I am continuing my review of chapter 5 of CG Hunter's Darwin's Proof. The mistakes are very dense in this part, so discussing only six pages of his book will require a lot of ink.
Hunter next deals with the evidence from genomic comparisons. He first tries his hand at the fact that humans and chimpanzees have 98.7% of our DNA in common. First, he says that this difference is too small to account for the difference between humans and chimps. Of course it can account for it. That has nothing to do with evolution. However humans and chimps differ, it is encoded in our DNA. However much our DNA differs, whether it is 99% or 50%, that accounts for how we differ, for creationists as well as evolutionists. Hunter seems to be denying that DNA actually determines the characteristics of humans and chimps. What does he suggest does account for our differences, if not our DNA? He points out that humans and mice have almost nearly 98% of our genes in common, just like humans and chimps, but here he is subtly switching comparisons. With the human mouse comparison he is simply comparing how many of our genes have an ortholog in mice. He is not comparing how similar those genes are, which is what is being compared in humans and chimps. When we do that comparison, mice and humans are 91.3% similar.
He then points out that evolution doesn't predict 98.7% similarity. Of course it doesn't. He is right that it could be 99% or it could be 80%. No one claimed evolution predicted that number a priori. The number isn't evidence for evolution. The significance of the number is that a relatively small difference in DNA can account for dramatic differences in the organisms; it was never given as evidence for evolution directly. The evidence is found in the pattern of similarities when we compare the DNA of many organisms, and their correlation with other characteristics. The high number just reminds us that we are not as unique as we think--in fact it is similar to the difference between a horse and a donkey (99%).
Next Hunter tackles what I think is wonderful evidence for evolution, the fact that coding regions are more similar between species than noncoding regions. This happens because the noncoding regions are unconstrained and selection will not remove mutations, but most mutations in coding regions are removed by selection. How does he address this? He points out that it can be explained by evolution only if the noncoding regions are nonfunctional, so that is an assumption we are making. No, it isn't. It has been tested. Certainly the fact that the third base of a codon is usually redundant is testable. The fact that pseudogenes aren't transcribed is testable. The fact that introns are degraded after excision and dramatic changes in their sequence has no effect is testable.
Then Hunter tries his usual gambit--he says evolution would have no problem if we didn't see that pattern, or any possible pattern. If we saw something different, we would just assume that the region that didn't change must have a function. No we wouldn't. We would assume that, based on evolutionary theory, but would have to find the function. If a conserved region was observed and we could not find a function, it would be unexplained by evolution. As it is, the opposite is true. Every time we find a conserved region, we soon find a function. It is a great way to identify importan cis regulatory elements, for example.
The next topic Hunter addresses are examples of small scale evolution, like the beaks of the finch or the peppered moth. He admits we have seen these, but says that these changes appears to have limits. He says this several times in various ways, but never cites any evidence at all that there are such limits. I would love to find them. In fact, I give students in any of my classes an outomatic A if they can find any evidence for such a limit. Hunter ignores the vast literature comparing the changes in the fossil record with observed changes. This literature has shown that the kinds and rates of changes that we see today are consistent with the changes in the fossil record. Hunter claims that small scale changes do not appear to be able to extrapolate to large changes when in fact the literature shows just the opposite.
Hunter quotes mines Ernst Mayr saying we don't know how genotype changes bring about the changes in phenotype. Mayr was pointing out that we didn't know exactly how genetic changes bring obout phenotypic changes. That is being very well addressed today by the field of evo devo, which is in fact filling in this last gap in our knowledge and showing how specific DNA changes lead to specific phenotypes. But even without these advances, this is at best an argument from ignorance.
Then Hunter moves the goal posts and points out that small scale evolution requires a reproductive apparatus in the first place, so how did that come about, huh?
Hunter's last section is on the origin of life. Yes, he has a section on the origin of life in the chapter on evidence for evolution. When has the origin of life ever been presented as evidence for evolution, rather than a question for evolution? Every biologists knows there is a lot we don't know about the origin of life. It is an unanswered question. No one has ever brought it up as positive evidence for evolution. Hunter already had two chapters on problems with evolution, and it would make more sense to put the origin of life there. In fact, he did. He just felt like bringing it up again here because he really likes the areas where we are less certain and he really likes to get back to ultimate origins as often as he can.
