Richard Dawkins is perhaps the world’s most famous atheist. He might not be the most insightful, or the most learned, however. Indeed, when it comes to fields beyond his own expertise, evolutionary biology, he is clearly out of his depth. Unfortunately, he seems not to realize it, making confident pronouncements on a range of subjects about which he knows very little. His so-called “analysis” of Thomas Aquinas’ Five Ways, found in his screed The God Delusion (hereafter, TGD), is a case-in-point: at just about every stage, he evinces a thorough lack of understanding of the good monk’s arguments for the existence of God. The mind boggles at how such an eminent public intellectual, who has enjoyed a glittering career communicating the seemingly opaque truths of science to a wide audience, can misread and misunderstand arguments that have been propounded, debated, analysed and scrutinised for more than 700 years. One wonders why the (now retired) Oxford academic couldn’t have simply walked across to, say, Blackfriars College and asked a resident theologian for an accurate summary of Aquinas’ thought. I’m sure they would have obliged.
Dawkins explores Aquinas’ arguments for the existence of God as a way of opening the chapter in TGD that purports to deal with the main rational and philosophical defences of theistic belief. Wading through his analysis makes one wish that he had excised this section from the book entirely; at least that would have been a more honest way of signalling to the audience that one is not interested in anything the opposing side might have to say. In any case, given my own interest in cosmological arguments, I thought that a sustained examination of Dawkins’ attempted critique of the Thomistic variety was appropriate – if for no other reason than to try and re-assure those who may have been overawed by his fame, and overwhelmed by his bluster.
In the relevant section of TGD, Dawkins attempts to critically analyse all of Aquinas’ Five Ways. However, only the first three “Ways” could be called cosmological arguments in any sense, and it is upon these that I shall focus. Dawkins simply conflates them, at one point writing that “they are different ways of saying the same thing” (emphasis mine); this, despite the fact that they are actually subtly different attempts to reason from various empirical observations, via metaphysical principles, and ultimately to God (perhaps the good Mr Dawkins isn’t really interested in subtle thinking – or charitable interpretation – when it comes to theological matters and theistic arguments. No…perish the thought!). There are several reasons why I think Dawkins’ arguments against Aquinas’ “proofs” fail (and do so ignominiously), leading one to the sobering conclusion that they are essentially worthless. One advantage of Dawkins’ sloppy conflation of Aquinas’ cosmological arguments is that my rebuttals are equally relevant to each one. As such, my remarks may be seen as generally applicable, though I will signal any specific references to one of Aquinas’ first three Ways. A quick caveat before we get started: this is not meant to be a complete exposition and defence of Aquinas’ cosmological arguments. Even though it will be necessary to explore elements of his metaphysics in some detail, my primary purpose is to expose the fallacies and misunderstandings contained within Dawkins’ attempted critique. It will be easiest if I simply enumerate my objections in sequence.
First, Dawkins seems, wrongly, to assume that Aquinas was interested in trying to argue against an infinite regress – at least, if one uses that term in a chronological sense, stretching backwards through time. He writes that Aquinas’ cosmological arguments “rely upon the idea of a regress, and invoke God to terminate it”. It seems clear that Dawkins employs the term in just such a chronological sense, because further on, he compares the Thomistic solution to the impermissibility of a regress with the Big Bang. The Big Bang is currently seen as the best explanation for how our universe began. But of course, it takes its place at the head of a causal sequence stretching across time, via the various stages of cosmic development. On several occasions, Dawkins refers to the supposed importance of this regress, and the claimed absurdity of such a phenomenon, to Aquinas’ argument(s). But this is quite misleading. Aquinas never relies upon the idea of an infinite regress, and whatever mileage one might get out of claiming that such a phenomenon is nonsensical is irrelevant to his purposes. Here is how W.P. Carvin, a philosopher of religion, describes Aquinas’ view, in reference to the argument for the Unmoved Mover, or First Way:
“…Aquinas was [not] thinking in terms of a ‘billiard ball’ theory of motion, where to say that ‘everything that moves is caused by something else in motion’ is to think of an endless chain of billiard balls stretching back into infinity, each one hitting the next forever. There would be no logical reason for there not to be an infinite chain of such billiard balls…”
