This will be just a brief note on part of the Stephen Law’s presentation on the strengths and weaknesses of The God Delusion, which, since it is related to remarks made by Peter Williams considered earlier, is worth while inserting here. First the six and a half-minute clip where he discusses Alvin Plantinga’s response to the central argument of The God Delusion.
I disagree with Law’s assessment of Plantinga as a very very good philosopher. But, having said that, notice that this is a bit like Williams’ argument about artifacts and artisans and then following up with necessary or impossible gods. How convincing is it? Stephen Law calls this the heavyweight argument, having already disposed of Swinburne and Craig. But how compelling is it? It may be true that explanations come to an end, but what Law does not seem to see is that if the argument is valid, it is an explanation for life on earth. In other words, Plantinga is presenting us with Intelligent Design theory. Does Law want to follow him along this particular train of argument?
For we already know, independently, that life on earth gives the appearance of design, and that there is no reason to think that it is actually designed by a designer. It is reasonable, based on what we know, to infer, from the existence of tractor-like machines on an alien planet, the existence of intelligent beings who designed and manufactured them. That is merely an inference based on experience. But, even if we are only trying to explain the complexity of life on earth — and Law is unclear as to what we are trying to explain here – there is no reason, based on our experience, to make the inference to the existence of a non-physical, purely spiritual, or mental, supernatural, designer.
Now, we have to remember what the God Hypothesis (the GH of that last screen) is. Here is Dawkins’ formulation:
… there exists a super-human, supernatural intelligence who deliberately designed and created the universe and everything in it, including us. [31]
So the God Hypothesis, for Dawkins, is, in fact, an ultimate explanation of organised complexity, not just for the existence of terrestrial life. It is not like coming upon something which appears to be designed and manufactured in ways familiar to us. Recall that in his Natural Theology, William Paley encouraged us to look at living things and manufactured things as being similar in precisely the way that Plantinga is suggesting. And so long as he keeps the description vague, he can almost get away with it, for, as we know, until Darwin, as the example of Hume makes clear, it was not easy to see the distinction, although it was obvious that Hume felt that there was a significant distinction to be made. Darwin’s theory made it clear what the distinction was, and why it was so important to make it.
Seeing that there is a distinction to be made between designed things (like watches), and things which appear to be designed (like animals and plants), but are instead the outcome of mindless physical processes, is vital to Dawkins’ argument. If the inference in the first case, say, to Williams’ artifacts from the second century BC[E], or Plantinga’s tractor-like objects, is reasonable, that is, if it is reasonable to infer an intelligent designer, it does not follow that it is reasonable in the second, for we now know that living things are not designed, but are the result of random variation and environmental selection. And talking about the necessity or the impossibility of gods, or about the fact that explanations must come to an end, does nothing to change this. That is just so much hand waving, once we see the trick that Plantinga is trying to play on us.
I conclude, therefore, that Law has not shown that there is a serious flaw in Dawkins’ central argument. It may be flawed in other respects, of course, but not in this one.
If you define god as a necessary being, then god by definition must exist. That’s it. I don’t get what is so special about this argument – it just seems to be defining god so that existence is part of the definition of god.
1. God is a necessary being.
2. Necessary beings exist.
3. God exists.
How do we know 1. is true?
Well, obviously you get 1 from 2 and 3. This imminently good reasoning also works for anything you’d care to try. It is handy, of course, to finally have an argument that covers every topic, real and imagined, with equal fidelity.
Not true. You can’t get 1 from 2 and 3.
What? How could you crush my hopes and dreams like that?
I suppose next you’re going to tell me that I have to stop using this argument for the Easter Bunny.
I am (hopefully) obviously being entirely facetious from the word go. But if these arguments can just be made up, then I should just get to make up stuff too.
The question that was asked, “How do we know 1. is true?” is imminently reasonable. But it won’t get a reasonable answer for all of the usual reasons.
My notes:
-Plaintinga is a fine philosopher, but he does not seem to be scientifically (or mathematically) literate enough to make the kinds of claim that he makes. When he talks about philosophy of mind, or evolution, this is an absolutely fatal flaw and immensely undermines his ability to move forward.
