13)
Why do many prominent evolutionists insist that macroevolution is a fact? After criticizing some of Richard Dawkins' reasoning, influential evolutionist Richard Lewontin wrote that many scientists are willing to accept scientific claims that are against common sense "because we have a prior commitment, a commitment to materialism." (Awake! September 2006, p.17)
Why do many prominent evolutionists insist that macroevolution is a fact? I suspect it is because they have seen and/or worked with examples of the following evidence. There is a lot more evidence out there than most people realize. We have seen the power of natural selection above, but that is a drop in the bucket. Let me simply list the basic classes of evidence that support evolution. We're not even talking about individual examples; we are talking about whole assemblages of facts! The following list was presented on the internet by Dr. Douglas Theobald (copyright 1999-2006). If you wish to explore them in greater detail, check out his site.
I A unique, historical phylogenetic tree
1. Unity of life
2. Nested hierarchies
3. Convergence of independent phylogenies
Statistics of incongruent phylogenies
4. Transitional forms (examples)
Reptile-birds
Reptile-mammals
Ape-humans
Legged whales
Legged seacows
5. Chronology of common ancestors
II Past history
1. Anatomical vestiges
2. Atavisms
Whales with hind limbs
Human tails
3. Molecular vestiges
4. Ontogeny and developmental biology (examples)
Mammalian ear bones, reptilian jaws
Pharyngeal pouches, branchial arches
Snake embryos with legs
Embryonic human tail
Marsupial eggshell and caruncle
5. Present biogeography
6. Past biogeography (examples)
Marsupials
Horses
Apes and humans
III Evolutionary opportunism
1. Anatomical parahomology
2. Molecular parahomology
3. Anatomical convergence
4. Molecular convergence
5. Anatomical suboptimal function
6. Molecular suboptimal function
IV Molecular evidence
1. Protein functional redundancy
2. DNA functional redundancy
3. Transposons
4. Redundant Pseudogenes
5. Endogenous retroviruses
V Change
1. Genetic
2. Morphological
3. Functional
4. The strange past
5. Stages of speciation
6. Speciation events
7. Morphological rates
8. Genetic rates
This is why scientists call universal common descent the "fact of evolution." (Theobald, Closing remarks)
"Mutations cannot transform an original species [of plant or animal] into an entirely new one. …" (Lönnig, as quoted in Awake! September 2006, p.15)
It is natural selection, and possibly other mechanisms, that lead to new species. Mutations merely supply new genes for the genetic pool. By themselves, they don't lead to new species. They are a necessary but not a sufficient condition for macroevolution.
Thus, the law of recurrent variation implies that genetically properly defined species have real boundaries that cannot be abolished or transgressed by accidental mutations." (Lönnig, as quoted in Awake! September 2006, p.15)
It seems that the "law of recurrent variation" is not a genuine law of science at all! It traces back to Wolf-Ekkehard Lönnig, a geneticist at the prestigious Max Planck Institute for Plant Breeding in Germany. It appears that Lönnig was censored for using the official Max Planck Institute website to advance his personal belief in Intelligent Design. Given his Internet "crusade" for ID, it is not surprising that he authored a paper around 2002 claiming that natural selection cannot give rise to new species. Far from being supported by a scientific law, that view is rejected by the vast majority of biologists.
On the Internet (Google) I found something like 8 distinct references to "law of recurrent variation," perhaps as many as 41 if duplication is counted. Those 8 were from creationist sites quoting Lönnig. Compare that to the "law of reflection," a real law of nature, one that has about 35,000 references on Google. Awake! has not done its homework.
As previously noted, the evidence from research strongly indicates that mutations cannot produce entirely new kinds of plants or animals. (Awake! September 2006, p.15)
Any production of entirely new kinds of plants and animals would begin with new species, species being the only objective taxa in nature. All others are arbitrary in some respects, having fuzzy boundaries. Polyploidy in plants, where a genetic error causes the doubling of genes, creates an instant species as polyploid offspring can no longer breed with the parent species. Such offspring are often more robust than their parents.
