[Published in American Catholic Philosophical Quarterly, Summer 1999]
EVOLUTION AND COMPLEXITY
T. Michael McNulty, SJ
Department of Philosophy, Marquette University, P. O Box 1881
Milwaukee, Wisconsin 53201-1881
It is not surprising that Darwin's theory of natural selection continues to engender controversy. It seems to challenge some cherished beliefs that we have about ourselves, such as that human beings are the pinnacles of creation, and that the natural world has transcended itself in the production of human consciousness. If, on the other hand, the existence of the human species is the result of undirected natural processes, we occupy simply the tip of one branch of the tree of life, undistinguished from any other by any special aim of nature. One need only add to this disanimating conclusion a challenge to belief in a benevolent deity (whose putative role as designer becomes otiose in such an unfolding of blind nature), and there is ample fertile soil for the seeds of conflict to germinate.
Defenders and critics of Darwin have in common a lack of shyness.(1) In fact, the level of testiness on all sides of the dispute has had the effect at times of obscuring both the complexity of the questions and the logical independence of the issues. The admixture of religious (or anti-religious) commitments further complicates the situation and makes philosophical consideration of the issues difficult. Recall, for instance, Richard Dawkins' claim that "Darwin made it possible to be an intellectually fulfilled atheist."(2) Referring to Michael Behe, whose argument against natural selection and in favor of an intelligent designer is the occasion for this paper, Dawkins is quoted as saying in an interview published in The Chronicle of Higher Education (Nov. 29, 1996), "It's a pathetic cop-out of his responsibilities as a scientist. He should stop being lazy and get up and think about it himself."(3)
In its position statement "Scientific Integrity," The National Association of Biology Teachers, in an attempt to exorcise ideology from the discussion, says the following: "[P]roposed causes and explanations must be naturalistic. Any attempt to mix or contrast supernatural beliefs and naturalistic theories within science misrepresents the scientific enterprise and debases other, nonscientific, ways of knowing."(4) The original text of the Association's "Statement on Teaching Evolution" contained the following statement: "The diversity of life on earth is the outcome of evolution: an unsupervised, impersonal, unpredictable, and natural process of temporal descent with genetic modification that is affected by natural selection, chance, historical contingencies and changing environments." On October 11, 1997, the NABT Board of Directors, by unanimous vote, removed the words "unsupervised" and "impersonal" from the statement. According to Wayne W. Carley, Executive Director of NABT, "The purpose, of course, was to make sure that we made only scientifically supportable (i.e. testable) statements and did not make any theological (or anti-theological) assertions."(5) The relevant part of the statement now reads: "The diversity of life on earth is the outcome of evolution: an unpredictable and natural process of temporal descent with genetic modification that is affected by natural selection, chance, historical contingencies and changing environments."(6)
It is into this discussion atmosphere that Michael Behe introduces an argument against the claim of neo-Darwinians that complex biological systems arose by natural selection. In brief, Behe argues that some kinds of complexity are irreducible and cannot be built up from incremental changes in precursor systems. They are therefore beyond the capacity of natural selection to account for. This kind of complexity manifests itself most clearly on the molecular level. In his book Darwin's Black Box,(7) Behe offers what purports to be an entirely naturalistic argument for the infirmity of natural selection to account for the biochemical complexity of the cell (the "black box" of the book's title). But Behe goes further: he claims in addition that the failure of natural selection to account for the etiology of irreducibly complex systems leads us ineluctably to the conclusion that these systems must be the work of intelligent design. Behe thus offers us a three-step argument:
1. Irreducibly complex biological systems exist.
2. The existence of such systems cannot be accounted for by natural selection
3. Therefore, the existence of such systems is the product of intelligent design.
It will be the contention of this paper that premises (1) and (2) of the argument are at least interesting and worthy of further study by both philosophers and scientists. The inference represented in statement (3), however, is a non sequitur: Behe's claim about natural selection is logically independent of his conclusion to intelligent design; one can, therefore, doubt the universal efficacy of the mechanism of natural selection without overstepping the bounds of naturalism.
