Part 3b: Dawkins’ ‘Official’ Argument Against ‘God’
For reference, I’m using Mariner Books paperback edition (ISBN-13: 978-0-618-68000-9). Unless otherwise noted, page numbers refer to this version of the book.
Alright, this is going to be a long one (sorry!), but it’s important to deal with Dawkins’ ‘official’ argument all at once, I think. The argument if summarized on pp. 188-189. Here, I will give Dawkins’ summary – with extraneous matter removed – and respond to each point (or set of points) in turn.
It’s worth noting one thing right off the bat: in many respects, this argument is a combination of the previously discussed A1 and A2, bringing together the question of ‘who designed the designer’, the issue of complexity and the ability to design, and the necessity of a Darwinian process to explain that sort of complexity. A1 and A2 end up reinforcing each other, as though they were part of a house of cards, leaning against one another in order for each to support the other.
Points 1, 2, and 3
One of the greatest challenges to the human intellect, over the centuries, has been to explain how the complex, improbable appearance of design in the universe arises.
The natural temptation is to attribute the appearance of design to actual design itself.
The temptation is a false one, because the designer hypothesis immediately raises the larger problem of who designed the designer… We need a ‘crane’, not a ‘skyhook’, for only a crane can do the business of working up gradually and plausibly from simplicity to otherwise improbably complexity.
These three points form the first phase of Dawkins’ argument. A couple of things deserve deeper investigation, but allow me to make one preliminary point: is it really true that explaining “the complex, improbable appearance of design in the universe” has been “one of the greatest challenges to the human intellect”? Many humans have simply inferred a designer in order to explain the design. Indeed, it could be argued that ‘design’ becomes a problem only if one posits that no designer exists – but that no designer exists (or that such a designer is improbable) is precisely what Dawkins is trying to demonstrate!
The easiest way out of this, of course, would be not to try to demonstrate that there is no designer, but that there is no design, something I think Dawkins moves toward on p. 161. Unfortunately, Dawkins’ argument about the lack of design turns upon the ‘flaws’ that exist in some natural things – largely organic structures. Certain questions crop up here, though: What makes a ‘flaw’ a ‘flaw’? Is the word ‘flaw’ even meaningful in the language of evolution? If so, what does it mean? Do designed things not have ‘flaws’? And so on. In essence, what criteria are we using to determine if certain structures are ‘flawed’, and does the discovery of ‘flaws’ really rule out a designer?
Fortunately, Dawkins does not rest with the mere assertion of ‘flaws’, but continues on.
Who Designed the Designer?
I’ve already covered the bulk of my concern with this sort of question, so a mere review should do: The question of who designed the designer only arises if we assume that the designer is something like the designed entity. ‘God’ has already been defined as ‘supernatural’, and presumably most of what we observe is ‘natural’. We know that what we see in the present-day natural world is the result of natural processes – of which Darwinian evolution is one (at least, to the best of our knowledge). The question is: where did nature come from? Why does nature exist? If we were to propose a natural entity as a solution to these questions, then, of course, that solution would require an explanation in turn – because it is like the thing we are trying to explain. In other words, if anything that has the quality of being natural requires an explanation, then any natural explanation will also need an explanation.
Dawkins, however, has already defined ‘god’ as supernatural. Do supernatural things require the same sort of explanation? If so, why – i.e.: what qualities are shared with the natural that cause supernatural entities to require an explanation? Also if so, what sort of explanation is required? Can supernatural things be self-caused? What sort of processes, laws, etc. are supernatural entities required to be the result of, obey, etc.? How would we verify or falsify any answers to these questions?
I suspect that there has been a shift here in Dawkins’ definition of ‘god’. That is, he has defined ‘god’ as supernatural – and suggested that the supernatural is distinct from the natural, and that it is distinct particularly (perhaps even only) in terms of provenance – but then treats ‘god’ as though ‘god’ were natural, needing to have a provenance analogous to that of natural entities.