Have new species arisen in modern times? To answer that question we must start with the definition of a biological species. A biological species is often defined by twentieth-century biologists as groups of actually or potentially interbreeding populations that are reproductively isolated from other such groups (Tobin & Dusheck, p.425, "Asking About Life - Second Edition" [a college textbook]). The answer is a resounding "Yes!"
In 1964 a few polychaete worms from the species Nereis acuminate, were captured near Long Beach, California and bred in captivity. After the captive population had grown to several thousand, four individuals were sent to Woods Hole Oceanographic Institute in Massachusetts. These worms were grown and used as guinea pigs in experiments for 20 years. (Glenn R. Morton, "Foundation, Fall and Flood" p.86)
In those 20 years the Woods Whole population had evolved into a new species; it could no longer interbreed with the Long Beach population!
In Hawaii, there are several species of moths in the genus Hedylepta. Some species feed on bananas while others feed on various other plants including the palm. Similarities in form show that the palm-feeding variety is the parent species to the banana-feeding moth. The reason it can be so categorically stated that the banana-feeders are the descendants of the palm-feeders is that, until a thousand years ago, when the original Hawaiians settled the island, there were no banana plants on Hawaii. Thus, within the past 1,000 years the moth has changed into two separate species, probably in a manner similar to that being taken by the fruit fly, Rhagoletis pomella.
(Glenn R. Morton, "Foundation, Fall and Flood" p.87)
Here's another story of evolution. A surprising discovery was made by geologists who drilled into the mud of Lake Victoria in 1995. They had expected to get a record of several hundred thousand years of lake deposition. At only 9 meters down (a mere 14, 500 years ago) they discovered that the deepest parts of that lake were then covered in grass! "It seems that during the Ice Age a cool, arid climate dried up the rivers that fed the lake, and its vast supply of water simply evaporated." (Carl Zimmer, "Evolution: The Triumph of an Idea" p.90) The ancestors of today's cichlids in Lake Victoria must have toughed it out in nearby streams, and when the water returned a single species of cichlid took over the lake. How do we know that?
The cichlids living in Lake Victoria today are all close relatives of each other, and only distantly related to cichlids in other lakes and rivers. In fact: "Their genes show that a single lineage of mouth-brooders came to the lake after it refilled, and then, in the time that it took for humans to build civilization, 500 species were born." (Carl Zimmer, "Evolution: The Triumph of an Idea" p.90)
If not by evolution, where did these 500 closely related species live before the lake refilled? If in nearby streams, why aren't any close relatives found there today? You would think that, out of 500 species, at least a few would still occupy those streams. Indeed, could surrounding streams even hold as many species as now crowd Lake Victoria? How is it that the cichlids of Lake Victoria are so closely related? Evolution is the one answer that makes sense.
A similar story can be told of the Galapagos Islands, Darwin's old stomping ground. Several millions of years ago there was nothing but ocean there, so where did Darwin's marvelous group of finches come from? Obviously, they must have come from the mainland. But the mainland species don't have the marvelous array of adaptations that the finches on the Galapagos Islands have. Darwin's finches are occupying niches normally taken by other birds back on the mainland. The only sensible answer is that, on one or more occasions, a few mainland finches made it to these volcanic islands sometime after they first arose from the sea. Without the usual competition, Darwin's finches evolved into various species that took advantage of the full range of available niches, including niches not available to them on the mainland.
