Mark Perakh

In 1996 a professor of biochemistry at Lehigh University named Michael Behe, who is an  advocate of Intelligent Design (ID),  published a book Darwin’s Black Box1, where he propagandized the concept of “irreducible complexity” (IC). Behe and his ID colleagues have claimed that IC is a strong evidence of “design” of biological systems. The concept of IC rapidly acquired the status of one of the main pillars of the entire Intelligent Design “theory” (which essentially is one of the recent versions of creationism).

The concept of irreducible complexity was in fact known for many years before Behe’s book. The Nobel Prize winning biologist Hermann J. Muller had already discussed it (under the slightly different name of “interlocking complexity”) in 1918.2  Some 10 years before Behe’s book the same concept was explored by A. Graham Cairns-Smith3 .  These predecessors of Behe, unlike Behe, did not claim that the concept in question was a great discovery on a par with those by “Newton and Einstein, Lavoisier and Schroedinger, Pasteur, and Darwin" (as Behe asserted in Darwin’s Black Box).  Neither did they claim that “irreducible complexity” was a “marker” of a supernatural design.  To the contrary, according to Muller, development of interlocking complexity in biological systems has to be expected from Darwinian evolution. Therefore the concept in question, as such, evoked no resistance from mainstream science.

This concept, however, in Behe’s rendition met a strong resistance from the overwhelming majority of mainstream biologists. Perhaps the strongest display of the rejection of Behe’s thesis was the statement by the prominent biologist and author of popular textbooks of biology Kenneth Miller (who incidentally is a practicing Catholic as is Behe) who referred to Behe’s irreducible complexity as “nonsense.”4

As an example of an allegedly irreducibly complex system, Behe suggested in his book a mousetrap.  Soon afterwards, in multiple publications by various Intelligent Design advocates, images of a mousetrap were endlessly reproduced with a nauseating persistence.

The mousetrap, however, was not accepted by the mainstream scientific community as a genuine example of IC.  For example, professor of biology John McDonald suggested5 an animated illustration of how, starting with just a piece of a hook-shaped wire serving as a primitive mice-catching device, a full-fledged mousetrap can be gradually built up via two-part, then three-part, etc. contraptions, improving its mice-catching ability at each step, thus showing how poorly Behe’s example of an alleged IC system was chosen.

Apparently finally realizing that a mousetrap was not a very successful choice for illustrating Behe’s IC concept, ID advocates largely switched to another example. As the epitome of IC the ID creationists chose a bacterial flagellum.6  

A flagellum is a “device” used by bacteria for motility6.  Like the mousetrap before it, around 2002 the image of a flagellum had become a ubiquitous accompaniment to ID advocates’ books, papers, lectures, etc. According to one of the main advocates of ID William Dembski, the flagellum had become the “mascot” of ID. The image of the flagellum appeared on the cover of Dembski’s highly acclaimed by other ID advocates (but also severely skewered by critics) book No Free Lunch7, on creationist blogs, etc.

Figure 1 (  shows an image of a flagellum as it appears on Dembski’s blog named Uncommon Descent (a very similar picture appears on the dust cover of Dembski’s book7.)  Notice the smooth surface of the depicted contraption, its perfect symmetry, its tightly fitting components – all those features we usually see in man-made machinery.  This image is a product of an artist’s imagination of how a flagellum “must” look. The artist obviously knew what was required of him by his customers and performed accordingly. The crucial question is: does this image truthfully represent the real flagellum?

Flagella are tiny organelles which can’t be seen directly by human eyes.  Their dimensions are measured in nanometers (that is in billionths of a meter). Modern versions of cryogenic electron microscopy and of X-ray techniques have, though, enabled scientists to form a pretty good understanding of flagellum’s structure and shape.

Figure 2 (   shows a schematic model of a flagellum’s structure.  This model8 (one of several published in scientific literature) is a theoretical interpretation of the data obtained via electron microscopy, and mainstream scientists construe it more as an idealized schematic than a true-to-life representation of a flagellum’s actual structure (which we’ll discuss a little later). ID advocates, however, in a pursuit of their goals, happily treat such images as if they are real replicas of the tiny flagella, usually providing no disclaimers as to the degree of idealization inherent in such images. 

In 2004, when Dembski debated Professor Niall Shanks (at University of California Los Angeles) he displayed a different image of a flagellum as one of the tools of his pro-design argument (Figure 3 - see ). Unlike Figures 1 and 2, Figure 3 is neither an artist’s rendition, nor a schematic theoretical model; it is a “real” electron-microscopic photographically obtained image. While produced by scientists, such images are often exploited by ID advocates who are fond of pointing out their striking similarity to man-made machines.

For example, on February 15, 2008 I happened to debate Michael Behe on a Larry Kane’s TV show on the Comcast network. While Behe was invited to Larry Kane’s studio in Philadelphia, I took part in the show from the Public TV studio in San Diego, via a TV link. Unfortunately, while I could hear Behe’s and Larry Kane’s voices, I could not see either them, or myself on the screen, nor the pictures Behe displayed on the studio’s wall. However, a couple of weeks after the show, its producer kindly sent me a DVD where I finally could see Behe’s illustrations.  What did I see there? The same images of a flagellum that have been reproduced endlessly in creationist books, blogs, propagandistic movies, etc.  Behe’s argument at the show in fact boiled down to the same stale asseveration which can be succinctly summarized as follows: “You see - it looks like a man-made machine!  If it looks like a duck…. etc., then it must be a duck! All machines we are familiar with have been designed. Therefore the flagellum must be a product of intelligent design!”