It will be worthwhile briefly to remind ourselves of the basic ideas of biological evolution: 1) descent with modification; 2) common ancestor; 3) natural selection. Of these three, the first is relatively uncontroversial. Even an extremely conservative interpretation of the fossil record would acknowledge that at least some modern animals had ancestors that differed from them in some ways. The claim of a common ancestor is supported by a wealth of biochemical data and should not cause any serious consternation among scientifically-literate audiences. It is the third idea, natural selection, that causes all the difficulty, because it claims to account for the existence of things that, in Dawkins' words, "give the appearance of having been designed for a purpose."(8) But natural selection has no purpose; it is, in Dawkins' provocative phrase, a "blind watchmaker." It is a principle that is not difficult to state succinctly, although its implications take some work.
Here is the basic idea: populations exhibit genetic variation among the individuals that make them up. Individual organisms carry genes that they pass on to their progeny if they reproduce. If those genes, because of their phenotypic effects (characteristics whereby the organism interacts with its environment, e.g. a longer neck), enhance the ability of those organisms to a) survive to reproductive age; b) reproduce successfully; and c) produce a larger number of viable offspring relative to the species average, they will pass those genes on to the subsequent generation in such a way that they will be relatively more prevalent than in the population of the parent(s) generation. Genes that have this effect are said to be selected. Genes selected in this way enhance the fitness of the individuals that carry them. Because of the relationship between genes and their phenotypic expression, fitness has to be understood in a dispositional or propensity way. Otherwise, we are in danger of collapsing the principle of natural selection into a tautology: the survival of the survivors. It is also important to realize that fitness is always relative to population and environment, and that the genotype itself is part of the environment contributing to the selection of particular characteristics. But for our present purposes we can ignore these complications.
How does natural selection account for adaptive complexity? A number of factors are involved and can interact with one another. Environmental changes may make relevant previously "neutral" genetic characteristics. Or, as genetic material is replicated in the normal life-cycle of organisms, sometimes a mistake is made. Or a random event like radiation or environmental chemical impact might alter the DNA of a gamete. Sometimes such a copying error or accident has phenotypic consequences that alter the fitness (i.e., potential reproductive success) of individual organisms to whom such genes are passed on. If the alteration is positive, the number of individuals who have inherited the altered genetic material increases and may come to be represented in the population with relative frequency at or near one, at which point the alteration is said to have become fixed. This is called microevolution. Natural selection therefore conserves genetic alterations that enhance fitness, and the accumulation of many such small changes over geological time accounts for the adaptive complexity of biological systems. This is macroevolution.
Three points are important here. First, natural selection has no foresight and is therefore always a local (in both time and space) phenomenon. Natural selection can only respond to present circumstances, and it cannot produce changes that require a temporary reduction in fitness, even if those changes would eventually enhance fitness. Second, natural selection has no direction. It does not necessarily select for increased complexity. Third, change takes place typically over geological time: The earth is about 4 ½ billion years old, and life has been present for at least ¾ of that time. The numbers of organisms involved is likewise enormous. Natural selection does not lack for material to work on.
