I really appreciate the excellent letter! I'm guessing you have spent some time on this subject, because most people I encounter don't discuss these topics as well as you.

You bring up some excellent points, and I agree that my position needs to be made a little more clear. Here goes:

1) The definition of "god" is, as you say, a big, big problem. Most definitions of god that I've encountered are either unworkably vague or have holes in them you can drive a truck through. When I think of the possibility of there being a god (or gods), I generally think in terms of "a conscious entity that willfully created the universe." A deity could have other attributes as well, but the arguments I've seen for assigning these attributes have all appeared to be rather weak.

2) I disagree with atheists who say "there is no god" because I don't see any way to prove that there is no god. For example, it seems impossible to disprove the possibility of a prime mover that left after the moment of creation, and such a prime mover would fit my definition of a god.

3) I think it's a bit conceited for an atheist to say that there is no possible proof for a god's existence that they haven't thought of. I'm sure there are plenty of things I haven't thought of.

4) I consider the possibility of proof of a deity's existence through personal revelation. An omnipotent deity could make me believe He existed and I would consider this sufficient proof (because I would have no choice). This proof would not be compelling to anyone but me, however (they'd probably think I was nuts).

5) I don't see the likelihood of either empirical or logical evidence for a deity that can get past Occam's razor. If there is such a thing, it'll have to be pretty spectacular.

6) I don't think that my admitting the possibility of a logical argument I hadn't thought of puts me on the same footing as a theist. First, the theistic position fails Occam's razor (at least for me). And second, theists generally don't believe in a deity because of the possibility of a proof -- they either think they have a proof or don't believe that proof is necessary. (As an aside, you're right about a theist and me having to agree on a definition of god before we could have any kind of agreement -- I haven't had much luck with that <G>.)

7) I do not reject mysticism. I also don't see any evidence for it.

I hope this makes my position more clear. Let me know if I missed anything.

I do indeed admit the possibility of an omnipotent entity, but only because I see no way to rule out the possibility of such an entity.

I did not mean to avoid the empirical/supernatural distinction, so I'm sorry that I gave that impression. I agree that any deity would be supernatural, and I believe that it is possible that the supernatural exists (for what it's worth, I also believe the possibility exists that I will only get heads every time I flip a coin even if I flip coins constantly for the rest of my life). You are absolutely right that it is difficult (if even possible) to qualify evidence for the supernatural. If evidence for something seeming to be supernatural appeared (such as evidence for psychic abilities), it might just be evidence for something non-supernatural that we do not yet understand.

So, how do I argue against the claim that God exists? I don't. I argue against evidence that God exists. That is why I don't say that I am an atheist because God doesn't exist and instead say that I am an atheist because I see no reason to believe that God exists.

I actually have a problem with denying the supernatural a priori. Doing so can lead one into certain intellectual traps -- for example, not testing something scientifically because the only explanation you can think of is supernatural.

Or look at it this way -- I can't think of any way that we could test for universes outside of our universe. If such universes exist, they are likely completely inaccessible to us. But would I therefore say that such things definitely do not exist? No, I would not.

I also don't think you should cringe at real and supernatural things in the discussion of complexity. Since I am not convinced that the supernatural exists, invoking it multiplies entities beyond necessity. Any supernatural explanation therefore fails Occam's razor (at least for me) because it includes the supernatural.

You ask why I should require a strong atheist to prove that there is no god when a strong atheist wouldn't accept the possibility of such a thing in the first place. I feel like this just pushes the "prove a negative" problem back a step: instead of proving that there is no possibility of god, the atheist must prove that there is no possibility of the supernatural. I don't know how one could do that.

By the way, there are a whole lot of non-supernatural explanations for the universe that I can't prove aren't correct: aliens created our universe, we all live in the Matrix, solipsism, I'm having an epic dream, etc. I rank the likelihood of these down there with the likelihood of the supernatural, but I'd be hard pressed to prove that any of them are not true.

Finally, regarding "everything is evidence for the existence of God." Okay, but it is not sufficient evidence. There are much more reasonable explanations.

This is a very interesting discussion. I look forward to your response.

Let's start at the end of your note and take as a given than God can't be an empirically bounded entity. I do not necessarily agree with this statement because it really boils down to how we define "God," and I think there are theoretical candidates that could be called "creator" but still be empirical. But taking this as a given will help move our discussion along.

Now, back to the top of your letter. I agree that humans would not be able to access this theoretical realm (I'll call it "God's realm"), and that any evidence for that realm must present itself empirically. I also agree that accepting or rejecting evidence from that realm boils down to a value judgment. But here's where you may be misunderstanding my position -- I think that's okay.

Value judgments in the form of "is proposition A more likely than proposition B" come up all the time in science. I grant you, most of them are not this huge, but I think the only difference is magnitude.

