Michael Behe and His Irreducible Complexity
Introduction: The Origin of New Genetic Code
It is common for creationists to deny that the information content of a genome can increase over time. This is after all a large part of the idea behind the watchmaker argument (see the informal essay Thoughts On The Intelligent Design Inference), the so-called thermodynamic argument based upon the second law of thermodynamics, and the more recent claims of William Dembski involving specified complexity. However, real scientists are well aware of the fact that there are different kinds of mutations which result in the enlargement of the genome and consequently increase the information content of the genome.
One of the most dramatic is when the entire genome will become duplicated - these are mutations which occur quite regularly in orchids even today. Phillip Johnson (often refered to as "the father of the intelligent design movement"), found it necessary to admit that such things occur but argued that it only happens in species capable of asexual reproduction.
Steven J. Gould critiques this in:
Impeaching a Self-Appointed Judge
by Stephen Jay Gould
http://www.stephenjaygould.org/reviews/gould_darwin-on-trial.html
from
The Unofficial Stephen Jay Gould Archive
http://www.stephenjaygould.org
Moreover, we have evidence that it occured fairly recently in mammals in the red vizcacha rat.
For those who are interested, please check:
Plant & Animal Genomes XIII Conference
January 15-19, 2005
W288 : Polyploidy
Tetraploidy And Apparent Hybridization In South American Rodents.
http://www.intl-pag.org/13/abstracts/PAG13_W288.html
from
Plant and Animal Genome Conferences
http://www.intl-pag.org
and
Polyploidy in Mammals: The First Tetraploid Rodent.
Milton H. Gallardo, Ricardo A. Ojeda, Claudio Bidau, and John A. Kirsch
http://www.cricyt.edu.ar/INSTITUTOS/iadiza/ojeda/grecia%202000.htm
from
Cricyt - Conicet | Centro Regional de Investigaciones Cientificas y Tecnologicas
http://www.cricyt.edu.ar
Likewise, there are instances of chromosomal duplication, segmental duplication and gene duplication. Then there are retroviral insertions and the well known phenomena of hypermutation by means of tandem repeat slippage. Then creationists will argue that this doesn't actually represent an increase in the amount of information in the genome - after all, the information has simply been duplicated. But once the genes have been duplicated, they can evolve along separate paths.
Michael Behe, Irreducible Complexity and Gene Duplication
To demonstrate this simply, we will focus for the moment just on gene duplication, and examine it in relation to Behe's concept of irreducible complexity. Michael Behe's original definition was, "By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning." (Darwin's Black Box, pg. 39)
Now assume you have a gene which codes for a promiscuous enzyme or other protein; in other words, it is multi-use. In this case, let's assume that the protein has two functions. Now make sure that those two functions are interdependent, such that each is required for the organism to survive or for a wider function to be performed. Now let us assume that the gene is duplicated. (Incidentally, there is nothing unusual in this -- retrotransposons naturally duplicate segments of DNA.)
Once the gene has been duplicated, both copies can undergo subfunctionalization -- meaning that the two copies of the original gene will mutate so that they each code for more specialized proteins. Why is that? The original protein was a generalist -- a kind of a jack-of-all trades. He knew how to do a little bit of everything, but didn't know how to do anything particularly well. Now that you have two proteins, each one may become more efficient at what it does -- and become finely tuned for performing its role in whatever system you have going. And why is that? Because those organisms in which such a division of labor is achieved simply as the result of random mutation will tend to be more successful and stand a better chance of leaving offspring that make it to the next generation.
However, now that you have two specialists, neither of them will be particularly good at doing what the other one does. But in practical terms, this loss of generalized functionality doesn't matter -- so long as you have both specialists. Moreover, if they are particularly good at what they do, you can expect the organism to put them to greater use -- in essence, counting on them performing at the much higher level of efficiency than that which the original generalist had. Why? Mutation and natural selection will tend to result in the adaptive "exploitation" of the opportunities made available by increased specialization.
But since the system will evolve so that it exploits the greater efficiency of each specialist, the removal of either specialist will result in the failure of the system. And there you have it: the evolution of irreducible complexity. Beginning with a gene that codes for a non-specialized (i.e., "promiscuous") protein which performs two interdependent functions poorly, gene duplication may result in two genes which code for distinct, more specialized proteins, each of which performs its own task more efficiently, but neither of which would work without the other.
Sometimes as a way of explaining irreducible complexity, someone will speak of the construction of a roman arch. In the final arch, if one removes any one stone, the entire structure will fall. Nevertheless, only one stone can be put in place at any given time, and the arch must remain stable while it is being constructed. The trick to building the arch is to begin with pieces in the center space of the arch, the part that will end up being the opening that people actually pass through/beneath, then put the individual pieces of the encompassing structure - the final arch - in place one at a time until one has finished the arch.
While constructing the arch, the pieces are supported by the stones where the hole will ultimately be. Once all of the stones of the final arch are in place, one can remove the center stones to create the walkway. The same principle applies in the case of the two interdependent proteins. The original protein acts as the central, supporting stones. Once it has been duplicated, each of the two resulting proteins support the function of the other. As they both gradually become more specialized, this in effect removes the "central stones."
