Irreducible complexity is an argument by proponents of intelligent design that certain biological systems are too complex to have evolved from simpler, or ‘less complete’ predecessors, through natural selection acting upon a series of advantageous naturally occurring, chance mutations.
The argument is central to intelligent design, and is rejected by the scientific community at large, which overwhelmingly regards intelligent design as pseudoscience. Irreducible complexity is one of two main arguments (both discredited) used by intelligent design proponents, the other being specified complexity (which singles out patterns that are both specified and complex as markers of design by an intelligent agent).
Lehigh University biochemistry professor Michael Behe, the originator of the term ‘irreducible complexity,’ defines an irreducibly complex system as one ‘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.’ Evolutionary biologists have shown that such systems can evolve, and that Behe’s examples constitute an argument from ignorance. It asserts that a proposition is true because it has not yet been proven false, it is ‘generally accepted’ (or vice versa). This represents a type of false dichotomy in that it excludes a third option, which is that there is insufficient investigation and therefore insufficient information to prove the proposition satisfactorily to be either true or false. In the 2005 ‘Kitzmiller v. Dover Area School District trial’ (the first direct challenge brought in federal courts testing a public school district policy that required the teaching of intelligent design), Behe gave testimony on the subject of irreducible complexity. The court found that ‘Professor Behe’s claim for irreducible complexity has been refuted in peer-reviewed research papers and has been rejected by the scientific community at large.’
Supporters of intelligent design use this term to refer to biological systems and organs that they believe could not have come about by any series of small changes. They argue that anything less than the complete form of such a system or organ would not work at all, or would in fact be a detriment to the organism, and would therefore never survive the process of natural selection. Although they accept that some complex systems and organs can be explained by evolution, they claim that organs and biological features which are irreducibly complex cannot be explained by current models, and that an intelligent designer must have created life or guided its evolution. Accordingly, the debate on irreducible complexity concerns two questions: whether irreducible complexity can be found in nature, and what significance it would have if it did exist.
A second definition given by Behe (his ‘evolutionary definition’) is as follows: ‘An irreducibly complex evolutionary pathway is one that contains one or more unselected steps (that is, one or more necessary-but-unselected mutations). The degree of irreducible complexity is the number of unselected steps in the pathway.’ Intelligent design advocate William Dembski (who holds PhDs in mathematics and philosophy from the University of Chicago) gives this definition: ‘A system performing a given basic function is irreducibly complex if it includes a set of well-matched, mutually interacting, nonarbitrarily individuated parts such that each part in the set is indispensable to maintaining the system’s basic, and therefore original, function. The set of these indispensable parts is known as the irreducible core of the system.’ This definition ignores the utility of individual components in unrelated systems.
The argument from irreducible complexity is a descendant of the teleological argument for God (the argument from design or from complexity). This states that because certain things in nature are very complicated, they must have been designed. English philosopher William Paley famously argued, in his 1802 watchmaker analogy, that complexity in nature implies a God for the same reason that the existence of a watch implies the existence of a watchmaker. This argument has a long history, and can be traced back at least as far as Roman statesman Cicero’s ‘De Natura Deorum ii.’
Roman physician Galen (b. 129 CE) wrote about the large number of parts of the body and their relationships, which observation was cited as evidence for creation. The idea that specifically the interdependence between parts would have implications for the origins of living things was raised by writers starting with French priest Pierre Gassendi in the mid 17th century and English clergyman John Wilkins, who wrote (citing Galen), ‘Now to imagine, that all these things, according to their several kinds, could be brought into this regular frame and order, to which such an infinite number of Intentions are required, without the contrivance of some wise Agent, must needs be irrational in the highest degree.’
In the late 17th century, English theologian Thomas Burnet referred to ‘a multitude of pieces aptly joyn’d’ to argue against the eternity of life. In the early 18th century, French philosopher Nicolas Malebranche wrote ‘An organized body contains an infinity of parts that mutually depend upon one another in relation to particular ends, all of which must be actually formed in order to work as a whole,’ arguing in favor of preformation, rather than epigenesis (unfolding development), of the individual; and a similar argument about the origins of the individual was made by other 18th century students of natural history. In his 1790 book, ‘The Critique of Judgment,’ Kant is said to argue that ‘we cannot conceive how a whole that comes into being only gradually from its parts can nevertheless be the cause of the properties of those parts.’