How would I argue against someone who said that God is the most reasonable explanation? Well, I'd probably ask some questions to find out if the person really believes this statement (that is, that they think God is really the most reasonable explanation and not just the explanation they like most). Next, I would ask the person whether they would change their mind if science produced evidence that (for example) life was actually not unlikely at all. If the person, after this discussion, still considers God the best explanation for the universe, then I will accept that. We simply disagree. Of course, definitely stating this deity's existence and assigning qualities to this deity are still a matter of faith in that they have not been proven.

You also point out that there is no counter to someone saying that they believe in God because they have experienced God's realm. That's true -- this is what I meant in a previous note when I talked about proof of God through personal revelation. I can't argue against such a thing, but it's also not compelling proof to those who have not had a revelation. The best I could do is argue that what they experienced is likely something more mundane than God's realm, but I can't prove it and such arguments generally go the same place that the anti-atheist argument "you know in your heart that God exists but deny it" goes (that is, nowhere).

Finally, let's look at whether one is justified in saying "There is no God" by the standards you propose. If the reason one is making that statement is that one considers the possibility of a deity to be vanishingly small, then I would not favor making that statement. The reason is that we are dealing with a subject that is not empirical, so I don't think we can trust empirical methods (such as scientific likelihood) to come to conclusions about it. When we come right down to it, if there is a deity or a God's realm, you and I don't know a thing about it -- any statements we make about its probability of existence are educated guesses, so I shy away from acting like they are more than that. At best, we can make statements about what such a realm is not (because certain qualities would lead to logical contradictions, for example).

This may all sound philosophically wimpy, but there are pragmatic reasons for my thinking in this way. For example:

1) If I say that there is no deity even though I can't "prove" it, it could be said that I have faith that there is no deity. I don't want to go there.

2) I place no value on proving that there is or is not a deity. However, I place a lot of value on helping people sort through their religious beliefs and make sure that those beliefs are consistent and non-contradictory. The possibility of a deity says almost nothing about that deity's characteristics, and it's in these details that I prefer to hold discussions with theists because such things can be logically tested with relative ease. It's a big jump from "there might be a deity" to "the Christian God exists," particularly since very few Christians I've met have a good working knowledge of their own concept of God (as evidenced in the difficulty they often had defining good and evil).

3) Many people who say that God is the most reasonable explanation don't have much familiarity with how Occam's razor works. One example of this is that they often think that Occam would prefer one all-powerful deity to many less-powerful deities (as in Greek mythology). Globally applying their misuse of Occam might lead one to believe that (for example) it is more likely that the pyramids were built by one giant guy than by thousands of normal guys. I think that helping people learn how to use tools like this does more to advance the cause of science than trying to argue against deity as a concept, and I secretly hope that as people learn to use these tools they will realize that God is not necessary.

4) Even if I agreed 100% with your argument that there is no God, I'd say that the argument is inaccessible to your average person without a lot of discussion, and that this discussion would ultimately almost never be beneficial to the theist. Since I say that I am not convinced there is a deity, it is up to religious people to prove to me that there is a deity, and I believe that the exercise of attempting this proof is excellent experience for theists, many of whom have never really thought about why they believe there is a deity.

5) The bulk of people I talk to have no logical reason to believe in a deity (even if they think they do). Their arguments are an attempt to justify their own beliefs, as opposed to the real reason that they believe in the first place. I think that there is huge value in getting people to admit that they believe in a deity only through faith. This essentially brings God into the camp of "personal truth" (as opposed to "compelling truth"), and it's a lot easier to build tolerance and acceptance of others when we realize that everyone has personal truths and that none of them are more valuable than anyone else's. (This is a topic we could go on at great length about, but I think you see what I mean.)

As a final point, I don't have anything against religion in general. It's intolerance and ignorance that I prefer to battle.

Whew! Big response from me. I hope it was worth the wait. I look forward to hearing from you!

Well, some atheistic beliefs are the dogmatic equal of theistic beliefs. But I argue against those, too <G>.

I agree that an empirically bounded deity wouldn't fit the general religious concept of a deity. I think that many religious people would accept a sufficiently powerful non-supernatural creature (such as the Q, if you are familiar with Star Trek) as god, if the creature presented itself as such (and I'll have more to say on defining God a little later).

I also agree that we should not structure our beliefs as a response to the arguments of others, but I think it is often good to express our beliefs with such arguments in mind.

My lenience toward proof through personal revelation is not a concession to religious belief. If there were a deity, and if that deity could change things in our world, and if that deity chose to make me completely believe that he existed, I would have no choice but to believe. Others might conclude that I suffer from some kind of delusion or mental problem, but I would believe, much as I might wish it otherwise. For this reason, if someone tells me that they believe in God because they can feel God's love within them, I do not feel justified in telling them that they are definitely wrong. To do so would be to presume knowledge that I do not have, since the thing they are claiming to know through supernatural means is something that I cannot test myself.