To start with, the argument about a duck is an obvious non-sequitur: there are numerous examples of objects whose appearance is deceptive. All the mimicry, so common in nature, shouts against the “duck” pseudo-argument. (For example, look up the article on “Mimicry” in Wikipedia where examples are described of animals looking like “twigs, bark, leaves, or flowers” etc., thus negating the “it must be a duck” conclusion). 

If, during Larry Kane’s show, I had seen Behe’s pictures, I could have simply repeated what I wrote in a post to the Panda’s Thumb blog on June 15, 2004 regarding the Dembski-Shanks debate. Here is a quotation from my post9:

 [Dembski] has devoted a considerable attention to the discussion of what he referred to as the mascot of intelligent design - the bacterium flagellum. He insisted that the flagellum is in fact a machine, and to support this statement, he displayed that standard picture where the flagellum is shown in a geometrically perfect shape, fully symmetric and consisting of geometrically perfectly formed parts. Of course, such a presentation was misleading as the real flagellum is far from having such a perfect geometric shape. Unlike machines, which may be close replicas of each other (say, all Jeeps of the same year have almost exactly the same shape) the real flagella, first, have shapes with many deviations from a perfect geometric symmetry, and, second, there are no two flagella exactly identical. Individual flagella differ in various respects, like the entire biological organisms vary from an individual to individual. If Dembski’s picture were closer to reality, it would be much less effective in supporting his claims. Since he did not offer a disclaimer pointing to the idealization used in his depiction of the flagellum, we are entitled to conclude that he was interested not in an honest discussion based on facts, but rather on winning the debate regardless of means.

The above words could be addressed to Behe as well as to Dembski. 

However, for the sake of discussion, let us now provisionally accept the argument so forcefully pushed by ID advocates, asserting that if a biological assembly looks like a man-made machine, then we have to conclude it was designed.  The question then arises: are indeed the images of biological systems, in particular of flagella, so closely reminiscent of man-made machines?

In fact the images Behe, Dembski, and their ID colleagues show are often not pictures of real flagella. Some of them are just products of an artist’s imagination (like in Figure 1); some others are computer-generated images of imaginary machine-like contraptions.

Since the image in Figure 1 is obviously a product of an artist’s imagination of what a flagellum “must” look like, it cannot be regarded as having evidentiary value. The schematics like that in Figure 2, while reflecting many actual features of flagella,  are  products of a modeling approximation which likewise can’t pretend to reflect adequately the actual structure of a tiny organelle.  However, some other pictures of flagella may indeed be “real” photographically obtained images (like Figure 3).  Are the images in the latter category adequate representations of the flagella structure?

So, let us look again at Figure 3, which ID advocates so triumphantly parade as an alleged true-to-life representation of a flagellum (as, for example, was done by Dembski in his debate with Shanks).  It is, at a glance, impressive.  Indeed we see here a contraption which is symmetric, its structure indeed machine-like, so it is easy to understand the satisfaction of the likes of Dembski and Behe at the sight of this contraption so neatly fitting in with their “design” hypothesis, as well as the delight of their often gullible audience.

What ID advocates forget to mention, are two important details.

The first “detail” is that the image in Figure 3 is in fact a composite photo. It means that it is a result of a superposition of many photos, of several flagella, made from various angles. This way the image in question is eliminating from view various imperfections which, unlike in man-made machinery, are inherent in every natural flagellum.  Moreover, the procedure of superposition of a number of photos eliminates from view the inevitable individual differences between various flagella, which radically distinguish flagella from “designed” machinery.   

The second “detail” is that the resolution of this picture is insufficient to see the flagellum’s intrinsic structure. To appreciate the significance of this “detail” recall, for example, the “human face” or a “female human figure” on Mars, shown on multiple websites10, or Lowell’s nonexisting Martian “canals.”  When the resolution is insufficient, we “see” nonexisting structures which at a closer look dissolve into actual, usually very different patterns. This is equally true for the images of very small objects perused under insufficient magnification and/or resolution.

Similar illusions are common in the earthly macroscopic world.  For example, at the Western edge of the Caucasian mountainous range there is a region named Dombai.  Among mountains surrounding the Dombai valley, one named Sulakhat stands out. From the Dombai valley it looks like an image of a woman prostrated on her back. However, having climbed up the slopes of that mountain, the climbers discover that the woman-like image is an illusion: there is no woman-like configuration there in reality, as it becomes obvious from a closer distance. Likewise, the images of flagella obtained with a higher resolution, and assisted by other modern sophisticated methods of investigation, reveal the actual configuration of flagella, demonstrating that the seeming machine-like appearance of the flagella in Fig. 3 is deceptive.

