An epistemic objection to religious belief is an objection that denies that religious belief is rational, justified, warranted, or can constitute knowledge. In short, religious belief is intellectually sub-par, cognitively third-rate, or epistemically challenged. There are other non-epistemic objections to religious belief (e.g., that it leads to immoral behavior, makes people intolerant or arrogant, or is psychologically unhealthy). We will not be considering these sorts of objections.

1. Evidentialism and the Evidentialist Objection

Evidentialism is the view that rational beliefs or knowledge require the possession of evidence of some sort. Applied to religious belief, evidentialism is the view that religious beliefs are rational (or constitute knowledge) only if a person has adequate evidence in support of those beliefs. "Evidence" in this context refers to propositional evidence, i.e., other rational beliefs or known propositions. Evidentialism originated in early modern philosophy under the influence of philosophers such as John Locke. It became an integral part of the Enlightenment and post-Enlightenment critique of religious belief. The evidentialist objection to religious belief is the view that religious belief fails to be rational or justified because there is not sufficient evidence for such beliefs.

The evidentialist argument against religious belief can be stated as follows:

[E1] Religious belief is rational only if it is based on other rational beliefs that provide adequate evidential support it.

[E2] There are no rational beliefs that provide adequate evidential support for religious belief.

From which it follows that

[E3] Religious belief is not rational.

Not all evidentialists accept [E2]. Many of them argue that [E2] is false. It is important to distinguish the evidentialist challenge and the evidentialist objection to religious belief. The challenge is [E1], the demand that one's religious beliefs have evidence. The objection is [E3], which is based, not merely on [E1], but the conjunction of [E1] and [E2].

It may be argued that [E2] is false, a matter to be investigated in the latter part of the course. Granted, it is not plausible to suppose that there is no conclusive evidence for God's existence, but there is hardly conclusive evidence for any of our beliefs. So that can't be the right sort of standard for adequate evidence. There are potentially plenty of rational starting points that provide support for God's existence: the existence of the Universe (cosmological argument), the orderly nature of the Universe and its fine-tuning (design argument), the existence of morality (moral argument), religious experience and miracles. Even if none of these arguments by itself provides adequate evidence, they might plausibly be construed as elements of a cumulative case argument. Of course these arguments are subject to objections, and the objections are subject to objections too. But one simply can't say "there is no evidence" or "there is no adequate" without at least engaging these arguments. Hence, the defense of the evidentialist objection would require a systematic and exhaustive analysis of theistic arguments.

That the evidentialist requirement is not applicable to all beliefs can be argued as follows. Some beliefs must be rational or warranted in a way that does not depend on evidence (in the form of reasons or arguments); otherwise no belief can be rational or warranted. Clearly, if a belief C is supported by another belief B, where B is a reason for C, B must also be a good reason for C. But in that case, it seems like B must be a warranted belief. But if all warranted beliefs require reasons, then there must be evidence for B too. And the evidence for B must be supported by still further evidence. And so on ad infinitum. But since no human person can possess an infinite number of beliefs, it would follow that no belief is rational or warranted. Of course, maybe one's evidence eventually loops back to a previous belief at some point. B is evidence for A and C is evidence for B and A is evidence for C. But then one's reasoning is circular. (Furthermore, this would entail that A is evidence for itself, but self-evidence does not involve a belief being evidentially supported by another belief, so this is actually inconsistent with an evidentialist requirement). To avoid the dilemma of infinite regress or circularity, there must be some rational beliefs that do not require reasons. So it appears that some beliefs must be properly basic, rational (or warranted) but not held on the evidential basis of other beliefs. This is the basic thesis of foundationalism.

Classical foundationalism maintains that an evidentialist requirement is appropriate for some beliefs, while other beliefs are properly basic. But the only beliefs that are properly basic are either self-evident (e.g., 2 + 2 = 4), about one's immediate mental states (e.g., I seem to see a tree), or evident to the senses (e.g., there is a tree outside). But since religious beliefs are neither self-evident, about one's immediate introspective experience, nor evident to the senses, they cannot be properly basic. Hence, the classical foundationalist has a criterion basis on which to apply an evidentialist requirement to religious belief but not all beliefs. Moreover, the evidentialist requirement for religious belief takes rationality to be deontological, a matter of being within one's intellectual rights. Hence, a belief is rational in this sense just if a person is within his intellectual rights in holding it. (See Handout II for discussion on deontologism). A belief is properly basic just if a person is within his intellectual rights in holding it in the absence of propositional evidence or reasons. The classical package then claims that a person would be violating some intellectual obligation if he held a religious belief without the appropriate propositional evidence. Such a belief would not be rational.

Plantinga cites the classical foundationalist principle as:

["Induction" and "abduction" are being used to refer to two kinds of probabilistic reasoning.]

1. Plantinga argues that classical foundationalism is self-referentially incoherent. If [CP] is true then no one is justified in accepting [CP], for [CP] is neither properly basic nor properly nonbasic.
2. Plantinga argues that classical foundationalism's criteria for proper basicality are inconsistent with paradigmatic cases of properly basic beliefs. For instance, we hold many justified basic beliefs that do not satisfy CP's criteria for proper basicality. Belief in other minds, the existence of the past, and the external world are rarely held on the evidential basis of other beliefs, but neither are they self-evident or about one's own mental life, and probably not evident to the senses either.
3. Plantinga argues that religious belief is properly basic for many people (i.e., it is deontologically rational in the absence of propositional evidence). A person can hold religious beliefs without evidence and without violating any intellectual duties. If upon investigating arguments pro and con for religious belief, a person is not impressed with objections to religious belief or impressed with the arguments of theistic philosophers, what more can reasonably be expected of the person? It seems that such a person would be deontologically rational in continuing to hold religious belief.