Rebuttals to Common Criticisms
of the Book Darwin's Black Box

Robert DiSilvestro, Ph.D. Biochemistry
Associate Professor, Human Nutrition,
The Ohio State University

DiSilvestro Dr Robert DiSilvestro holds a Ph.D. in biochemistry from Texas A&M University. He is currently professor of human nutrition at Ohio State University. Dr. DiSilvestro has published nearly 60 peer reviewed, life science-related research articles.

In the book Darwin's Black Box, Michael Behe contends that Darwinian evolution cannot explain the origin of complex biochemical systems. The central idea is that such systems could not evolve gradually because they need all their parts to function (a principle termed “irreducible complexity”). Behe proposes that intelligent design is needed to produce these systems. This idea has drawn many criticisms. Some are dressed up with impressive sounding biochemical details; others are based on more general reasoning. However, each major criticism can also be critiqued. These critiques leave Behe’s ideas of irreducible complexity and intelligent design still standing as viable options.


The book Darwin's Black Box, by biochemistry professor Michael Behe, has challenged the idea that Darwinian evolution explains many of the complex biochemical systems we see today. Instead, Behe proposes, these systems are the result of intelligent design. The book's message has been received with great enthusiasm by many people, including many evangelical Christians. On the other hand, much of the biological research community has dismissed the book. This dismissal is based on various criticisms. Sometimes, the criticisms are made fairly simply, while at other times, they are dressed up with many complex biochemical details. Whether the objections sound complicated or not, the most common objections distill down to about eight basic criticisms. I contend that these criticisms are flawed. Below, I provide a very short summary of the basic ideas of the book, and then briefly describe the eight main criticisms and give my rebuttals to them.

Basic Premise of the Book: When Darwin made his propositions, he did not know the degree of complexity involved in life's molecular systems. Now, we know that such systems are extremely intricate. Author Michael Behe says that many of these systems show "irreducible complexity." That means that if any one of the many components is removed, then the system breaks down. This goes against a basic idea of evolutionary theory. This idea says that systems start out in a simple form, and then, driven by natural selection, gradually get better and better. In contrast, irreducible complexity says that the system is useless until it reaches, or comes close to its final form. Thus, natural selection could not drive evolution "from scratch." A reasonable alternative is that intelligent design accounts for complex biological systems.

Basic Analogy. Behe compares the situation to a mouse trap. All the parts must be present to catch mice. One cannot start with the spring and catch some mice, then add the wood base and catch more mice, etc. Rather, there is a minimal structure required to catch any mice at all.

Criticism # 1. This is a new version of the God of the gaps. Once, when science could not explain how something in nature worked, people would just say: "God does it." Now, as science has progressed, we don't need God to explain such gaps.

Rebuttal to criticism # 1. There is a difference between explaining ongoing natural processes and the origins of natural processes. To use an analogy, just because a software program runs without the creator being present doesn't mean we should say there was no personal creator. Also, evolutionary theory has its own gaps that are assumed to occur without support by direct evidence. This is not to say that a broad theory of science has to explain every detail before being accepted. However, when it comes to evolutionary theory, far more gaps are accepted than are typical for other scientific theories.

There is another reason why Dr. Behe's ideas should not be equated to the God of the gaps idea. In the past, the gaps were generally due to lack of information about certain natural processes. In contrast, Dr. Behe's ideas involve processes where we do have information. That information, though not complete, is sufficient to indicate problems with postulating completely naturalistic explanations.

Criticism # 2. Gene duplication provides the complexity. According to this idea, an organism produces a copy of a gene it already has. The original gene keeps the original system running, while the newly copied gene evolves to eventually improve the system. This process is repeated often until there develops a complicated system involving many genes. For example, in today's human being, blood clotting, an example used in Behe's book, is a very complicated process. However, this process uses many genes, which have similarities in part of their structure. Therefore, at one time, blood clotting used just a few genes, but these duplicated and formed new genes that now work together.

