It seems so:

Although the Proper Basicality Thesis affirms:

The proper function rationality defeater principle affirms:

But then there seems to be the argument:

(1) If a person S acquired a defeater for theistic belief, then S’s holding theistic belief (or at least doing so firmly) would be PF-irrational and S’s not holding theistic belief (or at least not doing so firmly) would be PF-rational.

2. There are some people who acquire defeaters for theistic belief.

(3) There are some people for whom holding theistic belief (or doing so firmly) is PF-irrational and not holding theistic belief (or not doing so firmly) is PF-rational.

• (2) is true but (1) is false, so the argument is unsound.

If (2) is true, then this is only because of cognitive malfunction elsewhere in one’s cognitive system. But a defeater produces a PF-rational state of religious unbelief only if a person’s holding the defeater does not depend on malfunction elsewhere in the person’ relevant cognitive modules.

• (2) is false and so the argument is unsound.

A person has a defeater for theistic belief only if provision is made for this in the design plan, but we have no reason to suppose that such a provision has been made. Why? It is likely that a given cognitive faculty or module m is designed to yield a certain output only if another (related) cognitive faculty or module m* is also functioning properly - the two must function in tandem. Only then would the person’s cognitive faculties be functioning properly in the relevant aspects. So S may come to believe D and S’s noetic structure at t may includes B, but it is not the case that S’s cognitive faculties are functioning properly in the relevant respects, because the sensus divinitiatis must also be functioning properly and it is not (otherwise a person would hold a firm theistic belief).

Plantinga contends that the epistemic significance of a defeater for theistic belief and/or the resultant cognitive state of religious unbelief depends on the absence of cognitive malfunction in the relevant cognitive neighborhood. But that, according to Plantinga, is exactly what there is not in the case of theistic belief.

Plantinga’s argument depends on two crucial premises:

But. . . .

• Even if [P2] is true, [P3] might be false, in which case the argument is unsound.

A distinction is necessary. The depth of PF-rationality ingression can vary depending on the point at which cognitive malfunction enters into the path leading to a cognitive state (e.g., religious unbelief). But only if the PF-rationality ingression for a doxastic state is zero does it appear that such a state has no epistemic significance, for Plantinga admits that irrational beliefs can function as defeaters (and thereby affect the warranted status of a belief). But then, as long as a person’s state of religious unbelief has a depth of PF-rationality ingression greater than zero that cognitive state will have some degree of PF-rationality (even if it is not maximal). This is also consistent with the person’s having a defeater that has a zeroeth degree of rationality ingression, for it would not follow from that that religious unbelief has a zeroeth degree of PF-rationality ingression (since that cognitive state was produced by a properly functioning defeater system). So neither defeaters nor the state of religious unbelief must have a maximal degree of PF-rationality ingression to be epistemically significant.

• [P2] is not true and so Plantinga’s argument is unsound.

[P2] entails a different formulation of the proper basicality thesis:

However, [P4] is dubious for four reasons:

1. It is possible to acquire a defeater for a belief one holds very firmly, even if one does not give up that belief (or hold it less firmly), as is the case with optmistic overriders. So acquiring a defeater for theistic belief does not require a malfunctioning sensus divinitatis (hereafter SD).
2. If the SD is malfunctioning, then it may be that it is not functioning properly over the entire range of its operation (e.g., degree of belief). Since warrant does not require that a belief function properly over its entire range of operation, the sense in which acquiring a defeater for theistic belief depends on malfunction is not as significant as Plantinga suggests.
3. If the functioning of the SD is tied to the cultivation of certain affections, firm theistic belief may depend not only the proper functioning of the SD, but also on human actions that create the appropriate habits or affections. In that case, weak theistic belief may not be the result of SD malfunction but our failure to have cultivated the right sort of affections.
4. [P4] entails a significantly different formulation of the proper basicality thesis, one that does not include a situational or circumstantial index to the proper function clause. But the analogies Plantinga draws between basic theistic belief and other paradigmatic properly basic beliefs implicitly and explicitly affirm the circumstance relative character of the proper functioning of the SD.

