The God Delusion, chapter 3, part 1
One of my complaints about chapter 2 of The God Delusion was the sense of chasing rabbits. Early in chapter 2, Dawkins defined the “God Hypothesis” as the claim that “[t]here exists a superhuman, supernatural intelligence who deliberately designed and created the universe and everything in it, including us.” Much of chapter 2 was, strictly speaking, quite irrelevant to the God Hypothesis thus stated. The great bulk of chapter 3, however is directly on point. Here Dawkins considers, one by one, a variety of arguments in favor of God’s existence.
Thomas Aquinas’s “Proofs”
Dawkins is not particularly impressed with Thomas Aquinas’s five proofs of God’s existence from the Summa Theologica, and I can’t say that I blame him for this assessment. Dawkings groups the first three of Thomas’s “ways” together, since they all operate on basically the same principle of asserting God as the terminus to an otherwise potentially infinite regress. First, Thomas claims that everything in the world that moves has to be put into motion by a “mover,” but if you trace the chain of motion back far enough, you eventually have to come to a “first mover” (often called the “unmoved mover”), which according to Thomas “everyone understands to be God” (quoting directly from the Summa, not from The God Delusion). Second, Thomas claims that in the world we can apprehend with our senses, every effect has a cause, but if you trace the chain of causality back far enough, you eventually have to come to a “first efficient cause, to which everyone gives the name of God.” Third, Thomas claims that we readily perceive many things to exist that do not have to exist—at one time they did not exist, and at a future time they will no longer exist. However, anything that does not have to exist only comes to exist because of the actions of something that already exists. Therefore, Thomas reasons, there must be a necessary being—a being that has to exist and cannot fail to exist—which “all men speak of as God.”
In Dawkins’s view, Thomas’s three regress-style proofs (which are really just variations on a theme) “make the entirely unwarranted assumption that God himself is immune to the regress” (2:37:53). Dawkins regards the “conjuring up [of] a terminator to an infinite regress” as entirely arbitrary, but I am not sure that Dawkins is correct do to so. If one grants to Thomas that every effect needs a cause, then one must, at some point, entertain the notion of an uncaused cause. It is well nigh impossible to conceive of a chain of causation that stretches to the infinite distant past with absolutely no beginning point. On the other hand, Dawkins is spot-on when he objects that asserting a terminator and pasting the name “God” on it in no way gets you close to a personal god of the type asserted in the Abrahamic religions.
At this point in the chapter, Dawkins pauses to take a swipe at the concepts of omniscience and omnipotence. Before analyzing Dawkins’s comments, I should reveal that I have no great love for either term. The concepts of omniscience and ominpotence are, from my point of view as a student of the Hebrew Bible, Johnny-come-latelies who have little to do with the biblical portrayal of God and virtually everything to do with creating a philosophically respectable image of God in late Christian antiquity. That is, the concepts do not derive from the exegesis of biblical texts—the Bible doesn’t use these terms, and even the term “Almighty,” which you will find throughout English translations of the Hebrew Bible—is debatable as a translation-equivalent for shaddai (to me, it seems more of a substitution than a translation). Rather, the concepts derive from Christian attempts to define God’s character in a way that philosophically astute Christians in late antiquity would find to be worthy of worship. With that bias out on the table, you will understand why I am not going to try to defend the concepts of omnipotence and omniscience as such against Dawkins’s snarky “critique.”
Neither, however, am I going to affirm that Dawkins has put forward a strong critique of either notion. He hasn’t (at least, not at this point in chapter 3). Dawkins’s entire aside consists of a thin claim that omniscience and omnipotence are mutually incompatible. “If God is omniscient, he must already know how he is going to intervene to change the course of history using his omnipotence. But that means he can’t change his mind about his intervention, which means he is not omnipotent” (2:38:26). That’s the entire substance of Dawkins’s argument against these concepts (at least by this point in chapter 3). It is by no means incumbent upon Dawkins to oppose these concepts at all in order to disprove (or stack up large probabilities against) the God Hypothesis, since the God Hypothesis as Dawkins has stated it does not apply either label, much less both, to the creative intelligence that it does posit. Here again we have a red herring. On the other hand, if Dawkins is going to take swipes at specific theological dogmas, he needs to understand them. The philosophical-theological literature teems with controversy over just how omnipotence and omniscience are to be understood. The incompatibility that Dawkins asserts is only a genuine incompatibility if one defines omniscience as a factual knowledge of every actual state of affairs, past, present, and future, and if one defines omnipotence simply as the power to do anything at all. Some theologians have written entire books objecting to one, the other, or both of these simplistic definitions. The comment I just made remains within the purview of classical Western Christian philosophical theology; the entire situation becomes even more complicated—and Dawkins’s charge of incompatibility even more difficult to sustain—when one takes into account those varieties of Christian theology that do not press claims of divine omniscience, or of omniscience, or of either. Neither process theology nor open theism, for example, would be susceptible to Dawkins’s peremptory charge of incoherence.
