Flew: The Presumption of Atheism

Flew's article, "The Presumption of Atheism", attempts to state how we should conduct debates about God. As I stressed in class, Flew is not arguing for atheism in this article. Instead, he is trying to lay down some "ground rules" for how religious debates should proceed.

The Presumption of Atheism

Flew claims that debates about God should begin from a “presumption of atheism” (which I'll abbreviate as the PoA.)
As he puts it, the PoA amounts to the claim that:

“The onus of proof must lie upon the theist”

As Flew explains, the PoA amounts to the claim that it is up to the theist to:

(i) Introduce and defend her proposed concept of  God
(ii) Provide sufficient reason for believing that  this concept does in fact have an application.

"Theism" and "Atheism"

Flew realizes that this statement of the PoA might be misleading, and he spends a good deal of the paper attempting to make it more clear. The first point of clarification involves the way he intends the word "atheism" to be understood. Flew defines his terms as follows:

“Theist”
• Someone who positively asserts that God exists.


“Negative Atheist”

• Someone who is not a theist.
• Someone who doesn’t assert that God exists.


“Postive Atheist”

• Someone who positively asserts that God doesn’t exist.

The important distinction here is between the negative and positive atheist. When Flew claims that debate should begin from a presumption of "atheism", he only means that they should begin from a presumption of "negative atheism". (He specifically does not mean to claim that we should accept a presumption of "positive atheism".)

The Presumption of Innocence

Flew attempts to further clarify the PoA by comparing it to the legal presumption of innocence.

Presumption of Innocence:
• All defendants are presumed innocent until proven guilty.
• The onus of proof must lie upon the prosecution.

Flew is thus drawing an analogy between the PoA and the legal presumption of innocence. The main point of this analogy is Flew's claim that both the PoA and the legal presumption of innocence are procedural. That is, both presumptions specify how to proceed. (In the case of the PoA, it is intended to tell us how to proceed in debates over God, and in the case of the presumption of innocence it tells how to proceed in a legal trial.)

Flew then uses the analogy to make four points:

(i) In both cases, the required “proof” encompasses “any and every variety of sufficient reason”.
(ii) Both presumptions are defeasible.
(iii) Just because the presumptions are procedural, this doesn’t mean that it is a trivial matter whether to adopt them.
(iv) Defeats of either presumption in particular cases do not show that the general presumption was a bad one.

The Case for the PoA

Everything that Flew has said to this point has merely been an attempt to spell out what he means by the Presumption of Atheism (PoA). He has not yet given us any reason for thinking that we should, in fact, accept this presumption when debating about God. In the final section of his paper, he attempts to give a case for the PoA.

As I pointed out in class, Flew's argument is a pretty difficult one to understand. His discussion is very brief and demands a lot from the reader. In what follows, I have attempted to highlight what I take to be the central ideas in this argument. (I have not, as you will see, attempted to fully re-construct his argument in premise-conclusion form. Instead, I encourage you all to attempt such a re-construction.)

As Flew structures his argument, the case for the PoA comes in the form of  two "moves". (An interesting and important issue for understanding Flew's argument is to see how these two moves are related. Does the second move build on the first, or are they two entirely seperate ways that one might try to argue in favor of the PoA?)

(i) First Move:
As an "obvious" first move, Flew discusses the attempt to defend the PoA by appeal to the legal axiom: "Ei incumbit probatio qui dicit, non qui negat." Flew gives a couple of different translations of this axiom, in an attempt to find a version of it that will lend support to the PoA. However, his comments a fairly negative. He finds repeatedly that no construal of this axiom will, by itself, show that we should accept the PoA.

Given the negative conclusions Flew reaches, we might wonder what his point is in describing this "first move". As I read him, his reflections on the "first move" yields the following, very modest, conclusion:

 
Acceptance of these presumptions (the PoA, or the presumption of innocence) amounts to the adoption of a policy.
I think this conclusion then sets Flew up for his "second move".

(ii) Second Move:
Now that Flew has shown (or thinks he has shown) that to accept the PoA would be to adopt some policy, he is in a position to argue that this policy is the best one to adopt. But how does one argue in favor of the adoption of a policy? According to Flew:

 
To show that we should adopt some policy, we need to show that the policy furthers some aims or goals that we have.
And this seems right. In the more familiar case of the presumption of innocence, our legal system has adopted this policy because it furthers our goal of convicting as few innocent people as possible. (We are more worried about sending innocent people to jail than of letting guilty people go free, so we adopt the presumption of innocence. If our goals were different, and we wanted to ensure that no guilty people went free, we would've adopted a presumption of guilt.)

So, to find out whether we should accept the PoA, we need to examine what our aims and goals are. Flew claims that, in the case of theistic debates:

 
Our aim is to achieve knowledge (or reasonable belief) about whether God exists.
That's why we're debating after all. If we didn't care whether our beliefs were reasonable or knowledgeable, then why have an intellectual discussion about them?

Flew then states what he takes to be a fairly trivial condition of knowledge/reasonable belief:

 
To have a knowledgeable or reasonable belief about something, we must have sufficient reason for believing it.
And now Flew is ready to conclude his case in favor of the PoA. We desire knowledge, which requires sufficient reason. To best ensure that we arrive at knowledge when debating about God, we should accept the PoA, which places the burden of proof on the theist. To put it a slightly different way:
 
We should adopt the PoA to ensure that we never arrive at an unreasonable belief that God exists.