Michael J. Behe
The Edge of Evolution
2007, Free Press, 305 pages
I previously discussed reviews of Michael Behe’s book Darwin’s Black Box in 1999. Behe is a creationist, but not one in the traditional sense. He accepts the age of the universe and the fossil record, but he has issues with Darwin and the idea that random mutation and natural selection can account for biological evolution. In his previous book Behe argued that Darwin’s concept of evolution cannot explain the origin of a number of biochemical processes, the study of which is Behe’s professional field. Back then I summarized his idea:
Chemical processes that control such diverse life functions as blood clotting and disease immunity are exceedingly complex. Additionally, such processes are constructed like a house of cards in such a way that one missing card would bring down the whole business. Behe calls such systems “irreducibly complex.”1
The problem is Behe’s idea is considerably at odds with some known science. I noted some disagreements, including remarks by Donald C. Lindsay:
Behe doesn’t seem to be up to date. Although he implies on page 114 that he is expert at computer searches for scientific articles, he somehow managed to not find pretty well the entire literature on biochemical evolution. I personally own a textbook entitled Molecular Evolution, despite his claim that no such book exists.2
DBB was not our first encounter with Michael Behe. We met him in March 1992 when he participated in a conference at Southern Methodist University titled “Darwinism: Science or Philosophy.” Unfortunately I was absolutely clueless at the time and failed to recognize the crystallization of Intelligent Design that was unfolding before my eyes. Behe was completely beneath my radar on that day.
Thankfully for skeptics, Behe is back again, this time with a new book about an old idea. Will our cup ever run dry?
In DBB Behe pushed the idea that Darwinian evolution, as it manifests for cell chemistry, is a black box. The term black box relates to any mechanism whose external appearance and actions are well known, but nothing is known about its interior workings. In computer science a software process is typically designed as a block box. Its functions and its interfaces are carefully defined, but details of how the code performs its tasks are left up to the designer. Design, again.
Behe was not so much stuck on the black box concept in his previous book as he was on irreducible complexity. The biochemical processes Behe championed were deemed to be so critically constructed that they would not have been viable in a more primitive form. Therefore they could not have evolved by random mutation coupled with natural selection.
With DBB Behe made a big splash with creationists. Not such a big splash, however, where the rubber meets the road. When showdown time came, and he testified for Intelligent Design in the 2005 Kitzmiller creationism trial, he was forced to admit under cross examination he had not bothered to read the many books and scientific publications refuting his DBB claims.
DBB was round one. EoE appears to be round two.
I will not present an original review of this book. Others better at the matter are doing an excellent job of that. I will present some knowledge gleaned from existing reviews, and I will throw in some thoughts of my own.
The complete title of Behe’s new book is The Edge of Evolution: The Search for the Limits of Darwinism. Behe has not abandoned the black box in EoE. In his new book Behe even resuscitates the irreducibly complex bacterial flagellum. The bacterial flagellum that other scientists have explained appears not to be so irreducibly complex. Which explanations Behe fails to appreciate in his continued arguments for Intelligent Design and which explanations were explained to him again during his Kitzmiller cross examination.
What Behe is now beating the drums about is the limits of Darwinian evolution. The Intelligent Design movement casts a broad net to pull in support from religious fundamentalists, so it will come as a surprise to these creationists that Behe actually believes in evolution. In fact, he emphasizes his support for evolution a number of times in EoE. And that’s about as far as it goes.
There are limits to Darwinism, Behe asserts. Here is an example:
The structural elegance of systems such as the cilium, the functional sophistication of the pathways that construct them, and the total lack of serious Darwinian explanations all point insistently to the same conclusion: They are far past the edge of evolution. Such coherent, complex, cellular systems did not arise by random mutation and natural selection, any more than the Hoover Dam was built by random accumulation of twigs, leaves, and mud.3
Here is the basis for Behe’s main argument, and it’s an essential point of traditional evolutionary theory. Please refer to the figure below. The line represents the plot of an arbitrary mathematical function. Engineers, computer scientists, and others are often interested in extreme points on the graph. Of particular interest are greatest and least values of the function.