As we transition to the 19th century, we find references which relate to evolution. Paley’s ‘Natural Theology’ discusses at length what he called ‘relations’ of parts of living things as an indication of their design. French naturalist Georges Cuvier applied his principle of the correlation of parts to describe an animal from fragmentary remains. For Cuvier, this was related to another principle of his, the ‘conditions of existence,’ which excluded the possibility of transmutation of species.
Charles Darwin identified the argument as a possible way to falsify a prediction of the theory of evolution at the outset. In ‘The Origin of Species,’ he wrote, ‘If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find out no such case.’ Darwin’s theory of evolution challenges the teleological argument by postulating an alternative explanation to that of an intelligent designer—namely, evolution by natural selection. By showing how simple unintelligent forces can ratchet up designs of extraordinary complexity without invoking outside design, Darwin showed that an intelligent designer was not the necessary conclusion to draw from complexity in nature. The argument from irreducible complexity attempts to demonstrate that certain biological features cannot be purely the product of Darwinian evolution.
In the late 19th century, in a dispute between supporters of the adequacy of natural selection and those who held for inheritance of acquired characters (a hypothesis that physiological changes acquired over the life of an organism may be transmitted to offspring), one of the arguments made repeatedly by English biologist Herbert Spencer (who coined the concept ‘survival of the fittest’), and followed by others, depended on what Spencer referred to as co-adaptation of co-operative parts, as in:
‘We come now to Professor Weismann’s endeavour to disprove my second thesis — that it is impossible to explain by natural selection alone the co-adaptation of co-operative parts. It is thirty years since this was set forth in ‘The Principles of Biology.’ In §166, I instanced the enormous horns of the extinct Irish elk, and contended that in this and in kindred cases, where for the efficient use of some one enlarged part many other parts have to be simultaneously enlarged, it is out of the question to suppose that they can have all spontaneously varied in the required proportions.’ The history of this concept in the dispute has been characterized: ‘An older and more religious tradition of idealist thinkers were committed to the explanation of complex adaptive contrivances by intelligent design. … Another line of thinkers, unified by the recurrent publications of Herbert Spencer, also saw coadaptation as a composed, irreducible whole, but sought to explain it by the inheritance of acquired characteristics.’
American geneticist Hermann Muller, in the early 20th century, discussed a concept similar to irreducible complexity. However, far from seeing this as a problem for evolution, he described the ‘interlocking’ of biological features as a consequence to be expected of evolution, which would lead to irreversibility of some evolutionary changes. He wrote, ‘Being thus finally woven, as it were, into the most intimate fabric of the organism, the once novel character can no longer be withdrawn with impunity, and may have become vitally necessary.’
In 1974, Young Earth Creationist Henry M. Morris introduced a similar concept in his book ‘Scientific Creationism’ in which he wrote; ‘This issue can actually be attacked quantitatively, using simple principles of mathematical probability. The problem is simply whether a complex system, in which many components function unitedly together, and in which each component is uniquely necessary to the efficient functioning of the whole, could ever arise by random processes.’
A book-length study of a concept similar to irreducible complexity, explained by gradual, step-wise, non-teleological evolution, was published in 1975 by University of Illinois biologist Thomas H. Frazzetta: ‘A complex adaptation is one constructed of several components that must blend together operationally to make the adaptation ‘work.’ It is analogous to a machine whose performance depends upon careful cooperation among its parts. In the case of the machine, no single part can greatly be altered without changing the performance of the entire machine.’ The machine that he chose as an analog is the Peaucellier machine (a mechanical linkage invented in 1864). A biological system given extended description was the jaw apparatus of a python. The conclusion of this investigation, rather than that evolution of a complex adaptation was impossible, ‘awed by the adaptations of living things, to be stunned by their complexity and suitability,’ was ‘to accept the inescapable but not humiliating fact that much of mankind can be seen in a tree or a lizard.’