What, then, is Behe's problem? First we ought to make it clear that Behe accepts what we might call the standard scientific cosmology:
"For the record, I have no reason to doubt that the universe is the billions of years old that physicists say it is. Further, I find the idea of common descent (that all organisms share a common ancestor) fairly convincing, and have no particular reason to doubt it.... Although Darwin's mechanism--natural selection working on variation--might explain many things, however, I do not believe it explains molecular life." (p. 5)
When it comes to the basic mechanisms of cellular life, then, Behe claims that the situation is fundamentally different:
"Thus biochemistry offers a Lilliputian challenge to Darwin. Anatomy is, quite simply, irrelevant to the question of whether evolution could take place on the molecular level. So is the fossil record.... Neither do the patterns of biogeography matter, nor those of population biology, nor the traditional explanations of evolutionary theory for rudimentary organs or species abundance. This is not to say that random mutation is a myth, or that Darwinism fails to explain anything (it explains microevolution very nicely), or that large-scale phenomena like population genetics don't matter. They do. Until recently, however, evolutionary biologists could be unconcerned with the molecular details of life because so little was known about them." (p. 22)
But the biochemistry of the cell is where the action is, and it is precisely here that natural selection must inevitably fail as an explanation. The reason is that many biochemical processes at the cellular level are what Behe calls "irreducibly complex":
"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. 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. An irreducibly complex biological system, if there is such a thing, would be a powerful challenge to Darwinian evolution. Since natural selection can only choose systems that are already working, then if a biological system cannot be produced gradually it would have to arise as an integrated unit, in one fell swoop, for natural selection to have anything to act on." (p. 39)
Behe's key claim is that irreducibly complex systems cannot be produced incrementally by natural selection, since natural selection can only select for functions that already exist. As a homely example, he offers the mousetrap as an irreducibly complex system. Base, catch, holding bar, spring and hammer all have to be present--none can function without the others: remove one part and the system is not merely impaired; it will no longer work at all (unlike a car, which will still work with a burnt-out headlight). Unfortunately, the mousetrap is not a very apt example. For one thing, it is clearly designed for a particular purpose by an intelligent agent. For another, it is not as clearly irreducibly complex as Behe thinks. John H. McDonald has proposed a scenario in which parts from a mousetrap can be successively removed in such a way that while the efficiency of the mousetrap is adversely affected, its basic functionality is not: it remains capable of catching mice.(9) At best the mousetrap example serves to illustrate the interdependence of complex systems. Much more effective is his description of the long and complex biochemical process that produces a signal in the optic nerve from a photon impinging on a retina cell. (pp. 18-21) The process is multi-stage and requires an equally complex recovery cycle. Comparing this to Dawkins' description of the possible evolution of the eye (photo-sensitive pigmented spot, evolving to spot in a depression, then depression filled with jelly, etc.),(10) which follows closely Darwin's own, one can understand Behe's impatience. This is not a description of microevolutionary change, and it becomes difficult to see how it can occur by incremental change, one random event at a time. The problem is this: natural selection is blind: it has no idea where it is going. Every step must have immediate (i.e., this-generational) fitness enhancing effects. There is no planning for the future. The problem with seemingly irreducibly complex biochemical systems is that they include intermediate steps (often of mind-numbing complexity themselves) that have no fitness-enhancing function except as intermediates. Stopping anywhere between the beginning and the end of the process seems to have no fitness-enhancing effect at all. For instance, the change caused in a molecule of retinal by an incident photon that causes rhodopsin to alter its shape is pointless except in the presence of transducin. But rhodopsin and transducin are different proteins, whose synthesis is mediated by different DNA. Therefore, the evolution of these two proteins should be independent. As the number of independent events that must be coordinated increases, the odds against their simultaneous appearance increases geometrically. This seems to be the crux of Behe's argument about irreducible complexity.
In addition to the vision system, Behe discusses cellular cilia and bacterial flagella, the immune system and the blood-clotting system, all of which he claims are irreducibly complex. Concerning flagella he refers to a paper by Lucy Shapero of Stanford University:
"A rotating propeller at the cell surface, driven by a trans-membrane proton gradient, provides many bacteria with the ability to move and thus respond to environmental signals. To acquire this powerful capability, the bacterial cell is faced with the challenge of building a tiny rotary engine at the base of the propeller. Although the motor is anchored in the cytoplasmic membrane, a significant portion of the entire mechanism extends into the cytoplasm and, at the other end, out into the environment.... At least 20 individual proteins are used as parts for the complex structure and another 30 are used for its construction, function, and maintenance....
"To carry out the feat of coordinating the ordered expression of about 50 genes, delivering the protein products of these genes to the construction site, and moving the correct parts to the upper floors while adhering to the design specifications with a high degree of accuracy, the cell requires impressive organizational skills."(11)
The case for complexity was never really in doubt. The question is whether these systems are irreducibly complex, and if so, whether that implies that they cannot have arisen by natural selection. Both possibilities have been addressed by Behe's critics. After agreeing with Behe in rejecting simultaneous mutation and borrowing from other systems, Allen Orr argues:
"Behe's colossal mistake is that, in rejecting these possibilities, he concludes that no Darwinian solution remains. But one does. It is this: An irreducibly complex system can be built gradually by adding parts that, while initially just advantageous, become-because of later changes-essential. The logic is very simple. Some part (A) initially does some job (and not very well, perhaps). Another part (B) later gets added because it helps A. This new part isn't essential, it merely improves things. But later on, A (or something else) may change in such a way that B now becomes indispensable. This process continues as further parts get folded into the system. And at the end of the day, many parts are required.