Rebuttal to criticism # 2. To develop the specialized functions, the duplicated genes still had to evolve structural changes. What drove the changes? In all likelihood, a number of specializations would have had to develop simultaneously to have any value. This brings everything back to the mouse trap analogy. The only refinement is that some parts of the mouse trap would have some structural similarities.

An additional concern here is the high probability of the evolving genes messing up the original system. This is very likely with an abundance of structurally similar gene products. If one of these gene products becomes nonfunctional, it could get in the way of the function of original gene product. This phenomena is readily observed today. For instance, the chemotherapy drug methotrexate looks like the B-vitamin folacin, but does not work like it. The drug will compete with the real vitamin for binding to functional sites, but will not actually function. This action kills cancer cells.

Another problem with gene duplication is that it doesn't account for all, or even most, of the complexity in many systems. For example, the complexities of oxygen transport involve many genes which are not structurally similar. This is obvious when one considers anemia, a breakdown in oxygen transport. When I teach nutrition courses, I sometimes ask: how many different mechanisms can cause anemia? There are many causes which involve molecules with little or no overlap in structure.

Criticism # 3. Evolution can create systems from genes that are already around for other purposes. One example is a gene for a specialized function in insects, which is similar to a gene used for a different reason in mammalian systems. By analogy, this is like a third world mechanic scavenging parts from various types of machines to create a "Rube Goldberg type" of motorized vehicle.

Rebuttal to criticism # 3. How did the genes evolve for the other purposes? This just pushes the question of irreducible complexity backwards from one system to some other. The machine analogy doesn't work because the original machine was designed, as is the new machine. Another problem with the insect/mammal example is that one gene doesn't equal a whole system. So what if one gene in an insect system looks like a gene in a different mammalian system? What about all the genes in the latter system that have no resemblance to genes from insect systems? Instead of the Third World mechanic analogy, I have an alternative analogy. An intelligent designer of nature is like a US engineer using a similar connector wire in both a lawn mower and a motorcycle.

Criticism # 4. Some steps of evolution are no longer seen, but were there before a system looked irreducibly complex. By analogy, once the scaffolding used to make a building is removed, you can't tell it was ever there. Another analogy says that final computer programs don't show the steps discarded before reaching final form.

Rebuttal to criticism # 4. In these analogies, the progress toward the final product is driven by a committed, intelligent designer. These analogies are irrelevant to the development of biological systems formed by undirected processes. There is also another problem here. One still has to address the origin of some minimal amount of complexity needed for the system to work at all. Some people counter this rebuttal with the next criticism.

Criticism # 5. Some seemingly complex systems initially worked at a simpler level, which eventually evolved to a more complex, even an irreducibly complex system. This criticism often includes the gene duplication idea discussed above. So, blood clotting could work at a simple level with just a few proteins.

Rebuttal to criticism # 5. These are just so stories (I'll tell you a story, but not give reasons why it happened). These stories require many jumps in complexity for which there is no rationale for how natural selection could work. Often, these stories are backed up by looking at interspecies variations in protein amino acid contents. Using this data, an evolutionary scenario is constructed for how these changes evolved. However, these scenarios do not typically provide evidence as to how and why these scenarios were likely to occur.

Sometimes, the "just so" stories are backed by computer models using random combinations and selection advantages. However, the computer models are always too simplistic compared to biological systems. Also, many of these programs rely on intelligent input that natural selection could not duplicate. These attempts at computer modeling also still don't address an old problem. Many years ago, this problem was noted by a group of mathematicians. They could not derive a probability pattern that would make natural selection a likely means for evolution of complex biological systems. This contention has never been overturned.

Criticism # 6. We have examples of criticism # 5 (simpler versions of more complex biological systems). For instance, simple eye-like visual structures are seen in lower life forms. Therefore, these simpler systems evolved into more complex versions.

Rebuttal to criticism # 6. In nature, simpler versions of more advanced systems are structural dissimilar in many ways. Thus, primitive eye-like structures in lower life forms are not scaled down versions of human eyes, but are totally different systems.

