November 9, 2009

This Post is Old!

The post you are reading is years old and may not represent my current views. I started blogging around the time I first began to study philosophy, age 17. In my view, the point of philosophy is to expose our beliefs to rational scrutiny so we can revise them and get better beliefs that are more likely to be true. That's what I've been up to all these years, and this blog has been part of that process. For my latest thoughts, please see the front page.

What Caused God?

In comments to my post on Dawkins and the Philosophers, atheist blogger Jonathan West has been pushing back against Michael Ruse's claim that Dawkins' prominent use of the "what caused God?" question is, as Jonathan puts it, 'fatuous.' Jonathan has also pushed this point in a recent blog post which considers this question in light of Swinburne's 'necessary being' arguments in The Existence of God. I will first make a few remarks about Swinburne's work in this area, and then proceed to show why the "what caused God?" question is indeed confused. To be fair, I admit that there are some other questions in the neighborhood of "what caused God?" which are neither confused nor easily answered. More on that later.

A quick perusal of Jonathan's blog (which I hope to read and respond to regularly in the future) will show that he has engaged quite seriously with Swinburne's book. (In his comments to my post he provides a table of contents.) For this I applaud him. It is true, as Jonathan notes, that Swinburne is "one of Britain's leading academic theologians." I would say "theistic philosophers" rather than "theologians," but the point stands. No one should be accused of attacking a strawman who is engaging Swinburne's arguments so seriously as Jonathan is. However, it should be acknowledged that this necessary being line is not Swinburne's strong suit. Swinburne is known for a posteriori natural theology, and especially arguments from miracles. In this area, he is especially known for his creative applications of Bayesianism. He is also well known for his defense of substance dualism (which, it should be noted, neither entails nor is entailed by theism or Christianity more broadly - this is one of the many reasons why we should call him a "theistic philosopher" rather than a "theologian"). Still I think most philosophers would agree that the best recent work on necessary being theology is due to Alvin Plantinga, rather than Swinburne, so while Swinburne is certainly no strawman, his is also not the best view on the market.

Second, we should note that there is no principled reason to restrict ourselves to living proponents when classical theism is a tradition twenty-some-odd centuries long. (Note that the development of classical theism begins earlier than Christianity; most of the basic tenets were around in, e.g., ancient Stoicism, and there are elements to be found in both Plato and Aristotle, as well as even some of the Presocratics.)

Now, I am not familiar with Swinburne's book The Existence of God. As I hope the above indicates, I have great respect for Swinburne as a philosopher. Of the works of his that I have read, my favorite is Revelation: From Metaphor to Analogy. However, all this is simply to say that I don't think the quotes from Swinburne in Jonathan's post deal with the issue of divine necessity very well. In particular, I think Swinburne must be mistaken in his claim that "To say that 'God exists' necessarily is ... to say that the existence of God is a brute fact that is inexplicable - not in the sense that we do not know the explanation, but in the sense that it does not have one." The reasons why I think this is a mistake will become apparent later on. As such, I will give my own account of why the 'what caused God?' question (hereafter 'The Question') is confused, and will not base it on Swinburne.

There are three reasons for supposing that The Question is confused: God is non-physical, God is eternal, and God exists necessarily. I will run quickly through the first two because, while I think they do have some plausibility, the third is surely the most sophisticated and compelling reply. At the end, I will discuss some related questions that are left over from this line of reply.

God is non-physical. Causation, many people (especially philosophical naturalists) suppose, is a concept from physics, or perhaps from natural science more generally. On this view, if God is non-physical, then The Question doesn't make any sense. Many atheists do believe the premise of this line of argument, and some of them use this premise to argue against cosmological arguments: since physics begins from the Big Bang, and causation is a physical concept, it doesn't make sense to inquire about the cause of the Big Bang.

Still, this line of reply from the theist is not very good. This is because theists generally believe that God is the cause of the world, and this means that the concept of causation can be applied to non-physical entities and, specifically, to God. So the atheist can still propose The Question as a non-fallacious ad hominem - that is, the atheist can argue that from within the theist's own system The Question needs an answer.

God is eternal. This reply is, I think, a little better. Causation, according to this reply, is a temporal concept. That is, it is part of the concept of causation that the cause precede the effect in time. Since God is eternal, The Question doesn't make sense. Again, some atheists use this type of reasoning against cosmological arguments: since nothing precedes the Big Bang, it doesn't make sense to ask about its cause.

