Overwhelmingly strong proofs of intelligent and benevolent design lie around us ... the atheistic idea is so nonsensical that I cannot put it into words. --Lord Kelvin (1824-1907)

The Myth of the Blind Watchmaker

Richard Dawkins is widely regarded as the leading popularizer of the Neo-Darwinian Theory of Evolution today. His books on evolution are widely praised, both by evolutionist scholars and the news media. The Economist called his book The Blind Watchmaker (1986,1996), "As readable and vigorous a defense of Darwinism as has been published since 1859," and a prominent evolutionist called it, "the best general account of evolution I have read in recent years." Not surprisingly, Dawkins is also a radical atheist who gives lectures on British television bearing such provocative titles as "The God Delusion" and "The Virus of Faith."

The Neo-Darwinian Theory of Evolution is, of course, the modern theory that all life evolved by purely naturalistic mechanisms with no supernatural involvement or intelligent design of any kind. According to the theory, all life on earth originated from a single living cell. How that first living cell came to be is a continuing mystery but is technically outside the scope of evolution. The theory says that random mutations of DNA occur, and the mutations that happen to be beneficial are effectively "selected" by nature because they improve the organism's chances of survival. Thus, harmful mutations tend to culled out by death and extinction, but beneficial mutations are propagated through the eons to produce increasingly complex life forms. That, in a nutshell, is how the Theory of Evolution explains all life beyond the first living cell.

Chapter Three of The Blind Watchmaker is called "Accumulating small change." In this chapter, Dawkins attempts to explain how the amazing complexity of living organisms could have evolved by purely naturalistic mechanisms with no design. Obviously it could not have happened in a few large mutations. What is needed is a long series of small mutations.

To illustrate that point, Dawkins starts with a simple analogy of generating a short sentence by typing randomly. He chooses the sentence, "METHINKS IT IS LIKE A WEASEL" from Hamlet, which contains 28 characters. If we consider only upper case letters and the space, we have 27 characters to choose from. Now, if 28 characters are typed completely at random, the chances of getting exactly this sentence in one try are one in 27 raised to the 28th power (approximately one in 10 to the 40th power). If we keep trying, but start over from scratch each time, the chances of typing even this short sentence are virtually zero.

Ah, but that is not how evolution works, of course. It does not keep starting over; it builds on what is already established. So Dawkins refines his model to reflect this difference. The resulting simulation procedure is to start with the first random try, then produce copies, with a small random error introduced in each copy (i.e., a different error in each copy) to simulate mutations. The simulation then selects from each "generation" the mutated copy that matches the target sentence most accurately. Lo and behold, the simulation now converges on the target sentence after only a few dozen "generations."

The problem with this little pedagogical simulation should be obvious to the alert reader. Yes, of course it is way oversimplified, but that is not the fundamental problem. The fundamental problem is that the simulation requires a target state that is known in advance. In real evolution, where would such a target state be stored? A pre-specified target state implies an intelligence that is categorically ruled out by the Theory of Evolution. At this point in the book, the alert reader should be wondering in amazement if Dawkins could have really made such a monumental blunder!

After another page of rambling, Dawkins fesses up and admits that his simple simulation is "misleading in important ways," and that in reality, "Evolution has no long-term goal." In other words, his simplified simulation flagrantly violates the central premise of the Theory of Evolution, namely the lack of need for any intelligence. Of what value, then, is his little simulation? It's only function is to grossly and deceptively exaggerate the capacity of unguided naturalistic evolution to build up information gradually. Yes, Dawkins ultimately admitted in passing that his simulation was "misleading" (otherwise he would have made a fool of himself), but how many readers realize that it was rigged to violate the most important tenet of evolution?

Dawkins is just getting started. For his next trick, he programs another simplified pedagogical simulation, but this time the output consists of line drawings. He uses a recursive "tree-growing" procedure that starts with a single vertical line, then branches symmetrically into two lines, which each branch into two lines again, and so on. The parameters of the procedure, such as the depth of recursion, and the lengths and angles of the lines, serve as the "genes." Dawkins defines nine such parameters and shows examples of the resulting figures obtained by varying them. He expresses fascination with the diversity of the resulting figures.

