In Defense of Evolution
Dr. Kenneth Miller is as familiar as anyone in the scientific community with the intelligent-design movement and its attempts to undermine the theory of evolution. A professor of biology at Brown University and coauthor (with Joe Levine) of the standard high-school textbook Biology, Miller testified at the Dover trial as an expert witness for the plaintiffs, the Dover parents who brought suit against their town's school board. Here, Miller, who stresses that he is also a man of faith, talks about why evolution matters, what flaws he sees in the intelligent-design argument, and why the Dover decision hardly means the end of the controversy.
Faith and reason Q: Why is evolution so controversial?
Kenneth Miller: I think one of the reasons why evolution is such a contentious issue, quite frankly, is the same reason you can go into a bar and start a fight by saying something about somebody's mother. Evolution concerns who we are and how we got here. And to an awful lot of people, the story of evolution, the story of our continuity with every other living thing on this planet, that's not a story they want to hear. They favor an entirely different story, in which our ancestry is separate, our biology distinct, and the whole notion of our lineage traceable not to other organisms, but to some sort of divine power and divine presence. But it's absolutely true that our ancestry traces itself along the same thread as that of every other living organism. That, for many people, is the unwelcome message, and I think that's why evolution has been, is, and will remain such a controversial idea for many years to come.
Sean Pitman: I agree. All ideas that affect one's view of where one came from and why one is here on this planet are bound to be tied up with a fair degree of emotion - at least for most people. What is interesting is that scientists are not immune from this sort of emotional bias. Evolutionists, just like creationists and those who believe in some form of intelligent design or input into the origins of life, are often quite passionate about their respective positions on origins. Scientists are no more immune from this sort of bias than are philosophers, plumbers, or preachers.
Q: Where do you come from personally on this topic?
Miller: I think that faith and reason are both gifts from God. And if God is real, then faith and reason should complement each other rather than be in conflict. Science is the child of reason. Reason has given us the ability to establish the scientific method to investigate the world around us, and to show that the world and the universe in which we live are far vaster and far more complex, and I think far more wonderful, than anyone could have imagined 1,000 or 2,000 years ago. Does that mean that scientific reason, by taking some of the mystery out of nature, has taken away faith? I don't think so. I think by revealing a world that is infinitely more complex and infinitely more varied and creative than we had ever believed before, in a way it deepens our faith and our appreciation for the author of that nature, the author of that physical universe. And to people of faith, that author is God. Now, I'm a scientist and I have faith in God. But that doesn't make faith a scientific proposition. Faith and reason are both necessary to the religious person for a proper understanding of the world in which we live, and there is ultimately no necessary contradiction between reason and faith. "Whether God exists or not is not a scientific question."
Sean Pitman: I'm most intrigued by Miller's thoughts here. How is Miller's description of "faith" in God any better than wishful thinking or a child's belief in Santa Claus? I may be wrong, but it seems to me that Miller is talking about some sort of fantasy or concept of completely blind "faith" where one believes in this or that hopeful reality based on absolutely nothing but feelings or desire. In my opinion, those like Richard Dawkins are correct in becoming quite exasperated by such thinking and rightfully calling it "The God Delusion". While I personally do believe an intelligent Creator God, I do so because I think there is solid, testable, falsifiable evidence for a God-like higher power that goes far beyond human-level intelligence, power, and creativity. If I did not at least think I recognized such evidence, there is no way I would actually worship a God for which I saw no physical evidence of his/her/its existence or interaction with any aspect of nature.
Q: What's wrong with bringing God into the picture as an explanation?
Miller: Supernatural causes for natural phenomena are always possible. What's different, however, in the scientific view is the acknowledgement that if supernatural causes are there, they are above our capacity to analyze and interpret. Saying that something has a supernatural cause is always possible, but saying that the supernatural can be investigated by science, which always has to work with natural tools and mechanisms, is simply incorrect. So by placing the supernatural as a cause in science, you effectively have what you might call a science-stopper. If you attribute an event to the supernatural, you can by definition investigate it no further. If you close off investigation, you don't look for natural causes. If we had done that 100 years ago in biology, think of what we wouldn't have discovered because we would have said, "Well, the designer did it. End of story. Let's go do something else." It would have been a terrible day for science.
Sean Pitman: I see this argument all the time and am always amazed by how many otherwise intelligent men and women use it and/or are taken in by it. If a God or someone with at least high-level or God-like powers and/or intelligence decided to manipulate nature in any way, Miller and many other scientists actually argue that it would be impossible for humans to recognize any kind of manipulation of nature as being the result of deliberate intent or "artifact". Yet, when it comes to the detection of deliberate human activity, activity that is arguably far less intelligent than what anyone would call "God-like", scientists don't seem to have any problem detecting design.
Entire scientific disciplines are built up around the concept of detecting deliberate activity behind various phenomena in nature - - to include forensic science and anthropology. Of course, these disciplines are built around previous experience with and direct observations of humans in action. Yet, there are scientists who do in fact propose that highly intelligent activity, even superhuman-level intelligence, can be detected without any need for knowledge concerning the actual identity, motive, or method of the intelligent agents. These scientist spend their time searching for signs of intelligence coming from outer space - - as in the search for extraterrestrial intelligence or SETI.
The argument is, of course, that humans and alien intelligences living somewhere in outer space are "natural", not "supernatural", and can therefore be potentially detected by scientific investigation. So, what if someone with God-like intelligence decided to act in a similar way to manipulate nature in a way that would at least simulate what human or alien intelligences could or would do that would be detectable as artifact? Would it then be possible to detect such activity as at least intelligent or artifactual in nature? - - Rather than the result of some as yet unknown non-intelligent natural process?
Q: Does science have limits to what it can tell us?
Miller: If science is competent at anything, it's in investigating the natural and material world around us. What science isn't very good at is answering questions that also matter to us in a big way, such as the meaning, value, and purpose of things. Science is silent on those issues. There are a whole host of philosophical and moral questions that are important to us as human beings for which we have to make up our minds using a method outside of science.
Sean Pitman: It just so happens that I like vanilla ice cream. That's a fact. And, it didn't take any scientific investigation for me to discover this fact. It is not subject to testing or falsification by me or anyone else. It just is. It is an internally derived truth. Is it important to me? Well, I'm kind of glad I know it as a truth. It saves me a lot of time and frustration when I go to pick out an ice cream to buy at the store.