-I wish we could just stop with this dodge about God being simple. Either God is conscious and has a mind, or it is simple. Everything we know about minds, from cognitive science, philosophy, and so forth, suggests that what they are is a whole bunch of things working together in a complex way. Under the standard definition of mind, it is a logical impossibility that such a thing could be “simple” in a way that can satisfy Occam’s razor.
-Dawkins is evaluating God as a scientific hypothesis. That means that he is judging it, not as a philosophical idea, but according to the criteria that we use to evaluate scientific explanations. We know very well that a scientific explanation can only be successful if it explains data in a way that simplifies/compresses/unifies the previous understanding.
After all, you can fit a curve with perfect precision to absolutely any set of data whatsoever, but explaining 500 points of data with a 499 degree polynomial is worse than useless (and using a 1000 degree polynomial would be deliberate obfuscation). Any decent hard scientist would flunk any student who insisted on doing that. The arbitrary features in a model have to represent much less information than what is actually being explained by the model. This is one of many ways we use Occam’s razor in science.
-The only thing that the God Hypothesis is being invoked to explain here is abiogenesis. A single cell. Because of this, any explanation which invokes more complexity than that cell, and which we have no evidence at all for, other than the existence of that cell (solely, and in isolation from all the rest of physical reality), such an explanation must fail.
In fact, because GH only explains the existence of a cell, it cannot predict much of anything about why life is the particular way that it is. There’s no clear reason why God couldn’t have designed life that looked rather different at the biochemical level (even if there are some vague constraints on what kinds of replicator are physically possible). In fact, it seems to me that one should be able to specify an upper limit, in bits, to the complexity of God’s mind, in order for GH to satisfy Occam’s razor (although this strikes me as a difficult and pointless exercise).
-God therefore fails as a scientific explanation once you concede that the God concept introduces more complexity than a cell (as it seems he must have, in order to have a mind that can actually design one).
-The tractor analogy is rather poor. We already know that there are creatures that design tractors. We already (at least as a species) know how tractors are designed and built, and what they do. There are not a lot of arbitrary unknowns here, because we have lots and lots of outside evidence about what machines like this do and why they are built a certain way. So the datum to be explained here is not the existence of tractors, because we already have oodles of evidence and already-established explanation about them.
The datum to be explained is instead the presence of tractors on another planet. And we already know some very specific, non-arbitrary information about the tractors, such as that they must have been created to suit creatures of a particular size and and shape, based on our outside knowledge. We know that the tractors are probably not just sculptures, were probably designed with certain kinds of efficiency in mind, by creatures at a certain level of technological development, because of the outside evidence we have about tractors from our own planet. And we can use this knowledge to make predictions, to help us look for other evidence about the aliens on this other planet. The aliens may themselves be more complex than a tractor, but positing their existence does not introduce anything more complex than a civilization full of their artifacts, plus our own civilization full of people who create similar artifacts, which is quite a lot more evidence to leverage than “life exists”.
Life just isn’t in the category of things that we know of as being designed, and supernatural minds are not in the category of things we even know to exist. We can’t use the same argument here, not unless we ourselves gain the ability to create universes with life in them, in which case it would become much more plausible that our own could have been created in the same way (especially as we would then actually know enough to formulate and test such a hypothesis much more rigorously). Even then, we would not be establishing the existence of an infinite, supernatural theistic god, but rather of an alien designer with capabilities comprehensible to ourselves.
-When you ask “Why stop at God?”, then “Why not?” is a terrible reply. If you care only about whether or not something is logically consistent, then this is a sufficient rebuttal, but that cannot hold if you are thinking like a scientist. Explaining a tiny amount of data with an enormously arbitrary and complex idea is anathema to clear scientific thinking. Dawkins gets criticized (perhaps accurately) for not engaging with philosophy of religion. But it is equally the case that many philosophers are obviously incapable of engaging with, or possibly even oblivious to, scientific standards of evidence (or mathematically-driven theories of empiricism and probability in general).