So far his little procedure constitutes a highly simplified analogue of reproduction with genetic mutation, but it has nothing comparable to natural selection. Dawkins approaches that aspect by letting the user select which of the "children" of each generation will survive. Thus, the user (i.e., Dawkins) can guide the evolution process however he sees fit. That constitutes artificial rather than natural selection, of course, but Dawkins points that out clearly and explains why natural selection would be difficult for him to simulate. Fair enough.

What is the fundamental problem with this little simulation? No, the fundamental problem is not that it is oversimplified or that it uses human-guided artificial selection. The fundamental problem is that it simply does not model or test in any way the most important problems of evolution. It models competition for survival, but it completely ignores the more fundamental and infinitely more complex problem of basic biological viability. What if the metabolic system, or any of a dozen other major systems, has a glitch that causes extinction? A million things can go wrong and cause extinction before "competition" even becomes an issue, but Dawkins completely ignores those problems in this little simulation.

In Dawkins' simulation, global extinction cannot occur unless the user somehow neglects to select any survivors at some stage of the game. How could such an arbitrary selection procedure possibly model the problem of basic biological functioning -- eating, digestion, metabolism, respiration, circulation, perception, motor control, etc.? Obviously, it cannot. Nor can it model in any way the problem of a biologically viable transition path for macro-evolution. In the world of line drawings, any figure that can be drawn is "viable." Of what value, then, is Dawkins' line-drawing simulation? As with the earlier sentence-generating simulation, it's only function is to grossly and deceptively exaggerate the capacity of unguided naturalistic evolution to build up information gradually.

Are we trying to read too much into Dawkins' simulations? After all, they are merely oversimplified pedagogical models intended to illustrate that complexity can be built up gradually. The problem is that those models and simulations are able to illustrate the intended point only by violating the basic premise of naturalistic evolution or by completely ignoring the real problems. They mislead rather than illuminate. The deceptive tactics are rather obvious once they are pointed out, but Dawkins has nevertheless built a hugely successful career and reputation on the basis of such sophistry. Dawkins is either trying to fool others or has managed to fooled himself too with his phony little computer exercises.

How would one go about developing a simple but honest first-order analysis or simulation to test naturalistic evolution? According to the theory, beneficial mutations are "selected" by virtue of the fact that they improve the probability that the organism will survive to reproduce and propagate its genes. The other side of the equation is that harmful mutations obviously work against survival. So a key input to even the simplest evolution model or simulation would have to be the ratio of beneficial to harmful mutations (neutral mutations will be ignored here because, by definition, they have no significant effect). This ratio is analogous to the signal-to-noise ratio (SNR) of a radio signal. If the SNR is too low, no radio-frequency filter (tuner) can extract the signal from the noise. Similarly, if the ratio of beneficial to harmful mutations is too low, the natural-selection "filter" will not be able to produce biologically viable organisms.

Like most evolutionists, Dawkins barely mentions this ratio. He discusses the overall mutation rate extensively, but the significance of the ratio of beneficial to harmful mutations doesn't seem to particularly interest him. He never provides any actual estimate of the ratio except for the case of "small" mutations, which he asserts have equal chance of causing harm or benefit. But clearly the rate of harmful mutations exceeds the rate of beneficial mutations -- people do not deliberately expose themselves to mutagenic chemicals hoping the mutations will improve their health! Common sense suggests that the ratio of harmful to beneficial mutations is probably rather high. Imagine a random bit flip in the binary executable code of a computer operating system (e.g., Linux). What are the chances that it will improve the functioning of the system? Obviously very small. And what are the chances that it will be harmful? Obviously much higher. Perhaps several orders of magnitude higher.

Another important consideration is the degree of harm or benefit that a mutation can bring. A harmful mutation in a critical place can be disastrous, but a beneficial mutation is extremely unlikely to have a huge positive effect. And even if a wondrously beneficial mutation occurred, it would obviously not offset the effect of a disastrous mutation that causes or contributes to extinction.

The other consideration that is essential to an honest analysis or simulation of naturalistic evolution is some kind of estimate of the "sharpness" of the natural-selection "filter." A beneficial mutation may increase the probability of survival and reproduction only slightly (e.g., from 26.2% to 26.3%). The filter is then not very sharp, and it certainly does not guarantee the propagation of the mutation. Given the relative rarity of beneficial mutations to start with, the natural-selection filter may not be sharp enough to "amplify" the beneficial mutations above the noise level of the harmful mutations. The result could be global extinction -- or the absence of any evolution to start with -- which Dawkins' phony models and simulations do not even allow.