This reply will, I think, be strengthened if God is conceived as atemporal rather than (as Swinburne thinks) merely everlasting. Then a case can be made that the Big Bang does need a cause, but God doesn't. Or so one might think. I won't stop to defend this here, because, as I said earlier, I think the necessary being reply is better.

God exists necessarily. Nothing causes a necessary truth to be true. For instance, nothing is the cause of two twos adding up to four. Similarly, mathematical platonists would hold that nothing causes the number two to exist. (Plato's claim that numbers are caused by The One and the Indefinite Dyad - if he did indeed make that claim - should be attributed to the Greek word aitia having a broader semantic range than the English word 'cause'.) Asking for the cause of a necessary truth is a conceptual confusion.

Consider this simple argument, which is basically a standard cosmological argument:

  1. Every contingent entity or event has a cause.

  2. There are no causal loops and no infinite causal regresses.

  3. Therefore,
  4. There is at least one necessary (i.e. non-contingent) cause.

Call the necessary cause G. Now, even if this argument proves that G exists, it doesn't prove that G is God. (Depending on how you think the reference of the name 'God' is fixed, you might think this argument is enough to support the claim that 'God' refers to G, but even if that's the case, it leaves open the possibility that religious believes are radically deceived about the nature of God.)

Now, I want to note that, in addition to the platonist example above, there are many atheist philosophers who think that contingent beings, events, or facts need causes or explanations of a type that necessary beings, events, or facts don't. For instance, Peter Unger thinks that it is a point in favor of David Lewis's (in)famous modal realism that it makes the existence of a world like ours a matter of logical necessity (see sect. 2.7 of Lewis's On the Plurality of Worlds). Other atheists have thought that the laws of nature should be thought of as necessary rather than contingent so that they don't need a non-naturalistic explanation. (Sydney Shoemaker provides an argument that this might be the case in his "Causality and Properties", originally in van Inwagen, ed., Time and Cause, reprinted in Mellor and Oliver, eds., Properties. After concluding that the laws are necessary, Shoemaker notes that "to some this may be an attractive consequence" (sect. 9), but he himself seems mostly to regard it as an objection.)

However, many people think that there is something more problematic about God, a bona fide substance, existing necessarily. It is all well and good for facts, or perhaps abstract entities like numbers of properties (on Shoemaker's view, certain facts about properties ground the laws; the laws are not necessarily existing entities), to be necessary, but surely there are no necessary substances. (Lewis of course thinks that possibilia are bona fide substances, and that everything possible is necessarily possible, and everything that is possible in some sense exists, so in the broadest sense, everything exists necessarily. However, he would object to this language on the ground that when we say 'exists' we usually mean 'actually exists', and possibilia do not actually exist, because they are not part of the actual world - i.e. the one we inhabit.)

Well, perhaps G is not a bona fide substance. The argument above doesn't say that it is, it just says that G is a cause. Perhaps there are non-substance causes, and G is one of them. If so, then G is not the traditional God. I won't try to answer this objection here.

The view that what is necessary needs no cause is not a uniquely theistic idea. In fact, nearly everyone who believes in absolute (as opposed to hypothetical) necessities believes it. However, as I indicated above, this is not the end of the story. What I have been arguing is that if someone asks what the cause of two twos adding up to four is, I can answer that 2+2=4 is a necessary truth, and this will show that the question is confused - there is no cause. Similarly, if someone asks what I have been calling The Question - "what caused God?" - and I respond that God exists necessarily, I will have shown that The Question is confused. Necessities don't need causes. However, there is another question that needs to be asked: why is it necessary? In the case of 2+2=4, some philosophers and mathematicians do actually try to answer this question, and they may do so in terms of the concept of unity, or in terms of the von Neumann ordinals, or the definition of the successor function, which is used to define the + operator, and so on.