In 1981, Swiss American zoologist and creationist Ariel Roth, in defense of the creation science position in the trial ‘McLean v. Arkansas’ (a constitutional challange to an Arkansas state law known as the ‘Balanced Treatment for Creation-Science and Evolution-Science Act’), said of ‘complex integrated structures’ that ‘This system would not be functional until all the parts were there … How did these parts survive during evolution …?’ However, the law was overturned and the court concluded that ‘creation-science’ as defined in the Arkansas Act ‘is simply not science.’ In 1985, Scottish organic chemist Graham Cairns-Smith wrote of ‘interlocking,’ ‘How can a complex collaboration between components evolve in small steps?’ and used the analogy of the scaffolding called centering used to build an arch then removed afterwards: ‘Surely there was ‘scaffolding.’ Before the multitudinous components of present biochemistry could come to lean together they had to lean on something else.’
An essay in support of creationism published in 1994 referred to bacterial flagella as showing ‘multiple, integrated components,’ where ‘nothing about them works unless every one of their complexly fashioned and integrated components are in place’ and asked the reader to ‘imagine the effects of natural selection on those organisms that fortuitously evolved the flagella … without the concommitant [sic] control mechanisms.’
An early concept of irreducibly complex systems comes from Ludwig von Bertalanffy, a 20th-century Austrian biologist. He believed that complex systems must be examined as complete, irreducible systems in order to fully understand how they work. He extended his work on biological complexity into a general theory of systems in a book titled ‘General Systems Theory.’ After James Watson and Francis Crick published the structure of DNA in the early 1950s, ‘General Systems Theory’ lost many of its adherents in the physical and biological sciences. However, Systems theory (the interdisciplinary study of systems in general) remains popular in the social sciences long after its demise in the physical and biological sciences.
Michael Behe developed his ideas on the concept around 1992, in the early days of the ‘wedge movement,’ a broad political and academic agenda whose ultimate goal is to defeat materialism, naturalism, and evolution (a metaphor for a metal wedge splitting a log to represent an aggressive public relations program to create an opening for the supernatural in the public’s understanding of science). He first presented his ideas about ‘irreducible complexity’ in 1993 when the ‘Johnson-Behe cadre of scholars’ met at Pajaro Dunes in California. He set out his ideas in the second edition of ‘Of Pandas and People’ (a controversial biology textbook that espouses intelligent design) published in 1993, extensively revising a chapter, ‘Biochemical Similarities,’ with new sections on the complex mechanism of blood clotting and on the origin of proteins. He first used the term ‘irreducible complexity’ in his 1996 book ‘Darwin’s Black Box,’ to refer to certain complex biochemical cellular systems. He posits that evolutionary mechanisms cannot explain the development of such ‘irreducibly complex’ systems. Notably, Behe credits Paley for the original concept, not von Bertalanffy, and suggests that his application of the concept to biological systems is entirely original.
According to the theory of evolution, genetic variations occur without specific design or intent. The environment ‘selects’ the variants that have the highest fitness, which are then passed on to the next generation of organisms. Change occurs by the gradual operation of natural forces over time, perhaps slowly, perhaps more quickly, such as in instances of punctuated equilibrium (a theory in evolutionary biology which proposes that most species will exhibit little net evolutionary change for most of their geological history, remaining in an extended state called stasis; change occurs rarely and quickly). Evolution is able to adapt complex structures from simpler beginnings, or convert complex structures from one function to another (such as a spandrel, a characteristic that did not originate by the direct action of natural selection, that was later co-opted for a current use). Most intelligent design advocates accept that evolution occurs through mutation and natural selection at the ‘micro level,’ such as changing the relative frequency of various beak lengths in finches, but assert that it cannot account for irreducible complexity, because none of the parts of an irreducible system would be functional or advantageous until the entire system is in place.
Behe uses the mousetrap as an illustrative example of this concept. A mousetrap consists of five interacting pieces—the base, the catch, the spring, the hammer, and the hold-down bar. All of these must be in place for the mousetrap to work, as the removal of any one piece destroys the function of the mousetrap. Likewise, he asserts that biological systems require multiple parts working together in order to function. Intelligent design advocates claim that natural selection could not create from scratch those systems for which science is currently unable to find a viable evolutionary pathway of successive, slight modifications, because the selectable function is only present when all parts are assembled.