This is exactly correct. Substitute “discrete cosmic events” for “billiard balls”, commencing with either the Big Bang or God as the “first member” of that causal chain, and you’d have a fair interpretation of what Dawkins thinks Aquinas is saying. And if Aquinas was arguing against an infinite regress of the kind upon which Dawkins bases his critique, then the English anti-theist might have a point. There would be no clear reason to think that whatever initiated this causal sequence was God – or if it was, that this God was required for anything more than the first step in that process. Indeed, as Carvin asserts (perhaps over-stating his case), there would be no logical reason to think that this kind of causal series cannot stretch backwards into infinity. But, as Carvin also suggests, this fails to see what the good monk was attempting to do. For one thing, when he speaks of “moving” and “motion” in the First Way, Aquinas uses those terms in much the same way as we might employ the word “change”. As such, he is not simply referring to the type of movement to which the billiard ball analogy would be universally applicable. For another, Aquinas’ analysis in the First Way, in addition to analysing local motion (movement from one place/location to another) refers to simultaneous, instrumental change. He talks of a present regress of causes, going downwards, through ever deeper layers of reality – not a succession of discrete events, going backwards in time. Contrary to what Dawkins seems to imply, Aquinas is not analysing a sequentially ordered series of change and causation (at least in the First and Second Ways) in an effort to provide a metaphysical account of these phenomena. Take, for example, a person playing hockey. The competitor uses a hockey stick to guide and manipulate a ball to wherever he wants it to go. The hockey stick itself functions as an instrument in the player’s hands, driving the ball forward. Thus, the ball, the hockey stick and the hands are working in a simultaneous causal relationship. There is also a relationship of instrumental dependence existing between these members: the hand is driven (“actualised”, to use the appropriate jargon) by the muscles, which are in turn actualised by the player’s nervous system, which are upheld by his physical structure, which is itself upheld by its atomic structure, gravity, the weak and strong forces, and so on. All of this occurs in the present, simultaneously, with each member necessarily dependant on others for its causal power. What we have, then, is a seamless ribbon of activity and actualisation. Talk of infinite regresses, as least in a temporal sense, is wide of the mark.
Even worse for Dawkins is the fact that in none of the relevant Ways was Aquinas hugely interested in whether or not, say, the universe had a beginning in time. In fact, as some philosophers have pointed out (e.g., Edward Feser), he didn’t think that the temporal beginning of the universe was something that could be demonstrated philosophically – and argued that it would still require a cause, even if were itself infinite. That alone should call into question Dawkins’ invocation of the Big Bang as a possible solution to the problem Aquinas posed; to raise it as a possible alternative to Aquinas’ view is, apart from anything else, simply irrelevant, precisely because the putative beginnings of the universe, or the manner in which it came into existence, or whether it came into existence at all (as opposed to being a necessarily existent “object”) are not issues to which Aquinas is trying to provide an answer. Moreover, as the hockey player example suggests, Aquinas’ argument does not require something as grand as the universe for explanatory purchase; the everyday objects of our experience are sufficient as a basis for its deployment. Invoking the Big Bang as some kind of explanatory rival to God for the beginning of our universe is to commit a category mistake (and arbitrarily fails to apply the causal principle to that particular event). So again, it is quite clear that Dawkins simply hasn’t done his homework.
The regress to which Aquinas refers, and which forms basis for the second half of his Unmoved Mover argument (i.e., the First Way), is one that, as I have noted, goes downwards through the various strata, or layers, that compose the reality within which we exist. The hockey player example, which I outlined above, is once more apt. And because Aquinas is talking about a hierarchical-instrumental, and not a temporal-sequential, regress, he argues that it must, of necessity, terminate in a Prime or Unmoved (that is, unchanged and unchangeable) Mover. His argument is that every other so-called “cause” in a particular causal chain is derivative, in that each does not possess independent causal power. Strictly speaking (so Aquinas says), it is only the First Mover that really causes anything at all. Remove that cause, and the entire sequence fails to become extant (I should also point out that in using the word “First” to describe this Mover, I am not thereby implying that it is simply the first in an ordinal series. Rather, I am using it in the sense of “primary” or “most fundamental”). Every subsequent cause depends on that first “movement”, for in its absence, there would be no such causes.