Is a belief in the existence of God as a "truth" kind of like the truth that I like vanilla ice cream? Well, I may really like the concept or idea of a God or God-like being. It may really appeal to me. However, once I start suggesting how this God would act or actually did act or interact with the physical world that exists outside of my own mind, I have moved into the realm of science. Making suggestions or assertions about what God does or did outside of my own mind without at least some physical evidence to back up such assertions is like taking on a form of schizophrenia or deliberate mental delusions that are based on nothing more than mental projections or very strong mental or emotional desires - - which do not necessarily have anything to do with the reality that actually exists outside and independent of ones own mind.
Q: Can science prove or disprove the existence of a creator, of God?
Miller: Whether God exists or not is not a scientific question.
Sean Pitman: Actually, it is, or it at least could be a scientific question. It all depends on if God wishes to act in a way that is detectable as "artifactual" from a human perspective. If a God does actually exists that wishes or has actually acted in such a way, such actions could, theoretically at least, be detectable as "deliberate" and "intelligent" - - just as any alien intelligence could be detectable as such by SETI scientists.
Evolution in a nutshell
Q: What is evolution exactly?
Miller: Well, everyone knows that evolution, in a sense, is change over time.
Sean Pitman: Well, lots of things change over time. Even intelligent design advocates and creationists recognize this fact. The question is: How do living things change over time . . . and to what degree? For example, there is a distinct difference between the changes proposed by Gregor Mendel over time vs. those suggested by Darwin . Darwinian-style change cannot be explained using Mendelian-style change alone. Therefore, the changes over time proposed by Darwin require a different sort of mechanism that goes beyond the mechanism of Mendelian-type change.
But what few people understand is how straightforward the nature of this change is. It's important to understand, first of all, that individuals don't evolve. I'm not evolving into something else, and my dog isn't evolving into something else. I'm going to remain a human being, he's going to remain a dog. That's the way things are going to work. What changes over time are populations of individuals, for very straightforward reasons.
Sean Pitman: Strictly speaking, individuals do "change" over time. Parts of individuals even undergo what could be called Darwinian-style evolution over time - such as the human immune system which is often used as an example of functional evolution in action. Some individuals evolve entirely new proteins or chimeric protein combinations, such as the famous BCL/ABL tyrosine kinase protein seen in people who develop or evolve chronic myelogenous leukemia (CML). However, it is true that such "changes" aren't going to change a dog into a chicken, etc.
Number one, every species shows variation among individual members of that population.
Sean Pitman: This is true.
Number two, individuals in a population show what biologists call differential reproductive success. Some individuals leave more offspring than others. Some people have no children; some people have big families.
Sean Pitman: Also true. Finally, one of the factors that influences differential reproductive success is how well-suited individuals are to the present environment in which they find themselves—how good they are at obtaining food, defending themselves against their enemies, resisting disease, and finding and meeting a member of the opposite sex and raising offspring. All these things matter.
Sean Pitman: Right . . .
What Darwin appreciated is that nature herself selects from variants in the population for those that are best able to succeed in this race for differential reproductive success. Over time, and given a steady input of new variation into the population, that can change the average characteristics of a species, and it can split one species into two.
Sean Pitman: Absolutely. A good example would be horses and donkeys - - two "species" that clearly had a common ancestor, share the same basic "gene pool" and can interbreed to produce viable if not virile offspring (i.e., mules and/or hinnies). However, Mendelian variation can also change the average characteristics of a population over time via the guidance of natural selection. Yet, Mendelian variation cannot create truly novel gene pools with unique functional elements present in the offspring which were not already present in the parent population. Those species, those two groups, can then go on changing in different directions. That's what leads to the formation of yet more new species. Nature herself automatically selects for favorable variations, and this is the driving engine of evolutionary change. That, in a nutshell, is what evolution is.
Sean Pitman: This definition of evolution allows for more types of "change" that just Darwinian-style change over time. The real issue here is over the concept of Darwinian-style evolution where truly novel functional elements are added to gene pools over time. That is the definition of Darwinian-style evolution, in a nutshell.
The question then is: Can Darwinian-style evolution happen, and if so, does it have any evident limitations when it comes to the type or nature or degree of novel functional elements that can be produced?
Evolutionists strongly believe that given enough time and the appropriate environments or environmental changes or variations, the answer to this question is no - - There is no significant limit to the nature or degree of novel functional systems that can be added to any gene pool. It is this notion that creationists and intelligent design theorists wish to challenge. Many, even a number of very well educated scientists, to include several Nobel Laureates, are starting to question this particular claim of mainstream evolutionary theory.
Q: Why is evolution important? How does it affect people in their everyday lives?
Miller: We should care about evolution because it concerns who we are, where we came from, why we are the way we are, and maybe even where we're going.
Sean Pitman: That's true . . . The whole notion that biology is wrapped up in the idea of evolution is extremely important to experimental biologists, because otherwise, to paraphrase another scientist, biology is nothing but stamp collecting. It's an exercise in which you say, "Here's a worm and here's how worms work, and here's this type of cell and here's how this cell works. And here is a plant, and here is how plants work."
If they're all completely unrelated, then biology is not a unified science.
Sean Pitman: That's right . . . Clearly biological organisms are very "related". All living things share a great deal in common with each other. Clearly every living thing has some sort of common origin. This is not in question. What is in question has to do with what common origin is most likely? Is the common origin a common ancestor of all life that gave rise to all the various forms and functions of biosystem complexity that we see today via random mutation and natural selection acting over millions and billions of generations? Or, is the common origin found in a common intelligent Designer/Creator? Or, and I favor this option myself, is there very good evidence that both processes have been in play in the origin and diversity of living things?
But we know from a half century of biochemistry and molecular biology that all these living organisms, no matter how diverse they are, share certain common features, and those common features include the way in which they store and transmit and evolve information, and these common features tie all of life together. They help us to understand our own bodies and our own genomes in the light of the bodies and genomes of other organisms. So what evolution really does is to make sense of biology, and what biology does is to help us make sense of ourselves, our own lives, and the planet on which we live.
