A Long Review of Dawkins’ The God Delusion

This essay was originally written as a guest post for an agnostic friend’s blog.

Richard Dawkins’ book The God Delusion is immensely popular and has been hailed as a convincing defense of atheism. TheGodDelusionDawkins sets out not only to defend atheism but to portray its worldview as morally and aesthetically pleasing in a way that atheist thinkers of the past (say, Nietzsche) didn’t. In fact, the majority of the book is not actually an argument against the existence of God, but rather a polemic against the origins, abuses, and beliefs of religion (in Chapters 1,2,5-10). (At this point, let me briefly apologize to anyone reading this post who, at the hands of professed Christians, has experienced some of the hatred that Dawkins describes. It makes me very ashamed, not of Christ, but of those of us who follow him and bring his name into such ill-repute). However, since I have limited space, I’ve decided to focus only on the rational arguments for atheism since, to rephrase Dawkins: atheism’s (or religion’s) power to comfort (or offend) doesn’t make it true (or false).

Let me focus explicitly on the end of Chapter 4, since Dawkins presents in it what he calls “the central argument of [his] book” (p. 157; all quotations and page numbers are from the 2006 edition). His argument is as follows:

    • Life is too complex to have come about by pure, random chance
    • It is therefore tempting to believe that it was created by an “intelligent designer”(p. 157) like other complex things
    • However, this belief is false because a designer would be more complicated than the thing designed, and “the whole problem we started out with was the problem of explaining statistical improbability” (p. 158)
    • Darwinian evolution shows how life “with [its] spectacular statistical improbability” could have been produced (p. 158)
    • There is no analogous argument for physics, but the anthropic principle allows us to take “more luck” into account than we normally would in most arguments (p. 158)
    • Probably a better argument for physics does exist
    • Therefore, “God almost certainly does not exist” (p. 158)

I’d like to point out two central inconsistencies in this argument. In addition, I’d like to examine whether Dawkins’ arguments are purely empirical and derived wholly from scientific evidence and reason, or whether they contain an element of “faith.”

First, let’s note that Dawkins’ argument is essentially one of probability. What Dawkins has attempted to show is not that God’s existence is disproved but merely rendered very, very improbable. In the section Irreducible Complexity, Dawkins points out that “Chance is not a solution [to the problem of biological complexity], given the high levels of improbability we see in living organisms, and no sane biologist would ever suggest that it was.” (p 119-120) We need to be very careful here. Technically speaking, chance is a possible solution to the problem of biological complexity in the sense that it is physically possible that in 40 million B.C. a random fluctuation of molecules accidentally assembled the entire Eocene ecosystem. In the same way, a hurricane in a factory just might assemble a 747. There are no physical laws that are actually violated by either process (not even the 2nd Law of Thermodynamics; e-mail me later). But what Dawkins is saying is that no scientist in his right mind would believe a theory that depended on such a small probability. In contrast, says Dawkins, natural selection provides an elegant mechanism for the production of complex lifeforms: “natural selection is a cumulative process, which breaks the problem of improbability up into small pieces” (p. 121). In other words, given that some primitive form of life exists, natural selection provides a mechanism which ensures that the development of complex life is, if not guaranteed, at least very very probable.

What about the origin of life? Dawkins freely admits that “in once sense, it is a bigger gap” and that the origin of life may have been an “extremely improbable event” (p. 135). When he has to conjure up odds for the sake of argument, Dawkins throws out a truly improbable number (1 in a billion, p. 138), although he does say later that he “doesn’t for a moment believe the origin of life was anywhere near so improbable in practice” (p. 138). Doesn’t this mean that complex life existing at all is incredibly improbable? No, says Dawkins, because of the anthropic principle (Dawkins is actually invoking the weak anthropic principle as opposed to the strong anthropic principle). There are a billion, billion planets in the universe. Even if the chances of life evolving spontaneously on a random planet is one in a billion, that means that there are a billion planets on which life began, and given natural selection, nearly all of them will have evolved complex life. Of course we are on one of the lucky ones, because if we were on one of the unlucky ones, we wouldn’t be sitting here wondering why there is life on our planet.