Think of it this way, as Aquinas did. He argues that there are basically two types of causal series: ones that are accidentally arranged; and ones that are essentially arranged. An example of the former type would be the relationships forming a genealogical line of fathers and sons: Kent begets Wayne, who begets Clark, who begets Bruce, and so on. But Bruce’s ability to in turn beget a son does not depend on the existence or activity of Kent. Kent could have long since died by the time Bruce got around to having any sons at all (for that matter, Wayne and Clark could have long since perished, too; this would not prevent Bruce from having children). The various parts of this causal series are accidentally arranged – not in the sense that they have been randomly or haphazardly thrown together without deliberation or foresight, but because the latter members of the sequence do not depend on the earlier members for their causal power (in this case, the generation of children). The hockey player analogy is an example of the latter type of causal series, which is essentially arranged: each member would be bereft of any causal power at all if not for the earlier members (“earlier” in a metaphysical, rather than a temporal, sense). What power they do have is derived or secondary; remove (say) the hand gripping the hockey stick, and both the hockey stick and the ball cease their activity. Remove an individual’s atomic structure, and he, along with the ball and hockey stick, will all cease “causing” or “moving” anything at all. From his references to modern cosmology, it is apparent that Dawkins is implicitly thinking of the former type of causal series when he refers to an infinite regress. However, on the assumption that Aquinas is thinking of essentially ordered causal series – where concurrently extant causes form a simultaneous sequence in the present, such that the removal of the earliest member would lead to the collapse of the series in its entirety – it would appear to be the case that an infinite regress would not be possible, even in principle. The philosopher, Edward Feser, puts it this way:
“What Aquinas is saying, then, is that it is in the very nature of causal series ordered per se [i.e. essentially] to have a first member, precisely because everything else in the series only counts as a member in the first place relative to the actions of the first cause. To suggest that such a series might regress infinitely, without a first member, is therefore simply unintelligible”.
Similarly, David Oderberg, in commenting on Aquinas’ First Way, suggests that an essentially ordered series is one where there is, as I have noted, a relation of instrumental dependence, in which members in that series would not be able to do or to actualise anything apart from the causal power of earlier members (which, in the final analysis, would also be derived). A causal series of this nature would, for want of a better word, be “rootless”, in that it would have no metaphysical foundation upon which to become extant. Oderberg asserts that even if (say) the universe were eternal, and had no beginning, it certainly does not preclude the possibility of a first member, the existence of which is metaphysically necessary – not to mention metaphysically (as opposed to temporally) prior – for the existence of such a universe.
Second, Dawkins thinks that Aquinas is far too hasty in arguing that this Unmoved Mover (or, if one were to apply this to the Second Way, a First Cause) is to be identified with, or given the appellation of, God. According to Dawkins, Aquinas is simply “conjuring up a terminator…and giving it a name…” and of capriciously “calling it [i.e. the terminator] God”. Anyone reading Dawkins’ appraisal of Aquinas’ arguments would come away with the impression that the good monk has simply leapt on the apparent problem of an infinite regress (misconceived by Dawkins, in any case), and arbitrarily invoked an already worked-out concept of God. Now, I cannot avoid pointing out the asinine attempt to pit theism and the Big Bang against each other, as Dawkins does. It’s clear that he doesn’t understand at all the difference between these two types of explanation. The former is a metaphysical argument; the latter is a probabilistic hypothesis that seeks to explain physical phenomena, and which is itself in need of explanation (as I noted above). As a result, Dawkins’ comparison is entirely unwarranted, since the two explanations function in different ways, and at different levels. Be that as it may, Dawkins is simply wrong to think that Aquinas has hastily, and arbitrarily, invoked God as an explanatory agent. According to Aquinas, whatever else God is, he is that which the cosmological arguments he uses say he would have to be. In the case of the First Way, for example, the Unmoved Mover is a being of pure actuality (to massively simplify this concept, it is the argument that the Unmoved Mover contains within himself no potentiality – in other words, no change, and nothing that needs to be actualised by something else. It is “pure act” in the sense of being fully and completely actualised, possessing no “latent” qualities or potential for change). It is the only way, according to Aquinas, that “movement” (i.e., change) can take place in the world. In saying that this is what all people know as God, Aquinas is suggesting that whatever else people mean by the label “God”, he would have to be (a) Being of pure act – changeless in himself, but effecting and guaranteeing change in other, contingent, things. The conclusion of, say, Aquinas’ First Way leads one to an Ultimate Mover; and it is this “Mover”, being ultimate, that Aquinas argues is identified with God.