If the explanation “must stop somewhere” (though not everyone agrees that it does, as in coherentism or infinitism), then so be it. But there’s no reason to stop it at God, or beyond God. You can stop it at physical reality, or at something other than God (like a multiverse, or a simulated world, or what have you). And what Occam’s razor can tell us is that, whether or not we know where a good place to stop is, God most certainly is a bad place to stop. Simply put, the onus is on the believer to show that the God-creating-life stuff is not just a huge load of random made-up bullshit (and thus ludicrously unlikely right out of the gate), rather than on everyone else to prove that it is logically impossible that some mysterious thing in some mysterious reality for some mysterious reason used some mysterious process to do something that seems to be not so unlikely using non-mysterious naturalistic processes.
I think I’ve taken this apart enough now.
Hi all. This isn’t a response to this particular OP, but a broader response, saying what I think is wrong with Dawkins’ argument against God.
One major problem with Dawkins’ argument is that it’s muddled, leaving readers unclear just what the argument is supposed to be. Different critics and defenders have interpreted it differently. I should state up front that it’s been a long time since I read TGD, so I may be overlooking or misremembering something. However, Eric’s quotations and commentary have not resolved the confusion in my opinion.
I don’t reject Dawkins’ argument altogether. Parts of it are pointing in roughly the right direction. But it’s muddled, some of it is misguided, and it’s missing significant steps.
The main part of the argument seems to be along the following lines. Any designer capable of creating the organised complexity of the universe must be even more complex than the universe. So invoking such a designer is useless as an explanation of the organised complexity of the universe.
I have some reservations about the claim that a designer must be more complex than his creations. In particular, the argument seems to ignore the possibility that a designer created a relatively simple universe (at the Big Bang) in the expectation that great complexity would then evolve naturally. Is Dawkins arguing only against a creationist God who designed living organisms directly? I think his aim was higher than this. Having said that, this isn’t my main objection to this argument. My main objection is that, even if we accept the validity of this argument, it doesn’t get us to Dawkins’ stated conclusion. It only gets us to the conclusion that God is a bad explanation for organised complexity, not to the conclusion that the existence of God is improbable. There’s a significant step left that Dawkins hasn’t made.
At times Dawkins seems to be making another argument, along the following lines. God is complex; therefore God is statistically improbable; therefore it’s improbable that God exists. This argument (if it’s being made) is fallacious. It conflates two different types of improbability: the “statistical improbablity” of an entity and the improbability of the truth of a hypothesis about that entity. As far as I can tell, the term “statistical improbability” (at least in Dawkins’ sense of it) is his own invention, and not a standard technical term. Dawkins appears to mean the improbability of that entity appearing through a purely random (uniform probability) combination of its component parts. Elsewhere he reasonably argues that living organisms are statistically improbable, but that their occurrence is not improbable once natural selection is taken into account. So statistical improbability does not by itself imply improbability of occurrence, let alone improbability that the entity exists.
Dawkins may argue that God is supposed (by believers) not to have evolved, so natural selection is not applicable. But it remains for him to argue as to why we should reject the possibility of other ways in which an entity could have high probability of occurrence despite statistical improbability. One possible alternative is that there could have been such a vast (or infinite) number of opportunities for God to assemble spontaneously that he was likely to have done so eventually despite the vast improbability at each opportunity.
Moreover, God is generally supposed (by believers) not to have occurred. God is supposed to be some sort of eternal and/or uncaused entity who never began to exist. So arguments about the improbability of God occurring are insufficient.
I think Dawkins makes his argument a little better in this article than in TGD: http://richarddawkins.net/articles/126-who-owns-the-argument-from-improbability
In that article he insists that our explanations must ultimately be rooted in simplicity. I think that’s closer to the right argument than he gets in TGD, but is still incomplete. I would like him to have given a reason why this should be so.
@Michael
“How do we know 1. is true?”
When presented in this form, it’s patently a ridiculous argument. However, when Plantinga makes his argument, he’s using S5 modal logic and that’s a weird, but apparently consistent thing. To quote Wikipedia:
“[U]nder S5, if something is possibly necessary, then it is necessary – i.e. what is possibly necessary is necessary in at least one possible world, hence it is necessary in all possible worlds and thus is true in all possible worlds. This reasoning underpins at least one of the ‘modal’ formulations of the ontological argument for the existence of God.”