I did not mean to imply, by the way, that skepticism about souls is equivalent to a "you know God exists but deny it" argument. I meant to say that "your personal revelation should not convince you that god exists" is a similar argument to the latter statement. I don't feel that, in most cases, we can meaningfully argue about what another person feels deep within themselves. (But we certainly can discuss how someone should act on those feelings.)

One point where I either disagree or completely misunderstand you is where you say, "Access to God's realm is only possible if we have a medium within us that connects it to our empirical reality, a soul by any other name." I don't see why, if a non-empirical deity existed, we would have to have something within us that is outside the natural in order for us to experience what I'll call divine intervention. This sounds to me something like (if you'll pardon a seriously flawed analogy) someone saying that fish can't interact with things on dry land unless they have some spark of dry land within them.

We, obviously, couldn't reach out to sense or interact with a non-empirical realm (by definition), but I don't see any reason to deny a deity the power to interact with our realm, particularly since this deity would have created empirical reality. A deity could cause things to happen that we could sense or test (such as resurrections or personal revelations). For all I know, God makes the wind blow. It's vanishingly unlikely, but I don't feel that I can say it's impossible.

About the sun rising. I'd say that it is an expression of faith to say that the sun will rise tomorrow, but since it's a faith (in the consistency of the laws of physics) that we all share it goes unspoken and doesn't need to be pointed out. That is a nitpicky, vocabulary-based point, but it does point out the correctness of your statement that people speak differently when they speak about religion than they do when speaking about science.

Now we get to what may be the real sticky point. "I think it is intellectually dishonest, at best, to let anybody define a thing such that it's existence cannot be disproved and then concede them the right to accuse anyone dismissing that thing of having some kind of negative faith." I have two thoughts about this. I'll give you the short one first. If something is defined such that I can't prove it doesn't exist, I don't feel I can say it doesn't exist. All I can do is respond to arguments that it does exist.

My second thought is, I think, far more important. I think the crux of the statement is the phrase "let anyone define a thing such that its existence cannot be disproved." You already know that I won't say it's impossible to prove that a deity exists because there may be proofs I haven't thought of. But that aside, let's look at what kind of definition of a deity I allow.

Most theists don't think that it is impossible to prove that there is a god. Many of them think that they have proofs, many others just haven't given it sufficient thought. When I have a conversation with a theist, one thing I like to do is find out whether they believe they can prove that their god exists. If they don't and simply believe on faith, then that part of the conversation is over. They can't convince me with their faith, so I don't believe in their god and they have no right to accuse me of negative faith. If I were, instead, to say that their god does not exist after they admitted that their belief is based on feelings alone, then I either need to be able to prove that there is no god or I have to say that my disbelief is based on a feeling that I think trumps their feeling. I think that's a philosophical mistake.

Now let's take the much more common case where the person thinks that the existence of God is provable. I would proceed to ask what proof there is that God exists. You and I both know that whatever evidence they have will have to either fail under examination or turn into a statement of faith. If it fails, they haven't proven their case. If it turns into faith, we're back to the previous paragraph.

So what this boils down to is, if someone admits that the existence of a deity can't be proven or disproven, then there is no burden of proof. If they define god as provable, then the burden of proof is on them. So the only way the burden of proof falls on me is if I say that there is no deity. I see no philosophical necessity for inviting that burden.

But, in a way, much of this discussion falls away when we look at what you and I are talking about when we use the word "God." At the beginning of your letter you touched on the subject of what concept of God would be satisfying to a religious person, and this brings us to another reason that I do not say God doesn't exist (and it's a reason that I don't usually like to get into because religious people tend to freak out about it for some reason).

By and large, when people use the word "God," they really don't know what they are talking about. They can't define God's properties in a coherent, consistent way. So how can I say that "God" doesn't exist when, frankly, we don't know what "God" is in the first place? You and I are talking about God, but we're using a definition most religious people likely wouldn't accept, so we haven't exhausted the topic.

This is why I sometimes like to ask people to define God for me, particularly if their proof for his existence is something that can't be empirically tested. It generally leads nowhere.

So, to sum up big time: 1) I don't say it would be impossible to prove that a deity exists because the concept of deity is gigantic and there might be some way to prove that something we'd refer to as a deity or creator exists. 2) I say "I am not convinced that there is a god" (instead of "god does not exist") because I cannot rule out the possibility of the existence of a supernatural being that a) might not be detectable and b) we haven't even really defined.

I will, however, gladly say that many, many, many specific attempted descriptions of God are of entities that don't exist. And I will also say that, even if there is a deity, the bulk of religion on this planet is likely very close to entirely incorrect, so far as metaphysical matters are concerned.

This may be a discussion for another time, but I don't know if I agree with you that there is a difference in the word "faith" when used to refer to deities or sunrises. For example, to me it is a matter of faith -- something believed without proof -- that the laws of physics do not change and that my perceptions are (at least in some way) a reflection of reality.

