- STAFF PICKS
- GIFTS + GIFT CARDS
- SELL BOOKS
- FIND A STORE
Used Trade Paper
Ships in 1 to 3 days
More copies of this ISBN
The Edge of Evolution: The Search for the Limits of Darwinismby Michael J Behe
Life on earth developed over billions of years by utter chance, filtered through natural selection. So says Darwinism, the most influential idea of our time. If a rare random mutation in a creature's DNA in the distant past helped the lucky mutant to leave more offspring than others of its species, then as generations passed the species as a whole would have changed. Incessant repetition of this simple process over eons built the wonders of biology from the ground up, from the intricate molecular machinery of cells up to and including the human mind.
That's the claim, at least. But is it true? To answer that question, Darwin's theory has to be sifted carefully, because it isn't just a single concept — it actually is a mixture of several unrelated, entirely separate ideas. The three most important ideas to keep straight from the start are random mutation, natural selection, and common descent.
Common descent is what most people think of when they hear the word "evolution." It is the contention that different kinds of modern creatures can trace their lineage back to a common ancestor. For example, gerbils and giraffes — two mammals — are both thought to be the descendants of a single type of creature from the far past. And so are organisms from much more widely separated categories — buffalo and buzzards, pigs and petunias, yaks and yeast.
That's certainly startling, so it's understandable that some people find the idea of common descent so astonishing that they look no further. Yet in a very strong sense the explanation of common descent is also trivial. Common descent tries to account only for the similarities between creatures. It says merely that certain shared features were there from the beginning — the ancestor had them. But all by itself, it doesn't try to explain how either the features or the ancestor got there in the first place, or why descendants differ. For example, rabbits and bears both have hair, so the idea of common descent says only that their ancestor had hair, too. Plants and animals both have complex cells with nuclei, so they must have inherited that feature from a common ancestor. But the questions of how or why are left hanging.
In contrast, Darwin's hypothesized mechanism of evolution — the compound concept of random mutation paired with natural selection — is decidedly more ambitious. The pairing of random mutation and natural selection tries to account for the differences between creatures. It tries to answer the pivotal question, What could cause such staggering transformations? How could one kind of ancestral animal develop over time into creatures as different as, say, bats and whales?
Let's tease apart that compound concept. First, consider natural selection. Like common descent, natural selection is an interesting but actually quite modest notion. By itself, the idea of natural selection says just that the more fit organisms of a species will produce more surviving offspring than the less fit. So, if the total numbers of a species stayed the same, over time the progeny of the more fit would replace the progeny of the less fit. It's hardly surprising that creatures that are somehow more fit (stronger, faster, hardier) would on average do better in nature than ones that were less fit (weaker, slower, more fragile).
By far the most critical aspect of Darwin's multifaceted theory is the role of random mutation. Almost all of what is novel and important in Darwinian thought is concentrated in this third concept. In Darwinian thinking, the only way a plant or animal becomes fitter than its relatives is by sustaining a serendipitous mutation. If the mutation makes the organism stronger, faster, or in some way hardier, then natural selection can take over from there and help make sure its offspring grow numerous. Yet until the random mutation appears, natural selection can only twiddle its thumbs.
Random mutation, natural selection, common descent — three separate ideas welded into one theory. Because of the welding of concepts, the question, Is Darwinism true? has several possible answers. One possibility, of course, is that those separate ideas — common descent, natural selection, and random mutation — could all be completely correct, and sufficient to explain evolution. Or, they could all be correct in the sense that random mutation and natural selection happen, but they might be inconsequential, unable to account for most of evolution. It's also possible that one could be wholly right while the others were totally wrong. Or one idea could be right to a greater degree while another is correct to a much lesser degree. Because they are separate ideas, evidence for each facet of Darwin's theory has to be evaluated independently. Previous generations of scientists readily discriminated among them. Many leading biologists of the late nineteenth and early twentieth centuries thought common descent was right, but that random mutation/natural selection was wrong.
