The ‘Diderot Objection’ to Plantinga's Reformed Epistemology

Abstract

In response to Pascal's famous wager argument for adopting Christian belief, Denis Diderot noted that ‘An Imam could just as well reason this way’. In this article, I will show how Diderot's observation about Pascal's argument can legitimately be made about Alvin Plantinga's Reformed Epistemology (RE) and its use in defending the rationality of Christian belief. Plantinga's RE can, with some minor adjustments, easily be adopted by Muslims. I shall argue that an Islamic analogue of Plantinga's Christian RE presents an undercutting rationality defeater for Christian belief for those reflective Christians who adopt Plantinga-style religious epistemology. I call this defeater the ‘Diderot Objection’ to Plantinga's RE. As part of my discussion, I will consider how Plantinga attempts to respond to this sort of objection and will show why his response runs into difficulties.

Introduction

My principal aim in this article is to develop and defend a specific line of criticism of Alvin Plantinga's version of Reformed Epistemology (RE) as presented primarily in his important and influential work Warranted Christian Belief (2000; hereafter abbreviated as ‘WCB’). I shall argue that, given this criticism, Plantinga's RE fails as a defence of the rational acceptability of Christian belief for the reflective Christian. By a ‘reflective Christian’, I mean the Christian variant of what Philip L. Quinn calls the ‘intellectually sophisticated adult theist’ (Quinn (1985), 470). Such a theist, as Quinn explains, is aware of several challenges to the reasonableness of theistic belief in our culture and would have to be epistemically negligent not to be in such a position (ibid., 481). The specific line of criticism of Plantinga's RE that I will pursue focuses on the challenge posed by the problem of religious diversity, with a focus on and reference to the Islamic religion. Many critics of Plantinga's RE have observed that it is flexible enough to be easily deployed by adherents of different religions, including Islam; this fact, it has been argued, undermines the rational acceptability of Christian belief for those Christians who recognize it and who adopt Plantinga-style religious epistemology (e.g. see Baldwin (2006)). Little attention has been given, however, to developing such criticism of Plantinga's RE specifically in the restricted and concrete context of Christian–Muslim religious disagreement. It is in this context that I will pursue criticism of Plantinga's RE. As a tip of the hat to Denis Diderot, who responded to Pascal's well-known wager argument for adopting Christian belief by noting that ‘An Imam could just as well reason this way’ (Hájek (2017)), I will call my criticism the ‘Diderot Objection’ to Plantinga's RE.

Approaching Plantinga's RE with the Diderot Objection

In the preface of WCB, Plantinga makes it clear that the intended audience of his RE includes reflective Christians. As he notes, he is interested in the rational acceptability of Christian belief for ‘educated and intelligent people living in the twenty-first century’ (WCB, viii). By ‘Christian belief’, Plantinga means the conjunction of generic theism and a uniquely Christian component (WCB, vii). This Christian component, according to him, states

that we human beings are somehow mired in rebellion and sin, that we consequently require deliverance and salvation, and that God has arranged for that deliverance through the sacrificial suffering, death, and resurrection of Jesus Christ, who was both a man and also the second member of the Trinity, the uniquely divine son of God. (WCB, vii)

In this brief description of Christian belief, one sees several claims that are incompatible with Islamic belief, such as claims about the divinity of Jesus, the sonship of Jesus, and the Trinity. The Qur'an unequivocally denies all these doctrines (see 4:171, for example). Most Christians and Muslims acknowledge the doctrinal incompatibilities between Christianity and Islam.¹ These incompatibilities were certainly not lost on John Calvin, a central and inspirational figure for Plantinga's RE, who offers these remarks on why, according to him, the Jewish and Islamic conception of God is deficient:

[T]hey who form their ideas of God in his naked majesty apart from Christ, have an idol instead of the true God, as the case is with the Jews and the Turks [i.e., Muslims]. Whosoever, then, seeks really to know the only true God, must regard him as the Father of Christ; for, whenever our mind seeks God, except Christ be thought of, it will wander and be confused, until it be wholly lost. (Calvin (1840–1857))²

Incompatibilities such as those between Christian and Islamic beliefs regarding fundamental matters of doctrine pose an especially difficult problem for Plantinga's RE. One reason for this becomes apparent when we consider Plantinga's insistence that the only serious type of question regarding the rational acceptability of Christian belief concerns religious knowledge.