Perhaps sensing something of the visceral appeal of regress-type arguments, Dawkins asserts that “it is more parsimonious to conjure up, say, a Big Bang singularity or some other physical concept as yet unknown” (2:39:07). However, Dawkins makes no actual argument in favor of this bald assertion. Occam’s Razor notwithstanding, the most parsimonious explanation is not always accurate. Even if it were, Dawkins provides his readers with no reason (other than his say-so) to accept his claim that “a Big Bang singularity or some other physical concept” really is more parsimonious than a supernatural intelligence as the terminator to the chain of physical causes. Indeed, one might well argue the reverse. A physical terminator to the chain of physical causes would be, in a sense, “inside the system,” and would still present the problem of its own causation. (Actually, in a sense, the Big Bang singularity would not be inside the system, but rather the system would be inside the singularity—and indeed, still is inside the singularity.) Neither a Big Bang singularity nor a supernatural intelligence completely evades the question “Where did that come from?” While some people, Dawkins included, find it easier to conceive of a spontaneously existing, spontaneously expanding singularity than of an eternally existing supernatural intelligence, many others find just the opposite to be the case. Parsimony may well be in the eye of the beholder, and if so, it is too subjective a criterion by which to discriminate between possible terminators to the chain of causation.
Thomas’s fourth and fifth ways of proving God’s existence depart from the regress-termination model, but Dawkins is no more impressed with these than with the first three. As Thomas describes it in the Summa:
The fourth way is taken from the gradation to be found in things. Among beings there are some more and some less good, true, noble and the like. But “more” and “less” are predicated of different things, according as they resemble in their different ways something which is the maximum, as a thing is said to be hotter according as it more nearly resembles that which is hottest; so that there is something which is truest, something best, something noblest and, consequently, something which is uttermost being; for those things that are greatest in truth are greatest in being, as it is written in Metaph. ii. Now the maximum in any genus is the cause of all in that genus; as fire, which is the maximum heat, is the cause of all hot things. Therefore there must also be something which is to all beings the cause of their being, goodness, and every other perfection; and this we call God.
Dawkins’s initial reply to this proof is the incredulous exclamation, “That’s an argument? You might as well say that people vary in smelliness, but we can make the comparison only be reference to a perfect maximum of conceivable smelliness, therefore there must exist a pre-eminently peerless stinker, and we call him God” (2:41:03). Again, I’m not especially interested in defending Aquinas’s fourth way. To me, it just smacks far too much of Platonic idealism. But Dawkins has either misunderstood or intentionally misrepresented Thomas’s argument. Thomas does not assert that God is the maximum conceivable exemplar, and therefore cause, of every known property. Thomas might well agree that “there must exist a pre-eminently peerless stinker,” or at least, that there is a pre-eminently peerless exemplar of whatever is in the genus of stinkiness. But Thomas would not call this exemplar of maximum stinkiness “God” any more than he calls the exemplar of maximum heat “God.” I am not really convinced that Thomas’s fourth proof really proves anything—certainly Thomas is wrong about heat, so why should we think he is right when he applies the same principle to God? Yet despite this demurrer, it remains the case that Dawkins’s snide reply does not actually touch Aquinas’s argument at all. He mocks it, but does not engage it and certainly does not refute it with his mockery. The funny thing is, it’s not really that hard to refute Aquinas’s fourth way, precisely on actual scientific grounds. It just isn’t true that “the maximum in any genus is the cause of all in that genus; as fire, which is the maximum heat, is the cause of all hot things,” but this is the basis of Thomas’s fourth way. Why Dawkins goes in for a mocking attempt at reductio ad absurdum rather than simply refuting the flawed science underlying Thomas’s fourth proof is beyond me.
