Atheism versus theism: who bears the burden of proof?

In my last post I pointed out that atheism becomes very contentious when it is wedded to the naturalistic worldview. As a way into the discussion I summarized naturalism as follows:

Naturalism: "the view that the only things that exist are matter and energy and that which supervenes upon matter and energy."

Let's call this the "naturalism thesis" or "NT". I then argued that NT lacks justification: the atheist has no reason to believe it true.

With that background let's engage a couple of responses to this argument. First there is Ethan who decries my "weak argument" with the protest:

First of all, why should the naturalist have to prove that God, souls, angels etc. don't exist? What about leprechauns, the Loch Ness monster, etc, etc? Where does it end? Surely you're familiar with the concept of burden of proof.

I take Ethan to be offering a reductio ad absurdum here. In other words, he is claiming that if we insist that the naturalist must provide an argument for naturalism, then we must have evidence for the non-existence of Keebler elves as well. And surely this is absurd. To drive the point home, Ethan then chides me as being unfamiliar with the "burden of proof".

The burden of proof gets tossed around quite a bit in these debates, but rarely do people provide a systematic articulation of the nature and proper application of burden of proof. What is often missing from these discussions (and alas from Ethan's) is the appreciation that there are many burdens of proof here.

At the center of the discussion is NT. There are three basic epistemic stances toward NT:

(1) Affirming that NT is true

(2) Affirming thatn NT is false

(3) Refraining from affirming (1) or (2)

Who bears the burden of proof? The person who affirms (1), (2), or (3)?

As I address this question I know some eyes will glaze over because some people want a quick and dirty answer. They think every issue can be addressed in a thirty second sound bite. I would advise those folks not to read further. The rest of you are invited to follow the discussion.

Two Types of Burden of Proof

Let's begin by noting that there are two minimal senses in which one might bear a burden of proof. First, one might bear it in terms of oneself meaning that without evidence for a proposition one is fundamentally irrational or unjustified in accepting the proposition. Let's call this (rather inelegantly) a burden of proof qua oneself. Second, one could bear a burden of proof in terms of persuading somebody else to accept a given proposition. Let's call this a burden of proof qua another.

One may bear a burden in one of these instances but not another. For instance, let's say that I am charged with murder but I know I didn't do it. I don't bear a burden of proof for myself to believe "Randal is not guilty" but I do bear a burden of proof qua others: I must find some way to convince the jury.

One could also face a burden of proof to accept a proposition oneself and for anybody else to accept it as well.

With that in mind I will proceed by focusing primarily on the burden of proof qua another, while keeping in mind the question of burden of proof qua oneself.

Naturalism bears a Burden of Affirmation

When it comes to burden of proof the focus is typically on the burden of affirmation, and in Ethan's case on that burden as borne by the supernaturalist. In short, if a person believes that there is a supernatural realm then they have a burden of proof qua others to defend that belief.

I accept that the theist bears a burden of proof qua another. But as the last post made clear, the naturalist, that is the atheist who affirms NT, is also making a positive claim about reality and thus likewise bears a burden of proof.

Let us be clear on this point: both the supernaturalistic theist and the naturalistic atheist bear a burden of proof qua the other.

(My last post also made clear that the naturalist has not even begun to meet the burden of proof qua himself/herself. That is, he or she has no personal reason to believe NT.)

Naturalism bears a Burden of Denial

When we affirm one proposition, we are logically denying others. Thus the theist who affirms a supernatural theist actually has a dual burden of proof qua the other: argue for supernaturalism and against naturalism. Likewise, the naturalist has a dual burden of proof: argue for naturalism and against supernaturalism.

What about agnosticism? Is there a burden here too?

Many atheists who are aware of these burdens find it appealing to retreat from claims to knowledge altogether. Thus, you find atheists who now define atheism in the way that agnosticism has been historically defined. And this brings me to Beetle496 who responded in the last thread to my definition of atheism as the view that there is no God: 

Atheism is quite well understood to merely be lack of belief in a personal god. This fact has been spelled out to you in no uncertain terms, and you stopped making this particularly blatantly false assertion for long enough now that it was reasonable for me to assume that the point had gotten through to you. Are you hopeless in this regard, or merely dishonest? (emphasis added)

Ouch!

Fortunately I need not concede a personal character defect. Instead, let me quote from the Oxford Companion to Philosophy (a standard reference work in the field), specifically from the entry "atheism and agnosticism":

"Atheism is ostensibly the doctrine that there is no God. Some atheists support this claim by arguments. But these arguments are usually directed against the Christian concept of God, and are largely irrelevant to other possible gods. Thus much Western atheism may be better understood as the doctrine that the Christian God does not exist.

"Agsnoticism may be strictly personal and confessional--'I have no firm belief about God'--or it may be the more ambitious claim that no one ought to have a positive belief for or against the divine existence."

Now the fact that Beetle496 and some of his friends choose to define atheism the way that philosophers have traditionally defined agnosticism may be confusing, but it is a free country. He can do that if he likes.

If Beetle496 merely holds his agnosticism/atheism in a weak sense, namely that he does not know if there is a God, then that is a mere confession of personal ignorance. If on the other hand, he wishes to argue that nobody can know, well now we're again in the realm of claims to knowledge and evidential burdens qua another: how does he know that nobody can know?

Same thing with naturalism. The person who afirms (3) might merely be in a state of personal ignorance. But the person who claims that we can never know whether naturalism or supernaturalism might be true minimally faces an evidential burden qua another just like the person who affirms (1) and (2).