Cranes and Skyhooks
Dawkins seems to think that any recourse to ‘god’ as an explanatory hypothesis is a recourse to a ‘skyhook’, when what we need is a ‘crane’. The crucial difference here seems to be that a crane explains complexity via recourse to the simple – i.e.: simpler materials and interactions result in complex entities (p. 99), one example of such a crane is Darwinian evolution where simpler entities and natural selection account for the complexity of modern life forms. A skyhook does not rely on simpler entities and interactions. Rather, it relies on some entity that demands even more explanation (p. 99). This seems a bit of an oversimplification, however: a screwdriver is simple, and yet we know if we encounter one that it was designed and created by something more complex. Fortunately, we also know that the human who created the screwdriver is the result of a crane. We also, however, know that the creator of the screwdriver is probably natural. Again, we run into the problem of the distinction between natural and supernatural.
Dawkins asserts that ‘god’ must be complex, and hence not be a possible crane. But, we presume that ‘god’ is complex? We can go with Dawkins’ argument that anything intelligent enough to design something must be complex (p. 52). However, this assumption comes from our knowledge of natural entities – indeed, from our experience with the few sorts natural entities that we know of that design (or might design) things. Yet again, we run into that distinction between the natural and supernatural.
Second, we could go with the argument that anything “capable of sending intelligible signals to millions of people simultaneously, and of receiving messages from all of them simultaneously, cannot be, whatever else he might be, simple” (p. 184; similar ideas appear on p. 178 and p. 185). On the one hand, this attribute of ‘god’ is not part of the God Hypothesis, Dawkins is adding something to the definition. On the other hand, we can again note the distinction between the natural and the supernatural. Why can we assume that the rule that applies to natural things – that entities that can communicate, or at least that we think of as communicating are necessarily complex – also applies to a supernatural entity?
Finally, we could go with the idea that whatever is complex has a “heterogeneity of parts”, a sort of “functional indivisibility” (p. 179). This might be a good time to remember the quote from Atheism: A Very Short Introduction ‘cited’ on p. 34:
What most atheists do believe is that although there is only one kind of stuff in the universe and it is physical, out of this stuff come minds, beauty, emotions, moral values – in short the whole gamut of phenomena that gives richness to human life.”
Obviously, it is true that natural, physical things that are complex have a heterogeneity of parts – that’s the definition of ‘physically complex’! But, if Dawkins subscribes to this particular form of naturalism – this sort of physicalism or materialism – then, presumably, it’s possible that the supernatural is, in some sense, not physical (or, at least, not physical in the sense that we understand physical). What might it mean for something non-physical to have a heterogeneity of parts? Or even, simply, parts? We might more easily be able to think of this if we think about things that are not normally thought of as consisting of parts: emotions, for example. Does love have parts? Does anger? Do sadness, happiness, elation, depression, etc.? What would it mean for them to have parts. Now, clearly, non-physicality isn’t an explicit part of Dawkins’ God Hypothesis, but it is implied as a possibility in his description of the distinction between the natural and supernatural.
To sum up, I’m simply not convinced that the distinction between ‘crane’ and ‘skyhook’ isn’t one of fitting into a naturalistic framework. That is, that ‘cranes’ fit and that ‘skyhooks’ don’t. Methodologically, this is fine, and we can simply say ‘since science works on a naturalistic basis, science cannot describe/understand non-natural (including supernatural) entities’. We might even make the claim that the non-natural is not comprehensible or only comprehensible in part or only comprehensible symbolically (a claim made by more than a few theologians about ‘god’). Dawkins, however, seems to what to make one of two claims, though I’m not sure which:
- An epistemologically closed naturalism in which nothing non-natural is knowable at all, or
- An ontological naturalism in which nothing non-natural exists, or, at least, in which being non-natural somehow decreases the probability of existence
The first of these would make Dawkins’ book rather superfluous, since we couldn’t know anything at all about the non-natural, so my guess is he wants to go with the second claim. We’ll get to the problem with that claim shortly.