Sean Pitman: Darwinian-style evolution does indeed make sense of a number of interesting aspects of biology. However, the Theory of Evolution, as it currently stands, does not recognize certain abundantly obvious limitations. These limitations make it impossible to explain, from the standpoint of the entirely non-deliberate proposed mechanism, of random mutation and function-based selection, the existence of certain kinds of functional systems that exist in every living thing.
For example, while there are many examples of evolution in action producing novel biosystems that have few minimum structural threshold requirements (i.e., less than a few hundred specifically arranged amino acid residues), there are no examples of evolution in action producing a novel functionally beneficial biosystem that has a minimum structural threshold requirement of more than 1,000 specifically arranged residues - - not one example in all of scientific literature.
Now isn't that most interesting? Why might there be such a stalling-out effect illustrated by known examples of evolution in action? I suggest that the reason for such a stalling out effect is due to the dramatic decrease in the ratio of potentially beneficial sequences/structures in sequence/structural space as once considers biosystems with greater and greater minimum structural threshold requirements. I term this problem the exponentially expanding non-beneficial gap problem.
Intelligent design
\Q: What is intelligent design?
Miller: My understanding of intelligent design is that it is the argument that the structures, features, organs, and biochemical pathways that we find in living cells are so complex that they could not have been produced by natural processes such as evolution and that they would require the intervention of an intelligent designer outside of nature to bring them into existence.
Sean Pitman: Close, but not quite right. Intelligent design theory does not care if the proposed designer is "natural" or "supernatural". The intelligent agent could be a very smart alien from a galaxy far far away from the perspective of ID Theory. ID Theory says absolutely nothing about the actual identity or nature of the intelligent agent beyond the notion that the agent is actually intelligent. To suggest that ID is proposing that the intelligent agent is also "Supernatural" is a mischaracterization of the basic tenet of ID Theory.
Now, it is true that those who carry the title of "IDists" have actually gone about proposing who they think their proposed designer actually is. This fact notwithstanding, the very basic notion that one can detect deliberate artifact without any additional knowledge of who, why, or how is still true and has at least the potential to be used as a valid scientific hypothesis to explain many types of phenomena observed in nature.
"I often hear people say that they're not descended from monkeys. Well, they're right."
Sean Pitman: And, descent from a pre-monkey pre-human common ancestor is supposed to make them feel better? ; )
Q: [Phillip Johnson, the father of the intelligent-design movement] likens this process to flipping a coin: if it lands and it's not heads, it must be tails. He says that evolution can't account for the diversity of life, therefore it's got to be something else. The only other thing it could be is an intelligent designer.
Miller: It's a negative argument in the sense that their proof of the existence of a designer is the alleged inadequacy of evolution to account for these complex features. What's wrong with that explanation is that it's a contrived dualism. It's an argument that says, "Either evolution can explain everything, or we can invoke an intelligent designer." What it amounts to, for example, is the claim that the moon is made of green cheese, and someone else says, "No, I think it's made of granite." Then we go to the moon, we bring back samples of rock, and we say, "You know what? They're not made out of granite." Does that mean we now have definite proof for the green-cheese explanation? Of course not. The whole idea of intelligent design is a confession on the part of its advocates that they actually can't get any evidence at all in favor of a designer. So what they resort to is the notion that it's either evolution or it's design. And if evolution right now, today, cannot explain everything, that lack of a complete explanation amounts to evidence for the other side. Well, it doesn't. What it really points out would be the current inadequacy of science to explain everything. And science, as any realist knows, is necessarily incomplete. On the day when we have a complete scientific explanation for everything in nature, it'll be time to close every science department of every research institution in the world, because all questions will have been figured out. I don't expect to see that day. But that doesn't mean that the incompleteness of science is an argument for a supernatural alternative like intelligent design.
Sean Pitman: Much, if not all, of science is based on the potential for falsification. If a hypothesis or theory cannot be falsified, then some would say it isn't a scientific hypothesis or theory. Many scientific theories are in fact set up in a rather dualistic way, such that if X is not true, then Y most likely is true. Again, SETI science is also based on this very same "contrived" dualism. To quote Seth Shostak, senior astronomer at the SETI Institute:
"Perhaps the extraterrestrials will preface their [radiosignal] message with a string of prime numbers, or maybe the first fifty terms of the ever-popular Fibonacci series. Well, there's no doubt that such tags would convey intelligence." (Link)
Sean Pitman: This is an interesting statement. Why is it that such a numerical pattern, carried in a medium of a radiosignal, would so clearly indicate deliberate artifact? By Miller's argument it would be at least possible that some as yet unknown non-deliberate natural phenomenon may have been responsible. In fact, to ever stop looking for such a non-deliberate phenomenon and just give up after even a protracted search to conclude the action of some unknown intelligent agent would simply be anti-scientific - - at least according to Miller's argument. It seems that Miller would have one always conclude a non-intelligent agent no matter what and that one should keep up the search for a non-intelligent agent forever.
Is this a reasonable position? How long is it actually reasonable to search for a non-intelligent answer for a given phenomenon before the hypothesis of deliberate artifact or design gains a reasonable level of credibility? While no scientific hypothesis can ever be fully confirmed in that there is always a possibility of being wrong or of having one's theory falsified, is there a point at which the weight of evidence favors even the hypothesis of deliberate design in certain cases beyond a "reasonable" doubt?
Q: What's the harm in introducing intelligent design into a science classroom?
Miller: One could very well say that a God, a designer, a supernatural force was responsible, let's say, for an event that happened in your life or my life, or was responsible for our ability to meet the challenges of life. I don't see anything wrong with that, and it might be a valid explanation in many cases. But pretending that that explanation is a scientific one is a violation of everything we mean and understand by science.
Sean Pitman: Let's say that Miller travels to some alien planet, like Mars, and while walking around on this seemingly barren planet comes across a highly symmetrical polished granite cube that measures one meter on each side and is parallel and perpendicular to within 0.01% of perfect symmetry. In the center of each cube face is a highly symmetrical carved triangle that measures 10 cm on each side and is carved to a depth of 1 cm.