Since macro-evolution has never been directly observed either in nature or in the laboratory, a simulation test could potentially be useful. Like many evolutionists, however, Dawkins provides no indication that he has any clue about the most basic concepts involved. Evolutionists have a habit of simply declaring that Intelligent Design is "unscientific," or that "no evidence exists" in support of it, apparently thinking that arrogance trumps common sense. So why would they think evolution even needs to be tested? If a simulation showed that it is not viable, you can be sure they would simply reject the results anyway, or fudge it until it corroborated their preconceived notions. Such is the closed-minded mentality of all too many evolutionists.

In his book Not By Chance!, Lee Spetner presents an illuminating mathematical analysis of evolution. Spetner is a professor emeritus of physics from MIT who specialized in information theory. In this book, Spetner points out that, in order to build up information in small steps, each step must add information on average. But few if any mutations have ever been discovered that add information. Virtually all known beneficial mutations in bacteria, for example, reduce sensitivity to antibiotics by actually losing information. Spetner's mathematical analysis from first principles contrasts sharply with Dawkins' method of working back from a preconceived conclusion (or, equivalently, working forward from the premise of naturalism). And Spetner's analysis demonstrates that the Neo-Darwinian Theory of Evolution is not even close to mathematical viability. "The deck is stacked," as he puts it. But don't hold your breath waiting for evolutionists to concede that a mere mathematical analysis can trump their naturalistic dogma.

Chapter 6 of The Blind Watchmaker is called "Origins and miracles," and here Dawkins addresses the origin of the first living cell. As mentioned earlier, this topic is technically outside the scope of evolution, but it is obviously critical to the notion of pure naturalism. Why should intelligent design be ruled out for evolution after the first living cell if it cannot be ruled out in explaining the origin of that first cell? As an atheistic naturalist, Dawkins cannot concede the need for intelligent design at any stage in the origin and evolution of life. The problem for Dawkins is that the simplest known living cell is extremely complex, perhaps surpassing the complexity of all modern technology combined, and "conventional" biological natural selection does not apply because reproduction does not begin until that cell exists. Modern science is not even close to explaining how the first living cell could have come to be by purely naturalistic mechanisms.

Dawkins is nevertheless undaunted. He spends much of Chapter 6 citing speculation about how natural selection at the chemical level might have gradually built up the staggering complexity of the simplest known living cell. Speculation is perfectly reasonable, of course -- even wild speculation, which is what Dawkins engages in here. He certainly has very little if any actual evidence to support his speculation, but lack of empirical evidence is no problem for Dawkins. By the end of the chapter, he confidently proclaims that, "This chapter has had the modest aim of explaining only the kind of way in which it must have happened." In other words, we have no plausible naturalistic explanation for the origin of life, but Dawkins knows for fact that it "must have happened" without any intelligent guidance. And how does Dawkins know that? Because his faith in atheistic naturalism trumps the empirical evidence, of course.

Dawkins continues to say that, "The present lack of a definitely accepted account of the origin of life should certainly not be taken as a stumbling block for the whole Darwinian worldview, ..." And why shouldn't it? He is minimizing the problem because the lack of a naturalistic explanation for the origin of life is an embarrassment to his atheistic worldview. The problem is not that we just "don't know yet" how the first living cell originated; the problem (for atheists) is that we virtually know that it couldn't have originated by purely a naturalist mechanism. Yet for some ideological reason we are required to believe that the development of life after the first living cell was purely naturalistic. And if you don't believe it, you will be ridiculed by Dawkins and his followers.

Many great scientists of the past, including Newton, Pascal, Maxwell, Faraday, Henry, Kelvin, and Pasteur, were devout Christians who believed that the job of a scientist is to understand the natural laws and designs of the Creator. In Dawkins world, however, life itself is fundamentally nothing more than a complicated mechanism by which genes propagate themselves, as he explains in his book The Selfish Gene. In the end, Dawkins' radical atheism renders him incapable of objectively evaluating the Theory of Evolution. The fact that such a huckster is so revered today is a sad commentary on the state of modern science.

The more I study nature, the more I am amazed at the work of the Creator. --Louis Pasteur (1822-1895)

Suggested Reading