When discussing deities, I think we have to adjust our vocabulary to accommodate those to whom we are having a discussion. Otherwise we could find ourselves in a discussion of corn where I mean maize and someone else means wheat. When a Christian says "God," I assume they mean the Christian god and are saddling themselves with all the attached philosophical baggage. When you say "God," I assume you mean something much more narrowly defined. Being able to make statements about a deistic god doesn't help us when discussing a Christian god.

You are right that access to a god that is not accomplished by a connection to the supernatural (on the human's part) must be done through the empirical realm. You are also right that you can legitimately dispute such proofs. The problem is that if one says that God acts through nature (as Catholics do), then proving that what a religious person claims is an act of God is just an act of nature does nothing to disprove their point. We're back to the "everything in nature is proof of God" argument and what an individual will accept as sufficient proof of the supernatural.

You say that we "usually try to help people make the distinction between what they imagine and what is real" and you are right. The problem is that feelings of faith are generally closer to raw emotions than they are to obvious delusions. It is possible to argue that Sally doesn't really love Bob because Bob's a jerk, and maybe you could convince Sally that she is just mistaking need for love, but then we get into the mess of trying to define what love is in the first place. The feeling of experiencing God's love or of "knowing" that there just has to be a creator might be the same sort of beast, but I don't know because I don't experience it so I have trouble arguing that others aren't experiencing it.

If we could identify the state of mind that equals feeling God's love, then you are right that we could probably recreate it medically. But this does not prove that feelings of God's love are not caused by God any more than a magician bending spoons proves that Uri Geller is a magician and not a psychic (although it sure would lead me to suspect that).

I'd say that a personal revelation of god is indisputable only to the person who experiences it and only in the sense that there is no evidence I can produce that will change their mind. It is not immune to criticism, but such criticism will have no impact on the person who received the revelation. Can this kind of thing be said of mental illnesses as well? Sure.

I agree that I cannot think of any empirical proof for the existence of a supernatural creator that would be unquestionable. I don't agree that the laws of logic arise from the physical laws, because logic can exist without the physical (as can mathematics). You are also right that the deistic god is unprovable (and, I would add, non-refutable).

The deistic god piece is worth looking at. It seems to me that if I can't prove or disprove a god that took a hike right after creation, I also can't prove or disprove one that left a year, or a million years, or a billion years after creation. If god decided his work was done and left in 2004, I would be similarly out of luck. Just a thought.

Of course, anything related to a deistic god doesn't do us much good today. There aren't a lot of deists around (which is too bad, because I think deism is pretty defensible). Bummer.

More good stuff!

Let's start with the definition of "faith." I'd say that faith is "belief without proof," not "belief without evidence," because a great many people who have faith would say that they have evidence for that faith. For example, I have a lot of evidence that the laws of physics are immutable, but I can't prove it.

About you being a "strong" atheist with respect to the Christian God: this is a broad field, because even Christians can't agree on the nature of God. I think you and I technically agree on this, but for different reasons and with different means of expression. In that I have yet to hear a good, coherent, complete definition of God from a Christian, I am not willing to say that "God" (whatever that is) does or does not exist, but the being described by the definitions of God I've heard is either so vaguely defined that I can draw no conclusions or does not exist for the reasons you state.

I do think that "proof through revelation" is less disputable than "proof through burning bush," if only for matters of convenience. More importantly, I think that revelation is more compelling to the individual who experiences it than a burning bush. If there is an all-powerful god and that god wants me to believe, I will believe -- I have no choice. For that matter, if super-advanced aliens want to adjust my brain chemistry so I think I'm a carrot, I'll have to believe that, too. (Perhaps making me a theistic carrot.)

Certainly we can argue against and examine proof through revelation. I just think it's generally pointless because, ultimately, you can't prove there was no supernatural cause. Similarly, Catholics believe that God works through science and directs evolution, and there's no way to disprove that. At best, you can say that God is not a necessary part of the equation.

I agree with you about the Santa Claus replacement. I think there can be value in getting people to consider the possibility that their "revelation" of God's existence might have other explanations. But this neither proves nor disproves the existence of a deity (and, in any case, is pretty low on my list of priorities -- I think getting people to behave well within theism is more important right now than trying to convince people to give up their faith).

I am curious why you would say that logic or mathematics could not exist without the physical. Do you mean that there can be no number "1" without anything to be counted? This may be simply a semantic disagreement.

Deistic gods: a deistic god is not provable because it doesn't do anything anymore. It is possible that a theistic god is provable, because it might chose to do something to prove its existence. Aside from changing my brain chemistry, I don't know what that "something" might be, but I can't say that I'm smart enough to have thought of all the possibilities. Do I find this troubling? Not really. Frustrating at times, maybe. I would much rather be able to find out the truth behind everything, but some things are beyond investigation.