Perhaps a couple of examples will suffice to demonstrate the shallowness of Dawkins’ critique at this point. Edward Feser, for instance, has said that Dawkins’ complaint is akin to someone who, when confronted with a three-sided polygon, suggests that it’s arbitrary to call such a figure a “triangle”, and that it would be more parsimonious to “conjure up a point or a line” instead. For a three-sided polygon just is a triangle; it is simply the nature of things for such a state of affairs to obtain. It’s not gratuitous to apply this term to such a shape, any more than suggesting that an Unmoved Mover would have to be identified with the notion of God as the ultimate explanation of (in this case) the various contingent instances of change or causation in the world.
My own example might also help to shed light on Dawkins’ error. Suppose that your mother is the only parent in your life, and that it has been this way for as long as you can remember. Suppose, too, you know that, although your mother bore you, she did not generate you by herself; there was another individual involved in that process. Let us also assume, for the sake of argument, that you do not have (complete) knowledge of your own origins, as well as the biological processes by which human individuals come into being in the first place. Now, you might be encouraged to embark on a journey of discovery: reading whatever material you can lay your hands on in an effort to discover how it is that people come to be, and even attempting to find the mysterious individual who was personally responsible (along with your mother) for your generation. Imagine that you find this person, and, having arrived at some understanding of how human beings are generated, recognize him as the person who was a partner in your creation – in other words, the one whose own genetic material was combined with material from your mother to eventually form you. What would you call this man (and a man he would have to be)? You would, of course, call him father, for that is what he would be. Your journey of discovery leads inexorably to the conclusion that, whatever else this individual might be – short, skinny, and battling a nasty case of halitosis – he just is your father. By definition, a father is a man who has a child; apart from any other qualities, this is the necessary, essential relationship he would have to the individual he helped generate. You could hardly be accused of conjuring up the figure of a father, or of using the word indiscriminately, in order to explain the role this man played in your generation. The nature of things means that it could not be otherwise. And anyone arguing that you’re guilty of haste, arbitrariness or caprice would be, shall we say, a little wide of the mark.
Hopefully, it’s clear by now where I am going with this. As you might have come to the inexorable conclusion that the man involved in your generation is necessarily your father; and just as you would unavoidably have to call a three-sided shape a triangle; so Aquinas’ cosmological arguments – to the extent that they are sound – inevitably lead to the conclusion that, whatever else one means by the appellation “God”, identifying it with the so-called Unmoved Mover (to refer to the good monk’s First Way) is not arbitrary, but metaphysically required. This Unmoved Mover – this ultimate explanation for why change occurs at all – is by definition God, in the same way that the man who helped create you just is your father. “God,” it should be said, is not a proper name, any more than “father” is. Rather, it is meant to capture the notion of an ultimate ground of being, change and causation; the final (or first) foundation upon which all reality sits. Aquinas’ arguments, to the extent that they succeed, lead to this Being as the referent to which “God” – whatever else people may imagine when they use that term – points.
Third, Dawkins’ own conception of God, which he uses to try and argue against the likelihood of the “God hypothesis”, is, as we shall see, hopelessly confused. The problem begins when he discusses Aquinas’ Ways and the issue of the infinite regress. There, he argues that defenders of theism “make the entirely unwarranted assumption that God is immune to the regress”, presaging a fuller treatment of what he considers his signature anti-theistic argument. It makes its appearance later in the book, and the English anti-theist even helpfully distinguishes it in the index of TGD (see ‘argument, author’s central, 188-9’). The basic thrust of Dawkins’ thesis is that any God capable of creating the cosmos would have to be at least as complex as the results of his creative activity, and as such, would have to be at least as improbable (if not more so). In discussing the so-called “Goldilocks Zone” within which the emergence of life is made possible, Dawkins writes that any god capable of calibrating the many variables of our cosmos to produce such a habitable zone “would have to be at least as improbable as the finely tuned combination of numbers itself [the so-called “six numbers” that refer to cosmological constants necessary for life], and that’s very improbable indeed”. Later, he repeats this assertion, arguing that “God, or any intelligent, decision-taking, calculating agent, would have to be highly improbable in the very same statistical sense as the entities he is supposed to explain”. And in challenging theologians and theistic philosophers to answer his seemingly insuperable charge, Dawkins asks how they cope with the argument that this god must be supremely complex and unlikely: “need[ing] an even bigger explanation than the one he is supposed to provide?”