Summary
Just a brief note here, as it will come up again later: at least one hinge of Dawkins’ argument seems to be the distinction between natural and supernatural. If Dawkins is taking the position that the non-natural (including the supernatural) does not exist as a premise, then he is engaging in a fallacy of petitio principii, as what he is trying to demonstrate is that the supernatural does not exist (or, at least, is very improbable).
Points 4 and 5
The most ingenious and powerful crane so far discovered is Darwinian evolution by natural selection… We can now safely say that the illusion of design in living creatures is just that – an illusion.
We don’t yet have an equivalent crane for physics. Some kind of multiverse could in principle do for physics the same explanatory work as Darwinism does for biology. This kind of explanation is superficially less satisfying that the biological version of Darwinism, because it makes heavier demands on luck. But the anthropic principle entitles us to postulate far more luck that our limited human intuition is comfortable with.
Darwinian evolution by natural selection is, indeed, a powerful crane, and it does offer an explanation, with predictive power, of how the incredible diversity of life on our planet could have arisen from a single common ancestor. Whether it is necessarily the case that this rules out any design is open to debate… in large part because it depends on what, exactly, we are saying is designed and what, precisely, we are claiming constitutes design. I’ll let this pass, however, and address the second point.
When Dawkins says “we don’t yet have an equivalent crane for physics” I rather suspect that what he means, given our previous discussion of ‘cranes’ and ‘skyhooks’, is that ‘we don’t yet have a naturalistic explanation of the existence of the universe’. But why should we assume that such an explanation exists? Remember that Dawkins is trying to argue that the supernatural (‘god’) is highly improbable, so he can hardly legitimately take as a premise that the supernatural does not exist! One could even ask what it means for the universe to have a ‘natural’ explanation – our understanding of the ‘natural’ is based upon our observations in our own universe, so would an explanation of our universe, based on something outside our universe, constitute a ‘natural’ explanation? Isn’t something outside of our universe by definition unnatural?
No matter, I suppose, since Dawkins provides us with what he claims is a sort of crane for physics: a combination of the Anthropic Principle and the Logic of Really Big Numbers, specifically in the form of the multiverse.
The Anthropic Principle
Dawkins states the Anthropic Principle thusly:
The anthropic argument is usually applied to the cosmos, and I’ll come to that. But I’ll introduce the idea on a smaller, planetary scale. We exist here on Earth. Therefore Earth must be the kind of planet that is capable of generating and supporting us, however unusual, even unique, that kind of planet might be. (p. 162)
Here, Dawkins states the Anthropic Principle perfectly, and I will only make it a bit more general by restating it: ‘wherever an entity exists, the conditions necessary for the existence of that entity must also exist.’ In a sense, this principle stands as a corrective to our awe that we live on a planet with all of the necessary conditions for our existence, and in a universe with all of the necessary conditions. Of course we do – if those conditions did not exist, we wouldn’t be here to observe them! From a logical viewpoint, the Anthropic Principle is simply a modus ponens argument: if p (human beings) then q (the conditions necessary for human beings); p; therefore q. We could, in fact, continue down that path, arguing that, if the conditions necessary for human beings exist, then the conditions necessary for those conditions also exist, and the conditions for those conditions, and so on down to some sort of first set of conditions. In principle, with an adequate knowledge of the universe, we could deduce, from our own existence, precisely what the initial conditions of the universe would have to be.
