Irreducible Complexity

Questioning the Existence of Evolutionary Pathways

Behe’s argument from irreducible complexity is often construed as deductive, but it is actually a two-part argument. The deductive first part is his criticism of what he terms

“direct evolutionary pathways”. However, I will argue that Behe’s criticism of “indirect evolutionary pathways” is probabilistic, not deductive. Behe admits Darwinian evolution as an in principle possible explanation for biological machinery, but argues that empirical details pose great difficulties for this Darwinian explanation. Instead, he argues that the details of biology strenghted the design inference.

I return to Dawkins' argument from The Blind Watchmaker (1989), since it provides the backdrop to Behe's argument. Dawkins argues that it is highly rational to believe in the capacities of the process of mutation and natural selection to explain even complex structures. Consider the evolution of the mammalian eye, a highly complex structure that was already used as an example of design by William Paley.⁷²⁰ Dawkins argues that

716 Behe 2007a.

717 Wiker & Witt 2002.

718 Meyer 2009, for Dembski’s comments see Barham 2012.

719 The implication is that if a system is necessary for life, then their evolution is a part of the problem of the origin of life, and thus not part of the explanatory scope of Darwinian evolutionary theory. (Barham 2012.) The naturalistic answer to this is that although these systems are necessary for the current form of life, they may not have been necessary for some unknown form of life which preceded the current one, and in which the basic functions of life such as protein systems were handled in some other way. See Griesemer 2008 for a review article of the state of origin of life research.

720Natural Theology, chapter III.

evolution may have started with a small light-sensitive patch on the skin. When mutations improving the capability for vision appeared, they were beneficial and thus contributed to the fitness of the organism. Natural selection preserved these mutations, and accumulated them. Over a long period of time, the modern eye was evolved.⁷²¹ Dawkins presents the central assumptions of this evolutionary explanation as follows:

1. Could the human eye have arisen directly from no eye at all, in a single step?

2. Could the human eye have arisen directly from something slightly different from itself, something that we may call X?

3. Is there a continuous series of Xs connecting the modern human eye to a state with no eye at all?

4. Considering each member of the series of hypothetical Xs connecting the human eye to no eye at all, is it plausible that every one of them was made available by random mutation of its predecessor?

5. Considering each member of the series of Xs connecting the human eye to no eye at all, is it plausible that every one of them worked sufficiently well that it assisted the survival and production of the animals concerned? ⁷²²

Dawkins’ answer to the first question is negative: explaining the emergence of new complex organs through large random mutations is too improbable to believe. Dawkins answers the remaining four questions in the affirmative, however.

Behe regards the example of eye as misleading, because the small steps described by Dawkins are actually extremely large on the biochemical level. According to Behe, there is no such thing as a “simple” light sensitive cell. Rather, the biochemical basis of vision is highly complex, an irreducibly complex cascade of proteins.⁷²³ So, explaining the evolution of the eye in Dawkins’ manner is, for Behe, akin to explaining the production of a stereo set by saying that you just first make a cassette player, then add a cd-player, then loudspeakers, then a remote controller and so on. The small leaps described by Dawkins are, for Behe, giant leaps between canyons.⁷²⁴

Behe’s argument from irreducible complexity is directed against the idea that all biological structures can be explained by the accumulation of small, useful mutations. Behe’s critique of Dawkins’ logic is directed against the fifth point of Dawkins’ argument: the plausibility that all mutations necessary for the evolution of the current form are useful. With his argument, Behe is attempting to answer Darwin’s challenge, as set out in the Origin of Species: “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.”⁷²⁵ If some biological organ could not be developed through the

721 Dawkins 1991. See also similar arguments in Freeman & Herron 2007, 98-99.

722 Dawkins 1991, 77-81.

723 Behe 2006a, 38.

724 Behe 2006a, 22.

725 Darwin 2008 [1859], chapter VI. The basic idea remains part of the mainstream of current evolutionary theory.

As Jerry Coyne (2007) puts the point: “It is indeed true that natural selection cannot build any feature in which

mechanism of natural selection and mutation, then evolutionary theory’s claim to explain life’s order would be jeopardized. Behe argues that “irreducibly complex” machines in cells satisfy Darwin’s challenge.⁷²⁶

Behe assumes that each mutation in the evolutionary series has to be useful or at least neutral in order for natural selection to help generate new biological structures. Otherwise, natural selection will act against its preservation. The basic idea is common in explanations of evolution: As Denis Noble states, if we compare the design of an organism to an aircraft, evolution must modify and improve the aircraft without foresight while the aircraft is in flight and all systems are in use.⁷²⁷ The basic idea of Behe’s argument is that this sort of useful modification and building up of new structures is very difficult or impossible in the case of irreducibly complex systems.