1. Plantinga explains the operation of the SD in functional terms. Like sensory perceptual beliefs, the SD takes certain experiential input and yields theistic belief as an output.
2. In chapter 11 Plantinga rebuts the objection that theistic belief, if basic, would be beyond defeat, rational evaluation, or objection. Plantinga’s counter-argument depends on analogies draw from testimony, sensory perception, and memory. In all these analogical cases, a belief gets defeated in a way consistent with the proper functioning of the faculty that produced the original belief. So the analogs to basic theistic belief can be defeated by maximally rational defeaters. To then deny this of theistic belief seems to destroy a crucial point of similarity between basic theistic belief and other paradigmatic basic beliefs.

1. [P4] entails that there is no maximally PF-rational defeater for theistic belief. But beliefs that cannot be defeated by a maximally rational defeater are so by virtue of possessing certain epistemic immunities (e.g., beliefs about one’s current mental states). Theistic belief lacks that essential property, and so does not fit the class of beliefs for which one cannot acquire a maximally rational PF-defeater. The charge in (iv) was that [P4] is inconsistent with a crucial analogy between basic theistic belief and other properly basic beliefs. Here [P4] is inconsistent with a crucial disanalogy between basic theistic belief and basic beliefs with epistemic immunities.

According to Plantinga’s second argument, since it is likely that many of the modules of our cognitive establishment are designed to work together, the epistemic standing of a deliverance of one module, such as a the defeater system module, might have no epistemic standing if there is a failure in another module (designed to operate in tandem with the defeater system). The cogency of this argument in relation to theistic belief rests on two assumptions: (i) the design plan specifies that a person’s defeater system function this way in relation to the SD and (ii) acquiring a "putative" defeater entails malfunction in the SD.

But. . . .

• There is not adequate evidence for (i)

We know that the design plan sometimes specifies a certain doxastic output from the module(s) governing the defeater system, where the input to the module(s) includes irrational beliefs (i.e., beliefs produced as the result of malfunction in some other cognitive modules). So why should matters be any different for theistic belief (or religious unbelief)? This needs to be argued. Surely it is possible that the design plan makes the specifications Plantinga suggests for theistic belief. But since we know that there are cognitive states that have an epistemic status even though they depend in part on a malfunctioning module (working together with a properly functioning defeater system), some reason must be provided to suppose that this does not also hold true for theistic belief.

• Assumption (ii) is false

1. I have already argued above that acquiring a defeater for theistic belief is consistent with the proper functioning of the SD, so I see no need to assume - as Plantinga does in this argument - that the SD is malfunctioning. We actually have significant reason to suppose the opposite.
2. Given the analogies Plantinga draws between basic theistic belief and other properly basic beliefs, the specifications of the design plan with reference to acquiring defeaters can and should be spelled out in terms of more general principles that are applicable to theistic belief. It seems that this would be the result of an inductive procedure working from paradigmatic cases of properly basic belief.

Plantinga wants to maintain that the proper basicality of theistic belief does not entail that it is beyond rational evaluation or defeat. But isn’t this claim severely compromised by Plantinga’s arguments?

• If no defeater for theistic belief can be maximally rational, then it seems that one has good grounds for not attempting to answer any objections to theism (having understood the irrational source of all such objections). It would be a little like taking seriously a defeater a person proposes for my own existence. Perhaps I would even be suffering from some kind of dysfunction if I took this sort of thing seriously. So why answer objections to theistic belief? I think this reveals a tension in Plantinga’s treatment of properly basic theistic belief, a tension between wanting to locate it in the class of basic beliefs (along with sensory perceptual, memory beliefs, etc), while at the same time wanting to give it some privileged status by virtue of a theistic design plan, but where such a status seems appropriate only for beliefs with certain epistemic immunities that theistic belief does not have.
• If it is not possible to acquire a defeater for theistic belief, then it seems rather obviously to follow that theistic belief cannot be defeated, and any belief that theistic belief has been defeated would itself be the product of cognitive malfunction. But Plantinga wants to argue in chapter 11 that theistic belief, like other basic beliefs, can be defeated. But if it can be defeated, it must be possible to acquire a defeater for it. Isn’t this a reason for supposing that the design plan makes the relevant specification for religious unbelief?