Two Types of Atheism

• atheism from burden-of-proof skepticism
• atheism from natural worldview (naturalism that entails atheism)

20th century atheism was almost entirely the former.

The 20th century champion of atheism from skepticism was Antony Flew. The position he advocated consisted of two questionable assertions. (1) that the burden of proof is on those who say there is a God, and (2) that theists had failed to meet that burden.

Shortly before he died, Flew changed his mind about number 2 and decided that theists had indeed met the burden—at least for a God like Aristotle’s. Prior to this, however, Flew was probably the most prominent atheist arguing from skepticism. His approach was flawed from the beginning, but is pretty representative of a lot of “thoughtful” atheism you will run into today. (I put thoughtful in quotes because although it may be thoughtful, it is not sound. As I will explain.)

So why is Flew’s “burden of proof” assertion unsound?

Let’s start by presenting his argument. He takes the concept of burden of proof from the legal system. When someone is accused of a crime, they are presumed innocent until sufficient evidence is presented by the prosecution to show that they are guilty as accused. In other words, the burden of proof is on the prosecution, not the defense.

And there is a sound reason for this: presenting sufficient evidence that you did not commit a specific crime is often difficult if not impossible. Perhaps you have a strong alibi—disinterested witnesses who can testify that you were elsewhere at the time of the crime, for example. But if you have a good alibi like this, the police probably already weeded you out as a suspect. Police are almost always going to arrest suspects who don’t have a strong alibi and who can be imagined as having motive.

But does not having an alibi make a suspect guilty? Would anyone consider it reasonable to assert that whoever can’t provide proof that they were elsewhere Friday at 3 pm when the bank was robbed, must therefore be the robber? Of course not. We require evidence that links suspects to the crimes we suspect them of. Otherwise we will jail a lot of innocent people.

So the burden of proof rightly falls on the prosecution, not the defense, in a criminal trial.

Flew argued that a similar burden of proof applies to propositions about God. In his analogy, asserting that God exists is like asserting that a crime occurred and that the burden of proof lies on the person making the assertion.

It is unreasonable, he goes on, to expect the other side to have to prove a universal negative. If you claim that a pink unicorn exists, it’s got to be up to you to provide evidence for that assertion. It’s not on me to survey the entire world in order to prove you are wrong (and allow you to say, nice try, but you overlooked the unicorn because it didn’t want to be seen).

Therefore Flew, and many atheists today, assert that the burden of proof must be on whoever asserts that something exists. Claim that God exists? Then it’s on you to prove it.

So why is this mistaken?

I think the best way to demonstrate the problem with Flew’s burden of proof argument is to point out that this is not actually the way burden of proof works in the sciences. In the sciences, burden of proof rests on whoever is contesting the scientific consensus. When flat space was the consensus, the burden was on Einstein and others to cast doubt on the consensus and to provide a more convincing alternative.

If he had looked to science instead of using the legal system for his burden of proof analogy, Flew would have more reasonably placed the burden of proof on whoever kicks against the consensus. And it makes practical sense: if most people disagree with you and you want to change their minds, it’s on you to present the evidence and arguments that will change their minds.

As long as the context of the conversation about God is an intellectual discussion and not a criminal courtroom, the burden of proof naturally falls on those with the minority position.

(Though I would argue that intellectual honesty means that it falls on all of us regardless of the popularity of our positions.)

Flew’s 2nd assertion—that theists had not met the burden of proof—was wrong-headed as well. Of course we shouldn’t forget that Flew changed his mind at the end, yet that was equally wrong-headed for much the same reason.

What is wrong-headed about it is that is pretends that there is only one side to the story (a proposition) which is either sufficiently proved or supported by evidence, or not. Flew thought the only proposition in question was “There is a God” and that either its proponents presented sufficient proof—or they did not.

Again, the sciences don’t work this way: in reality there are always competing propositions, and evidence which supports one proposition may be reinterpreted to equally support a different proposition. If you pretend that there is only one proposition in question, and good evidence is presented for it, you may entirely miss the alternative that is also consistent in its own way with that same evidence.

There is also a more obvious mistake that arises, in this instance, from atheism based on burden of proof skepticism. This is the assumption that if logical arguments for God’s existence are unsuccessful, the burden of proof is not met and therefore we should not assent to believing in God. But this is specious. Logical arguments for God can never be successful precisely because they are logical arguments.

God’s existence is a factual question, and logical constructions cannot adjudicate factual claims. It’s as simple as that.

The relationship between minds and the world minds exist within is necessarily empirical. There are two types of knowing—rational and empirical—due to the reality that minds evolved later. Minds produce logical, rule-based, coherent understandings (rational knowledge), but have no way to determine the usefulness (truth) of what they know except via interaction with the surrounding world.

Constructions of knowledge therefore must be selected based on their empirical usefulness within the world. And importantly, they are tested not against the world (there is no singular way to do that) but against other possible constructions of knowledge.

When Karl Popper says that knowledge can’t be verified, only falsified, this is what he means (or at least what he should have meant). If not-A (or B, C , D or other subset of not-A) is more useful within the world than A is, then A is falsified, and not-A (or the specific subset that was more useful) must get our focus.

Human knowledge constructions are infinite—we are capable of thinking of anything minds like ours can construct—because the world we exist within is fundamentally a black box, and what is most useful could turn out to be anything. This is why our knowledge capacity evolved to be so flexible. It has to be, for our survival.

Or rather, species with inadequate mental flexibility won’t reproduce as successfully, and will be replaced by natural competitors.

Why are we the only surviving hominid species? This might be the reason: we were more flexible empiricists. We could create a greater variety of rational explanations than other hominids, and reject less useful explanations more readily than others hominids could. This, perhaps, was our superpower.

This superpower developed in us (I believe) thanks to synesthesia enabling extensive language development. This would have started with synesthesia-style associations of unique visual objects with unique human utterances—in short, with the innovation of uttered names for everything important we saw or interacted with in our world. The brain’s capacity for synesthesia fused the name with the visual object, and thus the underlying basis for extensive language and knowledge-building was set in motion.

But we needed to be able to flexibly change our opinions as we interfaced with the changing world, or as we migrated into new parts of the world. Homo Sapiens was better at this than homo Erectus, and probably better than homo Neanderthalensis. And yet we’ve never been perfectly efficient at changing our beliefs as we interact with the world, and the persistence of outdated religions are evidence of that.

Which brings us back to topic (before this side-excursion). Logical constructions can never settle factual claims about the world.

Example: All valid logical arguments for God contain premises that belong to a supernatural worldview and that do not belong to a natural worldview, so such arguments always presume what they are meant to prove.* Thus their inevitable failure.

But this means that the inevitable failure of Aquinas’ Five Ways is not grounds for skepticism. Logical arguments can never prove a factual claim anyway, so failure of any and all logical arguments for God is beside the point.

For theists and atheists this is both good news and bad news.

The bad news is that most of the “thoughtful” arguments about God are a waste of time, but the good news is that this colossal waste of time is an important teachable moment which can redirect us to the real issues and factual evidence pertaining to the question of God’s existence.

What we really ought to be doing is comparing natural to supernatural worldviews so we can determine which is a better fit with the world around and within us. Which approach best fits the evidence we have?

This makes for a substantially different conversation than the ones atheists and theists have had in the past.

But there is a barrier preventing us from having this new conversation. Almost nobody understands what a natural worldview entails. We are experts on supernaturalism, familiar with it from every angle, but of naturalism we understand almost nothing. This makes a judicious comparison impossible.

So naturalism and its underlying premises need exposition.

Which is the purpose of Preface to Atheism.