The problem here, is that, regardless of how far back we go in terms of the conditions that must exist, we are always left with the question of why that set of conditions exists as opposed to some other set. Why are the initial conditions of the universe such that we can exist? Note that no one objects that they are such that we can exist, the question is one of why this is the case. Dawkins himself points this out in relation to the idea that any universe capable of existing must be able to produce intelligent life:
But why did that one way [of having a universe] have to be such a set-up for our eventual evolution? Why did it have to be the kind of universe which seems almost as if, in the words of the theoretical physicist Freeman Dyson, it ‘must have known we were coming’? The philosopher John Leslie uses the analogy of a man sentenced to death by firing squad. It is just possible that all ten men of the firing squad will miss their victim. With hindsight, the survivor who finds himself in a position to reflect upon his luck can cheerfully say, ‘Well, obviously they all missed, or I wouldn’t be here thinking about it.’ But he could still, forgivably, wonder why they all missed, and toy with the hypothesis that they were bribed, or drunk. (p. 173)
Here, Dawkins tells us why the Anthropic Principle isn’t really an explanation – and I sincerely doubt that it is a hypothesis (which it is referred to as on p. 165) or an answer (p. 172). The Anthropic Principle, as Dawkins states it, is merely a restatement of the initial question. That is, it responds to the question ‘why are the conditions just right for our existence?’ with ‘well, they obviously they are just right for our existence, since if they weren’t we wouldn’t be here to observe them!’ Well, obviously, but that doesn’t answer the question posed. Nobody is contradicting the idea that conditions are such that we exist, only that it seems incredibly improbable that they are, and that such improbability, as Dawkins rightly states elsewhere, needs an explanation. Stating that the conditions are obviously there – as the Anthropic Principle does – does not provide that explanation. For that, Dawkins needs the Logic of Really Big Numbers.
The Logic of Really Big Numbers
Dawkins invokes the Logic of Really Big Numbers a couple of times, although only one of them is really crucial to our discussion. The first appearance of this Logic comes to the planetary problem (which Dawkins addresses on p. 165): namely, that it seems incredibly improbable that a planet should exist with just the right conditions that we, in turn, should exist. Suppose, though, that there are billions and billions of planets in the universe. Now, suppose that the odds of life occurring are one in a billion. If there are billions and billions of planets, then there is still a reasonable chance – though still a very small one – that life will have appeared on at least one of them… and obviously it did, as we are here to ponder this.
The second appearance has to do with the natural moving back of this question: we accept that there are billions and billions of planets, and that life arose on at least one of them (obviously it did), but why are conditions such in the universe that there even are billions and billions of planets, and that the conditions necessary for life as we know it exist? Dawkins’ rather unfortunate answer is the multiverse (pp. 174-176). I call this unfortunate because it simply moves the problem up one level, rather than providing us with an answer. Here’s the argument: Suppose that, just as there are billions and billions of planets in our universe, there are billions and billions of universes (a multiverse), and suppose that only a fraction of these are capable of supporting life, then there is still a reasonable chance – though still a very small one – that life will have appeared in at least one of them… and obviously it did, as we are here to ponder this.
One might answer: okay, we accept that there are billions and billions of universes, and that life arose in at least one of them (obviously it did), but… can you anticipate the question?… why are conditions in the multiverse such that there even are billions and billions of universes, and that the conditions necessary for life as we know it exist? We could then propose some sort of hyperverse, but this, clearly, would raise the same question again! This is why the Logic of Really Big Numbers is so easy to dismiss in this case: while large numbers increase the probability of the desired result, they do not explain why that probability is higher that zero in the first place!
An illustration might be in order here. If you flip a coin, the odds of it coming down heads is one in two. As you flip more and more coins, of course, the odds of each individual coin coming down heads is still one in two, but the odds of at least one coin coming down heads increases. If you flip enough coins, it would incredibly unlikely that no coin in that series would come down heads. None of this, of course, would answer the question of a.)why any particular coin came down heads (indeed, the odds for this particular coin coming down heads remain exactly the same, 1 in 2) or b.) why it is possible for a coin to come down heads at all. Likewise, the existence of billions and billions of planets, universes, multiverses, hyperverses, etc. does not answer either a.)why this particular planet (or universe, etc.) exists (though it would be interesting to know if Really Big Numbers would decrease the odds against this planet, universe, etc.) or b.)why it is possible for this universe to exist.