I should also bring up the point that it may be possible to find proof of even a deistic god after death. I doubt it, and I'm not in a hurry to try, but I think it worth mentioning the possibility. Not that it does us any good at the moment.

I sometimes talk about UFOs and alien abductions as a bit of an analog to religion, and in the case of our discussion there are some interesting parallels. If aliens once visited Earth but no longer do so, I can't disprove their visits and there is no good evidence for their visits, but I wouldn't say aliens have never visited our planet. There are literally millions of people who have had a UFO experience of one sort or another, but even though I think the possibility of extraterrestrial life is quite high, I think it likely that all those millions of people are wrong. If aliens do visit us one day, I think we'll find that they are much different from anything current UFO-fans have in mind. My suspicion is that, in the unlikely event that there is a deity, we will find it completely alien to current common beliefs. And that makes discussion of such an entity's properties (including existence) very difficult indeed.

You say, "I think most theists would admit that they'd still believe in god even if they had no evidence of his existence, which is what makes such a belief faith." I'm not completely sure that's universally true. I can think of a couple of people who say that they believed in God because a creator was the only thing that made sense, but whom later became atheists when their knowledge of science improved. I also think it's possible that you're right in general, but that in many (perhaps most) specific cases theists will not say that they have belief without evidence, even if to you and me they do. For example, some people would call "feeling God's love" evidence.

You are right that one needs only a single contradiction to disprove something, and you are right that one does not need an exhaustive definition of god before contradictions have to be looked for. But one cannot disprove all definitions of "god" by finding contradictions in one definition. I think the problems here are that there are many possible definitions and that a great many theists speak of "God" without even having a real working definition (although they think they have one).

Fortunately, when theists (at least Judeo-Christian theists) start working on a definition they generally run into the kind of problems you speak of pretty quickly. Buddhists and deists, for example, often do better because their deity is more detached from empirical reality.

So, summing up my position, I have no problem arguing against an individual's definition of God, but I hesitate to assume that I can argue against all possible definitions of a deity.

Your thought that there are no numbers without at least one countable is interesting. Off hand, I don't think there is anything wrong with this, but it is a new concept to me. As an aside, would you say that mathematical principles (such as pi) are invented, where I would say they are discovered? Just curious.

I think we got our wires crossed on the subject of deistic vs. theistic (meaning non-deistic) gods. I think I know what you mean when you say that a theistic god is a deistic god and more -- in the same way that (to oversimplify) a scooter might be seen as a skateboard and more, right? I don't know whether I agree with this statement, because if the deistic god does not interfere with reality because of either moral or physical (for want of a better word) restraints, then it is probably significantly different from the theistic god and not just a theistic god that does not interfere.

You are right that there probably isn't a useful argument to prove that an unprovable god precludes a provable one. Sorry if I implied that I thought there was. I would say that disproving a theistic god does not necessarily do anything to disprove a deistic god. And I would say that it might be possible for a theistic god to permanently detach itself from reality at any point and effectively become deistic, possibly rendering its existence impossible to disprove.

There really was a point to my bringing up aliens that had nothing to do with bugging you <G>. I agree that aliens are nothing like a supernatural deity, but the way I argue against statements of their existence is very similar to the way that I argue against statements of the existence of deities.

You say, "If aliens exist, it is perfectly possible for me to go and look at them, smell them, touch them, hear them, and taste them." I think it is this statement that might be at the root of our disagreement, because I wouldn't make it, at least not in so many words. Your statement has an implicit assumption that aliens are accessible, and I am not willing to make that assumption even though I understand that you are speaking in a philosophical sense (because aliens are, in principle, accessible if they exist, even if access to them is not practical).

Let's assume there are aliens (meaning some kind of extraterrestrial life) so I can demonstrate how this has anything to do with anything.

I think that the probability of alien life is very high. I think that the probability that any of it has been observed by humans is very low. So if someone tells me that they tasted an alien (and it tasted like alien chicken), I think they're probably wrong. If they describe the alien in a way that is not logically coherent, then I can prove that they are wrong. If their alien encounter is more easily explained by more normal occurrences (you were sleepwalking and ate some leftover chicken), I consider that more likely. But in my mind, this has nothing at all to do with real alien life out there on some planet somewhere. Disproving every supposed alien encounter I run into would not prove that a) aliens don't exist, b) none of the people imagining that they encountered aliens got some of the details right, c) nobody I haven't run into really did encounter aliens, or d) if aliens wanted to hide from us, we could detect them. And this is assuming that my disproofs are valid.

There are people who say, with complete conviction, that aliens are out there, even if we can't prove it. I can't argue against that, because it's a statement of faith. And I won't agree with it, even though the probability is, for me, high that it is a true statement.