In the past hundred years science has advanced enormously; what do the results of modern science show? In brief, the evidence for common descent seems compelling. The results of modern DNA sequencing experiments, undreamed of by nineteenth-century
scientists like Charles Darwin, show that some distantly related organisms share apparently arbitrary features of their genes that seem to have no explanation other than that they were inherited from a distant common ancestor. Second, there's also great evidence that random mutation paired with natural selection can modify life in important ways. Third, however, there is strong evidence that random mutation is extremely limited. Now that we know the sequences of many genomes, now that we know how mutations occur, and how often, we can explore the possibilities and limits of random mutation with some degree of precision — for the first time since Darwin proposed his theory.
As we'll see throughout this book, genetic accidents can cause a degree of evolutionary change, but only a degree. As earlier generations of scientists agreed, except at life's periphery, the evidence for a pivotal role for random mutations is terrible. For a bevy of reasons having little to do with science, this crucial aspect of Darwin's theory — the power of natural selection coupled to random mutation — has been grossly oversold to the modern public.
In recent years Darwin's intellectual descendants have been aggressively pushing their idea on the public as a sort of biological theory-of-everything. Applying Darwinian principles to medicine, they claim, tells us why we get sick. Darwinian psychology explains why some men rape and some women kill their newborns. The penchant for viewing the world through Darwinian glasses has spilled over into the humanities, law, and politics. Because of the rhetorical fog that surrounds discussions of evolution, it's hard for the public to decide what is solid and what is illusory. Yet if Darwinism's grand claims are just bluster, then society is being badly misled about subjects — ranging from the cause of illnesses to the culpability of criminals — that can have serious real-world consequences.
As a theory-of-everything, Darwinism is usually presented as a take-it-or-leave-it proposition. Either accept the whole theory or decide that evolution is all hype and throw out the baby with the bath water. Both are mistakes. In dealing with an often-menacing nature, we can't afford the luxury of elevating anybody's dogmas over data. The purpose of this book is to cut through the fog, to offer a sober appraisal of what Darwinian processes can and cannot do, to find what I call the edge of evolution.
On the surface, Darwin's theory of evolution is seductively simple and, unlike many theories in physics or chemistry, can be summarized succinctly with no math: In every species, there are variations. For example, one animal might be bigger than its brothers and sisters, another might be faster, another might be brighter in color. Unfortunately, not all animals that are born will survive to reproduce, because there's not enough food to go around, and there are also predators of many species. So an organism whose chance variation gives it an advantage in the struggle to survive will tend to live, prosper, and leave offspring. If Mom or Dad's useful variation is inherited by the kids, then they, too, will have a better chance of leaving more offspring. Over time, the descendants of the creature with that original, lucky mutation will dominate the population, so the species as a whole will have changed from what it was. If the scenario is repeated over and over again, then the species might eventually change into something altogether different.
At first blush, that seems pretty straightforward. Variation, selection, inheritance (in other words, random mutation, natural selection, and common descent) seem to be all it takes. In fact, when an evolutionary story is couched as abstractly as in the previous paragraph, Darwinian evolution appears almost logically necessary. As Darwinian commentators have often claimed, it just has to be true. If there is variation in a group of organisms, and if the variation favorably affects the odds of survival, and if the trait is inherited, then the next generation is almost certain to have more members with the favorable trait. And the next generation after that will have even more, and the next more, until all members of the species have it. Wherever those conditions are fulfilled, wherever there is variation, selection, and inheritance, then there absolutely must be evolution.
So far, so good. But the abstract, naive logic ignores a huge piece of the puzzle. In the real world, random mutation, natural selection, and common descent might all be completely true, and yet Darwinian processes still may not be an adequate explanation of life. In order to forge the many complex structures of life, a Darwinian process would have to take numerous coherent steps, a series of beneficial mutations that successively build on each other, leading to a complex outcome. In order to do so in the real world, rather than just in our imaginations, there must be a biological route to the structure that stands a reasonable chance of success in nature. In other words, variation, selection, and inheritance will only work if there is also a smooth evolutionary pathway leading from biological point A to biological point B.