In short, neither the Anthropic Principle nor the Logic of Really Big Numbers give us an answer to the question ‘why is it that conditions happen to be just right for our existence (i.e.: why does the universe have the ‘appearance of design’)?’ I suspect that Dawkins realizes this, which is why he sneaks in a third possibility.
Luck
As we’ve seen, the argument from the Anthropic Principle and the Logic of Really Big Numbers doesn’t really demonstrate anything – the Anthropic Principle only restates the problem, and the Logic of Really Big Numbers only move the problem, as it were, up one level. Dawkins, in other words, still hasn’t demonstrated why it is that whatever set of initial conditions that are necessary for our existence exist. There is, however, another appeal that he makes, albeit in passing: luck. As Dawkins says,
We can deal with the unique origin of life by postulating a very large number of planetary opportunities. Once that initial stroke of luck has been granted – and the anthropic principle most decisively grants it to us – natural selection takes over: and natural selection is emphatically not a matter of luck. (p. 168)
And,
Nevertheless, it may be that the origin of life is not the only major gap in the evolutionary story that is bridged by sheer luck, anthropically justified. (p. 168)
Other possible candidates for ‘bridged by sheer luck’ might include the appearance of the eukaryotic cell and the appearance of consciousness (p. 168). We might also add the appearance of whatever initial set conditions must have existed in the universe (or multiverse) in order for life to arise at all, giving us a total number of potential candidates for ‘bridged by sheer luck’ to four.
Now, in fairness, ‘luck’ is a kind of answer to the question ‘why is it that the initial conditions of the universe (or multiverse, or whatever) are such that life can exist?’ But a few things are worth pointing out. First, luck is not anthropically justified. That is, we do not know that a lucky event occurred simply because we exist. Rather, we know that an event (or series of events) occurred. We cannot know from the Anthropic Principle that they were sheer luck. In short, the Anthropic Principle allows us to postulate conditions, not luck. Second, luck is not an explanation of anything. Indeed, one might argue that it is the opposite of an explanation, a one word of equivalent to ‘we don’t know and it’s useless asking questions’. Third, it leads to a possibility that I sincerely doubt Dawkins would like: if on any improbable one-off occurrence we can suggest luck as an answer to ‘why did that happen’, then there’s no reason we can’t suggest it for ‘god’. Does it seem incredibly improbable that ‘god’ should exist? Well, no matter, there’s only one, so it’s just an improbable one-off event that ‘god’ exists… I guess it’s luck.
Finally, I want to focus on the fact that luck assumes that there is no alternative explanation – or, at least, none that we can discover. Even if the Anthropic Principle allowed us to postulate luck, it would only do so assuming that an alternative explanation (‘god’) did not exist or was necessarily more improbable than the event itself. I strongly suspect that the only reason it can be assumed that ‘god’ is more improbable than the initial conditions of the universe (or multiverse or whatever) is if one assumes some sort of naturalism. There is a potential petitio principii here… again.
Summary
Again, not much to say here. It appears to me that the Logic of Really Big Numbers (even in the form of the multiverse) and the Anthropic Principle really don’t provide an explanation for why it is that conditions are such that life can exist (or anything else). Rather, they simply end up restating the problem and, possibly, pushing the problem up one level or more. In short, they are not even ‘superficially less satisfying’ – they dodge the questions at hand. In short, instead of providing an alternative to the God Hypothesis – despite the fact that he claims to provide such an alternative – Dawkins postulates luck. The problem here is that the Anthropic Principle and the Logic of Really Big Numbers does not allow us to postulate luck unless we already assume that there is not an alternative explanation.
Point 6
[A better crane might arise in physics.] But even in the absence of a strongly satisfying crane… the relatively weak cranes we have at present are, when abetted by the anthropic principle, self-evidently better than the self-defeating skyhook hypothesis of an intelligent designer.