So why is God's existence necessary? Even if the cosmological argument succeeds, it leaves this question open. While most philosophers don't think the ontological argument can be used to prove that God exists, it is my understanding that a few (Alvin Plantinga and James F. Ross) think that it can provide the kind of explanation that is wanted here. The strategy here would be to use a cosmological argument like the above to show the existence of G, and then use the ontological argument to show that something like the traditional God is a good candidate for being G. (Plantinga and Ross both think, if I understand them, that the ontological argument can be used to non-trivially derive necessary existence from the definition of 'God', which shows that the existence or non-existence of God is a matter of necessity, so that the possibility of God entails his actuality. Thus God is, in the Scholastic phrase, a being whose "essence includes existence.") I will admit to not quite understanding how this is supposed to work, so I will stop here, at least for now.

Posted by Kenny at November 9, 2009 10:20 AM
TrackBack URL for this entry:
Sobel's Logic and Theism: An Introduction to my Forthcoming Discussion
Excerpt: Some time ago, I promised that I would take time on this blog to seriously engage with some recent work arguing in favor of atheism. The book chosen, mostly on Brandon's recommendation, was Jordan Howard Sobel's 2003 Logic and Theism: Arguments For and...
Tracked: August 17, 2010 8:22 PM
A Genuine Dialectical Problem for Ontological Arguments
Excerpt: Sobel spends much of the third chapter Logic and Theism evaluating the dialectical status of ontological ar...
Tracked: September 11, 2010 4:32 PM


I look forward to your other posts on this topic.

I confess that I have serious difficulty myself taking Swinburne seriously; enough so that I wouldn't recommend to any of my atheist friends that they bother him. It's not that he's stupid or uninteresting, but I can count on one hand the arguments I've read by Swinburne that I think even plausible, and the reason is always that Swinburne tends to take his interesting but often highly controversial views about the philosophy of explanation (and his currently less controversial but still quite controvertible views about philosophical Bayesianism) and apply them directly to these matters as if they resolved anything -- which they don't, unless you happen already to have been convinced by Swinburne's very particular and very controvertible views of explanation. I honestly think that thirty years from now people will find Swinburne about as interesting as we find Joad now. But I admittedly have some biases here; I thoroughly dislike epistemological Bayesianism, of which Swinburne is one of the prominent defenders, and I can't help but feel like strangling someone everytime I come across it.

Posted by: Brandon at November 9, 2009 7:37 PM

Brandon - I do think that your remarks point out something potentially problematic, but also quite interesting in Swinburne's work: he likes to borrow from topics in mainstream (secular) philosophy in making his theological arguments. I think this often makes his work interesting, and I also think that he writes very well on these topics. For instance, the first half of The Christian God gives a very intelligent and understandable (albeit one-sided) view of a number of metaphysical issues suitable for the non-specialist.

As you point out, though, this has the potential to become a corresponding weakness, because these topics are controversial, and fads come and go in them. This probably undercuts Swinburne's chances of any kind of enduring relevance.

The only book-length works of Swinburne's I'm very familiar with are The Concept of Miracle, Revelation: From Metaphor to Analogy, and The Christian God. As should be clear from the Leibniz paper I recently posted, I think Swinburne's work on miracles is totally misguided, but I still think it's interesting. I thought The Christian God was hit and miss - some parts very good, some parts not so much. I thought the first half of Revelation was excellent, and it has had a pretty significant effect on my views in that area.

Posted by: Kenny at November 9, 2009 8:07 PM

I'll leave aside the "God is non-physical" and "God is eternal" arguments. Although in my opinion they are non-arguments, they are not central to the question of necessary existence. We can discuss them some other time perhaps.

Regarding necessary existence, your brief statement of the cosmological argument is logically valid, in that if the two premises are correct, the conclusion logically follows. But there are two problems with this. First, we don't know whether the two premises are true. Second, even if the premises are correct and the conclusion therefore valid, the mere fact that there is a first cause does not mean that it has the other properties (omniscience, reading minds, answering prayers etc) we traditionally ascribe to God. The capabilities God is believed by religions to have involve very great complexity and heterogeneity of parts (even if they are non-physical parts). Quite why such complexity makes God a good candidate for "G" is not explained.

Swinburne wishes this away by declaring by fiat that since God is made from a single substance (presumably a non-physical substance, whatever that might be), God is therefore simple and therefore a good candidate for "G". That just won't do. Also, Swinburne claims that infinities and zeros are both simpler things to explain than any finite value or quantity of something, and so a God of infinite power and knowledge is again a good candidate for "G". But again, he doesn't justify why infinities are simpler, and why a single God is simpler than an infinity of Gods. You may not think Swinburne's arguments in this area to be particularly appealing, but the fact is that he is expressing fairly conventional views of theistic philosophy.