Lönnig's bogus "law of recurrent variation," which implies that properly defined species have real boundaries that cannot be abolished or transgressed by accidental mutations, is refuted by ring species. Herring gulls are an excellent example of a ring species. The range of these gulls form a ring around the North Pole, and within that ring neighboring populations can mate with each other even though they look slightly different. But the birds at the two ends of the ring--the Herring gull and the lesser black-backed gull--are so distinct that they can't mate even though they live side by side. That is to say, they are legitimate species. Thus, we have the spectacle of two species that grade into each other! Obviously, these two species don't have fixed boundaries. (Carl Zimmer, "Evolution: The Triumph of an Idea" pp. 83-84)
Another example of a ring species is the Ensatina salamander of California. Populations run along the coast, from San Diego to the northern border, and then across the top of the state and down the Sierra Nevada. The southern portion connects to a slight gap, once occupied by a population that is now extinct, and that area connects to a population just northeast of San Diego. The central valley is a barrier to these lizards. Thus they inhabit an area that could be described as a tall, slim "n"-shaped range. The two subspecies at the bottom of this "n"-shaped range, the Monterey salamander and the yellow-blotched salamander, do not interbreed. That is, they are different species. Yet, they are linked by a continuum of subspecies that do interbreed. Once again, we see that the supposedly fixed borders between two species has dissolved! (Tobin & Dusheck, "Asking About Life - Third Edition" p.355)
Therefore, we may dismiss the notion that species boundaries are fixed for all time.
Indeed, Darwin's finches are not becoming "anything new." They are still finches. (Awake! September 2006, p.16)
If one looks at 20 years worth of evolution one will see 20 years worth of change, which is not much. How can one know if those finches are not going anywhere? Are they supposed to turn into snakes? Those few finches that managed to reach the Galapagos Islands from the mainland a few million years ago have evolved (dramatically) into numerous species exploiting the various niches. Tortoises occupying the various islands of the Galapagos, being fairly isolated from each other, have also developed distinct varieties (if not actual species). By looking at their shells, an experienced observer can tell which island they came from.
Peter and Rosemary Grant's study of finches showed natural selection in action: Drought favored larger beaks on Darwin's finches. The return of normal weather once again favored smaller beaks. That, also, is natural selection in action. The path of evolution is rarely a straight line. It is bushy and sometimes even runs backwards in places. But natural selection does work, and that's the point. And, there are many variables that affect natural selection, not just the amount of rain. Each niche--and some are rather permanent--acts in favor of those finches that can best take advantage of it. Natural selection tends to mold new species around niches. Hence, the conclusion that evolution is going nowhere in Darwin's finches is short-sighted. Even if the rains conveniently waxed and waned every few years--for millions of years on end--plenty of other factors would have steadily driven the evolution of Darwin's finches.
In 2004, National Geographic described the fossil record as being like "a film of evolution from which 999 of every 1,000 frames have been lost on the cutting-room floor." Do the remaining one-in-a-thousand "frames" really document the process of macroevolution? (Awake! September 2006, p.16)
Absolutely! Any major evolutionary change could easily involve millions of generations. If we only had one out of every 1,000 "frames" by random draw, we would have such a detailed sketch of evolution that paleontologists would shake their heads in amazement. Consider this: Of the millions of "frames" for some evolving creature, say the horse, we only need the key connections. A handful of fossils from each of the major stages will do nicely in showing the major pathways of evolution.
Niles Eldredge, a staunch evolutionist, admits that the record shows that for long periods of time, "little or no evolutionary change accumulates in most species." (Awake! September 2006, p.16)
Eldredge is talking about "punctuated equilibrium." Punctuated equilibrium predicts that many (not all!) species will be fairly stable for long periods of time and then, perhaps due to a dramatic change in environment, undergo rapid evolution. Darwin's finches might even be an example of this. Quite likely the parent finch species on the mainland were stable for a long period of time. After reaching the Galapagos Islands, they went wild in an explosion of evolution. Someone looking at the fossil record millions of years from now, when the Galapagos Islands might not even exist, would see a gap in the fossil record after which a whole bunch of finch species suddenly emerged. Such gaps are statistical, not absolute, and they are usually found at the species or genus level. (The higher taxa often merge seamlessly, so as to require an arbitrary cutting point in shades of gray.) That is to say, we have clear examples of evolution lacking gaps, even at the species or genus level. Even one exception does not bode well for the idea of created kinds.
Punctuated equilibrium does well in explaining the fossil record of higher animals, but is much weaker for invertebrates. A continuous evolution often applies to those creatures. That is, punctuated equilibrium may correctly explain some aspects of the fossil recorded while traditional evolution correctly explains other aspects of it. Accepting one does not entail the rejection of the other.