What would Miller instantly assume if he were to find such a granite cube on Mars? I'm pretty sure that even Miller would very quickly propose some sort of artifactual origin - - even without knowing the actual identity, motives, or methods of his proposed intelligent agents.
Now, why might that be? Thought it is impossible to know for sure, one could still be pretty confident that such a cube is highly unlikely to have been formed by any known non-deliberate force of nature. So, what is the most likely "contrived dualistic hypothesis"? Even Miller would no doubt proposed, contrived and dualistically not withstanding, the hypothesis of deliberate artifact. Just don't call the designer "God" is all.
Bringing that idea into the school classroom seems innocuous enough, because all you would do is tell students, well, there's either the evolution explanation or the design explanation. But consider the implications of that. If we present the idea of intelligent design as an alternative to evolution, students, who are very bright, are going to understand something right away, and that is, basically, you've got your atheist theory over here and you have your Bible or God-friendly theory over there.
Sean Pitman: I'm afraid very bright students would be just a bit brighter than to simply accept such a mischaracterization without question. If anything is an unwarranted dualistic hypothesis, this is it. Although the Theory of Evolution does not require philosophical atheism, what it does do, at least according to those like Richard Dawkins, is "Allow one to be an intellectually fulfilled atheist." (Richard Dawkins, The Blind Watchmaker, p. 6). Without the theory of evolution, being an intellectually fulfilled atheist is just a bit more difficult is all. It certainly was for most, even within the scientific community, before Darwin came along.
What it does is to falsely cast evolution in light of an inherently atheistic idea. This is the goal of the intelligent-design movement, indirectly to tell students that either you turn your back on the faith that you've been brought up with in order to embrace the scientific mainstream, or to be true to your faith you have to reject modern science. That's a false choice. It does disservice to religion, and it does disservice to science, and I think it is a terrible way to proceed with scientific education.
Sean Pitman: In a way there is a choice. Both the Christian religion and mainstream science say things about human origins as well as the origins of all living things. Not all of the various views of Christianity or science compliment each other. Many are completely dichotomous. The student is therefore faced with a choice. Many religious ideas the student may have had will most definitely have to be discarded if the student decides to accept the evolutionary story of origins - - and visa versa. This is no false choice or dichotomy. There is a difference between being religious and believing in some Santa Claus version of God and being religious where one's religion proposes very specific views about how the physical world/universe works and how God actually interacts with that world/universe.
Common ancestry
Q: People often say, "I'm not descended from a monkey." What's the true relationship there?
Miller: Well, I often hear people say that they're not descended from monkeys, and they would defy me or anybody else to show that they are. Well, they're right, they're not descended from monkeys. They're not descended from chimps or monkeys or gorillas or any other living organism. The essential idea of common ancestry is that ultimately all living things on this planet share common ancestors if we go far enough back into the past. So, for example, to take the case that people talk about all the time, we share a common ancestor with all primate species. This means that we're related, by having a single ancestor somewhere in the past, to monkeys, gorillas, chimpanzees, and so forth.
Sean Pitman: Which I'm sure makes those who do not want to be descended from a monkey much more relieved! ; )
But the idea of common ancestry goes way deeper than simply saying we're related to monkeys. We're in fact related to all mammals. You go farther back, we are related to all vertebrates. And, ultimately, we are related, if you go far enough back, to every living thing on this planet. The almost universal nature of the genetic code, the fact that all life depends upon DNA, all of these things are evidence of this commonality of ancestry, if we go far enough back in time.
Sean Pitman: Or, at least some of these similarities could be evidence of common design - - that is if the certain features of living things could be shown to be clearly beyond the powers of random mutation and natural selection acting over the course of hundreds of millions and even many billions of years.
Q: One of the lines of evidence that you pointed out at the Dover trial is the organization of our own chromosomes. How is that evidence for common ancestry?
Miller: We've known for a long time that we humans share common ancestry with the other great apes—gorillas, orangs, chimps, and bonobos. But there's an interesting problem here. We humans have 46 chromosomes; all the other great apes have 48. In a sense, we're missing a pair of chromosomes, two chromosomes. How did that happen? Well, is it possible that in the line that led to us, a pair of chromosomes was simply lost, dropping us from 24 pairs to 23? Well, the answer to that is no. The loss of both members of a pair would actually be fatal in any primate. There is only one possibility, and that is that two chromosomes that were separate became fused to form a single chromosome. If that happened, it would drop us from 24 pairs to 23, and it would explain the data. "The closer we look at our own DNA, the more powerful the evidence becomes for our common ancestry with other species." Here's the interesting point, and this is why evolution is a science. That possibility is testable. If we indeed were formed that way, then somewhere in our genome there has to be a chromosome that was formed by the fusion of two other chromosomes. Now, how would we find that? It's easier than you might think. Every chromosome has a special DNA sequence at both ends called the telomere sequence. Near the middle it has another special sequence called the centromere. If one of our chromosomes was formed by the fusion of two ancestral chromosomes, what we should be able to see is that we possess a chromosome in which telomere DNA is found in the center where it actually doesn't belong, and that the chromosome has two centromeres. So all we have to do is to look at our own genome, look at our own DNA, and see, do we have a chromosome that fits these features? We do. It's human chromosome number 2, and the evidence is unmistakable. We have two centromeres, we have telomere DNA near the center, and the genes even line up corresponding to primate chromosome numbers 12 and 13. Is there any way that intelligent design or special creation could explain why we have a chromosome like this? The only way that I can think of is if you're willing to say that the intelligent designer rigged chromosome number 2 to fool us into thinking that we had evolved. The closer we look at our own DNA, the more detailed a glimpse we get of our own genome, the more powerful the evidence becomes for our common ancestry with other species.
Sean Pitman: Case closed! - right? Just because one feature can be easily explained by a particular mechanism doesn't mean that all features can be as easily explained. The problem with the proposed evolutionary mechanism of random mutation and Natural Selection is that Natural Selection only recognizes functional genetic differences. Nature can only select, in a positive manner, those differences that are functionally more beneficial than what came before. Pointing out numerical similarities or differences between the genomes of great apes and humans is interesting and can certainly be explained by non-deliberate processes, but it really doesn't say anything about the functional differences involved. The real question is: Can the proposed mechanism explain the functional differences that exist between one creature and another? The overall pattern isn't a problem. The problems arise when one starts to consider functional differences at different minimum structural threshold requirements.