The reference to at least one of the modal formulations is probably that of Plantinga.
Hello, Richard:
I wanted to take a moment to ask you a question and maybe offer up a line of thought you could find useful.
1.) What bad explanation for an event do you suppose is nevertheless the likeliest explanation? That is to ask: what step do you think is missing between cause x is a bad explanation for event y, and therefore, x is probably NOT the cause of y?
The argument that Dawkins is making, it is important to remember, is not an isolated argument invented out of whole cloth. It is key to the book making any sense in the first case, to always bear in mind that he is responding to an argument. Everything he writes, therefore, should be taken in that light.
That said, when Dawkins is making the point about the statistical improbability of the designer, it is specifically in relation to the ability of that conjecture to account for the thing it is being invoked to explain. To shortcut the discussion a little, the probability of an event having happened after it’s already happened is 1. To make any other argument is on its face invalid. However, if we grant that it’s a potentially meaningful discussion, and we’re met with an argument about how statistically unlikely it is that the universe could have come into being sans creator, then we can also compare that against any competing suggestion.
So, in this case we have a universe that is, it is argued anyway, statistically improbable to come into existence all on its own. Fair enough. Since we’ll accept, in arguendo, that this is a valid conjecture, we are to ask if not by itself what then is its cause? In that case, we have the designer idea to entertain. Everything we know about designers implies that they are, at base, a locus of some degree of complexity. A freshly starting universe, no matter how simple or complex one wants to argue it is, is therefore a consequence of some intelligent agent that is causing its creation. Whatever else you can say about this designer, the one thing is necessary is that it is a locus of complexity.
Suggesting a complex entity as the simplest explanation for the creation of a universe (remember that complexity of the designer must always be compared to the complexity of the beginning universe since that’s the temporal, causal relationship) is to posit an entity that shouldn’t exist early in the universe in the first place. That is to say that everything we know about the kind of complexity necessary for “mind” to have the sophistication required to go about creating things like, say, watches (let alone universes) comes late in the slow, incremental stages of evolution. Therefore, positing an entity that is unlikely to exist in a universe that young as the cause of that universe is a bad postulate.
It is a bad postulate because the necessary traits of the designer shouldn’t exist at the point in “time” when they had to necessarily exist in order for the designer to be a viable cause for the event. Therefore, this candidate hypothesis fails to do the work it would need to do in order to remain a candidate hypothesis.
So, that’s the book. The rest of it is commentary.
To address your last bit there, which confuses me (in terms of being a response to Dawkins). If you’re going to argue that perhaps there are many, indeed maybe infinitely many, ways in which a god could spontaneously assemble, then you’re forgetting a rather important concept in science: lex parsimoniae. In order to explain the designer of a universe (which is only necessary if you decide that universes don’t just spontaneously come into being), you are arguing that we should entertain the idea of a conscious entity spontaneously assembling. Thank you, but no. If I’m going to posit such an event, I’m going to posit one for the least complex event that I can. A universe is a simpler concept than a universe-creating designer, and thus, I’m going to remove the extra step. After all, if we’re going to exist in a universe where there exist some entities that can self assemble, why add extra ones? Carl Sagan covered this in the “Cosmos” series some (oh my gosh, I’m getting old!) almost 30 years ago.
Dawkins was not writing a lecture meant to explain everything. But here’s why a simpler explanation is more useful: it’s easier to justify as it has less complexity. Consider, I will presume that you have parents. Fairly vague, but probably apt statement. Now, I’m going to start postulating extra information and see how long it takes until I go astray. One is male, and one is female. His name is Joe, and her name is Anastasia Beaverhousen. He’s a Sagittarius, and she’s a Cancer, and culturally Mormon. She wears the magic underwear out of solidarity with her peeps, but it holds no mystical import for her life.
How’d I do? Incidentally, these are all very plausibly true (though together it is almost certainly not) when taken individually. After all, all of those things are known to exist, and they’re not very rare events. But I am quite confident that I have accurately portrayed neither of your parents. That’s why in science and maths, we don’t get to just add in whatever feels good. Every piece of data, every postulate, every assumption must be carefully accounted for, clearly stated and minimized. Indeed, if you can get rid of it altogether, you have a much better chance of making progress.
