More Recent Comments

Wednesday, December 18, 2013

Elliott Sober illustrates (inadvertently) the problem of definitions

One of the problems in most debates and discussions in the problem of definitions. It's common for two opponents to end up talking past one another because they don't agree on what they are arguing about. That's why an important component of critical thinking is to define your terms so that everyone knows what you are defending (or attacking).

But there's more. If you are going to be a good critical thinker, then you have to be aware of other points of view. If there are other, equally valid, definitions out there then you MUST acknowledge them and incorporate them into your argument. You can't, for example, just make up your own definition of words like "noncoding," "junk," or "function," and declare that you are right. Since you know that there are other definitions out there, you are obliged to show why YOUR definition is the only correct one. That's a crucial part of the debate.

If you don't even know that there are other valid definitions then you are not an expert and you should not be pretending to be an authority on the subject. This is why I object to people who argue against the Central Dogma of Molecular Biology without understanding what Francis Crick actually said.

Let's look at a video of Elliott Sober lecturing on "Some Questions for Atheists to Think About." He begins by asking the members of his audience whether they are atheists or agnostics. Apparently, most members of the audience are atheists and ony a few are agnostics.

Next, he defines his terms ...
Theism = God exists
Atheism = God does not exist
Agnosticism = We don't know whether God exists
The lecture is about something called "evidentialism." Elliott Sober claims that the following proposition is true ...
For any proposition, you should believe it only if you have evidence that it is true and you should disbelieve it only if you have evidence that it is false.
He then goes on to show that we can never have evidence that God does not exist. Therefore, "If 'God exists' is untestable, you ought not to be an atheist. You should be an agnostic."

He suggests that all the atheists in the audience should become agnostics because of evidentialism. If I had been in the audience, I would have pointed out that MY definition of atheist is that an atheist is someone who doesn't believe in god(s). My definition is such a common definition that it's part of the Wikipedia article on Atheism. Since I do not need "evidence" to not believe in something, I'll remain an atheist, thank-you very much. My position is perfectly consistent with the proposition about evidentialism.

Here's the problem. Elliott Sober is a prominent philosopher. Doesn't he realize that his argument relies entirely on his definition of "atheism"? Doesn't he realize that his argument is completely useless if an atheist is simply someone who doesn't buy into the God delusion? This sort of thing makes me livid and it makes me wonder whether there's something seriously wrong with modern philosophy.1



1. I am not suggesting that Sober's definition is wrong. He should not be ignoring the fact that many members of his audience don't agree with his definition and that's why they are atheists.

44 comments :

Eamon Knight said...

A professional philosopher like Sober is unaware of the old strong vs. weak usages of "atheist"? (Of course the latter arguably subsumes agnosticism)

What level of evidence do we need to reach a conclusion in good faith? I think the evidence is very good that the gods of the prominent orthodox religions don't exist. I think the only not-definitely-falsified gods are the rarefied ones of the philosophers and apophatic theologians, who seem to keep themselves carefully hidden (the gods, I mean). But it's a bit perverse to say I'm "agnostic" about propositions which I have no a priori reason to accept, and which are untestable (in many cases, apparently contrived to be so). There's an infinity of such propositions, which makes me agnostic about a lot of things, philoso-gods among them.

The term "ignostic" seems apropos here: I'm ignoring the question as not worth my time.

Rob said...

There's also not enough definition there. Define God.

Philosophically, do consider myself agnostic (6.9+ on Dawkins scale)
But, when you start asking about *specific* gods, that changes, a lot. Specific gods are falsifiable, and have pretty much been falsified. Abrahamic, Hindu, Greek, Norse gods? I'm atheist, 7 on Dawkins scale.

Anonymous said...

I think that at some point, lack of evidence itself becomes evidence.

Take the example of the Sasquatch or Bigfoot. The "best" evidence has been shown to be faked. The remainder is totally ambiguous or the sincere reports of people who probably did see something, but saw it too briefly anyone else to be certain it wasn't a bear, a deer, a person, a tree, or even a large floater in their eye. There are a priori reasons to believe there wouldn't be large non-human apes in the Pacific Northwest, too, though they could be trumped by evidence if there were any. One can do theology-like reasoning about why large mammals could be here and yet no bodies and no bones have ever been collected, but that hardly constitutes evidence that Bigfoot actually exists. I suppose one could be agnostic about Bigfoot, but at this point I just have to say, "There is no Bigfoot." I could be wrong, of course, but until incontestable evidence shows up, there is no Bigfoot.