Let me try to summarize Dawkins’ argument thus far: given the (weak) anthropic principle, and natural selection, it is not at all surprising (i.e. it is probable) that there is a planet (perhaps many planets) somewhere in the universe which contain complex, sentient life like humans; there is no need to invoke a designer. Now we come to the problem: what Dawkins has presented thus far is not an argument, but a framework. He set out to show that there is a natural and probable explanation for the origin of complex life in the universe. If P is the probability for the existence of sentient life somewhere in the universe, then he claims that P is large (say > 50%), so we need not look for a creator God. According to his argument, P = p * N where p is the probability of spontaneous biogenesis and the subsequent evolution of life on a random planet and N is the number of planets in the universe. Since astronomers and cosmologists tell us that N = 10^20, the final, conclusive step in his argument is to provide an estimate of p and to show that p * N is large. So what is the probability that Dawkins calculates? He doesn’t provide one. Although this number is the cornerstone of his argument, he makes absolutely no attempt to calculate it.

Since this number is such a crucial piece of his argument, let’s try to estimate it using Dawkins’ (admittedly low) number 1/10^9 for the probability of the spontaneous genesis of life on a random planet and his estimate of the number of planets in the universe, 10^20. If these numbers are correct, then the probability that sentient life evolved somewhere in the universe is essentially 100%. But are we missing anything? Later in the chapter, Dawkins mentions that “it may be that the origin of life is not the only major gap in the evolutionary story that is bridged by sheer luck, anthropically justified. For example, my colleague… has suggested that the origin of the eukaryotic cell was an even more … statistically improbable step than the origin of life. The origin of consciousness might be another major gap whose bridging was of the same order of improbability” (p. 140). But if we take Dawkins at his word, something interesting happens. If -as he suggests- each of those steps were equally unlikely (1/10^9), then the probability of overcoming all three would be 1/10^27. Given that there are 10^20 planets, that leaves only a one in ten million chance that there is any planet, anywhere in the universe that contains sentient life like us.

Let me be clear that I am not a biologist, nor am I claiming that the probability of spontaneous biogenesis is one in a billion or one in a trillion, or any other number (if any molecular biologists are reading this, I would be very interested to know your estimate; I’ve asked biologists that I know and there doesn’t seem to be a consensus). My point is that Dawkins does not provide any number at all because he is taking his argument the wrong way around. If you are trying to prove that P is large and find that P = p * N, the next logical thing to do is to estimate p and N using what we know about physical laws from astronomy and biochemistry (see p. 137). It is a specious argument to instead assert “since we know P is almost 1, we can estimate p.” Unfortunately, this is precisely what Dawkins does. On page 140, at the end of his argument about biology, he says “The anthropic principle states that, since we are alive, eucaryotic and conscious, our planet has to be one of the intensely rare planets that has bridged all three gaps”. But the anthropic principle (as Dawkins is using it) doesn’t exactly say that. It says that we have a certain number (10^18) of planets to work with. If the probability of conscious life evolving spontaneously is greater than 1/10^18, then whatever our theory of biogenesis is, it is a probable one. But conversely, it also says that if the probability is significantly less than 1/10^18, then our theory is very unlikely indeed. It simply does not say “since we’re here, we must be a very probable event” (the strong anthropic principle does make this argument, but Dawkins doesn’t invoke it, presumably because it undermines his argument that there is a probable, natural explanation for the universe). Dawkins has constructed an elaborate framework, but has left out the final step which is the very crux of his argument.

My central objection to Dawkins’ reasoning is essentially this: he has mistaken one of his assumptions for a conclusion. What was his assumption? That there is a natural, probable explanation for the origin of life. If this statement is accepted as a postulate then, and only then, does his reasoning make sense. If there is a natural, probable explanation for the origin of life, then we can assert (indeed, must assert), as Dawkins does, that “our planet has to be one of the intensely rare planets that has bridged all three gaps” (p. 141). However, if we are trying to determine whether there is a natural, probable explanation for life, we certainly cannot use this reasoning. Well, why does Dawkins’ believe that there is a natural, probable explanation for life? I assert it is part of his faith in naturalism. At this point, this statement might appear a bit excessive, but I believe that further justification emerges when we examine Dawkins’ next argument regarding the values of the fundamental physical constants.

Dawkins points out that there are (at least) six  fundamental physical constants, which, if any of them were altered very, very slightly from their current values, would prohibit the existence of a life-supporting universe (usually due to the collapse of the universe within a few attoseconds of the Big Bang). Of course, this presents a similar puzzle as the origins of complex biological life and, in a sense, is a precondition for it: if these constants hadn’t lined up and the universe had collapsed, complex life wouldn’t exist.