So, applying this thinking to a deity:

Our disagreement over your implied assumption that aliens are accessible is similar to our disagreement over your definition of a supernatural deity as completely detached from the empirical. I would say that aliens are not accessible because, at the moment, we can't access them. I would also not define a deity as necessarily completely detached from the empirical, even though we could never prove its existence empirically.

I think that the probability of a deity is very low. I think that the probability that anyone has proof that a deity exists is low. If someone tells me that they have experienced God, I think they're probably wrong. If they describe God in a way that is not logically coherent, then I can prove that they are wrong. If they offer no proof, I still consider mundane explanations for their experience far more likely. But in my mind the fact that people have developed incorrect religions or that they mistakenly believe that they have interacted with a deity does nothing to do with whether there really is a deity of some kind. I can rule out specific descriptions of a deity, but I can't rule out all possible descriptions or invalidate claims that I can't investigate.

You are right that a big part of our disagreement is whether or not a supernatural entity can be observed. I still do not agree that a supernatural being must by definition remain entirely within the supernatural realm. I don't know what it would mean for God to send part of himself to Earth in the form of Jesus, because I don't know how a supernatural being would work, but my not knowing doesn't make it impossible. Seeing a Jesus-looking guy walking on water wouldn't be proof of the supernatural (because there are non-supernatural explanations that are more likely), but my having a better explanation is not proof that the Jesus guy isn't supernatural.

Looking at this another way, there's a movie in which God is sitting in a courtroom and the judge asks him to show him a miracle. God takes out a deck of cards, spreads, them and asks the judge to "pick a card." If God uses supernatural means to state what card the judge picked, we'd never know it because there are so many mundane ways to do card tricks. For all I know, there is a deity out there that interacts with the world but always does so in a way that appears to follow the laws of physics. There would always be a non-supernatural explanation for the behavior of such a deity.

One last thing, just to keep our conversation traditionally overlong.

Let's look at the Catholic belief of transubstantiation -- if you're not familiar with it, this is the belief and bread and wine become the body and blood of Jesus during a church ceremony. If you run tests on the food products before and after the ceremony, they will appear to be the same thing. If someone asked me whether the bread really became "the body of Christ" during the ceremony, I would not say no. Instead, I would say that I have no idea what transubstantiation really means and therefore can't make a statement. I feel the same way about most definitions of god.

I think we're getting very close to settling this here, which is kind of exciting. It's fine with me if we discard self-contradictory definitions of a deity.

It appears to me that there are a great many definitions of deity out there because, for example, different groups of Christians can't even necessarily agree on the nature of God (for example, does God consider good works to be important or only expressions of devotion -- and I'd say that these are significant distinctions in that they can be used as a foothold to examine other described properties of the deity). There is also the problem of whether we can meaningfully say that two religions who ascribe different activities to God (such as writing or not writing a holy book) believe in the same God. Christians, Jews, and Muslims will generally say that they do worship the same deity (the God of Abraham), but often their descriptions of God are so divergent that it seems to me to be overly charitable to grant this. Particularly when Christians can't even agree on who can be called a Christian. And when we look at the beliefs of individuals instead of those of organizations, definitions of God vary even more, even within Christianity (I say this based on personal observation only).

This is why I disagree that we are always talking about the same entity when we are discussing monotheism. Some of the definitions are mutually exclusive.

I agree that not all religions are equal, by the way. I also agree that deism and other forms of theism are not equal. An aside: there is also the point of view that all religions are reactions to the "experience of the divine" at some point in the past and therefore contain a grain of truth even if they are incorrect in all the details. I haven't a clue how I'd prove that isn't true, and I've heard it espoused by some people who say that they are Christians.

Now, for the rabbit in your pants (a sentence I never imagined I'd be typing). The pants rabbit is a different animal than aliens and god. You can check your pants for rabbits; humanity might some day be able to check every planet in the universe for aliens; we can't check for god in any way I can think of. So the idea of checking your pants for a variety of rabbit types is a bit of a false analogy, since we can't check for the existence of god. Or, rather, we can at best say what kind of animal isn't in your pants. And if you were to respond that the reason I couldn't find the rabbit was that it is an invisible, immaterial rabbit, I couldn't prove you wrong, although I'd suspect you were bananas.

You ask what I expect you to conclude from the fact that there is always a new possible definition of a deity. The conclusion I draw from this is that even if I can conclude that every deity ever described by humans does not exist, I can't say that there is no deity out there. This works for aliens the same way -- rule out every alien every described by a human, and there still may be aliens. Your rabbit does not fare so well -- check for him, and if he's not there, he's not there (assuming the normal definition of "rabbit").

You are right that I have a strong preference for Occam's razor. I agree that Occam's razor seems to completely rule out the necessity of a deity. However, this is not universally compelling proof that there is no deity because, a) the razor is a guideline, not a rule, and b) people with a different view of probability will apply it differently than I will.