So, so summarize, we don't know whether the cosmological argument is valid, and even if it is, we don't know why the first cause would result in God rather than something else. So, the question "Why is the first cause God?" is a very live philosophical question.

This is pretty close to asking "What caused God?", which Michael Ruse criticised Dawkins for asking. In fact, Dawkins in TGD discussed first cause arguments, and it is perfectly clear that he was talking in terms of asking "Why is the first cause God?".

So Ruse was criticising Dawkins by misrepresenting the way in which Dawkins phrased the question. That is one further misrepresentation to pile on top of all the others he made. I'm not surprised, though I am disappointed.

Posted by: Jonathan West at November 10, 2009 3:29 PM


I have given above a standard, mundane, conventional line of thought that is (at least) several centuries old. The point here is that anyone who has seriously studied the philosophy of religion should be familiar with it. This post was totally, completely unoriginal, and that is precisely the point. By the way, the locus classicus for this general line of thought is book one of Aquinas's Summa Contra Gentiles.

In light of the above, the question "what caused God?" is, strictly speaking, a conceptual confusion, and anyone who has seriously studied philosophy of religion should recognize that. However, there are two very similar questions that can be asked, and I think your comment confutes the two: first, what, according to this hypothesis, makes God's existence necessary? Second, why should we suppose G to be something like the traditional God? These questions are similar to "what caused God?" but differ in more than just phrasing.

If Dawkins intends to ask one of these two legitimate questions and uses the phrasing "what caused God?", then there are three possibilities:

(1) It could be that he is not sufficiently familiar with the tradition of philosophical theology (and of philosophy generally) to recognize the important distinctions at work here.

(2) It could be that he understands the difference between the illegitmate question and the two legitimate ones, but thinks that, because the legitimate questions have not been adequately answered (as, it seems, they are not adequately answered by Swinburne), the difference doesn't matter much and can be safely glossed over in a work intended for popular audiences.

(3) It could be that he understands the difference, but glosses over it because "what caused God?" has more rhetorical punch, and he does like rhetorical punch.

You seem to favor interpretation (2); Ruse favors interpretation (1). I would have to take a close look at TGD in order to have an opinion. However, any of these three possibilities is sufficient to persuade me that I, as a philosopher, need not take a professional interest in Dawkins' book. That is, the fact that he asks this question is sufficient in and of itself to show that his book is not a rigorous philosophical work. (Further evidence for this is that I spend most of my time around professional philosophers, very few of whom believe in God, and none of them has ever recommended that I read Dawkins, and some of those who have read Dawkins have expressed the same kind of embarrassment as Ruse.)

One thing this means is that we should hold Dawkins up to the standard of pop philosophy rather than professional philosophy. At this level, perhaps he will do well. After all, it doesn't take a lot of philosophical skill to refute the ridiculous things pop apologists for Christianity go around saying.

Posted by: Kenny at November 10, 2009 3:54 PM

By all means look at TGD and make your own view, but the point I was trying to make is that Dawkins is in fact asking "why should we suppose G to be something like the traditional God?" and that Ruse is either mistaken or deliberately misrepresenting him when turning the question into "What Caused God?"

Posted by: Jonathan West at November 10, 2009 4:14 PM

Hrm... A search on google books does not show the phrase "What caused God?" appearing anywhere in TGD. If Dawkins does not in fact use this phrase, then Ruse's sarcasm may be overblown or otherwise undeserved. On the other hand, Dawkins is a great one for hyperbole himself. I, for one, read Ruse as implying that Dawkins uses that phrase not once, but many times. Perhaps he has done so somewhere other than TGD. If not, we should interpret Ruse as sarcastically implying that Dawkins' arguments are no better than the 'what caused God?' question, and it is Ruse who has the overblown rhetoric here.

Posted by: Kenny at November 10, 2009 4:43 PM

Par for the course. Dawkins is well used to being misquoted and misrepresented, and all those who have read him are well used to seeing it.

Posted by: Jonathan West at November 11, 2009 1:10 AM


Thank you for your comments noting that I have addressed Swinburne's book seriously. I appreciate the fact that you recognise this.