I'm not agnostic about Bigfoot. I'm not agnostic about God any type of God that could be distinguished from the non-supernatural world, either. (And hypotheses of God that can't be distinguished from the non-supernatural world are appealing to me but not parsimonious.)

John Harshman said...

I think his definition fits mine pretty well, so I wouldn't complain about it. I would merely add that scientists, and in fact rational people, tend not to make absolute statements. So I'd define an atheist as a person who thinks that gods are very likely not to exist. I could even go with "almost certain".

But has Sober defined "god"? That seems to be a major sticking point. What is a god? What, consequently, would we expect to observe if this god existed that we wouldn't observe if it didn't? If he's including in his definition gods for which there could be no conceivable evidence, i.e. if their existence or nonexistence would make no discernable difference in any observation, then we are entering the realm of Jovian teapots, and there's no real way to speak meaningfully about belief or disbelief.

But of course nobody actually imagines that sort of god. All the gods people really believe in have consequences, and so there can be evidence for or against them. And thus his argument is moot, and it's rationally possible to be an atheist.

Diogenes said...

A god that is omnipotent and benevolent is a testable hypothesis because it makes predictions about observable events in the real world.

Let me test this by looking into today's newspaper, and examining human behavior and natural events in the real world...

AAAARGH! AAAARGH! AAAARGH! AAAARGH!

Kevin said...

I think you're misrepresenting Sober's argument. I don't recall hearing Sober state that evidentialism is true, or that atheism simpliciter is false. He recognizes that there are many God concepts, and that on evidentialism, you should be an atheist for some of those concepts, and on others you should be an agnostic.

Moreover, I took it that Sober's central point wasn't to establish any particular conclusion, but to encourage atheists to reflect on the relationship between evidentialism and their views.

Kevin said...

Sober does so at least twice. If you follow his presentation he recognizes that there are many different ways to define God. In fact, your post is pretty much in perfect agreement with Sober.

Hawks said...

Larry seems to get quite irate because Sober defines atheism differently from what Larry himself would do. Well yes, Sober did - and he did not acknowledge that there are other ways to use he term. But it was a 15 minute talk and the definition of the term atheism is really a side issue.

The point of the talk was really about evidentialism and when we should accept/reject a claim vs when we should suspend judgement. As he says, we can reject (be atheistic about) the claim that life on Earth is less than 50,000 years old but should suspend judgement (be agnostic) about the claim that "God is a mysterious force who gives life its meaning".

Personally, I would the terms in the same way that Sober did (although I must say that using "Eamon Knights" term ignostic is even better) - which would make me an atheist AND an agnostic. Or a strong atheist AND a weak atheist. Or a positive atheist AND a weak atheist. Sometimes explicit, sometimes implicit. And more...

So, while Larry's definition of an atheist as "someone who doesn't believe in god(s)" is perfectly fine, it misses out on an important distinction for WHY an atheist might choose to disbelieve. That hardly makes Sober's argument useless and it certainly gives you no reason to reject modern philosophy.

Faizal Ali said...

We can reject (be atheistic about) the claim that life on Earth is less than 50,000 years old but should suspend judgement (be agnostic) about the claim that "God is a mysterious force who gives life its meaning".

Why "should" we "suspend judgment" on that? Should we "suspend judgment" on the existence of Bigfoot, as well?

Faizal Ali said...

Or better yet, should we "suspend judgment" on the claim "The Kwidjibo is a mysterious force that makes bacon taste delicious"?

K said...

If a particular definition can never be ascertained, then doesn't that hint that there's a problem with how the definition is framed? In this case, if it can be that evidence in favour of God's nonexistence is something that is impossible, then what's the difference between no evidence in favour of God existence and God's nonexistence?

Alex SL said...

No, his definition of atheism is quite clearly wrong. End of story.

Larry Moran said...

Larry seems to get quite irate because Sober defines atheism differently from what Larry himself would do.

Nope. That's not at all what pisses me off. I'm irate because he believes that his definition is the only correct definition and doesn't acknowledge that fact that many people in his audience might define themselves differently.

... atheism is really a side issue.

LOL!

(You weren't begin serious, were you? Did you forget the title of his talk?)

Larry Moran said...

There will always be some sort of god(s) that resist detection. That means that you can never be an atheist (Sober's definition) about all gods if you accept evidentialism.

Do you really think that Sober is NOT an evidentialist? Why do you think he claims to be an agnostic and not an atheist?