(It can be noted that often legitimate scientists happen to use such terms as “machine” when describing various biological assemblies. This usage, however, unlike in case of ID advocates, is purely metaphorical, reflecting the superficial resemblance of certain biological structures to man-made machinery.  Scientists normally do not imply that biological entities are indeed intrinsically similar, not to mention identical, to man-made machinery.  Perhaps such a usage by scientists is not very fortunate given ID advocates’ misuse of the superficial resemblance between the designed man-made objects and natural biological entities. We have to realize, though, that scientists by and large are not aware of ID advocates’ misuse of such terminology, as only a small minority of scientists pay any attention to ID advocates’ actions.)  Perhaps a similar situation (in a certain sense in the opposite direction) exists in engineering where parts of machines and other constructions often are given names of parts of a human body. For example, elements of a construction that serve to support it often are named “legs,” or “feet” without implying that the parts in question are indeed intrinsically identical to parts of a human body.

Let us look at a few selected illustrations of my thesis.  

The detailed images of the flagella structure obtained via cryogenic electron microscopy combined with sophisticated X-rays techniques are exemplified in Figures 4, 5, and 6 (,, and   These images, showing the actual configuration of the flagellum, have been selected practically at random from numerous similar images available in the scientific literature.  Instead of tightly-fit machine-like parts, we see in these pictures convoluted garlands of protein molecules. These structures look similar to typical bacteriophage viruses6, and have nothing in common with any man-made machine. They vividly illustrate that the image shown in Figure 3 is deceptive and owes its machine-like appearance to the insufficient resolution (not to mention the utter artificiality of the artist’s renditions of flagella, whose variations serve as “mascots” of ID).  


ID advocates often point11 to the allegedly fraudulent “icons of evolution” utilized by the “Darwinists” for their nefarious purposes.  One of such allegedly fraudulent “icons” is that of embryos by Ernst Haeckel12. In fact, the faults of Haeckel’s embryos images (dated 1874) were revealed13 not by creationists but rather by the “Darwinists” themselves.  On the other hand, creationists of various hues, including ID advocates like Dembski and Behe, incessantly reproduce images of flagella which are often heavily doctored, without any disclaimers as to the great degree of idealization inherent in these images. Indeed, look again at the images of flagella’s actual molecular structure, as shown above in Figures 4, 5, and 6, and it becomes obvious that real natural flagella are far from looking like man-made machines.

An interesting question is: Why ID advocates and other creationists, who so eagerly and persistently display pictures like those in Figures 1, 2, and 3, never deign to show much more realistic representations of flagella structure like shown in Figures 4, 5, and 6?  If they are unaware of these better pictures, perhaps they should try to educate themselves regarding the entirety of the available information about flagella?  If, though, they are familiar with the images like those shown in Figures 4, 5, and 6 (which are freely available both in print and on the internet) could it then be that they are interested not so much in facts and truth as in winning the “cultural war” by any, even not always honest, means? 

We must conclude that the argument in favor of “design” of biological entities based on their alleged similarity to man-made machinery is not supported by evidence.

My thanks to Matt Young, Paul R. Gross, and Nicholas Matzke for pithy advice.  


1.     Behe, Michael, 1996. Darwin’s Black Box. Free Press, New York NY.

2.     Muller,  Hermann J. 1918. “Genetic variability, twin hybrids and constant hybrids, in a case of balanced lethal factors." Genetics 3: 422-499.

  1. Cairns-Smith, A. Graham 1986. Seven Clues to the Origin of Life: A Scientific Detective Story.  Cambridge Univ. Press, Cambridge, MA.
  2. Miller, Kenneth J. 1999. Finding Darwin's God. Cliff Streets Books, New York, NY.
  3. MacDonald, John. Online:, last accessed on June 20, 2008.
  4. Luria, Salvador E., Stephen J. Gould and Sam Singer. 1981. A View of Life. The Benjamin Cummings Publishing Co., Menlo Park, CA. 


  1. Dembski, William A. 2002. No Free Lunch: Why Specified Complexity Cannot Be Purchased  Without Intelligence. Rowman & Littlefield Publishers. Lanham, Md.

       8. Yonekura, K., S. Maki, D. G. Morgan, D. J. DeRosier, F.Vonderviszt,

           K. Imada, and K. Namba, 2000. “The Bacterial Flagellar Cap as the Rotary

          Promoter of Flagellin Self-Assembly.” Science 290: 2148-2152.

 9. Perakh, Mark. 2004. “Three SH’s and One D.” Online at Panda’s Thumb ; last accessed on June 20, 2008.

10. See, for example the striking photographs of objects on Mars’s surface creating an illusion (caused by insufficient resolution and/or magnification) of a human face ( ) or of a female human figure ( ).

11.  Wells, Jonathan. 2002. Icons of Evolution: Science or Myth? Why Much of What We Teach About Evolution is Wrong. Regnery Publishing, New York, NY.

  1.  See, for example, the article titled “Ernst Haeckel” in Wikipedia.
  2. Nic Tamzek (Nicholas Matzke). “Icon of Obfuscation”. In Talk Reason: (last accessed on June 20, 2008).