Analyzing The Outsider Test For Faith (Part 3)

As we saw in Parts 1 & 2, the Outsider Test For Faith is an objection to Christian belief that says (a) a stance of skepticism toward any particular religion is the only warranted position unless (b) the belief in question passes “intellectual muster.” Atheist freethinker John Loftus provides six considerations that, in his mind, warrant a position of skepticism toward any religion, including Christianity.

Let’s review them once again:

The amount of skepticism warranted depends on [1] the number of rational people who disagree, [2] whether the people who disagree are separated into distinct geographical locations, [3] the nature of those beliefs, [4] how they originated, [5] how they were personally adopted in the first place, and [6] the kinds of evidence that can possibly be used to decide between them. My claim is that when it comes to religious beliefs a high degree of skepticism is warranted because of these factors.

In Part 1 of this series we saw that his first skeptical consideration leads to skepticism only if Equal Weight Theory (EWT) is true, but, I argued, EWT is ultimately self-defeating. In Part 2 we saw that his second consideration is also doomed for failure. The fact that I probably wouldn’t believe in Christianity if I were born in Saudi Arabia doesn’t mean that I am unjustified in my belief. As a result of what I’ve argued in part 1 and 2, neither of Loftus’ first two considerations warrant skepticism.

In part 3 of the series, I will argue that the remaining considerations ([3]-[6] above) warrant skepticism only if it is known that Christianity is false. First, let me briefly lay out what is known as Reformed Epistemology.

Reformed Epistemology

In his work, Alvin Plantinga distinguishes between de jure and de facto objections to Christian belief. De facto objections are aimed at the existence of God. They aim to show that Christianity is false. De jure objections, on the other hand, are aimed at believers and their beliefs. These objections seek to show that Christian belief is unreasonable, unjustified, or unwarranted.

Here is an example of a de jure objection to Christian belief:

(1) It is rational to believe in Christianity only on the basis of a good argument.
(2) There is no good argument for Christianity.
(3) Therefore, it is not rational to believe in Christianity.

Plantinga responds to this objection by challenging premise (1). He argues that (1) is false because belief in Christianity can be “properly basic.” A belief is basic when it is not held on the basis of any argument. However, not every basic belief is properly basic. Properly basic beliefs are basic beliefs that are rational for a person to hold. No one is rational in holding the basic belief that your next gamble at the casino will make you a millionaire. On the other hand, when I jam my finger in a car door and believe that I feel pain, this belief is both basic and properly basic.

Here’s where it gets interesting. In his work, Plantinga argues that de jure objections reduce to de facto objections. This is because if Christianity is true, then it’s likely that Christian belief is properly basic, and second, if Christianity is false, then it’s likely that Christian belief is not properly basic. If this is right, then the answer to whether Christian belief is properly basic will depend on whether Christianity is true or false. In other words, the epistemological question is not separate from the metaphysical question.

With this in mind, let’s list each of the remaining skeptical considerations and see if they warrant skepticism.

The Nature of Christian Beliefs

His third skeptical consideration says that the nature of Christian beliefs warrants skepticism. Now, it’s not immediately obvious what Loftus means by “the nature of [Christian] beliefs.” What precisely about Christian beliefs warrants skepticism? Part of me thinks he sort of ran out of things to say and was merely summarizing his first two points. Another part thinks that maybe he’s trying to say that Christian belief is really complex or affirms things he finds implausible (e.g.: belief in God) and so Christian belief ought to be met with skepticism. Let’s take this second interpretation.

Suppose that for whatever reason Christian belief is super unlikely. Let’s assume this is true for the sake of argument. Does it follow that Christian belief is unjustified? Well, if Reformed Epistemology is true, then no, that doesn’t necessarily follow.

Here’s an analogy. The world around us is highly complex. Scientists have estimated that there are roughly 10^80 fundamental particles in our universe. These particles all obey a variety of physical laws, act in certain ways, occupy different regions of spacetime, and so on. Does this mean that, unless we have really good arguments and evidence, we can’t be justified in believing that the universe is real? Ask yourself, do you have good arguments and evidence that you aren’t in the Matrix or a brain sitting in a vat of chemicals?

Here’s the point: If Christianity is true, then God plays a causal role in the formation of Christian beliefs. However, Christianity is only irrational or unjustified if God doesn’t play a causal role in the process. In other words, Christianity is irrational only if Christianity is false. As with belief in the external world, argumentation isn’t always necessary [1].