I use a very similar argument with those who owe me money. If you possibly, necessarily owe me money, then you definitely necessarily owe me money.
They don’t seem to buy it either.
@Justicar Yellits
Oh, I don’t buy his argument. Even if I did, I don’t think it could be used to get you anywhere near a personal god like the Christian one. The best it (and Kalam, or any other philosophical “proof”) could get you is to the deist god.
On the other hand, while I’ll happily rip the cosmological argument made by Craig apart on multiple grounds from equivocation to begging the question, I don’t know enough about modal logic to dismiss Plantinga outright.
The ignorance of modal logic may be one of the reasons that Dawkins gets called “philosophically naive”. I wish there was somebody blogging that had the formal logic training tear apart the S5 thing once and for all. Dan Dennet or Sam Harris would seem like two who might have that sort of qualification, much like our host here has the training to be able to tear apart the woolly obfuscation of theologians.
Actually, Dawkins refers to himself as being known in philosophical circles as a “naive realist”. And it’s not an unimportant quote in this context.
Here’s what you need to know about modal logic with respect to the proposition it purports to bear on: it’s irrelevant.
No argument, however well-crafted, logically congruent, aesthetically pleasing can overcome one non-trivial fact of anything amenable to science: they must all stand subordinate to evidence.
Justicar, you seem to be denying that Dawkins was arguing for the non-existence of God (that God probably doesn’t exist). I don’t have TGD in front of me, but based on my memory and what I can currently read elsewhere on the web, I find that very hard to believe.
You mention the principle of parsimony. I certainly haven’t forgotten about that. I consider it important. But I was addressing Dawkins’ argument in TGD, and as far as I recall that argument doesn’t make an appeal to the principle of parsimony. If you can give me a quote to show that it does, I’ll be happy to stand corrected and that will probably improve my overall opinion of the argument.
I am, I must confess, taken aback. Richard Dawkins also did not present an argument for literacy. As I said, he wasn’t writing a book meant to encapsulate all that one requires to be able to, well, not get run over crossing the street, let alone become a working, fully competent science reader.
He also doesn’t have a section on understanding footnotes. Yet, he has footnotes. Some of them are possibly useful.
I have reread what I wrote, and perhaps I am not a good candidate to critically sum it up. I will therefore submit to the reasonable people who read it to see if the following statement bears some fidelity to what I wrote:
Richard Dawkins was not arguing for the non-existence of god [by demonstrating why the conjecture is not tenable in the first case].
If anything I said can remotely be reasonably taken to mean anything in the neighborhood of that, I should be immediately banned from the internet-at-large. Indeed, I think chapter four is titled something like ‘why there is almost certainly no god’.
John Shook has a page or two on why the modal logic version doesn’t get you anywhere in his book The God Debates. Somebody email him and ask him to weigh in here.
Well basically Shook says what Justicar says. It works at what it does but it’s supremely beside the point.
“Logically, this ontological argument at most warrants belief in certain most “perfect” things (perfect according to each person’s subjective notion of perfection) that actually happen to exist in the universe. There is no logical way for this argument to point to something outside nature, or to something completely perfect within nature.” [p 129]
Justicar, no doubt I misunderstood you, and I’m happy to accept your assurance that you were not making such a denial.
But I still don’t think you have correctly described Dawkins’ argument.
Crap. Now I’m in a pickle: do I double post, or trust Ophelia can track down my response?
I’ll double post. Sorry, Eric. (Please delete the earlier copy of which this shall be a duplicate). Bad netiquette, bad netiquette!
Copied and Pasted:
Thanks, Ophelia!
One of the joys of my life is being proved right, repeatedly, over and again. I mean, err, yeah, for all the interesting thing that come up in logical constructs, they’re just that with respect to this conversation: interesting things in logical constructs.