Dawkins goes on to dismiss the efforts of theistic scholars like Richard Swinburne, who he says unjustifiably invoke the notion of divine simplicity to try and avoid the problem he has posed. According to Dawkins, Swinburne (for example) has utterly failed to provide proof that God would have to be “simple”, and suggests that this simply (pardon the pun) couldn’t be the case. Sounding like a broken record, he insists that this God would need a “mammoth explanation” of his own. All of this is deeply relevant to Dawkins’ attempted critique of Aquinas, for two reasons: on the one hand (and as we have seen), he thinks that Thomistic arguments for the existence of God are just as vulnerable to his central thesis as are those of a more modern “bent”; and on the other hand, it is well-known that Aquinas was a great exponent of the doctrine of divine simplicity, which is precisely the doctrine Dawkins derides as a gratuitous non-solution to the “problem” of divine complexity and improbability.
In regards to the former point, I have already noted how Dawkins’ critique seems to trade on a very simplistic – nay, incoherent – concept of God. That he could even countenance the idea that this “god” would himself require an explanation shows us that Dawkins is very far from the final explanation that Aquinas’ “Unmoved Mover” or “First Cause” seeks to provide. If Aquinas is correct, then the kind of god that is reached by his cosmological arguments would necessarily be without external explanation. A god who is apparently vulnerable to the irresistible flow of, say, an infinite regress – something Dawkins thinks is a possibility – is no god at all. He might, at best, function as some kind of extra-mundane being, which possesses of a degree of causal power, but which is itself in need of an explanation for its existence. And a “god” like this is a contradiction in terms. Any such being is simply – necessarily – not God; its candidacy for that title would be automatically invalidated by its explanatory “parent” (which, on Dawkins’ logic, would have to cede its position to yet another causal agent, and so on). One ought to remember that for Aquinas, God is the ultimate cause of change in this world, and the ultimate reason for the existence of contingent things. Anything less, and you’d be left with a finite (though perhaps still powerful) entity, whose existence is itself contingent, and whose being relies on something more fundamental than itself. As such, the metaphysical ultimacy inherent within Aquinas’ basic conception of God (and which I would argue is an essential component of any such conception) is utterly absent in the kind of Platonic demiurge with which Dawkins seems to be toying.
Moreover, as Keith Ward has pointed out, Dawkins appears to be allowing a particularly virulent kind of dogmatic materialism to infect his understanding of God. How else does one explain his insistence that any god capable of creating this complex universe would have to be even more complex (and to that extent, improbable)? I couldn’t find any specific justification for this assertion, save for the latent presumption that materialistic reductionism is absolutely and everywhere applicable. It seems that Dawkins cannot avoid analysing reality through the prism of his own philosophical worldview – namely, that all things must have, and be reducible to, a complex physical basis (it’s unclear whether Dawkins thinks that God must, in some sense, be physical; either way, it’s clear that he depends on a thoroughly incoherent idea of what God is). Unfortunately, this is simply assumed, rather than argued, but Dawkins treats it as a universal axiom to which even God is subject. As such, we can add question-begging statements to incoherency on the charge sheet. In any case, it should be clear by now that any god that is mechanically complex in the way Dawkins thinks would have to be composed of parts (whether material and/or metaphysical). But if this is the case, then, as Edward Feser has perceptively suggested, this being “would be metaphysically less fundamental than [said] parts, and would depend on some external principle to account for the parts being combined in the way they are”. And a derivative being like that could not, even in principle, be considered God.