Belief Origins

Loftus’ fourth and fifth considerations have to do with belief origins. The question of origins–at least with respect to religious belief–falls under the jurisdiction of Cognitive Science of Religion (CSR). It’s undeniable that religious belief is widespread, even natural. Most cultures throughout the centuries have believed in God or gods. It’s safe to say that religious belief is both natural (biologically) and normal (statistically). But why is this?

The standard view in CSR is that religious belief is a byproduct [2] of a cognitive mechanism known as the Hyperactive Agent Detection Device (HADD). When you’re home alone at night and you hear a sound, chances are you’ve ascribed agency to that sound. Is there a burglar about to get me? Maybe that sound was a ghost? This is the HADD at work. Evolutionarily, we can see why HADD would be an adaptive trait to have (e.g.: it would aid in avoiding predators). The standard view in CSR is that religious belief originates, at least in part, from an overactive agency detection device; we just sort of naturally attribute agency to everything, including the universe.

A number of atheists (e.g.: Dawkins, Dennett, and co.) take this information and say that religious belief is therefore irrational or unjustified. But this is a mistake. On Christianity, Christian belief is not purely the result of a cognitive mechanism (even if it is in part). As mentioned previously, God plays a causal role in the production of Christian belief (see here for more). Without already knowing that Christianity is false, we don’t know that Christian belief is irrational or unjustified.


The sixth and final consideration from Loftus has to do with the kinds of evidence that can be used to decide between different faith traditions. Loftus seems to suggest that without propositional evidence, we have no way of “deciding” which religion is correct. Notice that this is basically just the de jure objection I’ve given above in the section on Reformed Epistemology. Here’s the argument once again:

(1) It is rational to believe in Christianity only on the basis of a good argument.
(2) There is no good argument for Christianity.
(3) Therefore, it is not rational to believe in Christianity.

Since I’ve already given Plantinga’s response to this argument, I’ll briefly summarize another approach. The Christian could deny premise two. She could argue that Christianity in particular has very good arguments in support of it. In defense of this claim, she might appeal to a cumulative case starting with cosmological arguments all the way to Bayesian-type arguments for Jesus’ Resurrection. I’ve argued for both of these in my own work (see the links).

So it seems to me that both premises of this argument are dubious; there’s no good reason to accept either one.

Final Thoughts

We’ve come to the end of my analysis of The Outsider Test For Faith. It’s apparent that none of Loftus’ considerations actually warrant skepticism. Epistemic disagreement doesn’t, geographical location doesn’t, belief origins doesn’t, the evidence doesn’t. Maybe Loftus has good reason to be skeptical about Christianity, but those reasons are not included in his infamous Outsider Test For Faith.


[1] For a more robust defense of this, see my 4-part series.

[2] The standard view in CSR is that religious belief is a byproduct of evolution. Religious belief isn’t itself adaptive, but it was formed as a side effect of traits that were adaptive. Here’s a quick illustration. My fingers are adaptive. I can pick up and grab things with my fingers, I can perform complex operations (like fashioning tools), and so on. Now turn your hand over and look at the wrinkles on the back of your knuckles. Those wrinkles exist as a byproduct of your fingers. They aren’t themselves adaptive. Religious beliefs are like the wrinkles on the back of your hand.

[3] I’m using the term atheism in the traditional sense. An atheist believes that the proposition “God does not exist” is true.

About the Featured Image

When one thinks of Jerusalem (or even Israel), this is the picture that comes to mind. Shortly prior to taking this, Brittany, my wife, had the time of her life riding a camel for the first time (that was probably her sole reason for wanting to go to Israel so bad). I want to go back.



Get updates on new posts, upcoming live discussions, and more.


  1. I think both (1) and (2) are false. (2) definitely so. It says a lot about the psychology of atheists that just because *they* are not convinced that they jump straight to accusing others of being irrational. That’s not rational argument, that’s straight out social engineering.

    We’ve lost the art of disagreeing: That you still have respect for people even though you have come to fundamentally different conclusions.

  2. How that heck does God make the HADD reliable? How can God do that?!! That’s like saying God can make it that the case that drinking gasoline is healthy.

Leave a Reply

Your email address will not be published. Required fields are marked *