Hunter doesn't even try in these chapters to address several other kinds of evidence, such as development or biogeography. In fact, I have never seen anyone deal with biogeography. Overall, his attack on the positive evidence is very feeble. He mostly grants the evidence, but then declares it unconvincing or ambiguous. Never does he hint at the true depths of evidnece or go into any detail.
Sunday, September 7, 2008
Wednesday, September 3, 2008
Darwin's proof, chapter 5
I am continuing to review Darwin's Proof, by C. G. Hunter. Here, I will review chapter 5, which, like chapter 4, tries to debunk the evidence for evolution. I cannot find a distinction between the material in chapters 4 and 5. It appears the material was separated into two chapters just to keep the chapter size smaller.
To understand Hunter's approach to evidence for evolution perhaps I can use a comparison. The same approach could be used to call into question any science. Let's look at Mendelian genetics. Geneticists claim that their theory predicts certain ratios in the offspring of crosses, such at a 3:1 ratio. However, there are many crosses in which we don't see that. Some crosses produce 2:1 ratios or 9:4 ratios. Sometimes the traits aren't even inherited from the father at all but only the mother. Geneticists will give various explanations for these anomalies, but they have so many excuses and explanations that any ratio at all is consistent with genetics.
That is what Hunter does. He gives a simplistic idea of what evolution should predict, then points out that there are many examples that do not fit this prediction. He admits that biologists have explanations for the exceptions, although he never bothers to tell his readers what they are so they can judge for themselves. He dismisses the explanations as ad hoc excuses and claims that because there are so many exceptions, any possible data will fit with evolution. Not once does he suggest that these explanations are actually tested. He implies that the explanations are accepted by biologists out of desparation, rather than being hotly debated.
I discussed the irony of this before--being able to fit any data describes intelligent design, for which all possible worlds really are consistent with it. Of course, any correct theory will explain what we actually see, but Hunter is wrong that all possible patterns can be explained by evolution, as I will discuss.
The first kind of evidence Hunter discusses in this chapter is the molecular clock. His first error is in thinking that the ticking of a constant clock is evidence for evolution, rather than a tool to determine relationships. The molecular clock was not even observed until a century after Darwin, and many evolutionary biologists have and continue to dispute it's reliability. Of course, Hunter quote mines these disputes to suggest the evidence for evolution is in question. The evidence for evolution is the nested pattern of similarities of DNA, not the fact that the changes occur at a constant rate. All biologists know that the clock isn't constant. He says that the fact that different parts of the genome tick at different rates is a problem for the theory, when it is not. This is one of those areas where we have well tested explanations for differences from his simplistic prediction. He says that a problem with the molecular clock as evidence for evolution is that we must first calibrate the clock with the fossil record. He again is unable to distinguish between molecular evidence for evolution and a clock as a tool to infer the time of speciation.
Hunter relies heavily on some papers criticizing molecular evolution by Christian Schwabe. It wasn't easy to find these papers, since they were published in the 1980s. Such outdated references should immediately cause suspicion. I eventually discovered that Schwabe is a chemist with some very radical views on evolution. He did publish some criticisms of molecular evolution 25 years ago, but all of his criticisms have since been addressed. Quite simply, Schwabe was wrong. Hunter never bothered to find more recent evidence or responses to Schwabe's criticism. Schwabe especially found problems with the evolution of the hormone relaxin, but a recent analysis has resolved the problems (Wilkison, TN, TP Speed, GW Tregear, and RA Bathgate. 2005. Evolution of the relaxin-like peptide family. BMC evol biol 12:5-14).
Hunter acknowledges that we have explanations for the descrepancies from perfect ticking of the clock (although he does not tell his readers what they are), but suggests that these are ad hoc rescues that could be used to explain any patterns, as I discussed above. These explanations are testable and are not accepted until they have been tested. We have evidence for lateral gene transfer, positive selection, etc. He implies that any pattern of molecular similarity would be consistent with evolution, but the truth is that we would have rejected common descent if genes of similar species weren't similar or if most genes did not fit in nested heirarchies.
Nezt, Hunter addresses the nested heirarchical pattern of evolution. This is good, because most creationists fail to even acknowledge this evidence. He says, correctly, that evolutionists say if we didn't find this pattern in nature, evolution would be falsified. He tries to avoid this by claiming that evolution in fact would explain any pattern, not just heirarchical ones. He says this would be true if evolution were very fast or non gradualistic. He pretends that disputes about gradualism within evolution include radical nongradualistic change that could produce non heirarchical patterns. I suppose if evolution were so fast that all evidence of ancestry were obscured, we would not see a heirarchy, or almost any other evidence for evolution. Such a radical theory would not be the theory we have. If organisms evolved that the current theory would be falsified, even if we replaced with a different theory of rapid evolution. Hunter thinks that if we replace one theory with another naturalistic theory, it is cheating. He would only be happy if evidence against evolution causes us to reject all possible naturalistic theories.