According to Behe, an irreducibly complex system is “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.”⁷²⁸ He argues that there are multiple irreducibly complex structures in cells. His basis for this is both empirical and conceptual.

First, biochemical experiments can be used to test the necessity of the parts of biochemical machines. If the functioning of the system ceases when a part is removed, this part can be judged essential for the functioning of the system. Second, upon understanding how a system works, we can determine that certain parts are crucial for its operation. Behe argues that a system for swimming, for example, requires a minimum of three parts: a paddle, a motor and a part connecting the two. These parts also have to be fine-tuned to fit each other, or “well-matched” as Behe defines.⁷²⁹

In the discussion on ID, the bacterial flagellum has emerged as the most often used example of biological irreducible complexity. The flagellum, according to Behe, is an outboard motor which propels bacteria forward. Behe argues that the flagellum requires tens of parts to function.⁷³⁰ Additionally, he argues that the flagellum is part of a network of interlinked irreducibly complex machines. For example, the building mechanism of the flagellum is argued to be irreducibly complex.⁷³¹

Some problems with Behe’s initial definition of irreducible complexity have emerged in the discussion. The definition of irreducible complexity assumes that all of the parts of a system are necessary for its function, but in practice no biological system is like this. For example, many parts of the bacterial flagellum are unnecessary for its functioning. Behe

intermediate steps do not confer a net benefit on the organism.” Like Darwin, Coyne argues that the existence of any such feature has not been demonstrated. Note, however, that evolutionary biology does not require all features of life to be selected for by natural selection. On this see the classic paper Gould & Lewontin 1979, as well as Pigliucci & Müller (eds) 2010.

726 Behe 2006a, 36. See also Behe 2001b, 2003a and2004.

727 Noble 2006, 109.

728 Behe 2006a, 39.

729 Behe 2006a.

730 E.g. Behe 2004a.

731 Behe 2007, chapter 5.

himself recognizes this already in Darwin’s Black Box, and admits that ciliar motors have many parts whose function is even unknown to us. However, for Behe, this does not mean that we cannot identify many other parts as crucial.⁷³² The adjustment by Behe and Dembski is that Behe’s definition applies only to the irreducible core of the system – there can be additional parts which make the system more robust, and which can be removed without necessarily losing the function.⁷³³

Behe argues that irreducible complexity is a barrier to “direct evolutionary pathways”.

By direct evolution, Behe means evolution which works by improving some existing function step by step, as in Dawkins’ depiction of the evolution of the eye, where each modification results in a slight increase of vision. A direct evolutionary pathway means the evolution of a system from humble origins by small improvements in the system’s “core function”. Irreducible complexity is, for Behe, impossible to evolve in this way, because the core function of the system emerges only after all of the necessary parts are in place. Natural selection, however, cannot select for a function that emerges only after all of the parts are in place. The system thus cannot be evolved by small beneficial steps through a direct evolutionary pathway. For example, the bacterial flagellum cannot have evolved from a simpler system for moving the bacterium.⁷³⁴ Instead, its evolution is – according to Behe – better explained by the actions of an intelligent designer, who can arrange parts to fulfil some future end.⁷³⁵

So, the idea is that the mutations leading up to an irreducibly complex system could not generate the benefit given by the whole system and thus the series leading up to the system would not fulfil Dawkins’ condition number five, and natural selection does not help explain the system. Most of Behe’s critics have admitted that the argument works against direct evolutionary pathways, but have still considered indirect evolutionary pathways possible.⁷³⁶ Behe’s argument is that the parts of an irreducibly complex system would not have the function of the system, and so would not be selected by natural selection. However, this does not mean that they could not have had some other function, and then only later been transformed into parts of the newly formed irreducibly complex system. The evolutionary history of Behe’s irreducibly complex systems, such as the bacterial flagellum,

732 Behe 2006a, 39, 72-73. Miller (2002, 141) interprets Behe without noticing these. According to Miller’s interpretation, Behe errs by assuming that all of the parts of the flagellum are essential for its function. But this is shown to be false by the fact that there are many different flagella with slightly different parts. However, Behe’s original argument includes the admission that a motor can be built in several different ways (Behe 2006a, chapter 3).