The question of the pathway is as critical in evolution as it is in everyday life. In everyday life, if you had to walk blindfolded from point A to point B, it would matter very much where A and B were, and what lay between. Suppose you had to walk blindfolded (and, to make the example closer to the spirit of Darwinism, blind drunk) from A to B to get some reward — say, a pot of gold. What's more, suppose in your sightless dizziness the only thought you could hold in your head was to climb higher whenever you got the chance (this mimics natural selection constantly driving a species to higher levels of fitness). On the one hand, if you just had to go from the bottom of a single enclosed stairwell to the top to reach the pot of gold, there might be little problem. On the other hand, if you had to walk blindfolded from one side of an unfamiliar city to the top of a skyscraper on the other side — across busy streets, bypassing hazards, through doorways — you would have enormous trouble. You'd likely stagger incoherently, climb to the top of porch steps, mount car roofs, and so on, getting stuck on any one of thousands of local high points, unable to step farther up, unwilling to back down. And if, just trying to climb higher whenever possible, you had to walk blindfolded and disoriented from the plains by Lubbock, Texas, to the top of the Sears Tower in Chicago — blundering randomly over flatlands, through woods, around canyons, across rivers — neither you nor any of billions of other blindfolded, disoriented people who might try such a thing could reasonably be expected to succeed.
In everyday life, the greater the distance between points A and B, and the more rugged the intervening landscape, the bleaker are the odds for success of a blindfolded walk, even — or perhaps especially — when following a simple-minded rule like "always climb higher; never back down." The same with evolution. In Darwin's day scientists were ignorant of many of the details of life, so they could reasonably hope that evolutionary pathways would turn out to be short and smooth. But now we know better. The great progress of modern science has shown that life is enormously elegant and intricate, especially at its molecular foundation. That means that Darwinian pathways to many complex features of life are quite long and rugged. The problem for Darwin, then, as with a long, blindfolded stroll outdoors, is that in a rugged evolutionary landscape, random mutation and natural selection might just keep a species staggering down genetic dead-end alleys, getting stuck on the top of small anatomical hills, or wandering aimlessly over physiological plains, never even coming close to winning the biological pot of gold at a distant biological summit. If that is the case, then random mutation/natural selection would essentially be ineffective. In fact, the striving to climb any local evolutionary hill would actively prevent all drunkards from finding the peak of a distant biological mountain.
This point is crucial: If there is not a smooth, gradually rising, easily found evolutionary pathway leading to a biological system within a reasonable time, Darwinian processes won't work. In this book we'll examine just how demanding a requirement that is.
As a practical matter, how far apart do biological points A and B have to be, and how rugged the pathway between them, before random mutation and natural selection start to become ineffective? How can we tell when that point is reached? Where in biology is a reasonable place to draw the line marking the edge of evolution?
This book answers those questions. It builds on an inquiry I began more than a decade ago with Darwin's Black Box. Then I argued that irreducibly complex structures — such as some stupendously intricate cellular machines — could not have evolved by random mutation and natural selection. To continue the above analogy, it was an argument that the blindfolded drunkard could not get from point A to point B, because he couldn't take just one small step at a time — he'd have to leap over canyons and rivers. The book concluded that there were at least some structures at the foundation of life that were beyond random mutation.