Piotr Gąsiorowski said...

Sober's justifiable atheism would be disbelief based on evidence of absence. It's a straw man, erected only to be knock down for rhetorical effect. Of course it's impossible to demonstrate the non-existence of anything as vague and sloppily defined as god(s) "in general", but who cares? Disbelief based on absence of evidence is a perfectly rational default position. What sort of problem could a philosopher, of all people, have with it?

Piotr Gąsiorowski said...

knock -> knocked. How I miss an "Edit" button!

Piotr Gąsiorowski said...

I'm not a philosopher, thank God, so I can only wonder why Sober treats disbelief (refusal to believe) as a kind of belief (so that he can apply his evidentialist principles to both). Belief requires convincing evidence -- yep, and where does it leave religion? Disbelief is the natural default position. It doesn't have to be specially argued for -- it's sufficiently justified by lack of evidence. Proponents of magic supernatural beings are welcome to falsify it.

Matt G said...

I, for one, did not choose to be an atheist - I just am one. By the same token, I cannot choose to be a theist.

JimV said...

I am also an atheist who defines atheism as lacking belief in any god. That way I can reply to those who say atheism is a religion with the famous saying, "Atheism is a religion in the same way that not collecting stamps is a hobby." Also because it makes sense linguistically to me: asymmetry - lacking symmetry; atheism - lacking theism.

However, I have had creationists get upset about this and tell me my definition is wrong. I think this is because it destroys their favorite argument against atheism (the one Sober makes, according to the description - I haven't watched the video)

I'm not going to watch the video because it sounds like old news. Creationist arguments never change, in my experience. Fallacies, lies, damned lies, and "you can't prove my god doesn't exist!"

Kevin said...

As to the first part, well, we haven't even begun to have a definition of the notion of evidence. So I don't know whether or not you can never be an atheist according to Sober's definition. It's certainly an intuitive conclusion from the evidentialist thesis as Sober presents it, but by no means is it the only possible conclusion. For example, some notions of evidence developed in response to Hempel's confirmation paradox would entail that there is evidence against deist gods.

Are there grounds for believing that Sober is an evidentialist on the basis of this talk? And is there a time in the talk where he states that his conclusion is agnosticism?

Pedro A B Pereira said...

"""That hardly makes Sober's argument useless and it certainly gives you no reason to reject modern philosophy"""

The problem with this line of "modern philosophy" is that, by extension, I could apply the same principle to any supernatural being and claim that I should be agnostic about unicorns, leprechauns, fairies, and all kinds of nonsense. The idea that "god" somehow is a philosophical concept deserving more serious consideration than any of those other supernatural creatures is arbitrary and unjustified. It's assumed and taken as granted simply because we live in the 21st century and people stopped believing in most of those others by now but it's still reputable to believe in god (specially if you want to be a US president).

Plus, I think there's a lot of indirect evidence that there are no gods, in no small part by the continuously shrinking space left by science where he/she/it could still hide. Even supposedly "sophisticated" versions of gods/designers are still no more no less than theistic apologetics. I fail to see why I should take the claim that "God is a mysterious force who gives life its meaning" any more seriously than leprechauns being real. Are you agnostic regarding leprechauns? Should I be agnostic regarding ANY claim for which I have no evidence against it's existence? And what exactly counts as evidence against?

I think Sober should have shut up instead of providing more fodder that will no doubt be used by some to show the purported useless nature of so much of "modern philosophy". And in this specific case, they're pretty much right.

SRM said...

I tend to like the idea "ignostic" here though from what I've read that term itself carries some variable meaning amongst philosophers.

In any case, even those people who firmly claim to believe in god seem not to have a clear idea of what this god is supposed to be and do and they may sometimes even retreat to a deistic position under heavy questioning even though they clearly favour a theistic god day by day.

So I consider myself an ignostic: unless a person can, in a clear and consistent way, explain to me what this god thing is (even in their own mind), it is not really worthy of debate or consideration.

Pedro A B Pereira said...

Some people will point out that the simplistic "absence of evidence is not evidence of absence" argument, though.