Despite the William Lame Craigs of the world, a good argument doesn’t actually dictate much about the universe. Of course, he knows this, and freely admits that even if his arguments weren’t torn asunder by superior arguments, it wouldn’t touch his impenetrable faith. Since there isn’t an argument, real or imagined, that can overcome his position, it kind of lets me know there isn’t an argument that supports his position. Let alone the pesky matter of evidence.
That’s how I know when they start talking to return the favor they always give non-believers: I stick my fingers in my ears and say la-la-la.
Before we can get anywhere, don’t we need definitions of complex and simple?
What’s this we? You have a rat in your pocket?
Or, you could do as I’ve done to duck the whole pointless issue, let it be defined in any fashion that’s convenient provided that whatever property it is that makes something simple, the property that makes it complex will that property that makes it simple plus anything else.
I thought I was the only one who knew about the rat.
Michael,
I think you’ll never get anywhere if you start by demanding definitions of every term. Better to take terms on an intuitive basis until you run into a problem, and only then start worrying about what they mean.
As it happens, IIRC, Dawkins has suggested in the past (in a previous book) that we take complexity to be roughly equivalent in meaning to “statistical improbability”.
What can I say? I am supremely observant, particularly in areas of cosmic importance and personal interest. I’ve said too much.
My pleasure, Justicar!
I learn so much about what I still need to learn by reading this blog and the associated comments. Thanks all.
Yeah, thanks Ophelia. Thanks to you too Justicar.
I guess I just don’t find the had waving away of the whole thing intellectually satisfying. I want to see it torn to shreds and defeated in detail.
To quote Bertrand Russell on the ontological argument: “The argument does not, to a modern mind, seem very convincing, but it is easier to feel convinced that it must be fallacious than it is to find out precisely where the fallacy lies”. I have no bother dismissing the earlier versions of it, but the modal version seems to make the quote from Russell resonate in my head for some reason.
From some further digging, though S5 is widely accepted, it’s not universally accepted. So I’m off to do some further reading in that direction. I may even resort to getting Shook’s book, though I was trying to avoid reading it.
Well, I can certainly understand the point about an answer not being quite satisfying; it’s an artifact of what got us this far in history. But, and I hate to say it this way, the complete refutation is that the argument is simply not applicable here.
Assume every possible harebrained conjecture about modal logic is true (ignore any logical incongruities – just accept that somehow they’re all universally true with no contradictions). It’s a complete theorem in other words. All of that being true, it would still not apply, and it can’t hope to apply because it just doesn’t simply have within its ambit the province of this topic.
It’s really much akin to shoving the square peg of an irrelevant idea into the round hole of a different problem.
Let’s say I wanted to argue against Dawkins using, oh, say, something interesting like a Tarski sphere. I could go on for hours and hours with extremely detailed proofs, maybe some fancy computer simulations, and all kinds of logically coherent, internally consistent and imminently sound arguments about it. After all of that, I then say, “now, refute that.”
For the exact same reasons as with modal logic, the proper answer is the same: even being entirely true and a complete proof, it doesn’t model this problem because it doesn’t address any predicate of this problem.
It might not be an elegant reason, but it is the reason; two separate issues are dealt with and neither implies anything about the other. *shrugs*
Oh, FFS. It logged me in under my master name on wordpress. Sigh.
Eric, I’m sorry; I am a webmaster’s nightmare today. If this keeps up, I’m going into a self-imposed exile until my fingers are more cooperative. Delete that one in the moderation queue if you’d be so kind; I am not trying to be difficult.
Here is what I originally wrote in response:
Well, I can certainly understand the point about an answer not being quite satisfying; it’s an artifact of what got us this far in history. But, and I hate to say it this way, the complete refutation is that the argument is simply not applicable here.
Assume every possible harebrained conjecture about modal logic is true (ignore any logical incongruities – just accept that somehow they’re all universally true with no contradictions). It’s a complete theorem in other words. All of that being true, it would still not apply, and it can’t hope to apply because it just doesn’t simply have within its ambit the province of this topic.
It’s really much akin to shoving the square peg of an irrelevant idea into the round hole of a different problem.
Let’s say I wanted to argue against Dawkins using, oh, say, something interesting like a Tarski sphere. I could go on for hours and hours with extremely detailed proofs, maybe some fancy computer simulations, and all kinds of logically coherent, internally consistent and imminently sound arguments about it. After all of that, I then say, “now, refute that.”