Now that you have seen that an atheist can make a serious and detailed analysis of a book written from a theist perspective, can I challenge you to do the same in reverse? How about it if you do a chapter-by-chapter analysis of The God Delusion, quoting relevant passages which you feel illustrate or encapsulate Dawkins' key points.

In the fourth chapter of The God Delusion, Dawkins describes the evidence and line of reasoning that leads him to conclude that "there is almost certainly no God". Since you are a Christian, you must think he is mistaken in that conclusion. Therefore, I would hope that in the course of your review you would be able to state what evidence he has missed or what error of reasoning he has made which caused his conclusion to be wrong.

Posted by: Jonathan West at November 11, 2009 12:10 PM


I am indeed interested in seriously engaging with atheistic arguments and points of view. However, for the reasons I have already given, I think that Dawkins would be a poor choice of atheist literature for me to spend my time with. The most important of these reasons is that my atheist colleagues would think I was attacking a strawman. (My impression that they would think this was supported by an informal poll I took in the philosophy department today: of six atheist philosophers, one had a mostly positive view of Dawkins, one had a strongly negative view, and the rest were neutral. The one who had a positive view said that he bought TGD "to support it" because he agreed with Dawkins' encouraging atheists to be more "loud and proud." However, even he didn't want to endorse the arguments in the book, and did not recommend that I read it.)

Perhaps in the near future I will spend some time with Hume's Dialogues on Natural Religion, or some of J. L. Mackie's papers.

Posted by: Kenny at November 11, 2009 5:58 PM

J. H. Sobel's Logic and Theism is a fairly good atheistic tome. Whoever was in charge of the editing should be shot, since the book is riddled with typos and misprints, but it's easily one of the best book-length philosophical works on atheism currently out there. (And certainly better than anything in Mackie, IMHO.) So very nice to have an atheist who actually discusses the modal issues in arguments from evil.

Strictly speaking, Hume's Dialogues wouldn't satisfy a requirement to comment on an atheistic work because it's not an atheistic work, and the evidence that Hume was an atheist (as opposed to a non-personalist theist) is almost nonexistent. Actually, another reason to recommend Sobel: he gets Hume more-or-less right on this point.

Posted by: Brandon at November 11, 2009 9:20 PM

I may look that up.

I am aware that Hume is not strictly an atheist, but he does produce some good arguments against more traditional forms of theism. If I recall correctly (and it's been a while) he does bring up some atheistic arguments, though, as you note, there is no evidence that he actually thinks they work.

Posted by: Kenny at November 11, 2009 9:58 PM

I think the point is that TGD is the one book on the subject from the atheist point of view that a large number of the public have read.

The perception will be that if you claim Dawkins' arguments are no good, but decline to state in what ways they are no good and go and review the arguments of some lesser-known atheist work instead, you will look like you are running away from the argument.

Posted by: Jonathan West at November 12, 2009 1:08 AM

Jonathan, I'm sure some people will feel that way (you might, for instance), but my impression is that most of my readers are seriously engaged with academic philosophy, and I see no reason to believe that those people will feel this way.

If I were to write a book for popular audiences, I should certainly have to deal with TGD for the reasons you state.

Posted by: Kenny at November 12, 2009 9:12 AM

The problem is that unless you can actually deal with issues on the Dawkins level, there's no reason to think that you can go higher.

You're probably aware by now of my opinion that Swinburne's quality of argument is execrable bordering on comical. He is a long-established and eminent theistic philosopher, and yet his arguments fall apart like wet tissue paper, such that even somebody like me without a professional training in philosophy can see right through them. And yet Swinburne is doing little more than rehashing conventional theistic arguments (e.g. the cosmological argument, the teleological argument) that have been round for thousands of years.

And that is part of the the problem. Since there has been no new evidence of God to work on, (rather, what was once accepted as evidence has greatly shrunk) theistic philosophy has had no basis on which to advance. Rather, the proliferation of inconvenient scientific evidence has resulted in it taking two mutually conflicting approaches.

One, which we can term the fundamentalist approach, is to set itself up in active denial of and opposition to scientific discoveries, proclaim the Bible inerrant and retreat from all new knowledge.

The other, which I will call the accommodationist approach, has been to accept scientific evidence, and redefine its conception of God by retreating into untestable propositions.