It’s true that we might, one day, find a better crane – or, given that I’m not sure that the multiverse and the Anthropic Principle provide a crane at all (that is, a process that moves us from simplicity to the complexity of our universe), a crane, period – for physics. Here, we might do well to remember Dawkins’ description of a ‘crane’: some sort of process that ‘ratchets’ from the simple to the complex. Darwinian evolution is one such crane, clearly, that allows us, in principle, to explain how all of the wonderfully complex life on our planet came about naturally from a single – and simple – common ancestor. I agree that it is completely possible that there is some such possible crane for physics – that is, a crane that explains the existence of the universe. Note that this in no way requires any sort of multiverse. It is completely possible that our universe began simply and that some sort of natural process led to the complexity necessary to bring us about. Note as well that we do not need to use the Anthropic Principle or any sort of Logic of Really Big Numbers for this: we are aware that we exist and that, therefore, the conditions that allow us to exist also exist, and the conditions for those conditions, and so on. The Anthropic Principle and the Logic of Really Big Numbers, I think, only muddy the waters – we need only recognize that there may be a natural process that led from an initial simple universe to the universe we know today, even if we do not know what that process is.
The problem for Dawkins, as already stated, is that such an initial set of conditions seems improbably in and of itself. Imagine, if it is possible, all of the possible universes that could have existed in which that particular set of conditions which existed at the beginning of our universe did not exist. That is, imagine every universe except ours: a universe where the gravitational constant was any other number, or where the strong force has any other value, and so on for every physical constant in the universe. We would have, were we capable of imagining that, have an infinite number of universes. Of course, we must also add to this the possibility of there being no universe – although it’s difficult to fathom just what that would mean. In short, it seems incredibly improbable that this universe exists. I don’t deny that it does exist, just that it is incredibly improbable. This is precisely why Dawkins need to turn to some amount of luck in order to ‘explain’ this incredible improbability actually arising.
The problems with this point can be stated very briefly: First, luck is not a crane, it simply does not explain anything. Indeed, it is at least as much an evasion of explanation as the God Hypothesis. Second, this non-crane is not self-evidently better that the skyhook of ‘god’. After all, at least ‘god’ provides an explanation, even if we must then hope to find some sort of supernatural crane – that is, a crane of the sort that would apply to ‘god’.
Summary
Alright, a (very) brief summary of why I think that Dawkins’ argument fails. We’ll return to some of these points later on:
- Dawkins assumes that the at least one of the properties that applies to the natural applies to the supernatural. Remember that the description of the ‘supernatural’ in The God Delusion relies on it not being of Darwinian origin (p. 98) and, possibly, not being physical (p. 34). That means that ‘god’ might have any other provenance. It would be a case of petitio principii for Dawkins to then assume that “entities that are complex enough to be intelligent are products of an evolutionary process.” After all, that very statement assumes that no counterexamples exist, when the improbability of a counterexample existing is exactly what Dawkins is trying to argue towards! Moreover, without a more robust description of the ‘supernatural’ we have no idea what sort of requirements are needed in order for a supernatural being to be intelligent. I strongly suggest that Dawkins is either a.) ignoring the fact that he has defined ‘god’ as supernatural and is treating ‘god’ as natural for the purposes of his argument, or b.) smuggling naturalism in as a premise.
- Dawkins hasn’t actually explained why ‘god’ is improbable, unless we assume naturalism. That is, we must assume that the nonnatural is improbable – which it obviously is by the standards of naturalism – in order to come to the conclusion that ‘god’ is improbable. The argument, then, is trivial at best: ‘Why is god improbable?’, ‘Because god is improbable.’ That is not an argument.
- Dawkins does not provide an alternative explanation of nature. It’s worth noting that it’s not actually necessary for him to do so – and we’ll get to why I suspect he attempts such an explanation. After all, if ‘god’ is improbable, then it is improbable on its own merits. However, Dawkins’ own suggestion relies on luck, which is no explanation at all. It is rather, the end of explanation. Now, it is completely possible that our ability to explain anything ends at existence of nature – that is, once we have nature, we can explain things within it, but we cannot explain nature itself. Dawkins’ argument would have been much stronger if he had made a case for this rather than attempting to explain the existence of nature by the non-crane of ‘luck’ – and especially if he hadn’t suggested that luck constituted some sort of crane.