Thomas’s fifth way is the teleological argument, which Dawkins glosses as the argument from design. Dawkins’s paraphrase of the fifth way goes something like, “Things in nature look designed. In our human experience, things that look designed are designed by intelligent designers. Therefore, nature was designed by an intelligent designer, whom we call God.” That’s the form in which the “argument from design” is normally seen these days, but it’s not exactly what Aquinas put forward. Rather, Aquinas’s argument is a bit more subtle (though not necessarily, by that fact, more convincing). Thomas argued that “things which lack intelligence, such as natural bodies, act for an end,” but
whatever lacks intelligence cannot move towards an end, unless it be directed by some being endowed with knowledge and intelligence; as the arrow is shot to its mark by the archer. Therefore some intelligent being exists by whom all natural things are directed to their end; and this being we call God.
Dawkins’s rejoinder to the teleological argument is simple and straightforward: biological evolution provides a remarkable demonstration and explanation of nonintelligent processes leading toward an outcome that appears to be designed, but is not. Naturally, creationists of all stripes, including religious intelligent design proponents (I know, I know, that’s a tautology), would object that this is precisely the question at issue, and that therefore Dawkins cannot rightly use evolution as evidence against the teleological argument. However, evolution by natural selection is, to be blunt, a biological fact, and it really does throw a major monkey wrench into teleological arguments for God’s existence. This is not to say that evolution automatically scuttles teleological arguments, but it does present a major challenge to the same.
Of Thomas’s five ways, the fourth is obviously based on deeply flawed premises, and really needs no further discussion. The fifth way, the teleological argument, is open to very strong challenges from science. I am not sure that it is completely decimated by the fact of biological evolution, and Darwin’s explanation of that fact by natural selection, but I have not yet had time to thing deeply enough about that question to comment at this particular point. I am less impressed by Dawkins’s attempt to refute Thomas’s first three ways by grouping them together under the rubric of infinite regress. I don’t mind the grouping at all, and in fact I think it’s fair to say that, at bottom, the Summa basically offers three ways rather than five. Yet I continue to harbor a suspicion that Dawkins has dismissed the argument from regress a little too quickly, and that a productive (but by no means airtight) argument may still lie at the intersection of the argument from regress and the teleological argument. I don’t yet know how to formulate these thoughts in a more rigorous fashion, so I’ll return to this question at a later point.
12 comments Christopher Heard | books, religion, religion and science

“late Christian antiquity”
I understand that you were probably referring to the OT, but for the record Revelation 19:6 uses “pantokrator” for “all powerful/mighty” or “omnipotent” with respect to Kurios o Theos (Lord God).
“certainly Thomas is wrong about heat”
I don’t see how. There are no 2 of anything exactly equal. Heat boils down to excitation, which can’t be narrowed down with exact/perfect precision.
Thomas is wrong about heat in the sense that he thinks all heat comes from (elemental) fire. You do not need fire to produce heat, and not all heat can be traced back to (elemental) fire.
I see. Thanks for explaining. I had misinterpreted your remark as being relative to his overall 4th “way/proof”. You’re correct that he was wrong about fire being the “maximum heat”, but would it not be fair to give him the benefit of implying that it was the “maximum heat known“ to him, rather than believing that he really thought it was the hottest thing in the universe?
Dawkins claims that omnipotence should mean that God should be able to change his mind. However, if God knows everything past, present, and future, then God can get it right the first time, and in his omnipotence, he would. So isn’t Dawkins complaining that God doesn’t have the power to fix a mistake that he wouldn’t have made in the first place?
I want to invite you and your readers to join us in reading and discussing Dawkins “The God Delusion” during Q1, 2007. I’m working on getting him in a live chat session for some time in March 2007, but nothing is set in stone. If this chat happens you are welcome to attend.
We had Richard Dawkins for a live chat back in 2003 where we discussed “Unweaving the Rainbow.”
http://www.booktalk.org/
Chris O’Connor
You say: “If one grants to Thomas that every effect needs a cause, then one must, at some point, entertain the notion of an uncaused cause. It is well nigh impossible to conceive of a chain of causation that stretches to the infinite distant past with absolutely no beginning point.”
I don’t see why this is any more inconceivable than to imagine the negative numbers strech back to minus infinity without a beginning point. I can see why some people might feel uncomfortable with the idea of an inifinite regress of causes or of explanations, but I have never seen a convincing argument that this is strictly inconceivable (expect perhaps in a weak sense of “unimaginable” -but that is irrelevant to proving actual impossibility).