In his 2008 book ‘Only A Theory,’ biologist Kenneth R. Miller challenges Behe’s claim that the mousetrap is irreducibly complex. Miller observes that various subsets of the five components can be devised to form cooperative units, ones that have different functions from the mousetrap and so, in biological terms, could form functional spandrels before being adapted to the new function of catching mice. In an example taken from his high school experience, Miller recalls that one of his classmates, ‘…struck upon the brilliant idea of using an old, broken mousetrap as a spitball catapult, and it worked brilliantly….It had worked perfectly as something other than a mousetrap….my rowdy friend had pulled a couple of parts –probably the hold-down bar and catch– off the trap to make it easier to conceal and more effective as a catapult…[leaving] the base, the spring, and the hammer. Not much of a mousetrap, but a helluva spitball launcher….I realized why [Behe’s] mousetrap analogy had bothered me. It was wrong. The mousetrap is not irreducibly complex after all.’
Behe’s original examples of irreducibly complex mechanisms included the bacterial flagellum of E. coli, cilia, coagulation, and the adaptive immune system. Behe argues that organs and biological features which are irreducibly complex cannot be wholly explained by current models of evolution. In explicating his definition of ‘irreducible complexity’ he notes that: ‘An irreducibly complex system cannot be produced directly (that is, by continuously improving the initial function, which continues to work by the same mechanism) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional.’
His theory has not found acceptance in US jurisprudence or the scientific community because it fails to account for shifts in the function of a trait during evolution (exaptation). For example, a trait can evolve because it served one particular function, but subsequently it may come to serve another. Exaptations are common in both anatomy and behavior. Bird feathers are a classic example: initially these may have evolved for temperature regulation, but later were adapted for flight. Interest in exaptation relates to both the process and product of evolution: the process that creates complex traits and the product that may be imperfectly designed.
Undeterred, Behe and others have suggested a number of biological features that they believe may be irreducibly complex. The irreducible complexity argument assumes that the necessary parts of a system have always been necessary, and therefore could not have been added sequentially. However, in evolution, something which is at first merely advantageous can later become necessary. Natural selection can lead to complex biochemical systems being built up from simpler systems, or to existing functional systems being recombined as a new system with a different function. For example, one of the clotting factors that Behe listed as a part of the clotting cascade was later found to be absent in whales, demonstrating that it is not essential for a clotting system. Many purportedly irreducible structures can be found in other organisms as much simpler systems that utilize fewer parts. These systems, in turn, may have had even simpler precursors that are now extinct. Behe has responded to critics of his clotting cascade arguments by suggesting that homology (characteristics of biological organisms derived from a common ancestor) is evidence for evolution, but not for natural selection.
The ‘improbability argument’ also misrepresents natural selection. It is correct to say that a set of simultaneous mutations that form a complex protein structure is so unlikely as to be unfeasible, but that is not what Darwin advocated. His explanation is based on small accumulated changes that take place without a final goal. Each step must be advantageous in its own right, although biologists may not yet understand the reason behind all of them—for example, jawless fish accomplish blood clotting with just six proteins instead of the full 10.
The eye is a famous example of a supposedly irreducibly complex structure, due to its many elaborate and interlocking parts, seemingly all dependent upon one another. Although Behe acknowledged that the evolution of the larger anatomical features of the eye have been well-explained, he claimed that the complexity of the minute biochemical reactions required at a molecular level for light sensitivity still defies explanation. Creationist Jonathan Sarfati has described the eye as evolutionary biologists’ ‘greatest challenge as an example of superb ‘irreducible complexity’ in God’s creation,’ specifically pointing to the supposed ‘vast complexity’ required for transparency. In an often misquoted passage from ‘On the Origin of Species,’ Darwin appears to acknowledge the eye’s development as a difficulty for his theory.
However, the quote in context shows that Darwin actually had a very good understanding of the evolution of the eye. He notes that ‘to suppose that the eye … could have been formed by natural selection, seems, I freely confess, absurd in the highest possible degree.’ Yet this observation was merely a rhetorical device for Darwin. He goes on to explain that if gradual evolution of the eye could be shown to be possible, ‘the difficulty of believing that a perfect and complex eye could be formed by natural selection … can hardly be considered real.’ He then proceeded to roughly map out a likely course for evolution using examples of gradually more complex eyes of various species.