"The point is there's no guarantee that improvements will remain mere improvements. Indeed because later changes build on previous ones, there's every reason to think that earlier refinements might become necessary.... [A]lthough this process is thoroughly Darwinian, we are often left with a system that is irreducibly complex. I'm afraid there's no room for compromise here: Behe's key claim that all the components of an irreducibly complex system 'have to be there from the beginning' is dead wrong."(12)
The problem with Orr's criticism is that it is not detailed enough to address Behe's claim, a point that Behe is quick to make.(13) Other criticisms addressed by Behe (pp. 90-97) have claimed that there are mechanisms, such as gene duplication, for building up "irreducibly" complex systems like blood clotting.(14) The most puzzling of these challenges to Behe is Doolittle's citation of an experiment breeding a strain of mice that lack two clotting factors (plasminogen and fibrinogen) but are basically normal. Although this seems at first glance to be a counterexample to Behe's claim that the blood-clotting cascade is irreducibly complex (if it is, no element should be dispensable), it is not yet obvious what is going on here, and there is certainly no clear theoretical model to explain it.
The picture that emerges from this discussion is that of a science in ferment, with a budget of unanswered questions. What would a definitive answer to Behe's challenge look like? Ideally, one would hope for a schema that would spell out, in some detail, how an irreducibly complex system can be built up incrementally in such a way that it can be demonstrated (or at least made plausible) that each increment confers selective advantage (neutral changes are by definition invisible to natural selection, but it should be remembered that neutrality is itself a local phenomenon, and "neutral" changes caused for example by genetic drift might subsequently be made visible by environmental changes). If it can be shown that no such schema exists, then it will be necessary to look for other mechanisms.
One suspects that had Behe limited himself to a critique of natural selection without offering intelligent design as an alternative, his views would have received a more measured (but less widespread) response. But Dawkins makes a very strong claim about the truth of natural selection:
"My argument will be that Darwinism is the only known theory that is in principle capable of explaining certain aspects of life. If I am right it means that, even if there were no actual evidence in favour of the Darwinian theory (there is, of course) we should still be justified in preferring it over all rival theories."(15)
Behe associates his position with the argument from design articulated by William Paley(16) (pp. 210-31), which Dawkins uses as his stalking horse to argue against intelligent design and for natural selection. Paley's argument is analogical. He reasoned that if we found a watch on the beach, we would infer from its intricacy of design and appropriateness of the interconnection of its parts to its manifest purpose, that it was designed by an intelligent being. By analogy, therefore, the intricacy of biological systems and their evident suitability for some purpose should lead us inexorably to the conclusion that they, too, were designed by an intelligent being.
The argument from design exerts a strong attraction, because it seems to accord so well with our experience. In his review of Darwin's Black Box, Robert Dorit illuminates the problem:
"I've often wondered why the argument from design so appeals to engineers and chemists. I suppose that the problem derives from the day-to-day experience of these professions. Engineers and chemists know that they do not get a desired outcome-stable bridge or purified compound-from random inputs, time and a statistical principle for differential representation. In these professions, there is no design without a designer, no desired outcome without careful and intelligent planning."(17)
Paley's argument is an example of what we would now call an "inference to the best explanation." According to Elliott Sober, prior to Darwin's proposal of natural selection, Paley's argument had some merit, and Hume's critique of the argument from design in the Dialogues concerning Natural Religion never establish that it is bad science. Darwin's accomplishment was to offer a better alternative.(18) It is the burden of Behe's argument that natural selection fails as a better alternative, and that the inference to the existence of an intelligent designer is therefore inescapable.