What makes Behe's argument from irreducible complexity plausible at first (at least to laymen) is that when we consider constructing something one step at a time, we tend to assume that the process is always additive - and thus that at no point is anything removed. If each piece depends upon all the other pieces in the final product -- given this assumption of pure additivity, then it seems impossible for something irreducibly complex to be constructed in the first place.
The Evolution of Behe's Irreducible Complexity:
Now I should point out the basic strategy to Behe's encounters with criticism. Behe will give examples of so-called irreducibly complex systems, systems which are sufficiently complex that it is difficult to identify the precise sequence of steps which lead to the creation of the system. Bacteria flagella and blood clotting are two of the examples he's used. As it becomes clear that such examples can in fact be the result of a natural evolutionary process, he will tend to resist this revelation until it becomes sufficiently obvious that his position is no longer tenable for that particular example. At this point, he will concede the fact that the example he used was not irreducibly complex, but then simply switch to another example, and so on and on, ad infinitum.
This suggests that his conception of irreducible complexity has itself evolved into a simple appeal to ignorance or personal incredulity, where if one cannot see how something may have been the result of a natural process, one must assume that it was designed, making "designed" the default position, and requiring proof that the system resulted (or at least could have resulted) from a natural process from anyone who claims otherwise. Then, so long as there are some instances of things which still can't be explained as the result of natural, causal processes, one can claim the absence of such explanations as evidence for and even proof of the existence of irreducible complexity.
In biological systems, it is not unusual for Behe to react to a given incremental explanation of the evolution of a presumably irreducibly complex sub-system by pointing out that the theorists who have expended considerable effort in arriving at a naturalistic explanation of that sub-system have as yet to identify all the possible ramifications of the sequence of steps which they have proposed. Alternatively, when shown how a presumably irreducibly complex system may have in fact resulted from a sequence of steps from a simpler system, Behe will sometimes claim that it wasn't actually the more complex system which was irreducibly complex, but rather the simpler system.
The effect of these two strategies is that of requiring a step-by-step explanation of the evolution of the entire organism before the claim of irreducible complexity is itself removed from the organism which possesses the sub-system, and which Behe originally claimed was irreducibly complex. But by that time he will have moved on to some other purported instance of irreducible complexity. Such moves on Behe's part - moves which require biologists to constantly explain progressively more and more in order to get Behe to remove a particular claim - are instances of a fallacy known as "moving the goal-posts". These claims of his require no real substantive work on Behe's part to actually demonstrate them - they are contrived so that both the burden of proof and the real effort lies entirely with those proposing genuine causal explanations. And if such persons meet any given burden, Behe will always be more than happy to give them another.
What Behe requires of biological science is the proof of a negative, that is, that there does not exist any system which is complex in such a way that it could not have arisen by means of naturalistic causation. However, the burden of proof lies with the individual who asserts the positive, that is, that unicorns exist, or for that matter, that such a complex system exists.
Moreover, given the ambiguity with which he now employes his notion of "irreducible complexity," it has become so ill-defined that one would not know what one was looking for in order to prove that it does not exist. The only criteria Behe ultimately has is that this system could not have arisen by means of naturalistic causes - which means that barring a fully detailed and explicit proof that everything which exists was the result of naturalistic causes, Behe will continue with his assertion. At this point, it seems almost superfluous to point out that there is no actual experiment or test which can be performed to eliminate his claim that at least some irreducibly complex systems exist.
For one biography of Michael Behe, please see:
Michael Behe
http://en.wikipedia.org/wiki/Michael_Behe
from
Wikipedia
http://en.wikipedia.org
For those who are interested in further analysis of Michael Behe's "irreducible complexity," you may want to see:
Part 08: Dr. Michael Behe, Dr. Robert Pennock Q&A
Transcript: American Museum of Natural History
April 23, 2002
http://www.ncseweb.org/resources/articles/923_part_08_dr_michael_behe_dr_10_31_2002.asp
from
National Center for Science Education
http://www.ncseweb.org
Irreducible Complexity Demystified
http://www.talkdesign.org/faqs/icdmyst/ICDmyst.html
from
Talk Design
http://www.talkdesign.org
Irreducible Complexity and Michael Behe
Do Biochemical Machines Show Intelligent Design?
http://www.talkorigins.org/faqs/behe.html
from
Talk Origins
http://www.talkorigins.org
A reducibly complex mousetrap
http://copland.udel.edu/~mcdonald/mousetrap.html
from
John H. McDonald
http://copland.udel.edu/~mcdonald/
The Flagellum Unspun
The Collapse of "Irreducible Complexity"
http://www.millerandlevine.com/km/evol/design2/article.html
from
Kenneth R. Miller
http://www.millerandlevine.com/km
Irreducible Complexity and Michael Behe on Intelligent Design
http://www.talkorigins.org/faqs/behe.html
from
Talk Origins
http://www.talkorigins.org
© Timothy Chase 2006