But what of the latter point, namely, that the notion of divine simplicity is some kind of ad hoc attempt to ward off the disarming power of Dawkins’ central thesis? Well, it should be borne in mind that the doctrine of divine simplicity – whether in its ancient or modern iterations – was not cooked up as a way of answering the challenges of Oxford atheists. Rather, it is presented as something that follows organically, so to speak, from (in this case) the basic groundwork laid down by Aquinas’ first three “Ways”. As I said, that God is simple is an idea for which Aquinas was well-known – though Dawkins does not seem to make the connection in his analysis. For Aquinas, to say that God is simple is to say that he is in nowise composed of parts (whether physical or metaphysical). Now, it is too difficult to outline the complete case for divine simplicity here; that would entail a detailed exposition of the other Thomistic “Ways”, which would take us too far afield. But even if we were to confine ourselves to the good monk’s cosmological arguments, it is still possible to see how they might lead one to justifiably conclude that God would have to be simple.
For instance, if God is purely actual (as the First Way suggests), then it would seem to follow that he is unchangeable. That is, if God is purely actual, having no unactualised properties within him, then it follows that he cannot undergo change, either of a metaphysical or a physical kind; for change, on a Thomistic view of reality, entails the ontological transition from potentiality to actuality. This implies that God could not be composed of a body, since parcels of matter – out of which bodies are composed – seem universally susceptible to change. This appears to be a constitutional entailment of compositeness, of bits and pieces of physical “stuff” being brought together in different forms. And if God were a composite (likely, on Dawkins’ account), then he would not only owe his existence to something more fundamental than himself, as Edward Feser’s quote above suggests; he would himself be susceptible to change, just as much as the features of his putative creation. Thus, since the First Way suggests God is unchangeable, it also leads one to the justified conclusion that he could not be a conglomeration of various physical elements (thereby ruling out the idea that God might be a complex entity – and to that extent, improbable). Similarly, if God is imperishable and necessarily existent (as the Third Way suggests), then he would also have to “exist” outside time – thus, being eternal. To exist within time is to be susceptible to change, generation, decay and corruption, and therefore, to lack necessary existence. This, too, seems to be something inherent in composite “things”. Although it represents a slightly different starting-point, the Third Way also appears to allow Aquinas to say that God is not (materially) composite, but simple.
But in saying (for example) that God is not capable of change, and that he is purely actual (as opposed to being a composite of act and potency), the good monk is also arguing that he, God, is metaphysically simple. It’s impossible to enter into a detailed discussion of this point, given that it deals in some fairly arcane medieval metaphysics. Suffice it to say, however, Aquinas’ conclusions regarding God’s metaphysical simplicity logically emerge from prior arguments regarding his existence – in our case, the first three “Ways”. The upshot of all this is that divine simplicity in no way constitutes some kind of magical invocation, without basis and without reason. Rather, it seems to be an inescapable concomitant of Aquinas’ basic arguments for God’s existence. What is more, denying this tenet leaves one with the kind of incoherent concept that characterises Dawkins’ understanding of the Divine (and which I have already spent time critiquing).
What more can one say? It’s clear that Dawkins has little idea when it comes to Aquinas’ natural theology in general, nor his cosmological arguments in particular. His errors are far from insignificant – as if he were guilty merely of getting a detail wrong here and there. Rather, Dawkins’ writings betray a fundamental lack of understanding, which undermines his attempted critique of Aquinas at a structural level. At one point, he confidently assures his readers that Aquinas’ proofs “are easily…exposed as vacuous”. Having now read his foray into medieval metaphysics and theistic philosophy, I think one would be hard pressed to find a more appropriate epitaph for the English atheist’s own efforts.
 Richard Dawkins, The God Delusion (London: Bantam Press, 2006).
 One of the main Catholic theological colleges at the university.
 The philosopher Edward Feser provides a comprehensive and detailed summary of the Five Ways in his book on Aquinas, called (surprisingly) Aquinas. See Feser, Aquinas (London: Oneworld Publications, 2009), 62-130, and esp. 65-99 for an exposition of the first Three Ways. See also Feser, The Last Superstition: A Refutation of the New Atheism (South Bend, Indiana: St. Augustine’s Press, 2008), 91-101.
 I place “proof” in scare quotes, not because I think Aquinas’ arguments are actually worthless, or do not constitute metaphysical proof for the existence of God, but because I am not assuming this to be the case.
 Of course, I do not think that God has a gender, for as God, he is beyond gender. I use this for the sake of convenience.
 Feser, The Last Superstition, 100. I have slightly changed the parameters of the example as Feser has used it.