That conclusion stirred a lot of discussion. In particular, a lot of heat was generated in the scientific community by my inference that the structures are intelligently designed. Many people are viscerally opposed to that conclusion, for a variety of reasons. In this book, although my conclusions are ultimately the same, and will undoubtedly be opposed by some, I spend the bulk of the chapters drawing on molecular evidence, genomic research, and — above all
— crucial long-term studies of evolutionary changes in single-celled organisms to test Darwinism without regard to conclusions of design. Readers who cannot accept my final conclusions should still be able to consider the evidence presented in the bulk of these chapters, before taking issue with my conclusions in the final three chapters of the book. As I will argue, mathematical probabilities and biochemical structures cannot support Darwinism's randomness, except at the margins of evolution. Still, as we seek to find the line marking the edge of randomness, there is no need to infer design.
Darwin's Black Box was concerned to show just that some elegant structures in life are beyond random mutation and natural selection. This book is much more ambitious. Here the focus is on drawing up reasonable, general guidelines to mark the edge of evolution — to decide with some precision beyond what point Darwinian explanations are unlikely to be adequate, not just for some particular structures but for general features of life. This can be compared to the job of an archeologist who discovers an ancient city buried under sand. The task of deciding whether random processes produced things like intricate paintings on walls of the city buildings (perhaps by blowing sand) is pretty easy. After all, elegant paintings aren't very likely to be made by chance processes, especially if the paintings portray not just simple geometric patterns, but images of people or animals.
But once the cherry-picking is over, the going gets tougher. Are the dark markings at the side actually a part of a painting, or just smudges? Is a pile of stones next to an exterior wall a table or an altar of some sort, or just a random collection of rocks? Is ground near the wall the remnant of a tilled field? Where lies the border of the city? Where does civilization stop and raw nature begin? Deciding on marginal cases like those is harder work, and the conclusions will necessarily be more tentative. But at the end of the study the archeologist will be left with a much clearer picture of where the city leaves off and random natural processes take over.
In a way, archeologists have it easy. Although they have to worry about the effects of physical processes on artifacts they study, they don't usually concern themselves much with biological ones. In puzzling out where might lie the far boundaries of Darwinism, uniquely biological processes of course come strongly into play. Random mutations of DNA might be likened to random accidents that befall inanimate objects. But plants and animals reproduce, stones don't. Natural selection works on living objects, not on nonliving ones. Darwin's theory claims that random genetic accidents and natural selection working over eons will yield results that don't look at all like the effects of chance.
Life has been on earth for billions of years. During that time huge numbers of organisms have lived and died. Fierce struggles between different lineages over the ages are supposed by Darwinists to have led to biological "arms races" — tit-for-tat improvements of the capacity to wage biological warfare, analogous to the sophisticated twentieth-century arms race between humans in the United States and the Soviet Union. Maybe the results of those biological arms races were sophisticated living machinery, far beyond what we would ordinarily think of as the result of chance.
That's the theory. But it has proven extremely difficult to test adequately. Modern laboratory studies of random mutation/natural selection have suffered from an inability to examine really large numbers of creatures. Typically, even with heroic efforts by the best investigators, only a relative handful of organisms can be studied, only for a comparatively short amount of time, and changes in a few chosen traits are followed. At the end of such studies, while some interesting results may be at hand, it's usually impossible to generalize from them. Although scientists would love to undertake larger, more comprehensive studies, the scale of the problem is just too big. There aren't nearly enough resources available to a laboratory to perform them.
So, in lieu of definitive laboratory tests, by default most biologists work within a Darwinian framework and simply assume what cannot be demonstrated. Unfortunately, that can lead to the understandable but nonetheless corrosive intellectual habit of forgetting the difference between what is assumed and what demonstrated. Differences between widely varying kinds of organisms are automatically chalked up to random mutation and natural selection by even the most perceptive scientists, and even the most elegant of biological features is reflexively credited to Darwin's theory.
Breaking the theoretical logjam would require accurate evolutionary data at the genetic level on an enormous number of organisms that are under ceaseless pressure from natural selection. That data simply hasn't been available in the past. Now it is.
Even just ten years ago any attempt to locate the edge of evolution with any precision would have been well-nigh impossible. Too little was known. But with the relentless march of science, especially in the past decade, the task has become feasible.