Anyway, one of the problems with Sober's argument is that he won't follow his own advice. If a stranger tomorrow tells Sober that his wife has been cheating on him with his best friend for the last 20 years he will probably just assume the stranger is lying for some reason. I doubt he will take a rational "agnostic" position on the matter. He will say that that he knows his wife and friend well, that he never caught her or him lying before, that he "feels" her and his behaviour as inconsistent with that idea, and that the accuser is a stranger. In other terms, he regards the lack of evidence of betrayal as evidence against it. However he will gladly forget about that line of reasoning when it comes to god. Suddenly, the lack of evidence is considered as justifying agnosticism. This is being philosophicaly inconsistent. The reality is that evidence comes in many different forms and not all propositions are equal. Sober is taking a dissapointingly simplistic argument. For a professional philosopher, I'd expect much more than sophomoric arguments.

Pedro A B Pereira said...

A different argument would be if Sober's best friend tells him that he went to the cinema yesterday. Applying Sober's line of reasoning he should be "agnostic" about that claim because he doesn't have any direct evidence for or against. This is obviously idiotic. On the other hand, if he told him that he went to bed with 5 women than Sober would surely think he was joking. Clearly, Sober's simplistic argument doesn't cut it because it fails to account for the fact that different claims have different thresholds for belivability and what constitutes evidence in one case doesn't constitute evidence in the other. Your prior knowledge of your friend's behaviour will influence your judgement. Sober's argument is clearly too simplistic to take seriously.

Hawks said...

Should we "suspend judgment" on the existence of Bigfoot, as well?

Bigfoot seems to be thought of as some sort of humanoid animal. I'm sure we can quite easily conjure up some evidence against that claim. So, the answer would be no.

Or better yet, should we "suspend judgment" on the claim "The Kwidjibo is a mysterious force that makes bacon taste delicious"?

We already have a hypothesis that you have chemical receptors on your tongue and in your nose that detect taste and smell via various molecules in bacon. As far as I know, "Kwidjibo" has no such auxiliary propositions and can't really be scientifically tested in the first place. So, if you reject "Kwidjibo", it is not because you have evidence against it. I.e. your rejection is philosophical rather than scientific.

Faizal Ali said...

What if someone ways Bigfoot is "supernatural" and so is able to hide the evidence of his existence?

So are you agreeing we should "suspend judgment" on the existence of the Kwidjibo? I guess that teaches me: A reductio ad absurdum doesn't work if your opponent has no objection to appearing absurd.

Hawks said...

Larry,

I'm irate because he believes that his definition is the only correct definition ...

You can read his mind?

Did you forget the title of his talk?

IF you accept evidentialism and IF you define atheism the way Sober did, then we should not be atheistic about claims that there is no evidence against. If you define atheism as simple non-belief then evidence doesn't necessarily have anything to do with your conclusion, so I guess you can stop thinking...

Hawks said...

Pedro,

I fail to see why I should take the claim that "God is a mysterious force who gives life its meaning" any more seriously than leprechauns being real. Are you agnostic regarding leprechauns?

I'm not taking either claims seriously either.

Should I be agnostic regarding ANY claim for which I have no evidence against it's existence?

IF you accept evidentialism, you should not reject such claims scientifically. Remain agnostic. Or ignostic (i.e. pfft).

And what exactly counts as evidence against?

If claims have observational consequences, they could potentially be scientifically tested. If not, then go ahead and laugh all you want. Just don't claim that you have scientific evidence against the claim.

Hawks said...

What if someone ways Bigfoot is "supernatural" and so is able to hide the evidence of his existence?

So are you agreeing we should "suspend judgment" on the existence of the Kwidjibo? I guess that teaches me: A reductio ad absurdum doesn't work if your opponent has no objection to appearing absurd.


Just don't claim that that you have scientific evidence against Kwidjibo. But rather that your rejection is philosophical. Or you could start arguing against evidentialism.

Larry Moran said...

Hawk asks,

You can read his mind?

Yes, in this case I think I can. There's nothing in his talk to suggest that he takes into account another definition of atheism.

Piotr Gąsiorowski said...

Hawks: IF you accept evidentialism and IF you define atheism the way Sober did,...

If evidentialism requires me to provide evidential justification not only for belief, but also for disbelief and/or withholding of belief, I don't accept it. I don't believe in telepathy because all attempts to demonstrate its existence have failed so far. I might change my mind if faced with incontrovertible evidence, so my disbelief is tentative, but absence of evidence is all that's needed to justify it.

...then we should not be atheistic about claims that there is no evidence against.

I'm damned if I understand what "being atheistic about a claim" means.

anthrosciguy said...

Well, he's obviously of the belief that he (or Sober) can just make up whatever word usage he wishes and you have to accept it. So statements like "I'm atheistic about lunch" or "kids, stop bring atheistic about bedtime and go to sleep" can be presented and you should just stop being so atheistic and accept them as not being gobbledegook.