Any theistic philosophy which expects to be taken seriously must avoid doing either of these. And if you cannot provide an argument against atheist even at the popular level of Dawkins, then there is little or no chance of theistic philosophy ever being taken seriously as a subject.

Your call.

Posted by: Jonathan West at November 12, 2009 4:28 PM

"unless you can actually deal with issues on the Dawkins level, there's no reason to think that you can go higher."

That much, at least, is true, but nothing follows from it. First, I'm not in the business of giving you reason to think that I can do anything. From time to time I present arguments on this blog, and some of them are about religious issues. I hope that you will stick around to read some of these, and criticize the arguments. As to your overall opinion of my abilities, that's not really my concern. If you happen to think highly enough of me to think my arguments are worth your time, we might both stand to benefit, but if you don't think that, then I don't think that it's worth my while to spend a bunch of time persuading you. Second, there is no need to prove oneself at every level along the way. For instance, we wouldn't think any less of a professional baseball player if we found out that he had never played Little League. I can assure you, spending my time refuting Dawkins will not make my colleagues, current professors, or the departments that may one day employ me think any more highly of my abilities.

As to the rest of your comment, it rests on a faulty understanding of the relationship between God and nature (I have posted a draft of a paper which discusses what I take to be the correct approach), and possibly on some residual positivism (a form of radical empiricism which was in vogue in the early-to-mid twentieth century, but was ultimately found untenable). Unfortunately, the Swinburne method of apologetics, not to mention the Intelligent Design movement, tends to encourage this kind of approach. However, the greatest historical Christian philosophers never rested their arguments on the kinds of empirical premises the ID/creationism folks use. What I'm saying is that when theism (and specifically Christianity) is properly understood, the kinds of premises science has refuted (e.g., a young earth at the center of the universe) never could have supported it in the first place. I'm not aware of Swinburne ever using this type of evidence, but in certain ways he takes the same general approach; for instance, he likes arguments from miracles. As I argue in my paper, he is mistaken in this because (among other things) he adopts a bad definition of miracle.

Finally, let me note that theism is essentially a doctrine of metaphysics. While empirical evidence does have relevance to metaphysics, one could argue that we are working with essentially the same evidence Plato had. Yet metaphysics has advanced considerably. Of course, you may be hanging on to some residual positivistic skepticism about the whole enterprise of metaphysics. If so, a good (secular) source on positivism and the reasons for its collapse is Scott Soames' two volume Philosophical Analysis in the Twentieth Century. (Unfortunately, the relevant part is across the end of the first volume and beginning of the second.)

Posted by: Kenny at November 12, 2009 6:48 PM

Well, if Dawkins is of no interest to you, then I am surprised to find you wasting time on him on your blog (and wasting time dealing with his atheist critics such as Michael Ruse). It seems to me that you fall into that large class of theistic commentators who is in that habit of taking ad hominem potshots at Dawkins without having the gumption to say in what way you think he's wrong.

I've taken an extremely brief look at your draft - I might choose to pick it apart at more leisure (and believe me, it can easily be picked apart) but it seems to me that a major weakness is that it concentrates entirely on a definition of miracles without offering any evidence that miracles as defined have ever occurred. One might as well theorise as to the precise colour of the Invisible Pink Unicorn, arguing for this or that Pantone shade.

Posted by: Jonathan West at November 13, 2009 2:06 AM

Well, if you would pay attention to the introduction of the paper, you would note that it is not intended to be anything like a proof of the existence of God. It is purely internal to classical theism. It does, however, discuss how the relationship between God and nature should be construed and how empirical evidence might bear on the question of God within this framework. That's a totally different matter than trying to show that the empirical evidence is actually there.

Posted by: Kenny at November 13, 2009 9:59 AM

And if you would pay attention to what I wrote in the comment, you would notice that it makes no mention of God, and concentrates solely on the matter of miracles, the fact that you define them but provide no examples.

Posted by: Jonathan West at November 13, 2009 12:13 PM

Just wanted to write a thank you, Kenny, for your blog. Always interesting to read, some fascinating insights and a calm tone, though I always come here hoping to see some idealism posts (not to mention arguments in favor of idealism).

And so as not to leave out Jonathan in terms of compliments... I like your last name. It is easy to spell and remember.

Posted by: Joseph A. at November 21, 2009 5:16 AM

Post a comment

Return to