Let’s stop for a moment at this point. We have been trying thus far to determine whether or not there is a natural, probable explanation for the existence of complex life somewhere in the universe. Let us assume that Dawkins’ argument about biology is correct: natural selection provides a mechanism that explains how otherwise highly improbable-looking evidence (biological life) has a very probable explanation. Dawkins takes great pains to show that the beauty of Darwinian evolution is that it provides such an elegant mechanism, without which the existence of life would be highly suspect. But what if we did not have an elegant theory like natural selection which purported to account for the complexity that we observe? Would not the existence of a finely tuned, complex ecosystem then be highly suspect?

That is precisely the case we find ourselves in when it comes to the fundamental constants. To quote Dawkins in what is a bit of an understatement: “we don’t yet have an equivalent crane [i.e. mechanism] for physics” (p. 158). In other words, given our current understanding of the laws of physics, there is no objectively verified theory which explains the coincidence of the fundamental constants. If they were determined by pure chance, then the probability that the universe would have been able to sustain life is ridiculously small (Roger Penrose apparently estimated the probability to be 1 in 10^(10^123) ). I think it is at this point that Dawkins’ presuppositions become most apparent. For instance, as far as I’m aware there is not a single piece of experimental evidence for a multiverse (see the recent review of Susskind’s book in Nature). In a preface to his treatment of multiverse theory in The Elegant Universe, Brian Greene states that “No one knows if these ideas are right or wrong, and certainly they currently lie on the outskirts of mainstream science” (p. 366). That is not surprising since the infinite universes postulated by multiverse theory are usually tucked away in black holes or in other dimensions where we can’t observe them. In the face of no concrete evidence and overwhelmingly negative odds, Dawkins states that “We should not give up hope of a better [mechanism] arising in physics” (p. 158). Perhaps we should not. But again, my objection is not about whether some alternate theory of physics exists that will explain life. My argument is that any belief that such a theory exists rests, as Dawkins says, on “hope” (p. 158), not on evidence.

A fundamental postulate of the naturalist (I use the word descriptively, not pejoratively) worldview which Dawkins espouses is that: “everything in the universe can be feasibly explained by natural laws”, a statement to which I think Dawkins would readily assent. But is this assertion based on solely on empirical, objective evidence? There is an easy way to find out. Can everything in the universe currently be explained by natural laws, as we now understand them? In the case of physics, at least, the answer is a resounding no. The immediate objection is that we would be able to explain these phenomena if we had the right theory. But how do you know there is such a “right theory”? Such an assertion merely brings us back to the original postulate. The assertion that “at some point in the future, we will have a theory of that explains everything” is no more or less evidence-based than the assertion that “at some point in the future, we will live on the moon”. Both of these statements are plausible; they may even be true. But they certainly are based, at root, on faith. That is, these statements form a set of axiomatic beliefs or presuppositions. We do not derive them from evidence; rather, they are part of our worldview.

Let me be clear that I am not disparaging Dawkins’ for having a worldview. I have one too. Everyone has one. You can’t do science or mathematics or anything unless you begin with a set of assumptions about the nature of reality. These assumptions may appear very reasonable and almost unavoidable, but it is important to recognize that they are assumptions, not conclusions. I also have deliberately avoided the question of God’s existence. I do happen to think that science gives us very clear reasons to question naturalistic assumptions and to believe in the God who has revealed himself in the Bible. Historical evidence regarding the life, death, and resurrection of Jesus Christ gives us even more. I also think that Dawkins’ philosophical arguments against God’s existence (for instance, his statement in Point 3 on page 158 that God is improbable because he is complex) are simply wrong. But for the purposes of this essay, I have limited myself to Dawkins’ scientific arguments in order to show that they are not as clear-cut as he claims and that scientific evidence does not necessarily lead us to naturalism. Rather, we import naturalism, or deism, or theism into our reasoning about evidence.

To restate my central objection, I believe that Dawkins is failing to distinguish between his assumptions and his conclusions. As a result, he is unable to see how much his worldview is coloring his interpretation of the evidence. When it comes to physics (and as a consequence to biology), the evidence we face is a set of fundamental constants which all conspire to permit the existence of life in a manner currently so improbable that it defies description. What is it that makes Dawkins so confident that such a coincidence has a natural explanation? What makes him sure that multiverse theory, or many worlds quantum theory, or a grand unification theory which so far have no objective justification will explain the universe? What makes him certain that, in the end, we will find a solution that does not involve a personal, omnipotent, creator? Faith. A set of basic, presuppositional, axiomatic beliefs through which we evaluate the evidence. Dawkins, like all of us, possesses faith. As human beings, we cannot decide whether to have faith; we can only decide what or whom to put our faith in.