A major difficulty of evaluating an evolutionary theory like Darwin's has been that, while we can easily observe large changes in animals and plants, the reasons for those changes are obscure. Darwin and other early scientists could examine, say, alterations of finch beaks, but they couldn't tell what was causing the modifications. Closer to our own day, mid-twentieth-century scientists could determine that some bacteria evolved resistance to antibiotics, but they didn't know exactly what gave them that power. Only in the past half century has science shown that visible changes are caused by mutations in invisible molecules, in DNA and proteins. The only way to get a realistic understanding of what random mutation and natural selection can actually do is to follow changes at the molecular level. It is critical to appreciate this: Properly evaluating Darwin's theory absolutely requires evaluating random mutation and natural selection at the molecular level. Unfortunately, even today such an undertaking is intensely laborious. Yet there is no other way.
The good news is that, with much effort and insight, modern science has developed the tools to do so. A triumph of twentieth-century science has been its elucidation of one requirement of Darwin's theory — the underlying basis of variation. We now know that variation in organisms depends on hidden changes in their DNA. (For a summary of DNA structure, see Appendix A.) What's more, scientists have catalogued myriad ways in which DNA can change. Not only can single units (called nucleotides) of DNA accidentally change when the DNA is copied in a new generation, but whole chunks of the double helix can accidentally either be duplicated or be left out. Very rarely all of the DNA in a cell is copied twice, yielding offspring with double the DNA of its parents. Other times active DNA elements resembling viruses can insert copies of themselves at new positions in the genome, sometimes dragging other bits of DNA with them. Opportunities for nature to alter an organism's DNA are virtually boundless.
Not only has the hard work of many scientists shown the underlying basis of variation, the rate of mutation has been worked out fairly well, too. As a rule, the copying of DNA is extremely faithful. On average, a mistake is made only once for every hundred million or so nucleotides of DNA copied in a generation. But there are exceptions. In some viruses such as HIV the mutation rate is speeded up enormously.
Another critical advance in our ability to properly test Darwinism has come from DNA sequencing. In the past few decades the amount of DNA sequenced has been growing exponentially, and the number of organisms studied by sequencing has been expanded. In the mid-1990s the first complete sequence of an organism's genome — a tiny bacterium named Hemophilus influenzae — was published. Now the sequences of hundreds of genomes are known. Not only whole genome sequencing, but the easy ability to sequence at least key pieces of an organism's DNA gives scientists the ability to nail down the molecular changes that underlie genetic diseases, or that cause resistance to antibiotics.
Yet all that scientific progress would still not be enough to draw reasonably firm conclusions about the abilities of Darwinian evolution if sufficient numbers of organisms couldn't be studied. The more organisms there are, the more opportunities random mutation has to stumble across a beneficial change and pass it on to natural selection, the firmer our conclusions about what Darwinism can do become. Studies of animals like finches can at best follow hundreds at a time. In the laboratory thousands of fruit flies might be examined. That's better, but still far from enough. With thousands or even millions of organisms, a mutation comes along relatively rarely, and few of the mutations that do come along are helpful.
The natural world of course teems with organisms. There can be billions of a mammalian species on the planet at a time, such as humans or rats. In the seas there are huge numbers of fish. And these represent just the larger forms of life. There are also untold numbers of microscopic entities such as bacteria and viruses. While laboratories can't grow enough creatures to get a reasonable handle on the abilities of Darwinian evolution, nature has no such problems.
Evolution from a common ancestor, via changes in DNA, is very well supported. It may or may not be random. Thanks to evolution, scientists who sequence human DNA and find mutations that are helpful — against, say, our natural enemies — are not just studying the DNA of one person. They are actually observing the results of a struggle that's gone on for millennia and involved millions and millions of people. An ancestor of the modern human first sustained the helpful mutation, and her descendants outcompeted the descendants of many other humans. So the modern situation reflects an evolutionary history involving many people. When scientists sequence a genome, they are unfurling rich evidence of evolution — Darwinian or otherwise — unavailable by any other method of inquiry.