733 Dembski 2002a, 285.

734 Behe 2006a, 39.

735 Behe 2006a, 36-39. The argument has a predecessor in Michael Denton’s (1986, 90-91) argument from

”integrated complexity”, which Denton traces back to biologist Jean Cuvier (1796-1832). Denton also uses the bacterial flagellum as one example of biological complexity which Darwinian evolutionary theory fails to explain (Denton 1986, 224-225). Forrest & Gross (2004) find similarities in Behe’s argument to creationist Ariel Roth’s argumentation (e.g. Roth 2001, 86-87).

736 E.g. Miller 2002, 132-136, Orr 1996.

could be very complex, with similar parts serving in many slightly different systems with different functions.

Evolutionary Co-Option as a Response to Behe

The adaptation of an existing biological structure for a new function is called evolutionary co-option. In responses to Behe, it has been illustrated with examples from the larger anatomical level. Behe focuses on biochemical phenomena, but critics have noted that his definition could just as well be applied to larger structures. It seems difficult to say that no system on the larger anatomical level requires more than one part to function. For example, paleontologist Alan Gishlick has argued that under Behe’s definition of irreducible complexity, a bird’s complex wing is irreducibly complex, requiring the coordinated action of nine “well-matched” parts to provide the capability of flight. However, Gishlick goes on to argue that the wing has evolved through Darwinian evolution. Based on the anatomical similarities of birds and theropod dinosaurs, paleontologists have concluded that birds evolved from flightless dinosaurs. Several intermediates have been postulated. For Gishlick, these provide evidence that the irreducible complexity of the bird’s wing can be explained through an indirect evolutionary pathway.⁷³⁷ The possibility of evolutionary explanations on this level thus provides reason to also accept them as possible for biochemical irreducibly complex machines.

Behe has dismissed the use of paleontological evidence by remarking that the fossils only provide evidence of change, but not of the mechanism which affected the change. Behe accepts common descent, but not the Darwinian mechanism of selection and mutation.⁷³⁸ However, it seems that the fossil record does provide evidence for the possibility of co-option, supporting it as a possible evolutionary explanation. Behe’s argument thus does not prove that Darwinian mechanisms are in principle incapable of generating irreducible complexity. Precursors to the bacterial flagellum, for example, may have had other functions.

Indeed, it has been argued that even the modern bacterial flagellum doubles as a secretory system, and that the metaphor of a motor thus leads us astray.⁷³⁹ The co-option -argument has been the most popular response to Behe’s argument from irreducible complexity.⁷⁴⁰ The

737 Gishlick 2006, 71.

738 Behe 2006a, 22.

739 Musgrave 2006.

740 E.g. Miller 2002, 151; Dawkins 2006, 143-146; Coyne 1996, 227; Kitcher 2007, 89. In addition, other scenarios for the evolution of irreducible complexity have also been proposed. Not all defenders of evolution find the co-option scenario plausible for all of Behe’s systems (Orr 1996; 1997; Forrest & Gross 2004, 303-304). For Behe’s comments on these other scenarios see Behe 1997a, 1997b, 2001a.

Interestingly, biologist Allen Orr argues against the co-option scenario on grounds quite similar to Behe.

According to Orr, adapting biochemical parts for another function is a good general solution to the problem of irreducible complexity: “we might think that some of the parts of an irreducibly complex system evolved step by step for some other purpose and were then recruited wholesale to a new function. But this is also unlikely. You may as well hope that

idea of the argument is that the precursors of irreducibly complex machines had other functions, and thus natural selection preserved them. Serendipitously, the parts initially useful in other machines were also useful for the construction of current irreducibly complex machines.

The co-option scenario can be defended with further biological evidence. The most important category of evidence is homology: the similarity of the parts of irreducibly complex machines with parts of other systems. Miller argues that an important part of the bacterial flagellum is similar to the type III secretory system also found in bacteria, and Musgrave argues that homological parts are known for around 90 percent of the parts of the flagellum.⁷⁴¹ Similar homological evidence can also be found for most of Behe’s other cases, strenghtening the evidence for the evolution of irreducible complexity.⁷⁴² It can thus be argued that we already have at least the beginnings of plausible evolutionary histories for many of Behe’s irreducibly complex systems.⁷⁴³

However, few critics have noted that Behe himself considers the co-option answer in Darwin’s Black Box, and presents another version of his irreducible complexity -argument against it. Directly after presenting his argument against the direct evolution of irreducible complexity, Behe argues as follows:

Even if a system is irreducibly complex (and thus cannot have been evolved directly), however, one can not definitively rule out the possibility of an indirect, circuitous route.