Alejandro, I do see a distinction in that numbers are merely concepts; they are not “causes” and “effects.” 2 is not caused by 1, nor is -1 caused by -2. If numbers were caused, one might even suppose 0 to be the terminator from which numbers flow in multiple directions.
If you go to http://www.firstthings.com/ftissues/ft0104/correspondence-oakes.html you will find some interesting discussion of Aquinas’s fifth way (in Oakes’s replies) which makes the point that the argument is not really an argument from design, but an argument from order.
Christopher,
I’m not sure why the (correct) distinction you point out is relevant for conceivability. I *think* I can conceive an infinite series of causes with no first cause, in a similar way as I conceive the negative numbers: thinking “there is always a previous cause(number) causing(before) this one”. How can you show my “conception” to be mistaken?
OK, just a few points that might not be obvious to people who don’t actually study medieval philosophy:
(1) Aquinas has no problem with infinite regresses. In fact, he argues that it’s logically possible for the universe always to have existed on the same reasoning Alejandro gives. The reason he has a problem with infinite regress of the particular sorts of causes he considers in the Five Ways is that he has reasons — several independent arguments, in fact — that such a regress logically implies a contradiction. These issues are discussed more fully elsewhere (in particular, in the Summa Contra Gentiles and the Commentary on Aristotle’s Physics); he doesn’t go into them in the Summa Theologiae since the Five Ways are not Aquinas’s arguments for the existence of God, but his bare-bolts summaries of them. The ST was an introductory work in theology for people who would already have been familiar with most of the philosophical issues raised in those arguments. In any case, Dawkins can very easily be shown wrong in his claim that Aquinas simply assumes that God is immune to the regress; Aquinas has a rather interesting and involved discussion of these issues with infinite regress in a number of places. In all these cases even a serious look at a very introductory work on Aquinas’s arguments would have sufficed for Dawkins to recognize this.
(2) Thomas’s First Way is not an argument about motion in our sense of the term; the problem here is that Aristotelians like Thomas were using the term in their sense long before our modern sense of the term came about. Likewise, it is not concerned at all with chains of causes stretching back in time; none of the Five Ways are.
(3) The Fourth Way is one of the hardest of Aquinas’s arguments to interpret; but it is worth noting that the claim about maximum heat is often misunderstood. Aristotelians assumed that heat must be a particle; they called this particle ‘fire’. To claim that fire is the maximum in the genus of heat is to say that there is no more fundamental cause of heat than the fire, the particle of heat. The Aristotelians were wrong, of course, in thinking that heat must be a particle; it took quite a bit of experimenting to lead people to recognize that the genus of heat was of a completely different nature than people had originally thought. It’s also worth pointing out that Aquinas’s argument doesn’t depend on this particular example; he develops other arguments for that premise elsewhere — it’s just that in an Aristotelian worldview, the case of fire, understood in their sense, was important enough to be a case easily recognizable to anyone, so, of course, there was no reason for him not to mention it.
(4) As Michael Kremer notes above, the Fifth Way is not a teleological argument. This is because the term ‘final cause’ or ‘end’ in thirteenth century philosophy means something rather different than it does with us; what we call final causes would be considered only one, very unusual kind of final cause in the medieval sense. Thomas’s sense of ‘final cause’ is that which makes it such that this cause has this effect rather than some other effect.
Dawkins’s objections show a problem that pervades his entire discussion, namely, his lack of even basic research; it takes hubris to think that you can just pick up an isolated passage in a colloquial English translation of a technical philosophical text written in thirteenth-century Latin, without any regard for context or what the author says elsewhere, and see the obvious flaws in the discussion. Most of the mistakes Dawkins makes in this chapter (which are legion) are rookie mistakes, easily avoided by consulting basic references and doing a little reading. Even where there are weaknesses in the arguments he considers in this chapter, he regularly latches on to things that are not weaknesses in the argument but in his oversimplistic interpretations of them (the infinite regress one is the one that keeps me laughing). It would be one thing if he were in a position, as most people are, where he could just go on what seemed offhand to be the case (until better information came along), but if you are writing a book called ‘The God Delusion’, which has a chapter purporting to show problems with arguments for God’s existence, to botch so laughably almost every single step of the argument hardly speaks well for one’s reasoning on the subject.
Alejandro, the point is that numbers don’t “cause” each other.
Brandon, thank you for these very helpful comments. As someone who has read only a little of Thomas Aquinas (the Summa, chiefly), I found them very illuminating.