Criticism # 7. Today, we can see examples of genetic evolution that support some of the above criticisms. One example is drug resistance development in bacteria. A drug can kill most of a group of bacteria, but the few survivors are genetically resistant to the drug. These bacteria multiply and produce a lot of bacteria which have drug resistance (a trait not common in the original bacteria population). Another example is changes in the colors of the peppered moth. Different colors have predominated at different times, depending on what color best concealed moths from their predators. At one time, the predators had a hard time finding moths of a color that blended well with foliage. Later, air pollution made a different color a better camouflage. Later still, as pollution receded, the first color again became advantageous.

Rebuttal to criticism # 7. None of this has anything to do with developing new complex systems. The change in moth colors does not involve the development of any new traits, only changes in proportions. Natural selection works here because there are already functional traits. This says nothing about the ability of natural selection to develop a complex system "from scratch." The same is true for drug resistance, which is another fine tuning of already developed systems. To make matters worse, drug resistance development requires the killing of millions of organisms just to make one change. This rules out application to almost any animal species, especially in evolution models where only a few members of a species evolve to a new one (ie punctuated equilibrium models).

Criticism # 8. "I don't know how it could happen" doesn't equal "It couldn't happen." Behe states that he can't imagine how some system evolved. However, that doesn't mean that the system couldn't have evolved. A scientist should not give up so easily on natural explanations, especially in light of powerful new technology.

Rebuttal to criticism # 8. The technology issue is at best, a break even situation. We don't know whether new technology will give evidence for or against Behe's ideas.

There is another problem with the basic assumption of criticism # 8. An "I don't know how" statement is sometimes considered impressive. For instance, I work with copper enzymes which take electrons from oxygen or superoxide. If someone proposed that some copper enzyme gives electrons to oxygen, I would say: "I don't know how that could work." That's a reasonable response because the proposal has some theoretical problems, and it goes against current observational precedent. The same can be said for the origins of complex biological systems. Saying that these developed by chance goes against the precedents we have from every day life. For example, we can tell the difference between a painting of an alligator and a cloud pattern that looks a little like an alligator. The degree of complexity tells us one is designed and one is not. Similarly, we know a rock bridge can form by natural processes alone, but the Golden Gate Bridge was designed.

Some will say that this analogy is true for art and engineering, but not for biological systems. Even though a fine painting or the Golden Gate Bridge obviously didn't just happen by chance, biological systems have ongoing processes that don't occur with paints or bridge components. I agree that this contention has some truth. However, as noted above, even in biological systems, undirected evolution of functional complexity still has theoretical problems (ie the math situation mentioned above). Therefore, I still contend that complexity, whether in biological or nonbiological systems, can give evidence for design.

One more comment. Some scientists simply say that an intelligent design hypothesis has no place in science. However, this assumption is generally made without scientific support as to why it is true. In fact, the search for intelligent design is used all the time in certain aspects of science. For instance, forensic science tries to discern if a possible crime was an accident or "designed." Similarly, archeological science looks for evidence of design and seeks the identity of the designers.

Some people counter this argument by saying that for archeology or crime cases, we know there could be an intelligent person behind the observation (ie we know that people commit crimes or make artifacts). On the other hand, in the case of living systems, we haven't ever seen one originated by a personal designer.

One flaw with this response is that the same criticism can apply to saying evolution produced very diverse creatures. We have never seen a new creature arise simply by random, natural processes. In making this statement, I am excluding small developments created by artificial breeding. This would include fruit fly "species" that cannot mate with each other, as well as some plant breeding phenomena. These developments involve very small effects compared to what must happen to start with slime molds and evolve biochemists and race car drivers.

There is another flaw with the "we never saw this" statement. Just because one has never seen a divine designer, doesn't automatically mean one does not exist. Unless there is reason to rule out the possibility of a designer for living species, then one is possible.

© Copyright 1999, Robert DiSilvestro. All rights reserved.