Now let's get into a really technical bit: "We seem to be in complete agreement on the problem a deity would have in showing himself through natural means, but I cannot figure out why you seem to regard this as a defense of the existence of a supernatural god." I don't regard this as a defense of the existence of a supernatural god. Rather, it is a defense of the impossibility of disproving the existence (as opposed to the likelihood) of a supernatural god. I don't think god exists, but it is, so far as I can see, impossible to prove that god does not exist, so I do not say that god does not exist.

Life would be greatly simplified if, when someone asked, "Do you believe God exists?" I could answer "No" because I could demonstrate that the deity the questioner has in mind could not exist. I do not feel justified in doing so because:

1) I don't know what god they have in mind, and often they don't know either,

2) When I argue against a specific definition of god, people very often just start refining their definition to get around my objections.

3) I don't think such a response would honestly convey my opinion.

Both #1 and #2 may be evidence of belief without a good basis, and I recognize that, but it doesn't make my philosophical standpoint any easier. You and I could pretty easily prove to each other that the likelihood of a deity is vanishingly small. But this proof might not be accepted by those who make different assumptions about reality than you and I do, and since I can't prove that my assumptions are correct (particularly since they include an assumption that there is nothing supernatural), I would feel arrogant acting as if I am definitely correct.

You sound like you do not have this discomfort, so you feel justified in saying that there is no god.

I think we may have come to the roots of our disagreement.

You are right that, in monotheism, "god" refers to a specific, individual entity. Where we differ is that, to me, all definitions of a monotheistic god are not equal. This is different than counting the claws on a rabbit, because in some cases different proofs are used to show that different definitions of god are problematic. For example, a discussion of whether God is "good" can go very differently when speaking with a Catholic than with a Lutheran. For the "claws on a rabbit" analogy to be true, there would have to be a proof that no possible deities exist, and I am not yet convinced that one does exist.

I would say that the definitions of "aliens" and "god" actually do have some similarities that are important to understanding my way of thinking about this problem. In our current context, "aliens" refers to two different things: A) aliens as described by people on Earth who say that they have experience with such things, and B) whatever extraterrestrial life is really out there. We have no idea what B is, but if it showed up, we'd call it an alien. I think of the concept of god in the same way. Protestants don't believe in the deistic god, but if there were some way to prove deism true, I'm guessing Protestants would still call this being "god" even though it was nothing like what they had been worshipping.

I think one of the reasons we keep going back and forth on some subjects is that I won't say "there is no god" just because some class of deity has been ruled out. Even if we were to completely agree that the monotheistic, non-deistic god didn't exist, I still wouldn't make that statement because polytheistic, Buddhist, deistic, and other concepts of god have not been ruled out. I think that these concepts are distinct enough from the god of Christianity to be considered different gods.

I am also extremely wary of an ontological argument against the existence of a deity. If we agree that humans know nothing definitive about the supernatural, then I am very uncomfortable making definite statements about supernatural aspects of supernatural things.

You ask how I can tell someone that I don't believe in god without defining the term -- whose god don't I believe in, mine or theirs? That's a good question. From my perspective, since I say that I don't believe in god since I have seen no compelling proof for the existence of a deity, I feel that I'm covering all possible definitions of god. It is a true statement that I have not seen any compelling proof for anything that I could meaningfully label "god."

"If it's impossible to prove that god does not exist, than it's impossible to prove that he does exist." Well, that's not true in and of itself, but it is true in the context of our conversation (ignoring personal revelation, which I think we've agreed to disagree on). You continue, "If the proposition 'God exists' is neither verifiable nor falsifiable, then it isn't meaningful." That depends on what you mean by "meaningful." Certainly the statement has important emotional meaning for many. From a scientific perspective, there are many things in science which can't be verified or falsified, only speculated on based on existing evidence (for example, other universes and some historical truths).

At most, I would say that deities do not appear to be necessary.

I think your final paragraph may sum up our disagreement perfectly. You say that there is nothing outside of space and time because humans cannot define anything outside of space and time. I say that the inability to define something does not prove that something does not exist, particularly when we're talking about my personal ability (which is all I have to work with). I also feel uncomfortable assuming that there will never be an idea or concept or means of proof that I haven't thought of but that would change my mind.

For me, this is a comfortable place. I am never put in the position of having to prove a negative. And from a pragmatic perspective (as we've discussed before) this allows me to bypass trying to convince people to give up something based on feelings alone and concentrate my efforts on aspects of religion that are more significant from a social perspective.

From a philosophical perspective, I would say that a minimum definition of "god" for the purpose of this discussion would be something that is not bound by the laws of nature, is not confined to our universe, and has or had the ability to interact with our universe. This definition doesn't cover everything that some people call god ("god is nature," "god is love"), but I consider those types of definition to be in a separate class, one that doesn't impact whether I call myself an atheist. We might be able to come up with a better definition if we wanted to work on it a bit.