According to the fossil record, all the major groups of animals appeared suddenly and remained virtually unchanged. (Awake! September 2006, p.17)
This is so misleading! Awake!'s readers must be thinking that all the basic kinds were there at the start, that evolution only played a very minor role in the form of microevolution. Hopefully, Awake! is referring to the so-called Cambrian explosion, which actually took place over a period of some 10 million years. (That's longer than the time it took for man to evolve from the apes!) The Cambrian "explosion" produced almost all of the major groups of life at the phylum level--the level of basic body plans. There are one or two exceptions, meaning that all phyla were not created at that time. Moreover, as the fossil record now makes clear, life had its roots in the Precambrian. The Cambrian represents some kind of turning point. Perhaps life had become armored at that time, thus suddenly leaving easily preserved fossils. Perhaps oxygen rose above some critical level. Whatever it was, life finally took off, but life--even multicellular creatures--were around before the Cambrian. Simple life was around billions of years before the Cambrian.
The claim that little change occurred since then is patently ridiculous. Consider the creatures that had yet to make the scene: Trees, grass and all other land plants, bony fish, the more primitive types of fish, whales, cattle, horses, mastodons, dinosaurs, mice, birds, snails, clams, insects and any other land animal, and anything in the water with a backbone. Obviously, a whole lot of change has occurred since that time. The Cambrian "explosion" was hardly the beginning of life as we know it today, nor was it the beginning of life at all. It can't be passed off as the Genesis creation account, when God supposedly created all life.
Lewontin writes, "we cannot allow a Divine Foot in the door."
In this regard, sociologist Rodney Stark is quoted in Scientific American as saying: "There's been 200 years of marketing that if you want to be a scientific person you've got to keep your mind free of the fetters of religion." He further notes that in research universities "the religious people keep their mouths shut," while "irreligious people discriminate." (Awake! September 2006, p.17)
Stark's complaint sounds to me like a case of sour grapes! Good scientists, and that includes many religious people, know better than to mix science and religion. God is kept out of the laboratory for good reasons. Science seeks physical explanations of the universe, the only kind that can be tested, and testability is the feature that has made science successful and useful. By analogy, an auto mechanic also seeks physical explanations, the only kind that allow reliable repair. If your mechanic told you that the problem with your car was a certain demon living in the alternator, how long would it take you to find a new mechanic? The fact that mechanics keep their religion and work separate doesn't make them atheistic. The same is true for scientists. Some people just can't seem to understand this!
A supernatural "explanation" cannot be tested. (How would you test the idea that God created the sun? Maybe it was the devil, or Zeus. Maybe it was a committee of gods long dead. Maybe it was entirely natural.) Moreover, a supernatural "explanation" does not connect with known knowledge, a fabric woven together by natural principles and history, the latter relying on natural principles for verification. Indeed, a supernatural event does not even connect to other supernatural events by any rules we can identify. A supernatural "explanation" does not explain things in the sense of moving from the unknown to the known. Instead, it gives us a story that cannot be checked out, and any story will do. There is no way to decide between conflicting supernatural accounts. Finally, a supernatural "explanation" cuts off further investigation by arbitrary decree. No scientific progress is possible.
Even if a genuine, supernatural event were somehow detected and verified, it could only be recognized as an isolated fact; it could not be a part of science since it could not be connected to anything else. Indeed, once we have abandoned natural principle, including cause-and-effect, there is no reason to attribute such an event to anything at all; it could just happen all by itself. In that light, I hope that you begin to understand why religion and science are not mixed. It makes for poor religion and horrible science. Logic dictates that we exhaust the common explanations before turning to the exotic ones. Only if methodological materialism (which must not be confused with atheism) is exhausted are we justified in recognizing that a new class of facts exist in the world of science that cannot be explained scientifically.
Dave Matson
Pasadena, California
September 24, 2006
|
|