But, what about the fact that the observed pattern is quite predictable from the perspective of common descent? That's true. The pattern is quite predictable given the hypothesis of common descent. However, such patterns where once thought to be right in line with the mind of a very orderly God who produced a very integrated, pattered, orderly creation. For example, Carl Linnaeus, the father of modern taxonomy and the father of modern ecology, was the man who laid the foundations of modern classification and naming method of living things according to a binomial nomenclature. Yet, Linnaeus was a firm believer in God and believed that the patterns and similarities he observed in living things were the result of the mind of an orderly God, a God who was interested in exploring the range of creative possibilities within various basic body shapes, plans, forms, and functions.
From this perspective, finding a spectrum of creatures with seemingly slight genetic and morphologic variations in a nested pattern or "hierarchy" is quite consistent and even predictable. Could I possibly argue that the fusion of a human chromosome relative to that of the great apes is actually predictable given the hypothesis of separate design? It really isn't that much of a stretch to suggest the possibility of separate design of both humans and apes with 48 chromosomes each followed by subsequent fusion of two of these chromosomes in the human generations during a population bottleneck.
Both options are at least reasonably possible given the truth of either the design or evolutionary hypotheses. So, how to judge which hypothesis is most likely true? - beyond a reasonable doubt? To be honest, the design hypothesis has the greatest burden of proof. So, if there is any evidence that could possibly tip the scales in favor of separate design of apes and humans, it would have to be rather overwhelming. So far, I haven't found such overwhelming evidence from genetics, morphology, or the fossil record to argue that humans and apes definitely show evidence of independent creation. I do believe, however, that such evidence may be found with more detailed investigation into the functional differences between apes and humans. Again, this is because the limits to the evolutionary mechanism are functional limits, not pattern limits. While these functional limitations have not clearly been crossed between humans and apes, at least not to my knowledge, they are crossed in other well investigated biosystems that happen to be a bit more humble - such as flagellar motility systems, DNA transcription and RNA translation, ATPase function, and a host of other such functional systems that require more than 1,000 specifically arranged codons of genetic real estate at minimum.
The process of evolution
Q: What do gaps in the fossil record represent vis a vis evolution? Why are such gaps not a problem for evolutionary theory?
Miller: It's important to appreciate that all historical records are necessarily incomplete. We don't have complete data for any historical process. I've tried to trace my own ancestry, and after about four generations, we lose bits and pieces of it. I don't think that means I don't have any ancestry. I think it means that some of the evidence is missing. The same is true for the study of history. We know, for example, when and where the Battle of Gettysburg took place in the Civil War. We know the opposing generals on both sides. But we don't know exactly what every soldier, by name, was doing at every moment during the Battle of Gettysburg. That doesn't mean Gettysburg didn't take place. It doesn't mean that the Union forces didn't win. It simply means we have more to learn about that battle. The same is true for the fossil record. We have an enormous amount of information as to what life was like in the past. That information tells us that life changed, that it changed in a particular pattern, and that the history of change is complete, with one example after another of descent with modification, an ancestor-descendant relationship between organisms. And in a few lucky cases, we can trace almost step by step the evolution of key organisms in the history of life. [See Fossil Evidence.]
Sean Pitman: The problem here is that what may seem like a very small step morphologically is not necessarily that small of a step genetically. Evolutionists love to talk about how this feature morphed into that feature without ever really getting into just what genetic changes would be required or the pathway that such genetic changes could possibly follow where each specific mutation would be functionally beneficial over that which came before. The really significant gaps are not in the fossil record, but in the genetics. Assumed evolutionary relationships based on morphology change all the time.
"The more similar two species looked, the more closely related they were thought to be. But looks can be deceptive. This became abundantly clear more than a decade ago, when molecular biologists began comparing small numbers of genes from various organisms and found that many species were not what they appeared. Hippos, for example, were once thought to be the kissing cousins of pigs, but genetic evidence revealed their closest living relatives to be the cetaceans (whales, dolphins and porpoises).
Without the insights of molecular analysis, traditional morphologists also had no way of knowing whether a particular feature had been lost in a given lineage, or had never been there in the first place. In line with the idea that things evolve towards increasing complexity, they tended to assume the latter, sometimes quite incorrectly. Take the sea squirt. Its larva swims around looking like a tadpole, with a nerve cord along its back, gill slits for feeding and a tail - all classic features of chordates, the large group of animals with backbones that includes us. Then, however, it stands on its head and turns into a sack of jelly, having first digested what it had of a brain. The adult looks suspiciously like a plant. For a long time it was considered to be one of the most primitive chordates because of its simple adult form - about as far from vertebrates as it was possible to get. In between were myriad other groups, including the lancelets - fish-like animals that hang on to their nerve cords into adulthood. Then molecular studies revealed that sea squirts are genetically closer to us than are lancelets, and the tree had to be reshuffled." (Link) So, the fossil evidence and morphologic evidence really isn't that reliable. Many assumptions are made that are not necessarily true, even from the evolutionary perspective. Again, it is all back to genetics when it comes to really evaluating the plausibility of all of the claims of the Theory of Evolution, its potential, and limitations.
Q: What about the claim that no one's ever seen a new species form?
Miller: Right now new species are literally in the process of forming in the state of California . For years David Wake of the University of California at Berkeley has studied different species of salamander that surround the Central Valley in California . When you look at the range of these species, what you discover is that the local variations at the very ends of the range are now so different from each other that if you capture them both and you put them side by side in a cage, any biologist would agree that they are distinct and separate species. Nonetheless, they have been produced in recent times simply by the spreading of salamanders over a geographic range. Many opponents of evolution will sort of retreat and say, "Well, okay, but those species are really similar to each other. Show us a species that is dramatically different." But that initial splitting, that's the phenomenon that actually drives evolution. You shouldn't expect to see a cat suddenly give birth to a dog or something along those lines. At the moment when one species splits into two, you should see two distinctly different species that still show the similarities that previously united them within a single classification. We see this happen all the time. The people who say that macroevolution, by which they mean really big evolution, has never been observed, inevitably cannot give you a strict and rigorous definition of what macroevolution is. They'll simply say it's the formation of new categories or evolutionary novelties. They're loath to put specifics on that idea, to tell you what percentage of the genes or how many base pairs of DNA have to change, because I think they know very well that once they make specific what they mean by macroevolution, some darn biologist is going to go out into the field or into the lab and follow exactly that rate of change and show that macroevolution really does occur.