For the exact same reasons as with modal logic, the proper answer is the same: even being entirely true and a complete proof, it doesn’t model this problem because it doesn’t address any predicate of this problem.
It might not be an elegant reason, but it is the reason; two separate issues are dealt with and neither implies anything about the other. *shrugs*
I think that the clearer way to put the argument is this:
Say that something can only be true necessarily (it might be the existence of God, or it might be some logical or mathematical truth, such as “All triangles have angles that add up to pi radians in a flat (Euclidean) space.”).
Within S5, you can deduce that such a proposition is either necessarily true (true in all possible worlds), or necessarily false (false in all possible worlds). This is a true dichotomy.
If you admit that God is possible, then he exists in at least one possible world. Therefore he exists in all possible worlds, and therefore he exists in this world.
One problem with this argument is that it works both ways. That is, if you admit that the non-existence of God is possible, then that means that he doesn’t exist in at least one possible world, and therefore that he does not exist in any possible worlds, and therefore he doesn’t exist in this one.
I’ve heard that Plantinga is aware of this problem, which is part of why he doesn’t use this argument more. Regardless, this is not a hand-waving objection. Anything can be proven in S5 by simply declaring that it can only be necessarily (not contingently) true, and then declaring that it is possible. Which is the modal version of the same problem as all the other ontological arguments have; you can substitute all kinds of things in for God, and get identical support from the argument.
It should be pointed out that this is a feature, not a bug, in S5. In all modal logics, “necessary” and “possible” have linked definitions. So if you accidentally use them in different senses (such as “metaphysically necessary” and “epistemically possible”), you are committing a fallacy of ambiguity.
I hate to weigh in so late, but this subject pisses me off.
The whole point of logic is that it’s a reliable method of teasing out the implications of a set of premises. If done right, it can’t deliver anything that wasn’t there to begin with.
If God is the conclusion of a logical argument, either God was implicit in the presumption or the argument doesn’t satisfy the expected constraints of logic.
Sean Carroll commented on this here.
But if you admit that complexity can come from simplicity naturally, you don’t need a supernatural agent anymore to explain complexity at all. Remember that Dawkins isn’t making his arguments in a vacuum, he is specifically arguing against people who invoke God as a good explanation for complexity.
The same goes for the argument of improbability. Dawkins is just responding to people who claim the universe is improbable (due to perceived complexity, or fine tuning), and therefore God must exist to make it more probable. But if God is more complex than the universe, by the same reasoning God is more improbable than the universe.
In that case, why not simply argue that the universe itself eventually spontaneously self-assembled?
No, it’s argument by assertion that is insufficient. You can’t just define your way out of a contradiction. You can’t just define that everything needs a cause, except God. You can’t just define that all complexity needs a designer, except God.
Deen wrote: “The same goes for the argument of improbability. Dawkins is just responding to people who claim the universe is improbable (due to perceived complexity, or fine tuning), and therefore God must exist to make it more probable. But if God is more complex than the universe, by the same reasoning God is more improbable than the universe.”
If that were a correct account of Dawkins’ argument, then he would not be attempting to make a sound argument for the improbability of God (because he doesn’t accept the premises), but only to show an inconsistency in the position of some theists.
Your replies to specific passages I wrote are not responsive, mainly because you’ve taken them out of context.
I just came across this interesting blog post by a defender of Dawkins who goes to some length to try to establish just what the argument actually is:
http://omnisaffirmatioestnegatio.wordpress.com/2010/06/12/dawkins-and-the-ultimate-747-gambit/
I don’t agree with his conclusions, but at least he recognises the problem and makes a serious attempt to resolve it.
I think the common theist argument that Dawkins is arguing against goes something like this:
1. The universe is complex
2. Complexity implies improbability
3. Improbability implies intent
4. Therefore, the universe must have an intelligent creator.
You claim Dawkins rejects 2, but I think Dawkins is accepting premise 2 here. He’s basically using the first two premises to arge that:
1′. The universe is complex
2′. Complexity implies improbability
3′. A creator is more complex than its creation
4′. Therefore, a creator is more improbable than the universe.
