Darwin's Black Box: The Biochemical Challenge to Evolution (1996, first edition; 2006, second edition) is a book written by Michael J. Behe and published by Free Press in which he introduces his notion of irreducible complexity and claims that its presence in many biochemical systems indicates therefore that they must be the result of intelligent design rather than evolutionary processes.
The book has been a source of controversy, as the scientific community at large considers intelligent design and its constituent arguments to be religious, creationism, and pseudoscience. Common criticisms were that Behe's ideas are not falsifiable, that his definition of an irreducibly complex system is ambiguous, and that he ignores previous work in biochemical evolution. Though influential within the intelligent design movement for several years, the book has lost some of its currency as more and more examples given by Behe as evidence of irreducible complexity have been shown to be explicable by known evolutionary mechanisms, something Behe conceded under cross examination while testifying as an expert witness on behalf of the defendants in Kitzmiller v. Dover Area School District.
Behe begins by reminding the general reader of paradigm shifts in the history of science, in which the foundations and assumptions of theories are examined, sometimes resulting in the rejection of an entire theory. Behe suggests that such a paradigm shift in biology (and particularly in evolution) is imminent due to recent discoveries (circa 1996) in biochemistry. Behe acknowledges acceptance of the Theory of Evolution by "the great majority" of scientists, and states that "most (though not all) do so based on authority."
Behe states that elucidations of the evolutionary history of various biological features typically assume the existence of certain abilities as their starting point, such as Charles Darwin's example of a cluster of light-sensitive spots evolving into an eye via a series of intermediate steps. He then points out that Darwin dismissed the need to explain the origin of the 'simple' light-sensitive spot, summarizes the modern understanding of the biochemistry of vision and claims that many other evolutionary explanations face a similar challenge.
Behe next introduces and defines the concept of irreducible complexity as a system with a series of parts in which the removal of any part causes the entire system to cease functioning, offering a springloaded-bar mousetrap as a familiar example. In the following chapters, Behe discusses the apparent irreducible complexity of several biological systems, including the cilium, the bacterial flagellum, blood clotting, the immune system and intracellular gated and vesicular transport. Behe claims the underlying complexity and biochemical mechanisms of the systems are vastly under-appreciated, and identifies other, similar systems.
Behe identifies one of the primary counter-arguments of irreducible complexity, gradual adaptation—that certain systems may have been co-opted from an original, unrelated role to assume a new function as an irreducibly complex system. He counter-argues that though it is impossible to consider all possible roles for any component, it is extremely implausible that components can fortuitiously change function within a complex system and that the focus of the theory changes from making to modifying components and recounts unsuccessful attempts to discover evolutionary pathways for complex systems within scientific journals. Behe states that though he did identify assertions that evolution had occurred, he found none that had been supported by experiment or calculation, and concludes the book by offering intelligent design as a solution to irreducible complexity.
Richard Dawkins criticized the book for being logically flawed by setting up a false dichotomy in which Darwinian evolution is rejected despite an enormous amount of positive evidence due to a single apparent failure to explain irreducible complexity. Dawkins further commented that it was an argument Darwin himself had anticipated, and that the example of a bacterial flagellum used by Behe had in fact been refuted by Kenneth R. Miller in Kitzmiller v. Dover Area School District.
Behe has responded to some of these criticisms.
In the same trial, Behe eventually testified under oath that "There are no peer reviewed articles by anyone advocating for intelligent design supported by pertinent experiments or calculations which provide detailed rigorous accounts of how intelligent design of any biological system occurred". The result of the trial was the Kitzmiller v. Dover Area School District/6:Curriculum, Conclusion that intelligent design is not science and is essentially religious in nature.