I’ll leave you, for now, with that. And, since this post was so long, and I’m trying to dig into Dawkins’ argument here, I’ll try to answer any questions that come up in the next post.
Excellent piece, and I intend to link it on my site.
Dawkins, like many, is simply incorrigible in his inability or unwillingness to accept (or admit) that atheism is no more testable (and therefore, scientific) than any of the varieties of theism we see, and that it is, shall we say, “unseemly” for a scientist to prattle on, no matter how eloquently, with explanations that cannot be proven and amount to lofty smoke and mirrors. On the one hand I find it amusing but on the other, offensive, because he does, or should, know better.
I allude to him and this penchant in the following two posts, which may be of interest:
http://www.johnbennett.info/index.php/2007/12/27/agnosticism-and-the-new-atheism/
http://www.johnbennett.info/index.php/2008/01/17/natural-law-the-atheists-anodyne/
Once again, very well done.
Cheers,
John
I was a graduate student in science, but was ultimately repelled by the sense of hubris which appears to afflict most scientist.
I therefore find it extremely amusing that the title of the book is “The God Delusion”, since this describes most scientists perfectly.
Cheers, I have read some of your other articles and admire them greatly.
Loren Chabot
[...] im Wesentlichen gegen die Idee des Intelligent Design. Seine Argumentation scheint jedoch nicht viel zu taugen. Ihr fehlt es an theologischer Tiefe, um nicht zu sagen, sie ist reichlich [...]
John Bennett:
I doubt you would disagree with the proposition that “there are no pink unicorns anywhere in the universe”, yet such a statement is also “not testable”. (in the warped sense you seem to be using the concept, see below)
Of course technically the statement should be rephrased as “there is no evidence for the existence of pink unicorns”. In _exactly_ the same sense, there is no _evidence_ for the existence of a supernatural eternal designer that listens to prayers and punishes people for their sins. Maybe god exists, maybe it doesnt, in either case its existence, as long as it isnt measurable (as evidence of some sort) is irrelevant.
So yes, atheism is “not testable”, but given the evidence, it is the _only_ rational position to take.
ps: the proposition “there is no god” IS testable, in the sense that, after precisely defining what you mean by “god”, you could look at the consequences that would follow from the existence of such a god, and “test” them to see if they are there.
For example, for the definition of “god” I gave above, you would expect at least two things. First, people that pray often should statistically be better off than those people that pray less or dont pray at all. Likewise, one would expect that those people that sin against this “god” would be worse off than those people that follow his rules (again, statistically).
Well you can measure such things, and, surprise surprise, there is absolutely no statistically relevant correlation found in the two examples above that would hint to the existence of such a god.
Such an experiment can of course only be arranged if what is meant by “god” is precisely _defined_
And even that sort of precision is lacking in religious circles (especially the ones that pretend to “refute” Dawkins, without even understanding the arguments.)
Carlos,
what a typically arrogant comment to say that your position is “the only rational position” – typical of Dawkins and his disciples.
The difference between God and pink unicorns is of course that no one has ever remotely claimed interaction with pink unicorns. The only people talking about pink unicorns are Dawkinites.
Also, while pink unicorns are utterly unsupported by the evidence (which does not mean that they might not yet exist), the existence of God is a philosophical necessity, as long as you accept the rationality of the world and the fact that the material universe once had a beginning.
And you claim that atheism is a testable hypothesis. Well, you happen to reject any evidence against there being no God (such as people having experienced him) and instead issue bogus tests like praying people should do well and sinners should suffer. This shows that you do not take the proposition of a personal God serious, a God that is not as petty as human beings might, a God that is well able to decide himself what to do. Of course that attitude bars you from ever testing your atheism hypothesis since it is you who lack understanding. No wonder, since your atheism is a faith too.