Of course, you can attack Dawkins’ argument at premise 1′ or premise 2′ (like you did), but I don’t think Dawkins need be too worried about it in this context, as it would also refute the theist argument above. Remember, in TGD, Dawkins is not trying to argue that all conceivable God-concepts are improbable, just the ones that most theists seem to subscribe to.
Dawkins himself doesn’t have to reject premises 1 or 2 at all to reject the first argument. He can simply reject premise 3, which he does by arguing that complexity (and hence improbable outcomes) can and does arise from simpler states by natural processes.
I fail to see where I took your other quotes out of context. Just asserting I did is also “not responsive”. But I hope I have at least been more clear about this bit.
Richard:
Where, and how, do you suppose that I have misrepresented Dawkins?
Specifically: what is it that you take Dawkins to be arguing, and what part of what I said is incongruent with that?
Justicar, I gave my own interpretation in my first post here. Sorry, but I’m not interested in having a back-and-forth discussion about what Dawkins really means. I’m prepared to consider alternative interpretations, but if they seem wrong to me, I’m not going to get into further discussion of them.
Well, thanks for being honest. It’s great to know whom I should ignore in the future.
“you’re wrong~”
“where, and about what?”
“i’m not interested in a discussion”
“Okie dokie – one more waste of time duly avoided”
About whether God is simple…
There’s a simple counting argument that says Gods not, if he/she/it knows a large number of distinct things.
For each distinct thing that anything knows, there must be a difference between that thing as it is, knowing the thing, and as it would be if it didn’t know that thing.
A god which is very knowledgeable would therefore differ in many ways from one that does not know those things, or one which knows many things, but different ones.
Things that can differ in many ways can’t be simple. They have to have parts in some sense, which enable them to be (potentially) in many configurations, i.e., states of knowledge.
This argument does not depend on God being natural and material, and thinking and knowing in the same way we do. It only depends on there being a difference between something that knows some thing and an otherwise similar thing that doesn’t know that thing.
A very simple God would be very ignorant.
If there’s no such distinction applicable to God, the concept of knowledge is not applicable to God, either—it doesn’t make sense to say that God knows something if his not knowing it would not be any different.
There’s a subtlety here in that there’s such a thing as implicit knowledge. There are many things I know implicitly that are not explicitly stored in my brain, but can be generated as needed from things that are. E.g. if I know that two countries are further apart than they are wide, I can easily infer that any given location in one country is further from any given location in the other than it is from any given location in the same country.
Still, being able to draw such inferences requires a certain minimal complexity, and such things only go so far in accounting for knowledge—e.g., even if I don’t have to explicitly represent each inter-location distance, I still have to represent the locations, and concepts such as location and distance that allow me to make the inferences.
No way around it, if God “has no parts” and is very simple, as many theologians claim, God is utterly stupid.
You could express your ideas entirely in terms of Kolmogorov complexity as a measure of information content. This measure takes into account possible redundancy in information.
At the very least, such a simple God would be a far cry from the personal God that is capable of communication with humans that most people believe in. Or from the “mysterious” God that nobody could ever understand that some “sophisticated” theology proposes.
I have thoroughly enjoyed this superb discussion. Thank you, everyone.
Justicar,
I’ve wasted a great deal of time in the past on internet discussions that go nowhere. I now try to save everyone’s time (but especially my own) by making a judgement early on about whether a discussion will be fruitful, and, if not, to end it quickly. That’s my judgement in this case. If as a result you choose not to respond to my posts in the future, that will suit me too.
Pingback: O Resgate de Dawkins: interlúdio « O Gato Précambriano
If God exsists, didn’t he create everyone? Yes. OK. So God makes no mistakes? So if God made every person who they are how they are as it had been proven that those who are homosexuals are born that way why is it God looks down on those with a different sexual-orientation. Why is it wrong in christianity and Islam and different religions to be homosexual if God made us all who we are without making mistakes? Surely God makes people Homosexual or Hetrosexual so why do people say God sees it as wrong? I just want to know the answer the question and I am not trying to use this against the God Theory.