Sean Pitman: This is a valid argument. The subjective nature of what defines a species has always been a problem - for evolutionists as well as creationists and IDists.
For example, scientists from Berkeley have noted that, "the planktonic larvae of many marine invertebrates are commonly described as separate species when they are first discovered in the ocean. Only later when they can be reared in the laboratory can the link to their adult form be recognized. Similarly, the different life stages of many fungi are given different names because they have different physical forms and hosts. Only through detailed inoculation studies can mycologists work out which forms are members of the same life cycle. Since some fungi may have more than five discrete life cycle stages, this can be a long process. Similar problems exist for some marine algae and multiple-host parasitic organisms of many kinds. "Even among well-studied vertebrates, some tropical birds have been described as separate species until they are observed to mate and rear young together." The naming of hominid fossils not immune from this subjective problem. In a March 2002 statement, Tim White, who co-directs the Laboratory for Human Evolutionary studies said, "There’s been a recent tendency to give a different name to each of the fossils that comes out of the ground, and that has led to what we think is a very misleading portrayal of the biology of human evolution… But when you find a fossil like this one so similar to Asian and European ones, it indicates the same species." "This whole species question is all about what you accept as a sharp enough distinction to tell you that it is a separate species," said Susan Anton, a Rutgers University anthropologist.The classifications of plants is classically prone to give different names to very similar plants or even parts of the same plant. Bill DiMichele, a paleobotanist, notes, "The problem of organ association is one of the reasons why paleobotanists insist on so many different names for isolated parts of the same whole plant. Furthermore, there are phenotypic convergences that can cause great confusion, such leaves of virtually identical morphology borne on ferns and seed plants. Separate names for each fossil plant organ can be carried to extremes, however, and not all paleobotanists, myself included, favor the attribution of separate names to organs otherwise known in attachment (yes, this is still done routinely, no kidding)." (Link) However, functional aspects of specific biosystems are more objectively defined. A functional system that definitely has a certain minimum structural requirement can be studied to a degree in which this minimum requirement can be known in sufficient detail to be fairly objective. As it turns out, all examples of evolution in action produce novel beneficial systems that require no more than a few hundred specifically arranged codons of DNA or amino acid residues. There are no examples in scientific literature producing any novel system of function that requires a minimum structural threshold of more than 1,000 amino acid residue building blocks or 1,000 codons of genetic real estate to code for it. There's not a single example beyond this threshold level - period. If Miller wants a precise definition of what constitutes "macroevolution", this is it. Evolution just doesn't happen at this level or beyond.
Q: Another criticism often made is that all this couldn't just have happened by random chance.
Miller: One of the great mischaracterizations of evolution is that it's driven by random chance, that things just happen. People like to say, "I don't like to believe that I'm just an accident." Well, you're not. What evolution says is that the variation that crops up in a species is indeed unpredictable. We can't be sure what will happen next. But that doesn't mean it's random. To me, the word "random" means anything can happen. But the reality is that evolutionary change is restricted. It's restricted by the laws of physics and chemistry. It's restricted by the nature of molecular biology. It's restricted by the constraints of developmental biology during development. Most importantly, evolutionary change is governed by natural selection, and natural selection is not a random process at all. Natural selection selects for successful phenotypes, for successful combinations of characteristics that actually work, and that's not random at all.
Sean Pitman: Darwinian-style evolution is supposed to use aspects of random and non-random forces. The reason why random mutations are called random mutations is because they are in fact random. During DNA replication, for example, any base can be miscopies and turned into any of the other 3 options (i.e., a T could get turned into an A, G, or C - or it could get deleted or added) at each genetic loci. Such changes are pretty random when they happen.
Of course, natural selection is supposed to come to the rescue as the non-random force of Nature. While it is certainly true that Natural Selection a non-random force, it is not true that Natural Selection overcomes the problems associated with non-directed random mutations or "random chance". The reason for this is that Natural Selection is very limited in what random changes it can actually recognize to base a selection on any more than random chance. To overcome random chance a random mutation must produce a functional genetic change that actually affects the creature's survival and reproductive success. And, in order for Natural Selection to select in a positive manner, the random genetic change must produce a functionally beneficial genetic change.
As it turns out, the majority of random mutations produce no detectable functional change. These mutations are called "neutral mutations." There is even a neutral theory of evolution. The problem is that neutral mutations are truly random since they cannot be guided by the non-random force of Natural Selection. This is a problem because a linear increase in the gap distance between a starting point and the next closest potentially beneficial genetic target equates to an exponential increase in the number of random mutations needed to achieve success. This is why there is a marked stalling out effect of evolutionary potential when it comes to finding "targets" with higher and higher minimum structural threshold requirements. The minimum gap distance increases. As a result, the average time needed to achieve success declines - - exponentially.
"Any theory that can stand up to 150 years of continuous testing is a pretty darn good theory."
Sean Pitman: There have been a lot of wrong theories that have lasted much longer. Besides, the evolutionary mechanism has not been tested beyond the very lowest rungs of the ladder of minimum structural threshold requirements. Well, that's not quite true. It has been tested, but it has never passed the test beyond the 1,000aa threshold level.
Q: I have heard critics say that mutation doesn't create information, it destroys it. Miller:
That notion is at variance with the facts. Four or five million years ago, for example, the Antarctic Ocean, which was warm at the time, froze over as a result of a kind of climate change on this planet. Well, to this day, there are fish that swim in the oceans of Antarctica . One of the interesting things about those fish is that even though the saltwater is actually below the freezing point—our own blood would freeze solid in that cold water—these fish don't. The reason they don't freeze solid is because their blood contains an antifreeze protein, sort of the biological equivalent of ethylene glycol in antifreeze. Well, how did they get it? It turns out that the antifreeze protein that is found in the blood of Antarctic fishes was the result of a digestive enzyme that was mutated, retargeted to the bloodstream, and then mutated again and again to enhance its antifreeze properties. All of these changes were the result of mutation. Now, that Antarctic fish has a kind of biological information that its ancestors didn't have. It has the ability to make a completely new protein that enables it to survive in very cold waters by preventing its blood from freezing. That's novel information, and it's information that was produced by the process of mutation.