Philosoblogger said...

I am with Kevin and Hawks; many of the comments here seem uncharitable to Sober and his goals. As for some issues raised (1) I recall Sober saying in Evidence and Evolution that evidentialism is a rather controversial thesis among philosophers, (2) since substantive debates rarely if ever hang on definitions, who cares? (I mean as long as we know how Sober is using the times, which aren't that uncommon by the way, then why not just evaluate his arguments?). (3), When we do that, can't we be a bit nicer along the way? Comments that start with 'thank God I am not a philosopher', do not exactly make for healthy dialogue between philosophers and scientists.

Hawks said...

Piotr,

If evidentialism requires me to provide evidential justification not only for belief, but also for disbelief and/or withholding of belief, I don't accept it.

Evidentialism sure allows you not to believe in something if there is no evidence (you just remain agnostic). Personally, I sometimes reject claims when there in no evidence against them - I just realize that my rejection is philosophical rather than scientific.

I don't believe in telepathy because all attempts to demonstrate its existence have failed so far.

Larry seems to be able to read minds. Is that evidence for telepathy? The level of evidence for telepathy seems to be on par for the evidence that Sober "believes that his definition is the only correct definition".


I'm damned if I understand what "being atheistic about a claim" means.

Rejecting.

anthrosciguy,

Well, he's obviously of the belief that he (or Sober) can just make up whatever word usage he wishes and you have to accept it. So statements like "I'm atheistic about lunch" or "kids, stop bring atheistic about bedtime and go to sleep" can be presented and you should just stop being so atheistic and accept them as not being gobbledegook.

No.

Larry Moran said...

Look, I would have had no problem with Sober's talk if he had started by asking how many in his audience were atheists then asked "what kind of atheist?"

He could then have said that for all of you who think that an atheist is someone who believes that god(s) don't exist I'm about to show you why you should be an agnostic and not an atheist.

Then he could have said that the other atheists can go home because these questions are not for those of you who think that an atheist is just someone who doesn't believe in god(s).

Joe Felsenstein said...

I wonder whether Elliottt is atheist or agnostic with respect to the Tooth Fairy. I positively don't believe in the Toot Fairy. I suspect Elliott draws the same conclusion. But then I wonder whether that is consistent with his argument.

Piotr Gąsiorowski said...

Personally, I sometimes reject claims when there in no evidence against them - I just realize that my rejection is philosophical rather than scientific.

"Rejection", like "belief" is never absolute in science, as I understand it. We believe or reject propositions based on the evidence we have (or lack thereof), and will gladly change our attitudes if new evidence comes to light. I prefer to think in terms of sound methodology and the burden of proof, which is normally on the "believer", and not the "disbeliever". I don't care for the philosophical idea than not only "beliefs" but all "doxastic attitudes" should be justified by positive evidence.

Hawks said...

This thread is probably not read anymore, but I missed this part of the conversation first time around and felt the need to comment.

Piotr,

Disbelief based on absence of evidence is a perfectly rational default position. What sort of problem could a philosopher, of all people, have with it?

Pedro,

Some people will point out that the simplistic "absence of evidence is not evidence of absence" argument, though.

Yes, "absence of evidence is not evidence of absence" is simplistic, as is the notion that disbelief based on the absense of evidence is necessarily rational (here I assume that by disbelief Piotr mean "atheism' rather than "agnosticism". Otherwise I fail to see why Sober would have a problem with it). One important factor disregarded is the probability that I should possess any evidence in the first place. Sober explains this in his paper "Absence of Evidence and Evidence of Absence: Evidential Transitivity in connection with Fossils, Fishing, Fine-Tuning, and Firing Squads". Might I suggest that you guys read it?

Pedro then proceeds in doing some telepathy. I.e. Sober would do this and that. Pedro, of course, has no evidence for his claims.

Sober's argument is clearly too simplistic to take seriously.

It was a 15 minute talk. But one thing is for certain: Your analysis of Sober's argument was too simplistic.

Philosoblogger said...

I suspect that (for some at least) the real issue is not whether there are other ways of defining some theoretical term X (in this case theism). If that were the issue, we could point out that in almost all 15 minute talks on epistemic matters, speakers define things in ways that others might dispute. So what else might be going on? I suspect that at least some people want to protect the term ‘atheism’ itself. Rhetorically, atheism sounds stronger than agnosticism – the latter somehow seems more open to religion. If someone says ‘you atheist ought to really be an agnostic’ it sounds like they are somehow implying ‘you ought to be more open to religion somehow.’ Yet, if atheism is just defined as a lack of belief in God it isn’t stronger than common ways of using the term agnostic.