The fact is we have seen a heirarchical pattern with non living things that evolve by common descent, even with fairly rapid change. This includes the evolution of languages, chain letters, and Biblical manuscripts. The only way that Hunter can pretend evolution will predict a non heirarchical pattern is by basically inventing a new theory. Perhaps he can at least admit that even a moderately gradual theory of evolution will predict only a heirarchical pattern.
Hunter thinks that highly conserved proteins, which evolve very slowly, are a problem for evolution, because they could not have arisen in the first place (since they evolve slowly). It never occurs to him that proteins can evolve rapidly at first, but once they get a critical function they evolve much more slowly. In fact, we know of such proteins.
He also tries to attack the strong correlation between mophological phylogenies and molecular phylogenies--the fact that we get the same tree using both methods. He acknowledges this would be good evidence if it can be shown that there is no functional reason for them to be similar--if for example the hemoglobin of a frog can substitute for the hemoglobin of a mammal. He then makes an argument from ignorance, saying we do not know if this is the case or not. In fact, there is a vast amount of evidence showing that many molecular changes are neutral and proteins are interchangeable. This is even true of the coding part of proteins, but it is more obviously true in the various kinds of silent DNA--introns, psuedogenes, etc. There is no functional reason the difference between these should correlate with the difference between anatomical traits.
Nelson next mines the literature for a few instances when DNA trees don't correlate with morphological trees. Yes, we know that both kinds of trees have uncertainty and they won't have a perfect match. Very often when they don't match, we understand why--some of his exmaples involve lateral gene transfer or phylogenies that separated nearly a billion years ago and will have long branch attraction. He must at least admit that these problematic trees are exceptions for a rule of strong correlation. Again, he admits that there are explanations for the mismatches, and then implies that these are ad hoc excuses.
There is more in this chapter, but I have gone on long enough. I will finish in another post.
To understand Hunter's approach to evidence for evolution perhaps I can use a comparison. The same approach could be used to call into question any science. Let's look at Mendelian genetics. Geneticists claim that their theory predicts certain ratios in the offspring of crosses, such at a 3:1 ratio. However, there are many crosses in which we don't see that. Some crosses produce 2:1 ratios or 9:4 ratios. Sometimes the traits aren't even inherited from the father at all but only the mother. Geneticists will give various explanations for these anomalies, but they have so many excuses and explanations that any ratio at all is consistent with genetics.
That is what Hunter does. He gives a simplistic idea of what evolution should predict, then points out that there are many examples that do not fit this prediction. He admits that biologists have explanations for the exceptions, although he never bothers to tell his readers what they are so they can judge for themselves. He dismisses the explanations as ad hoc excuses and claims that because there are so many exceptions, any possible data will fit with evolution. Not once does he suggest that these explanations are actually tested. He implies that the explanations are accepted by biologists out of desparation, rather than being hotly debated.
I discussed the irony of this before--being able to fit any data describes intelligent design, for which all possible worlds really are consistent with it. Of course, any correct theory will explain what we actually see, but Hunter is wrong that all possible patterns can be explained by evolution, as I will discuss.
The first kind of evidence Hunter discusses in this chapter is the molecular clock. His first error is in thinking that the ticking of a constant clock is evidence for evolution, rather than a tool to determine relationships. The molecular clock was not even observed until a century after Darwin, and many evolutionary biologists have and continue to dispute it's reliability. Of course, Hunter quote mines these disputes to suggest the evidence for evolution is in question. The evidence for evolution is the nested pattern of similarities of DNA, not the fact that the changes occur at a constant rate. All biologists know that the clock isn't constant. He says that the fact that different parts of the genome tick at different rates is a problem for the theory, when it is not. This is one of those areas where we have well tested explanations for differences from his simplistic prediction. He says that a problem with the molecular clock as evidence for evolution is that we must first calibrate the clock with the fossil record. He again is unable to distinguish between molecular evidence for evolution and a clock as a tool to infer the time of speciation.