Behe's conclusion to the existence of an intelligent designer, however, is a non sequitur. The problem with Behe's proposal is not just that it asks us to imagine a designer of this or that organ or biochemical cascade, but of complex biological systems in general. We know how the design of an artifact relates to the plan of the designer because we understand how the design gets translated into concrete systems. In other words, we can comprehend the mechanism of the making. When we transfer this way of thinking to the design of biological systems in nature, unfortunately, the mechanism falls away and we are left with a purported identity of argument that is essentially meaningless.
Behe claims that scientists in the past have been like the little boy Calvin in the comic strip "Calvin and Hobbes":
"Calvin is always jumping in a box with his stuffed tiger, Hobbes, and traveling back in time, or 'transmogrifying' himself into animal shapes, or using it as a 'duplicator' and making clones of himself. A little boy like Calvin easily imagines that a box can fly like an airplane (or something), because Calvin doesn't know how airplanes work." (pp. 23-25)
Ironically, Behe's inference from irreducible complexity to the existence of a designer falls into just this trap. He simply pushes the "black box" back from the biochemistry of the cell to the process of transforming design into reality: what could possibly be the engineering techniques, materials, tools, scaffoldings, that transform these designs into concrete reality? Two possibilities exist: either we can imagine (in however obscure and abstract a way) what they are, in which case the designer becomes simply another part of the natural world, itself in need of explanation; or we cannot, in which case Behe commits what D. Z. Phillips calls the "naturalistic fallacy of religion," attributing natural mechanisms to an ineffable being.(19)
In the case of Behe's argument, we are at a loss as to what criteria to use to establish that the term "design" can be used in the same way for a purported "designer" of biochemical mechanisms and the designer of a watch, to use Paley's venerable example. As Dorit says at the end of his review, "The hand of God may well be all around us, but it is not, nor can it be, the task of science to dust for fingerprints."
It does not follow from the failure of Behe's design argument, however, that his critique of natural selection is incorrect. And other biologists have found problems with natural selection as well. Stephen Jay Gould points out that Darwin himself did not consider natural selection to be the only mechanism of evolution. He has long championed the role of contingency in biological explanation against pan-selectionists like Dawkins and Dennett.(20) And building on the work of Barbara McClintock and others, James Shapero has suggested the possibility of non-Darwinian mechanisms:
"The existence of cellular biochemical activities capable of rearranging DNA molecules means that genetic change can be specific... and need not be limited to one genetic locus.... In other words, genetic change can be massive and non-random.... This means that major chromosome reorganizations can be present in multiple gametes. Consequently, the appearance of new genome architectures during evolution is not necessarily limited to isolated individuals."(21)
It is important to recognize that although these mechanisms are non-Darwinian, they are thoroughly naturalist. But we have no reason to think that nature must be simple:
"The very definition of complexity, and the rules that govern its emergence, are indeed critical issues in evolutionary biology. But the problem becomes deeply uninteresting if the only legitimate approach to solving it is the demonstration of 'a direct, gradual route [leading to] irreducibly complex systems.' We are still deciding how to measure complexity, debating whether the history of life shows a tendency toward increased complexity and arguing about whether biological organization is but a subset of the larger problem of order and complexity. We do not as yet know what form the answers will take, but mathematical models, computational simulations and, increasingly, experimental results suggest that complexity and organization may be inexorable outcomes in multicomponent systems. If our hypotheses about complexity are to be of any use, however, they will have to be materialist explanations grounded in material cause."(22)
The fact is, there is a huge amount that we don't know. Richard Dawkins, with his characteristic innocence of modesty, wants to persuade us "not just that the Darwinian world-view happens to be true, but that it is the only known theory that could, in principle, solve the mystery of our existence."(23) His case remains unproven. If advances in physics (not to say biology) have taught us anything in the last century, it is that the world is a lot weirder than it seemed in Darwin's time. It may even be weirder than we can know. But allowing the progress of science to be limited by any theory that "must" be true will not make the weirdness go away.
1 For contemporary restating of the neo-Darwinian synthesis, see Richard Dawkins, The Blind Watchmaker (New York: W. W. Norton, 1996) and Daniel Dennett, Darwin's Dangerous Idea (New York: Simon & Schuster, 1995). Vigorous contemporary critiques of Darwin can be found in Paul E. Johnson, Darwin on Trial, 2nd ed. (Downers Grove, IL: Intervarsity Press, 1993) and David Berlinski, "The Deniable Darwin," Commentary 101(June 1996): 19-29. One can get a sense for the tone of the debate by consulting the responses to Berlinski in the September 1996 issue of Commentary. Further discussion of these issues is available on The Talk.Origins Archive <http://www.talkorigins.org/>.