Sean Pitman: That's true. Those who argue that it is impossible for random mutation to actually get lucky once in a while and land on some beneficial target are mistaken. There are many such examples. Several are discussed above in this essay dealing with Miller's best arguments (Link). The problem here is that not all functions are created equal. Different types of functional systems have different minimum structural threshold requirements. The evolution of single protein enzymes that do truly novel functions, as in this case of antifreeze evolution and other cases such as lactase or nylonase evolution, are truly spectacular. However, none of these examples produces a protein-based system that requires more than a few hundred specifically arranged residues working in concert. That's the problem, because there are in fact many such systems in all living things that go well beyond this 1,000aa threshold. How did these systems come to be? Explaining how evolutionary processes produced such systems isn't nearly as easy as explaining novel relatively small single-protein enzymes and the like. In fact, higher level systems are exponentially more difficult to explain.
The test of time
Q: How do you answer the charge that evolution has never been tested?
Miller: Evolution is tested every day in the laboratory, and it's tested every day in the field. I can't think of a single scientific theory that has been more controversial than evolution, and when theories are controversial, people devise tests to see if they're right. Evolution has been tested continuously for almost 150 years and not a single observation, not a single experimental result, has ever emerged in 150 years that contradicts the general outlines of the theory of evolution. Any theory that can stand up to 150 years of continuous testing is a pretty darn good theory. We use evolution to develop drugs. We use evolution to develop vaccines. We use evolution to manage wildlife. We use evolution to interpret our own genome. Every one of these uses of evolution is a test, because if the use turns out to be inadequate, we would then go back and question the very idea of evolution itself. But evolution has turned out to be such a powerful, productive, and hardworking theory that it's survived that test of time.
Sean Pitman: Again, all of these examples of evolution being tested do not test evolution beyond the very lowest levels of novel functional complexity. None of these examples produces anything requiring more than a few hundred specifically arranged genetic codons or amino acid residues.
No one is arguing that evolution doesn't happen at all. The argument is that the evolutionary mechanism is significantly limited - i.e., limited to very low levels of functional complexity that never go beyond the 1,000aa threshold.
Q: So when they talk about teaching the strengths and weaknesses of evolution, what are the weaknesses?
Miller: Evolution has great strengths in that it unifies biology and gives us a coherent explanation. Its only weakness is that it hasn't explained everything yet. For example, we have great doubts as to what the evolutionary purpose of sex is. Now, sex is everywhere, not just in us, but also in trees and flowers and microorganisms. It's very difficult to understand exactly how sex first evolved, why there are only two sexes, and why things work the way they do. Evolution hasn't completely explained that yet. We also don't understand where the first living cell came from or how prebiological evolution took place. But most of us in science don't regard the inability of science to explain everything as weakness. We regard that as the unexplored territory that's going to keep most of us busy for the rest of our careers.
Sean Pitman: No one expects any valid scientific theory to explain everything. If everything could be explained, science would no longer be needed. Science is only needed because of an inability to perfectly known anything about the world and/or universe in which we live. Science produces predictive power that never reaches perfection. However, when a theory consistently runs into very clear limitations, those limitations may eventually undermine many of the claims that were once supported by it.
A complexity theory
Q: What is irreducible complexity?
Miller: Irreducible complexity is a term that was first used on behalf of the intelligent-design movement by Michael Behe, a biochemist at Lehigh University . What Behe observed is that living cells are filled with complex biochemical systems and that these systems have multiple parts. Dr. Behe has argued that systems like that are irreducibly complex. He says that all these parts are required for the system to function, and if you take even one away, it stops working. That means its complexity is irreducible. In other words, you need all the parts. If that were true, it would indeed be a powerful argument against evolution, because what evolution can only do is to produce these complex systems by putting together a few parts at a time. And if there is no function until all the parts are assembled, evolution's in trouble. That's the argument from irreducible complexity. In reality, these supposedly irreducibly complex systems are cobbled together by evolution from individual systems that have functions of their own.
Sean Pitman: Interesting that this cobbling never actually happens when the system in question requires over 1,000 specifically arranged amino acid residues or codons of DNA at minimum. Its like getting from one meaningful 3 letter sequence to another by using what already exists in a genome - - as in cat to hat to bat to bad to bid to did to dig to dog. Easy. This is because the ratio of potentially meaningful 3-letter sequences in the English language system is about 1 in 18. For 2-letter sequences it is about 1 in 9. However, for 7-letter sequences the ratio drops to about 1 in 250,000.
The continued decline of the ratio in such an exponential manner means that the odds that the gap between a starting point and the next closest potentially beneficial target is always going to be just one character change away drops exponentially as well. Pretty soon the minimum gap distance grows from 1 to 2 for a given "pool" of options. With a gap of 2 needed character differences, the question is, what are the odds that these two characters will exist somewhere in the pool of options preformed? And, beyond this, what are the odds that these two characters will get copied and pasted properly into the location where they would create a new beneficial sequence? This is exactly the same problem that exists in genetics. There is no fundamental difference. And, this problem only gets dramatically worse as one considers systems with greater and greater minimum sequence and/or structural threshold requirements. By the time the minimum requirements are up to 1,000 codons or amino acid residues, the likely minimum gap distance is not one or two or three needed characters, but dozens of needed characters. The odds that a gap distance of 50 or so needed characters exist preformed in a specific order anywhere else in a even a very large pool of genetic options are extremely remote. Without such a pre-formed subsequence or structure, no one mutation is going to be able to cross the gap between what already exists in a gene pool and what might be beneficial if it were ever found by pure random mutation.
Miller and others have proposed that the actual steppingstones needed to form higher level functional systems, like flagellar motility which requires well over 10,000 codons of DNA, are actually small steps. There are several facts that strongly question this assertion. Perhaps the most obvious problem is that not one of these proposed steps in the evolution of something as minimally complex as the bacterial flagellum has ever been shown demonstrated to evolve in nature or under laboratory conditions - not one step. The only evidence behind assertions that such steppingstones could have paved the way is story telling and hand waving. There simply is no experimental evidence for any proposed step that crossing the gaps between novel functional targets that exist beyond the 1,000aa threshold level.