Hunter relies heavily on some papers criticizing molecular evolution by Christian Schwabe. It wasn't easy to find these papers, since they were published in the 1980s. Such outdated references should immediately cause suspicion. I eventually discovered that Schwabe is a chemist with some very radical views on evolution. He did publish some criticisms of molecular evolution 25 years ago, but all of his criticisms have since been addressed. Quite simply, Schwabe was wrong. Hunter never bothered to find more recent evidence or responses to Schwabe's criticism. Schwabe especially found problems with the evolution of the hormone relaxin, but a recent analysis has resolved the problems (Wilkison, TN, TP Speed, GW Tregear, and RA Bathgate. 2005. Evolution of the relaxin-like peptide family. BMC evol biol 12:5-14).
Hunter acknowledges that we have explanations for the descrepancies from perfect ticking of the clock (although he does not tell his readers what they are), but suggests that these are ad hoc rescues that could be used to explain any patterns, as I discussed above. These explanations are testable and are not accepted until they have been tested. We have evidence for lateral gene transfer, positive selection, etc. He implies that any pattern of molecular similarity would be consistent with evolution, but the truth is that we would have rejected common descent if genes of similar species weren't similar or if most genes did not fit in nested heirarchies.
Nezt, Hunter addresses the nested heirarchical pattern of evolution. This is good, because most creationists fail to even acknowledge this evidence. He says, correctly, that evolutionists say if we didn't find this pattern in nature, evolution would be falsified. He tries to avoid this by claiming that evolution in fact would explain any pattern, not just heirarchical ones. He says this would be true if evolution were very fast or non gradualistic. He pretends that disputes about gradualism within evolution include radical nongradualistic change that could produce non heirarchical patterns. I suppose if evolution were so fast that all evidence of ancestry were obscured, we would not see a heirarchy, or almost any other evidence for evolution. Such a radical theory would not be the theory we have. If organisms evolved that the current theory would be falsified, even if we replaced with a different theory of rapid evolution. Hunter thinks that if we replace one theory with another naturalistic theory, it is cheating. He would only be happy if evidence against evolution causes us to reject all possible naturalistic theories.
The fact is we have seen a heirarchical pattern with non living things that evolve by common descent, even with fairly rapid change. This includes the evolution of languages, chain letters, and Biblical manuscripts. The only way that Hunter can pretend evolution will predict a non heirarchical pattern is by basically inventing a new theory. Perhaps he can at least admit that even a moderately gradual theory of evolution will predict only a heirarchical pattern.
Hunter thinks that highly conserved proteins, which evolve very slowly, are a problem for evolution, because they could not have arisen in the first place (since they evolve slowly). It never occurs to him that proteins can evolve rapidly at first, but once they get a critical function they evolve much more slowly. In fact, we know of such proteins.
He also tries to attack the strong correlation between mophological phylogenies and molecular phylogenies--the fact that we get the same tree using both methods. He acknowledges this would be good evidence if it can be shown that there is no functional reason for them to be similar--if for example the hemoglobin of a frog can substitute for the hemoglobin of a mammal. He then makes an argument from ignorance, saying we do not know if this is the case or not. In fact, there is a vast amount of evidence showing that many molecular changes are neutral and proteins are interchangeable. This is even true of the coding part of proteins, but it is more obviously true in the various kinds of silent DNA--introns, psuedogenes, etc. There is no functional reason the difference between these should correlate with the difference between anatomical traits.
Nelson next mines the literature for a few instances when DNA trees don't correlate with morphological trees. Yes, we know that both kinds of trees have uncertainty and they won't have a perfect match. Very often when they don't match, we understand why--some of his exmaples involve lateral gene transfer or phylogenies that separated nearly a billion years ago and will have long branch attraction. He must at least admit that these problematic trees are exceptions for a rule of strong correlation. Again, he admits that there are explanations for the mismatches, and then implies that these are ad hoc excuses.
There is more in this chapter, but I have gone on long enough. I will finish in another post.
Monday, September 1, 2008
Power question 4: where did the dinosaurs come from?
Wow, question four was anticlimactic (these are AIGs "4 power questions for an evolutionist" I have been reviewing them in the previous few posts). Riddle tells us that he sees a lot about dinosaurs and what they were, but nothing about where they came from. Did he bother to do the most basic research, or just see if it was presented to him when he watched a documentary? The origins of the dinosaurs is no mystery. We know the earliest dinosaurs were creatures like Compsagnathus. These evolved from a group of primitive archasaurs, which evolved from early diapsids. Question answered. There are many books out there that can list specific species and show cladograms of the appearance of dinosaurian traits.