A mathematician possessing a formula for a well-behaved function can quickly locate maxima and minima by applying differential calculus.
If there is not a convenient mathematical formula describing the function, the problem gets harder, and computer scientists use numerical methods to locate maxima and minima. A computer program computes values of the function within a small region and then determines the general trend. If the goal is a maximum point, and the function seems to be headed north on the right side of the region, then the program will direct its search to the right of the region. The process continues until the program cannot detect any change in the value of the function across the region, and victory is declared. The program has found a maximum point.
What all of this has to do with evolution is that evolution is often compared to this mathematical process. If the survival fitness of a population can be compared to some sort of mathematical function, then members of the population having a higher fitness value (higher point on the plot) will prevail in the competition for survival. If any genetic change between successive generations produces a horizontal shift in the plot, then the lucky (or unlucky) heir to these traits will move up (or down) the slope of the plot and will either win or lose the next round of the competition.
In this sense, population shifts due to random mutation and natural selection are comparable to this computer process. Random mutations produce horizontal shifts along the plot, and natural selection locks in any resulting upward movement on the plot.
My diagram shows a case for a single random variable, the horizontal axis in the plot. Of course, populations are driven by multitudes of variables, but the process is extensible to any number of random variables acting simultaneously.
Behe has a couple of nice examples of plots with two independent variables on page 115 of his book. The issues are the same, but this time the goal is to find peaks in a two-dimensional surface.
Now here is the rub, as Behe points out. Suppose a population finds itself somewhere on the slope of the first peak on the left. A computer algorithm set to find the maximum value of the function will climb to the top of the left-most peak and stop there, stuck forever, unable to climb down and never able find the top of the highest peak just to the right.
Behe argues that this process will stymie the advancement of any population seeking to advance through natural selection. Natural selection, he asserts, is inadequate. The inescapable conclusion, according to Behe, is there must be some other process at work. He hints broadly at this process throughout the book. Here is an excerpt from a section titled “How deep goes design?”
Up until now we have examined molecular structures and processes and have drawn a tentative line marking the molecular edge of Darwinian evolution. Most protein-protein interactions in the cell are not due to random mutation. Since cells are integrated units, it’s reasonable to view cells in their entirety as designed. But keep in mind that accidents do happen, so there are Darwinian effects, of some degree, everywhere. For example, just as automobiles may accumulate dents or scratches over time or have mufflers fall off, but nonetheless are coherent, designed systems, so, too, with cells. Some features of cells of course result from genetic dents or scratches or loss, but the cell as a whole, it seems, was designed.4
OK, maybe not so broadly.
What Behe seems to be saying here is cells were designed, and Darwinian evolution only contributed accidental defects. Design made it right, evolution damaged it.
I admit to reading Behe’s book from beginning to end just so I could say I had. I was curious about what process Behe would invoke to explain design in nature. Apparently I was not alone. One of the reviewers of EoE is the high-profile biology professor and blogger P.Z. Myers. He has this to say:
It’s true. Nowhere in the entire book does he offer a mechanism to resolve this disconnect. He claims things were “designed”, but doesn’t explain by who [sic], how, or when, and doesn’t even give a clear picture of what parts of evolution are designed, and which aren’t. It’s nothing but one long and almost entirely fallacious gripe about the insufficiency of natural mechanisms.5
This seems to be a perpetual problem with Intelligent Design. Let me summarize what Intelligent Design really says:
1. Natural processes alone cannot produce the life forms we see today.
2. Therefore some sort of design process is at work.
3. This design process cannot involve natural processes alone. Else statement 1 would not be true.
4. Therefore at some point in the evolution of life some natural laws must have been violated.
Intelligent Design proponents pointedly do not emphasize statement number 4. Were they to do so, they might then be obliged to describe a scenario involving a supernatural process.