For practical purposes, though, I'd say that "god" is any supernatural thing that many people would label as a deity if it existed. That's a much easier definition to work with, and it again puts the onus on the believer (where I think it belongs). It also gives some pretty lame things the opportunity to be called god, but I can live with that.

You talk about the inability to use the term "exist" in relation to something undefined. You are right and I should have chosen my words better. Perhaps I should have said that we cannot make definite statements about things completely outside of our experience. I would also say we can't say something doesn't exist just because we are unable to investigate it scientifically.

Another point, because I think I let our definitions drift a little -- I wouldn't say that we can't define existence outside our universe, but rather that we can't describe it. So you could counter this if you could show that "existence outside our universe" is meaningless in the same way that "temperature below absolute zero" is meaningless.

I think you're misunderstanding my position a little when you say, "The statement can be rephrased as "god cannot exist," because, as you yourself agreed, god does not exist within the boundaries of the natural world." I didn't say that a god would not exist within the boundaries of the natural world; I said that a god would not be confined by the laws of nature and the empirical world. This is a big difference, and one we keep coming back to. I feel that restricting supernatural things from interacting with empirical things by definition is arbitrary. Saying that we cannot venture into a supernatural realm does not imply that a supernatural realm could not interact with us, even if we would perceive those interactions empirically. Similarly, I would argue that our inability to interact with a fifth dimension of space would not prove that there is no such thing.

You ask if I would say that something doesn't exist just because it is absolutely impossible, ever, for us to investigate it. No, I wouldn't. It might be impossible for us to detect alternate universes, but I wouldn't conclude from this that they don't exist. If something came before the big bang (so far as that term is meaningful), it is impossible to investigate, but I wouldn't say that the big bag was preceded by nothing. I wouldn't even make the statement that the big bang was necessarily causeless (keeping in mind that there could be a natural cause, of course).

Is it a giant violation of Occam's Razor to postulate something nobody can ever experience as a casual agent? Well, I don't agree with the implication that a deity wouldn't be able to cause experiences, but in any case, you're right that a deity is, to me, a giant violation of Occam's Razor. But violating Occam's Razor doesn't prove that something doesn't exist, it just makes it the less favorable explanation (in this case, the much, much, much less favorable explanation). Also, someone with different knowledge or assumptions about probability might reach a different conclusion (as is often the case).

As for the sky clown, how are you defining it? I'm assuming that we can't go and check for the clown, so perhaps you are using the term "sky clown piss" in the same way that a Catholic uses the phrase "body of Christ" during mass. If so, then I can't prove you wrong, but we would have to spend some time defining your terms before I was sure you were even saying something meaningful.

What difference would the sky clown make to our lives? You're right -- none (at least for practical purposes). But that doesn't prove anything. Of course, if we're talking about God, whether or not he exists makes a lot of difference to a lot of people. In fact, if a deity of some sort exists it might make a big difference to you and me, too, whether we know it or not.

I see what you mean with the proof that existence outside space-time as meaningless, but I don't necessarily agree with your definition of existence. Also, allowing for other universes to be included within this definition because they have space-time (even though it might not be compatible with our space-time) seems to make the issue even more complex.

When I spoke of a deity not being bound by the laws of nature and the empirical world, I was talking about our universe's laws of nature and the world we can detect. I would not rule out the possibility that a deity exists within some kind of universe. For example, a higher-dimensional being that created and can interact with our universe would fit my definition of a deity, even though it would have an empirical world of its own. I could not describe the properties of such a being, and I can't imagine what it would be like, but it seems to me that such a concept is not logically impossible.

Now, if you would consider this higher-dimensional being not to be god because it has a space-time existence in a context of its own, and you cannot conceive of anything without a space-time existence being possible, then I think you might be justified in saying you believe there is no god. But you would be doing so with a definition of god that might not be universally acceptable.

I agree that I can't describe what aspaciotemporal (a cool word that I hadn't come across before) existence might be. But since I can conceive of a being in a separate space-time context that we would consider to be a deity, I'm comfortable with that.

Agreeing to disagree is an okay place to end up. In fact, that's where I end up with most deep-thinking theists as well. Even though I don't share your opinion, you have convinced me that someone can have valid reasons to call themselves a strong atheist.

Your position seems to be very close to the linguistic argument against the existence of God, which holds that God cannot be defined and therefore cannot meaningfully be spoken of. But where the linguistic argument ends with "we can't discuss God because we don't know what 'God' is" you are able to specifically state that there is no such thing as God because you can see no possible context within which such a thing would exist. I think that's quite clever and hadn't run into it previously. The closest I've come is the creationist argument that it is inconceivable that our universe came into existence without a context, and that context must be God, but the creationist argument doesn't hold up under Occam's Razor and yours seems to.