Since Darwin’s day, the eye’s ancestry has become much better understood. Although learning about the construction of ancient eyes through fossil evidence is problematic due to the soft tissues leaving no imprint or remains, genetic and comparative anatomical evidence has increasingly supported the idea of a common ancestry for all eyes. The current theory is that eyes evolved from pigment spots, into pigment cups, into the simple optic cup found in abalone, and finaly into complex lensed eye of the marine snail and the octopus. The eyes of vertebrates and invertebrates such as the octopus developed independently: vertebrates evolved an inverted retina with a blind spot over their optic disc, whereas octopuses avoided this with a non-inverted retina.
The specific chain of development imagined is that the eyes originated as simple patches of photoreceptor cells that could detect the presence or absence of light, but not its direction. When, via random mutation across the population, the photosensitive cells happened to have developed on a small depression, it endowed the organism with a better sense of the light’s source. This small change gave the organism an advantage over those without the mutation. This genetic trait would then be ‘selected for’ as those with the trait would have an increased chance of survival, and therefore progeny, over those without the trait. Individuals with deeper depressions would be able to discern changes in light over a wider field than those individuals with shallower depressions.
As ever deeper depressions were advantageous to the organism, gradually, this depression would become a pit into which light would strike certain cells depending on its angle. The organism slowly gained increasingly precise visual information. And again, this gradual process continued as individuals having a slightly shrunken aperture of the eye had an advantage over those without the mutation as an aperture increases how collimated the light is at any one specific group of photoreceptors. As this trait developed, the eye became effectively a pinhole camera which allowed the organism to dimly make out shapes—the nautilus is a modern example of an animal with such an eye. Finally, via this same selection process, a protective layer of transparent cells over the aperture was differentiated into a crude lens, and the interior of the eye was filled with humours to assist in focusing images. In this way, eyes are recognized by modern biologists as actually a relatively unambiguous and simple structure to evolve, and many of the major developments of the eye’s evolution are believed to have taken place over only a few million years, during the Cambrian explosion (the rapid appearance of most major phyla around 530 million years ago).
The flagella of certain bacteria constitute a molecular motor requiring the interaction of about 40 different protein parts. Behe asserts that the absence of any one of these proteins causes the flagella to fail to function, and that the flagellum ‘engine’ is irreducibly complex as if we try to reduce its complexity by positing an earlier and simpler stage of its evolutionary development, we get an organism which functions improperly. However, experiments have shown that many proteins can be deleted from the flagellar apparatus without destroying its function, even though its activity may be reduced in some of these cases.
Secondly, the basal body of the flagella has been found to be similar to the Type III secretion system (TTSS), a needle-like structure that pathogenic germs such as Salmonella use to inject toxins into living eucaryote cells. The needle’s base has ten elements in common with the flagellum, but it is missing forty of the proteins that make a flagellum work. Thus, this system negates the claim that taking away any of the flagellum’s parts would render it useless. On this basis, Brown University biology professor Kenneth Miller notes that, ‘The parts of this supposedly irreducibly complex system actually have functions of their own.’ Dembski’s critique of this position is that phylogenetically, the TTSS makes an unlikely precursor to the flagellum given that TTSS is found in a narrow range of bacteria which makes it seem to be a late innovation, whereas flagella are widespread throughout many bacterial groups, which implies it was an early innovation.
Niall Shanks and Karl H. Joplin, both of East Tennessee State University, have shown that systems satisfying Behe’s characterization of irreducible biochemical complexity can arise naturally and spontaneously as the result of self-organizing chemical processes. They also assert that what evolved biochemical and molecular systems actually exhibit is ‘redundant complexity’—a kind of complexity that is the product of an evolved biochemical process. They claim that Behe overestimated the significance of irreducible complexity because of his simple, linear view of biochemical reactions, resulting in his taking snapshots of selective features of biological systems, structures, and processes, while ignoring the redundant complexity of the context in which those features are naturally embedded. They also criticized his over-reliance of overly simplistic metaphors, such as his mousetrap. In addition, research published in the peer-reviewed journal ‘Nature’ has shown that computer simulations of evolution demonstrate that it is possible for irreducible complexity to evolve naturally.