My thought, then, is this: at least some atheists might like to take an epistemically weaker position (a lack a lack of belief in God instead of full out disbelief) for philosophical reasons) but to keep the term atheism (for political reasons). In other words, Dennet's category of belief in belief might apply here. Maybe I am wrong, but I can’t understand some of the heat on this topic otherwise.

El Schwalmo said...

From my point of view 'atheism' is a pragmatic position, 'agnosticism' an epistemic. You can live as an atheist, even if you cannot be sure if there is a god or not (I don't know any atheist who says he can prove that there is no god, because the 'god' in these definitions is the 'god of philosophers'). Agnosticism isn't fence sitting, you have to decide, how to live. Agnosticism goes with atheism (e.g. me) oder theism (e.g. 'negative theology').

That may be the reason, why e.g. Russell called himself 'agnostic'

Russell, B. (1947) 'Am I An Atheist Or An Agnostic? A Plea For Tolerance In The Face Of New Dogmas' URL: http://www.positiveatheism.org/hist/russell8.htm

Even Dawkins calls himself an agnostic in interviews in german journals, as far as I know exactly for the same reason as Russell did. I guess that

Le Poidevin, R. (2010) 'Agnosticism. A Very Short Introduction' Oxford, Oxford Univ. Press

shows very clear what's the difference between 'agnosticism' and 'atheism' from the viewpoint of a philosopher.

John Harshman said...

You should know that the Tooth Fairy and the Toot Fairy are two quite different entities. The former pulls teeth from under your pillow while the latter just wants you to pull his finger.

Allen MacNeill said...

I wondered when someone would bring up Russell. It might also be interesting to consider the fact that the inventor of the term "agnostic" was Thomas Henry Huxley, who defined it in a manner essentially identical to Russell. Sober also uses Huxley and Russell's definition of "agnostic," as does Richard Dawkins, but only when speaking to Germans. To me, this argues strongly for primacy for the term "agnostic" for what most people who call themselves "atheists." Indeed, it seems to me that there is no fundamental difference between "agnostic" and "ignostic," especially with regards to evidentialism. Oddly enough, it seems like philosophers might actually have a better handle on what they are talking about than people with little or no formal training in such subjects.

Larry Moran said...

Oddly enough, it seems like philosophers might actually have a better handle on what they are talking about than people with little or no formal training in such subjects.

Oddly enough, I disagree.

But it doesn't really matter whose definition of "atheist" is correct as long as you are able to recognize that there ARE different definitions. If a philosopher is so blind to reality that he/she doesn't even realize that there's another reasonable definition then they don't know what they are talking about.

If you insist on defining "atheist" as someone who is absolutely convinced that there is no god and is willing to defend that position in a philosophy class, then you are talking about someone who doesn't exist. There are hardly any nonbelievers who fit this definition. The word "atheist" becomes meaningless.

I wonder how many philosophers realize this?

I have a friend who is a Jesuit priest but he says he is an agnostic because he cannot prove there is a god beyond a shadow of a doubt. He's a theistic agnostic. Allen, do you have a word for am agnostic who doesn't believe in gods? I do, I'm an atheist agnostic.

Unknown said...


This debate seems to me somewhar fuzzy.Before the question of what an athëist is, there is the question of what "God" means. Otherwise the debate remains void... Without a preconceived notion of God, to say that He does not exist remains mere syntax. It has no semantic meaning. It says no more than: "There is no X".In short: an atheist just cannot believe in the God he imagines (just as a believer needs an idea of Him...).

As a mathematician I think that logical coherence ("plausibility") more than "evidentialism" characterizes rational thought. On the basis of pure evidentialism, much of mathematics loses its raison d'être. There is no "evidence", for instance, that such strange things as Lie groups or Hilbert spaces exist. But almost nothing of modern physics is understood without these non-obvious things.


For those who believe in God, the "Logos" (Jn 1.1), it is an illogicism that reason should come from dead matter. As Rémi Brague states [in his book "Modérément moderne"]: "Those who claim to be 'rationalists' today are not really so. For them, the reason is something secondary. The reason comes from the irrational."


And that, infact, is not very convincing.