Basically, this question is exactly the same as his previous question, where is the fossil evidence? He says we should have thousands of millions of transitions, which is the same thing he said before, exagerating the completeness of the fossil record. This really isn't a new question at all, just more specific and easier to answer.
Riddle again does a typical quote mine. He quotes several biologists as saying things like the origin of dinosaurs has been debated for decades, etc. Without even looking up the quotes, I can tell that these quotes are just introductory remarks to a detailed discussion of the evidence. The fact that there is some disagreement about exactly which species might be ancestral does not mean there is no evidence for the origin of dinoaurs.
Of course, another answer to this is, so what? If not the dinosaurs, he could find some other group that we are rather uncertain of their ancestry (the turtles come to mind). We will probabliy never have a complete ancestry of every group that has lived. Science doesn't have every answer, which is the basic idea Riddle doesn't seem to get. This question isn't even addressing the evidence for evolution, except indirectly on the issue of fossil intermediates. He again makes no attempt to engage in a discussion of the intermediates in dinosaur evolution or any other group.
Riddle's basic approach throughout his presentation is to point out places where there might be some uncertainty in science and say since they don't have every t crossed, evolution is false. He then says that creationists do have the answers, we know how and when and where things were created--basically God did it. He really does consider that an answer. When it comes to the question of how, he says the Bible tells us--God said it, and it was. That is how. He considers that a complete answer, but is bothered by the incompleteness of the scientific answer. The answer to all four of his questions is the same three words, God did it. Somehow, he doesn't see that this isn't an explanation at all and is simply an argument from ignorance. No attempt is made at all to actually engage or explain the patterns we see. God could have created any pattern for life. The fact that we see these patterns is just because.
Riddle does finish with the inevitable story of his being born again. It seems that every creationist was once an evolutionists before they saw the light. After carefully looking at the evidence, they were convinced by creation (it was an objective look, uncolored by their religious beliefs). And he sees the Bible as all or nothing: if you can't trust the first chapter of the Bible you can't trust any of it (and by trust he means read it litteraly).
I was disappointed by this video. It's not like I expected something new, but I thought if you were to condense it down to the four best questions, there should be something better than a few arguments from ignorance and outright denial of the fossil record.
Basically, this question is exactly the same as his previous question, where is the fossil evidence? He says we should have thousands of millions of transitions, which is the same thing he said before, exagerating the completeness of the fossil record. This really isn't a new question at all, just more specific and easier to answer.
Riddle again does a typical quote mine. He quotes several biologists as saying things like the origin of dinosaurs has been debated for decades, etc. Without even looking up the quotes, I can tell that these quotes are just introductory remarks to a detailed discussion of the evidence. The fact that there is some disagreement about exactly which species might be ancestral does not mean there is no evidence for the origin of dinoaurs.
Of course, another answer to this is, so what? If not the dinosaurs, he could find some other group that we are rather uncertain of their ancestry (the turtles come to mind). We will probabliy never have a complete ancestry of every group that has lived. Science doesn't have every answer, which is the basic idea Riddle doesn't seem to get. This question isn't even addressing the evidence for evolution, except indirectly on the issue of fossil intermediates. He again makes no attempt to engage in a discussion of the intermediates in dinosaur evolution or any other group.
Riddle's basic approach throughout his presentation is to point out places where there might be some uncertainty in science and say since they don't have every t crossed, evolution is false. He then says that creationists do have the answers, we know how and when and where things were created--basically God did it. He really does consider that an answer. When it comes to the question of how, he says the Bible tells us--God said it, and it was. That is how. He considers that a complete answer, but is bothered by the incompleteness of the scientific answer. The answer to all four of his questions is the same three words, God did it. Somehow, he doesn't see that this isn't an explanation at all and is simply an argument from ignorance. No attempt is made at all to actually engage or explain the patterns we see. God could have created any pattern for life. The fact that we see these patterns is just because.
Riddle does finish with the inevitable story of his being born again. It seems that every creationist was once an evolutionists before they saw the light. After carefully looking at the evidence, they were convinced by creation (it was an objective look, uncolored by their religious beliefs). And he sees the Bible as all or nothing: if you can't trust the first chapter of the Bible you can't trust any of it (and by trust he means read it litteraly).
I was disappointed by this video. It's not like I expected something new, but I thought if you were to condense it down to the four best questions, there should be something better than a few arguments from ignorance and outright denial of the fossil record.
Subscribe to:
Comments (Atom)