It is illustrative to compare a mousetrap with a cat, in this context. Both normally function so as to control the mouse population. The cat has many parts that can be removed leaving it still functional; for example, its tail can be bobbed, or it can lose an ear in a fight. Comparing the cat and the mousetrap, then, one sees that the mousetrap (which is not alive) offers better evidence, in terms of irreducible complexity, for intelligent design than the cat. Even looking at the mousetrap analogy, several critics have described ways in which the parts of the mousetrap could have independent uses or could develop in stages, demonstrating that it is not irreducibly complex.
Moreover, even cases where removing a certain component in an organic system will cause the system to fail do not demonstrate that the system couldn’t have been formed in a step-by-step, evolutionary process. By analogy, stone arches are irreducibly complex—if you remove any stone the arch will collapse—yet we build them easily enough, one stone at a time, by building over centering that is removed afterward. Similarly, naturally occurring arches of stone are formed by weathering away bits of stone from a large concretion that has formed previously. Evolution can act to simplify as well as to complicate. This raises the possibility that seemingly irreducibly complex biological features may have been achieved with a period of increasing complexity, followed by a period of simplification.
In 2006 a team led by Joe Thornton at the University of Oregon, using techniques for resurrecting ancient genes, for the first time reconstructed the evolution of an apparently irreducibly complex molecular system. The research was published in ‘Science.’ It may be that irreducible complexity does not actually exist in nature, and that the examples given by Behe and others are not in fact irreducibly complex, but can be explained in terms of simpler precursors. There has also been a theory that challenges irreducible complexity called facilitated variation presented in 2005 by biology professors Marc W. Kirschner of Harvard Medical School and John C. Gerhart of UC, Berkeley. In their theory, they describe how certain mutation and changes can cause apparent irreducible complexity. Thus, seemingly irreducibly complex structures are merely ‘very complex,’ or they are simply misunderstood or misrepresented.
The precursors of complex systems, when they are not useful in themselves, may be useful to perform other, unrelated functions. Evolutionary biologists argue that evolution often works in this kind of blind, haphazard manner in which the function of an early form is not necessarily the same as the function of the later form (exaptation). The mammalian middle ear (derived from a jawbone) and the panda’s thumb (derived from a wrist bone spur) are considered classic examples. A 2006 article in ‘Nature’ demonstrates intermediate states leading toward the development of the ear in a Devonian fish (about 360 million years ago). Furthermore, recent research shows that viruses play a heretofore unexpectedly great role in evolution by mixing and matching genes from various hosts.
Some critics have argued that the concept of irreducible complexity, and more generally, the theory of intelligent design is not falsifiable, and therefore, not scientific. Behe argues that the theory that irreducibly complex systems could not have been evolved can be falsified by an experiment where such systems are evolved. For example, he posits taking bacteria with no flagellum and imposing a selective pressure for mobility. If, after a few thousand generations, the bacteria evolved the bacterial flagellum, then Behe believes that this would refute his theory.
Other evidence that irreducible complexity is not a problem for evolution comes from the field of computer science, where computer analogues of the processes of evolution are routinely used to automatically design complex solutions to problems. The results of such Genetic Algorithms are frequently irreducibly complex since the process, like evolution, both removes non-essential components over time as well as adding new components. The removal of unused components with no essential function, like the natural process where rock underneath a natural arch is removed, can produce irreducibly complex structures without requiring the intervention of a designer. Researchers applying these algorithms are automatically producing human competitive designs—but no human designer is required.
Critics view irreducible complexity as a special case of the ‘complexity indicates design’ claim, and thus see it as an ‘argument from ignorance’ and ‘God of the gaps’ argument (a theological perspective in which gaps in scientific knowledge are taken to be evidence of God’s existence). Eugenie Scott, along with Glenn Branch and other critics, has argued that many points raised by intelligent design proponents are arguments from ignorance. Behe has been accused of using an ‘argument by lack of imagination,’ and Behe himself acknowledges that a failure of current science to explain how an ‘irreducibly complex’ organism did or could evolve does not automatically prove the impossibility of such an evolution. Irreducible complexity is at its core an argument against evolution. If truly irreducible systems are found, the argument goes, then intelligent design must be the correct explanation for their existence. However, this conclusion is based on the assumption that current evolutionary theory and intelligent design are the only two valid models to explain life, a false dilemma.