Thanks to its enormous population size, rate of reproduction, and our knowledge of the genetics, the single best test case of Darwin's theory is the history of malaria. Much of this book will center on this disease. Many parasitic diseases afflict humanity, but historically the greatest bane has been malaria, and it is among the most thoroughly studied. For ten thousand years the mosquito-borne parasite has wreaked illness and death over vast expanses of the globe. Until a century ago humanity was ignorant of the cause of malarial fever, so no conscious defense was possible. The only way to lessen the intense, unyielding selective pressure from the parasite was through the power of random mutation. Hundreds of different mutations that confer a measure of resistance to malaria cropped up in the human genome and spread through our population by natural selection. These mutations have been touted by Darwinists as among the best, clearest examples of the abilities of Darwinian evolution.
And so they are. But, as we'll see, now that the molecular changes underlying malaria resistance have been laid bare, they tell a much different tale than Darwinists expected — a tale that highlights the incoherent flailing involved in a blind search. Malaria offers some of the best examples of Darwinian evolution, but that evidence points both to what it can, and more important what it cannot, do. Similarly, changes in the human genome, in response to malaria, also point to the radical limits on the efficacy of random mutation.
Because it has been studied so extensively, and because of the astronomical number of organisms involved, the evolutionary struggle between humans and our ancient nemesis malaria is the best, most reliable basis we have for forming judgments about the power of random mutation and natural selection. Few other sources of information even come close. And as we'll see, the few that do tell similar tales.
(Caveat lector: Unfortunately, in order to fully understand and appreciate the difficulties facing random mutation, and how humanity's battle with malaria illustrates them, we have to grit our teeth and immerse ourselves in details of the battle at the molecular level. I make every effort to keep technical details to a minimum, and some of them are confined to the appendices. But there is no way around the fact that this subject requires technical details.)
Although the number of malarial cells is vast, it's much less than the number of organisms that have existed on earth. Nonetheless, as I will explain, straightforward extrapolations from malaria data allow us to set tentative, reasonable limits on what to expect from random mutation, even for all of life on earth in the past several billion years. Not only that, but studies of the bacterium E. coli and HIV, the virus that causes AIDS, offer clear confirmation of the lessons to be drawn from malaria. HIV, in particular, is something of a Rosetta stone for studying random mutation, because such viruses mutate at extraordinary rates, ten-thousand times faster than the mutation rate of cells. Viruses contain much less genetic material, but it mutates so rapidly, and there are so many copies of it, that HIV alone, in just the past fifty years, has undergone more of at least some kinds of mutations than all cells have experienced since the beginning of the world.
Most of this book will focus on the operations of cells and molecules, but in the last two chapters I go further. In recent years, as science has progressed at an amazing clip, some molecular details underlying the development of different classes of animals have come to light. I make some inferences about the limits of the use of random mutation to explain features of animal life. In the final chapter I show that the conclusions reached in this book about random processes in biology mesh well with recent results from other scientific disciplines such as physics and cosmology. Together they illuminate the role of chance in nature as a whole.
One difficulty of writing a book questioning the sufficiency of Darwin's theory is that some people mistakenly conclude you're rejecting it in toto. It is time to get beyond either or thinking. Random mutation is a completely adequate explanation for some features of life, but not for others. This book looks for the line between the random and the nonrandom that defines the edge of evolution. Consider:
Somewhere in the middle of such examples lies the edge of evolution.
Copyright © 2007 by Michael J. Behe
What Our Readers Are Saying
Other books you might like
Religion » Christianity » Creationism
Religion » Spirituality » Science and Religion
Religion » World Religions » Religion and Science
Science and Mathematics » Biology » Evolution
Science and Mathematics » Chemistry » Biochemistry
Science and Mathematics » Nature Studies » Evolution