2 Dawkins, 6.
3 For an example of the effect of these ideological elements one need look no further that the recent Firing Line Debate "Resolved: That evolutionists should acknowledge creation," aired on PBS on Dec. 19, 1997, in which William F. Buckley, Michael Behe, Phillip Johnson, and David Berlinski for the affirmative and Barry Lynn, Eugenie Scott, Kenneth Miller, and Michael Ruse for the negative attempted with less than overwhelming success to discuss the issues.
4 National Association of Biology Teachers [NABT]. Position Statements. 1997. <http://www.nabt.org/positions.html>. 1 Feb.1998.
5 Private correspondence, Jan. 30, 1998.
6 "Statement on Teaching Evolution," NABT Position Statements. The history of the change and the considerations that prompted it are detailed in Eugenie C. Scott, "NABT Statement on Evolution Evolves." <http://www.natcenscied.org/nabtart.htm>. 2 Nov. 1998.
7 New York: Free Press, 1996. This work will be referred to by parenthetical page numbers in the text.
8 Dawkins, 1.
9 "A reducibly complex mousetrap." 1 Dec. 1997. U. of Delaware. <http://udel.edu/~mcdonald/mousetrap.html>. 23 Jan. 1998.
10 The Blind Watchmaker, 85-86.
11 "The Bacterial Flagellum: From Genetic Network to Complex Architecture," Cell 80(1995): 25-27. The second paragraph is quoted by Behe in "The Sterility of Darwinism," Boston Review 22(Feb.-Mar. 1997), 24.
12 "Darwin v. Intelligent Design (Again)," Boston Review 21(Dec.-Jan.1996-97), 29-30.
13 "The Sterility of Darwinism."
14 See R. Doolittle, "A Delicate Balance," Boston Review 22(Feb.-Mar. 1997), 28-29. G. Acton, "Behe and the Blood Clotting Cascade." Feb. 1997. The Talk.Origins Archive. <http://www.talkorigins.org/origins/postmonth/feb97.html>. 14 Jan. 1998. K. Robison, "Darwin's Black Box: Irreducible Complexity or Irreproducible Irreducibility?" Dec. 1996. The Talk.Origins Archive. <http://www.talkorigins.org/faqs/behe.html>. 14 Jan. 1998.
15 Dawkins, 287.
16 Natural theology, 12th ed. (Charlottesville, VA: Ibis Pub., 1986), ch. 1.
17 Robert Dorit, Review of Darwin's Black Box, American Scientist 85(1997), 475.
18 The Nature of Selection. (Cambridge, MA: MIT Press, 1984), 18-19.
19 The Concept of Prayer (London: Routledge and Kegan Paul, 1965), 101-102. More general, though related issues, are raised by Ludwig Wittgenstein in his Philosophical Investigations (New York: Macmillan, 1958), par. 349-50, where he talks about what might be called the "5 o'clock on the Sun" fallacy.
20 See S. J. Gould, and R. C. Lewontin, "The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme," in Conceptual Issues in Evolutionary Biology, 2nd ed., E. Sober (ed.) (Cambridge, MA: MIT Press) 73-90; S. J. Gould, Wonderful Life (New York: W. W. Norton, 1989); "The Evolution of Life on Earth," Scientific American 271 (Oct. 1994): 85-91; "Darwinian Fundamentalism," The New York Review of Books 44 (June 12, 1997): 34-37.
21 A. Shapero, "Natural genetic engineering in evolution," in Transposable Elements and Evolution [Contemporary Issues in Genetics and Evolution vol. I], ed. J. F. McDonald (Dordrecht, Netherlands: Kluwer Academic, 1993), 325-47; "A Third Way," Boston Review 22(Feb.-Mar. 1997), 32-33.
22 Dorit, 475.
23 Dawkins, xiv.