As the complexity of an interacting system increases, though, the likelihood of such an

half your car's transmission will suddenly help out in the airbag department. Such things might happen very, very rarely, but they surely do not offer a general solution to irreducible complexity.”(Orr 1997)

Orr’s argument against the co-option scenario is the same as Behe’s: the applicability of parts with one function to a different function is too unlikely. Orr’s own solution to the problem of irreducible complexity is to defend the possibility of direct evolution. Evolution might work by adding parts improving an existing function, like motion.

At first, these parts simply complement the function, but over time, they may become essential to the system’s function. Parts which at first interact haphazardly are formed into a seamless whole, where the removal of one part destroys the function. Thus irreducible complex systems can evolve from simpler beginnings. Behe’s response is that this scenario needs detailed evidence to make it plausible. (Behe 2000b, 3-4; 2001a, 692-695.)

741 Miller 2004, 85-87; see also Musgrave 2006, 81.

742 Miller 2002, 152-161.

743 In Darwin’s Black Box, Behe argued that the biological research literature does not contain any good evolutionary explanations for irreducible complexity. Behe admitted that the literature contains many examples of homology in biochemical systems, but mere similarity, for Behe, does not prove the mechanism which created this similarity (Behe 2006a, 175-176). Behe’s assertion has provoked much debate. Some of Behe’s critics have agreed with him about the lack of explanations (Harold 2001, 205, Coyne 1996, 227, Behe 2001a, 686). However, others have vigorously disputed Behe’s assertion, arguing that the literature contains many examples of how biochemical complex systems can evolve (e.g. Miller 2002, 147-152; Forrest & Gross 2007, 304; Inlay 2007; Weber 1999). See also Venema 2012.

In the Dover trial, where Behe testified about his views on irreducible complexity, the opposing attorney was able to present Behe with a stack of peer-reviewed papers and books which were claimed to present evidence for the evolution of irreducible complexity. Based partly on this evidence, the judge concluded that Behe’s argument was not good science (or actually scientific at all). However, the evidence wasn’t actually reviewed or discussed in the trial. (Jones 2005, 78; Behe 2006a, 7.) Behe responds to criticisms in e.g. Behe 2000c; 2000d; 2000e; 2000f and 2000g.

indirect route drops precipitously. And as the number of unexplained, irreducibly complex biological systems increases, our confidence that Darwin’s criterion of failure has been met skyrockets toward the maximum that science allows.⁷⁴⁴

Here Behe is admitting the point of his critics already in the course of stating his original argument. Irreducible complexity can in principle evolve through an indirect route, by which Behe means the co-option account. However, Behe argues that as the complexity of the system increases, the probability of such evolutionary accounts decreases to extremely low.

Behe’s argument against Darwinian evolution is thus probabilistic, just as his argument for design is also probabilistic. In Behe’s argument, as the amount of apparently purposeful, irreducible complexity increases, so the probability of design goes up and the probability of naturalism goes down.

Niall Shanks is one of the few critics who have noted this statement from Behe.

However, Shanks does not recognize any arguments from Behe to support this conclusion, and does not comment on this. Most critics of Behe’s irreducible complexity have bypassed this argument, and stated the co-option account without referencing Behe’s critique.⁷⁴⁵ The fragmentariness of Behe’s argument may be partly to blame for the misunderstanding. Behe does not state his argument against indirect evolution concisely, but elaborates on it further only as he analyses his examples of irreducible complexity in more detail. Behe has further

However, Shanks does not recognize any arguments from Behe to support this conclusion, and does not comment on this. Most critics of Behe’s irreducible complexity have bypassed this argument, and stated the co-option account without referencing Behe’s critique.⁷⁴⁵ The fragmentariness of Behe’s argument may be partly to blame for the misunderstanding. Behe does not state his argument against indirect evolution concisely, but elaborates on it further only as he analyses his examples of irreducible complexity in more detail. Behe has further

In document Intelligent Design : A Theological and Philosophical Analysis (sivua 182-190)