Q: Dr. Behe has pointed to the bacterial flagellum as a good example of irreducible complexity. Can you explain why you think it isn't?
Miller: Well, the bacterial flagellum is this marvelous little machine that consists of about 30 or 35 individual proteins, and the argument is if you take even one part away, the flagellum doesn't work anymore. So evolution couldn't possibly have produced it, because evolution is blind. Evolution couldn't say, "Well, we've got 20 parts for the flagellum. Next year we'll evolve the 21st part, and then 22 and then 23, and maybe in 10 million years, we'll get the 30th part, and everything will start working." Evolution doesn't work that way. When you look at the experiments that biologists and biochemists have done on the bacterial flagellum you discover that little clusters of proteins in the flagellum, in other bacteria that don't have flagella, are busy doing other functions. "Not a single scientific paper has been published that supports the notion of irreducible complexity." For example, about 10 of those proteins in the base of the flagellum form a little machine called the Type 3 Secretory system. It's kind of like a molecular syringe that bacteria use to pump poisons into cells they're attacking. This system, this little syringe, is found in bacteria that don't have flagella. The very existence of this little subset of parts, just 10 parts, with a perfectly good function of their own, shows that the idea of irreducible complexity is wrong. And when you take the flagellum apart, you discover that virtually every protein in there is related to another family of proteins that performs a different function somewhere else in the cell. So the prediction of evolution, which is that these complex systems are actually slapped together by scavenging pieces of different systems, turns out to be true. And the prediction made by irreducible complexity that none of these proteins would have any function until they're all put together and all work, that prediction turns out to be wrong. In the 10 years since Professor Behe first advanced the idea of irreducible complexity, not a single scientific paper, even from his own lab, has been published that supports the notion of irreducible complexity for any of the systems that he described, and that's why the scientific community simply has not embraced this idea.
Sean Pitman: Miller doesn't seem to understand the concept of irreducibility. Irreducibility means that a system requires a certain minimum number of parts in a certain arrangement for that particular function to work. It doesn't matter if some subsystem would still work in some other capacity if parts were removed. This doesn't change the fact that a minimum part and arrangement requirement is still required for the function in question to be realized.
For example, consider the powered motility function of a car. This function requires a certain minimum number of parts in a particular arrangement. Now, let's say that I take away the drive shaft. What happens to the motility function of the car? The motility function instantly disappears - right? But the lights and radio still work. Does this mean that the car's motility function is therefore "reducible"? Of course not.
All functional systems are irreducible in this sense. And, all functional systems that I can think of have subsystems that could be used in beneficial capacities as parts of other systems. Longer genetic sequences no doubt have shorter subsequences that make sense in many other capacities. For example, consider the Shakespearean phrase used by Richard Dawkins, "Methinks it is like a weasel". By itself it may have a beneficial function in a certain environment. Remove a few characters though and that function may be completely lost. Does this mean that all function will be lost? Not at all. The sub-phrase "it is like" might still be quite useful in various contexts. Even a single word or a single letter would no doubt be quite useful in many different contexts.
The same thing is true of biosystems. All functional elements have minimum structural requirements to perform their specific functions. Might their sub-elements have other beneficial functions? Of course. A single protein, like lactase, might have subsections that perform other useful functions in various other systems within the organism. But, that subsystem has its own minimum structural threshold requirement and its own subsystems that could perform other useful, though more basic, functions in the organism - - all the way down to single amino acid building blocks. These single amino acids also have minimum structural requirements. They also have subsystems in the form of different kinds of atoms that individually can perform many useful functions as parts of other systems or molecules within the organism. Again, it is like the difference between 3-character, 7-character, and 1,000 character sequences. Raising the minimum part requirement doesn't remove the potential for functionality of lower-level systems. What it does do, however, is to increase the likely minimum gap distance between what exists and what might exist, in a beneficial manner, within the gene pool.
Q: In the trial, both Michael Behe and Scott Minnich [a microbiologist at the University of Idaho who is a proponent of intelligent design] claim that intelligent design is testable, but then they say that they don't conduct those tests. What does that indicate to you?
Miller: One of the biggest problems with intelligent design is it's not empirical. It doesn't feature any testing. The advocates of intelligent design are not experimentalists. They're not going out in the lab and doing experiments to see this. Both Michael Behe and Scott Minnich have said that one could disprove intelligent design by taking a bacterium in the laboratory that didn't have a flagellum and evolving a flagellum in it. Well, that's a ridiculous proposal for an experiment for two very simple reasons. First of all, the experiment would probably take 10 to 100 million years to carry out, and it's kind of hard to get funding for that long. The second reason is that what they propose is to retrace the path of an existing sequence of evolutionary changes. Evolution doesn't repeat itself like that. So even if we were absolutely certain the flagellum had been produced by evolution, we wouldn't expect the same sequence of events to happen again. That's a critical point.
Sean Pitman: The test wouldn't have to produce an entire flagellum from scratch. Miller and many others have argued that the relevant steps between beneficial steppingstones along the pathway toward flagellar motility are closely spaced. All they have to do is demonstrate that one or two of these proposed steps are in fact crossable by random mutation and function-based selection. If these steps are in fact as close together as Miller claims, then the experimental setup and confirmation should be no problem.
The truth of the matter is that Miller and other scientists prefer to stick with their just-so-stories at such levels of complexity, even when it comes to demonstrating just one of their proposed steppingstones, because that is all they have. Statistically each one of their proposed steppingstones would require not 10 or 100 million years, but trillions upon trillions of years. That is why it is much easier to sit back and tell these oh so plausible stories as long as they can keep their audience from actually considering the statistics involved.
A better test for the whole notion of irreducible complexity is just to compare various bacterial genomes and see if their arguments are correct. Their arguments are that none of the genes that produce the proteins of the flagellum are used for any other purpose in any other organism. Well, that test has been done, and it turns out their premise is not correct, that these individual proteins and individual genes are used for other purposes in other organisms, which is the direct prediction of evolution.
Pitman Responds to