That seems to be the case with Behe in EoE. The book’s index includes only two links to the word “God,” involving only four pages. He will certainly not identify the God of Abraham as the designer. Creationists have been down that road before.
When school board member William Buckingham and other creationists lurched into promoting creationism in the Dover, Pennsylvania, science curriculum, they (figuratively) held the banner of God out in front. Even conservative federal judge John E. Jones III recognized this as a step toward a state religion and slapped the Intelligent Design movement down in a stinging 139-page decision.
Sensibly, Behe’s references to God in his book are less committal than would have been comfortable for Buckingham. Here is an example:
To reach a transcendent God, other, nonscientific arguments have to be made-philosophical and theological arguments. It is not my purpose here to rehearse what has been said over the millennia on that score, or to say why I myself find some of those arguments persuasive and others not. Here I’m content to “take purposeful ‘designer’ in a very broad sense.”6
Disregarding God, what mechanism does Behe propose to replace natural processes? First, he states he is not required to propose a mechanism. He does, however, make an attempt at supplying some detail:
…If random mutation is inadequate, then (since common descent with modification strongly appears to be true) of course the answer must be nonrandom mutation. That is, alternations to DNA over the course of the history of life on earth must have included many changes that we have no statistical right to expect, ones that were beneficial beyond the wildest reach of probability. Over and over in the past several billion years, the DNA of living creatures changed in salutary ways that defied chance.7
This explanation is not very soul-satisfying. Behe wants to hide God within the vagaries of chance. Not a good hiding place. Statistical probabilities may be the only place where pure mathematics and physical analysis truly intersect. Statistical probabilities explain why we must place a pot on a stove-top burner to cook food rather rely on heat from the air to suddenly migrate into the food. And that is the only explanation. When we find statistical probabilities being skewed, we wisely look for an underlying cause.
Permit me to provide a non-scientific example. In my favorite classic movie, Casablanca, casino owner Rick Blaine (Humphrey Bogart) wants a young refugee from Nazism to win at roulette. He tells the player to bet on 22, and he winks at the croupier. The player wins on the first spin, and Rick tells the player to let the pot ride for another spin. “Vingt et deux!” the croupier exclaims as 22 wins again. The croupier merely looks at Rick and shrugs. Do we suspect something nefarious is involved? Does Rick ever say “Here’s looking at you, Kid?”
The contention that we can hide purpose and design within mathematical probabilities is simply appalling. Paraphrasing Slim Pickens in another classic movie, “I’ve been to one world fair, a picnic, and a rodeo, and that’s the stupidest thing I ever heard coming from an educated scientist.”
In last month’s issue I discussed the philosophical weakness of the design argument. The point was, the only thing that seems to drive purpose and design is competition for survival within a population. It is purpose and design that have come about by random mutation coupled with natural selection. It is not the features of living organisms that have developed because of purpose and design.8
Scientists who have reviewed EoE have been more pragmatic. I shopped around my references for a quote that summarizes the assessment by mainstream science of Behe’s argument. I found this by Sean B. Carroll writing in the 8 June 2007 issue of Science:
Behe’s chief error is minimizing the power of natural selection to act cumulatively as traits or molecules evolve stepwise from one state to another via intermediates. Behe states correctly that in most species two adaptive mutations occurring instantaneously at two specific sites in one gene are very unlikely and that functional changes in proteins often involve two or more sites. But it is a non sequitur to leap to the conclusion, as Behe does, that such multiple-amino acid replacements therefore can’t happen. Multiple replacements can accumulate when each single amino acid replacement affects performance, however slightly, because selection can act on each replacement individually and the changes can be made sequentially.9
There are numerous serious reviews of EoE, and Wikipedia is a good place to start looking. The site offers numerous links to critiques of the book, both by mainstream scientists and by creationists. Here is the page:
2. Donald C. Lindsay at http://www.best.com/~dlindsay/creation/behe.html
3. EoE, page 102.
4. EoE, page 171.
6. EoE, page 229.
7. EoE, page 165.