The consensus among contemporary epistemologists is that a person, S, knows some proposition, p, only if S's belief that p is true. Suppose Peter, a Christian, believes that Jesus is God, and Khadijah, a Muslim, believes that Jesus is merely a human prophet. Now, if Peter knows that Jesus is God, then this entails that Jesus is, in fact, God. And in that case, Khadijah's belief that Jesus is merely a human prophet is false. And from this, it follows that Khadijah does not know that Jesus is merely a human prophet, since truth is a necessary condition for knowledge. This must be the case, even if Khadijah appears to have grounds like Peter's (e.g. religious experience, authority, etc.) for thinking that her religious beliefs are true. Suppose further that Peter is a reflective Christian and is aware of this point. Peter sees that there are reflective Muslims like Khadijah who make contrary claims about religious knowledge. Such Muslims also appear to have grounds of a similar kind to his own on which they affirm the truth of their beliefs. Peter also understands that if Khadijah knows that Jesus is merely a human prophet, then he does not know that Jesus is God, even if it appears to him that this is so and that there are grounds for affirming the truth of Christ's divinity. Given his recognition of the similarities between his situation and Khadijah's, how can Peter be confident that he knows that Jesus is God while Khadijah is mistaken in thinking she has religious knowledge about Jesus being merely a human prophet? This, in brief, is how the Diderot Objection may be stated in a preliminary fashion.

Plantinga is aware of the general problem raised by the Diderot Objection, although he does not provide any discussion in WCB regarding specific religious incompatibilities, such as those between Christianity and Islam. Nevertheless, in the early pages of WCB, he affirms that a ‘recognition of the variety and importance of religions incompatible with Christian belief’, something that can reasonably be expected of a knowledgeable Christian, does not constitute a defeater for such belief (WCB, xiii). In what follows, I will briefly explain Plantinga's RE and his attempt to provide a defence of the rational acceptability of Christian belief. Following this, I will argue that Plantinga's thinking can be applied, mutatis mutandis, to Islamic belief in a way that constitutes a defeater for Christian belief in the framework of Plantinga's RE. The result, I shall then argue, is that Plantinga's RE fails as a defence of the rational acceptability of Christian belief for the reflective Christian, pace Plantinga's claims to the contrary. Unpacking these key points will constitute a further detailing of the Diderot Objection.

Plantinga on Warranted Christian Belief

Plantinga's project in WCB is predicated on a fundamental distinction between de facto and de jure objections to Christian belief. De facto objections effectively claim that Christian belief is false, whereas de jure objections express concerns along these lines:

Christian belief, whether or not true, is at any rate unjustifiable, or rationally unjustified, or irrational, or not intellectually respectable, or contrary to sound morality, or without sufficient evidence, or in some other way rationally unacceptable, not up to snuff from an intellectual point of view. (WCB, ix)

In WCB, Plantinga examines a few different ways a de jure objection to Christian belief may be formulated. He maintains that the three main interpretations of such an objection involve the epistemic notions of ‘justification’, ‘rationality’, and ‘warrant’ (WCB, x). According to Plantinga, the de jure objection to Christian belief cannot be reasonably stated in terms of either justification or rationality; the only serious type of de jure objection, he maintains, concerns warrant.

Plantinga introduces his understanding of warrant through a discussion of the writings of Sigmund Freud and Karl Marx on religious belief. In essence, Freud considers religious belief to be the result of wish fulfilment, and Marx thinks it arises from cognitive dysfunction. Plantinga explains that the key idea behind these criticisms is that religious belief ‘is not produced by properly functioning truth-aimed cognitive faculties or processes’ (WCB, 151). For Plantinga, Freud's and Marx's criticisms of religious belief can be distilled into the claim that it lacks warrant, which he understands as ‘that further quality or quantity . . . enough of which distinguishes knowledge from mere true belief’ (WCB, 153). More specifically:

[A] belief has warrant for a person S only if that belief is produced in S by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S's kind of cognitive faculties, according to a design plan that is successfully aimed at truth. (WCB, 156)³

In Plantinga's view, the claim that religious (Christian) belief lacks warrant in this sense is the only viable kind of de jure objection that may be levelled against it (WCB, 161–163).

By way of response to the de jure objection that religious belief lacks warrant, Plantinga presents two models for consideration that are inspired by the writings of Thomas Aquinas and John Calvin. These models are designed to show how theistic belief, particularly Christian belief, can be warranted. According to the first model, what Plantinga calls the Aquinas/Calvin (A/C) model, human beings possess a cognitive mechanism – a sensus divinitatis. Under certain circumstances, this mechanism produces beliefs about God. As Plantinga notes, the sensus divinitatis is analogous to other belief-producing faculties or mechanisms, such as those involved in perception. Of the several features of the A/C model that Plantinga discusses, I will highlight two here. First, on the A/C model, theistic belief is produced immediately in a non-inferential manner. It is, in other words, a basic belief (WCB, 175). Second, the basic theistic belief produced by the A/C model enjoys warrant and, if held strongly enough, constitutes knowledge (WCB, 179).

In proposing the A/C model, Plantinga notes that Freud's and Marx's criticisms of religious (theistic) belief tend to presuppose that it is false. Both thinkers assume that there is no God and no sensus divinitatis. Nevertheless, Plantinga agrees that they are probably correct if theism is false (WCB, 186–188). If, however, theism is true, then theistic belief is probably warranted through something like the A/C model. This is because, in Plantinga's estimation, the natural thing to think is that a loving God would probably want us to know about Him. He would, therefore, probably create us in such a way that the cognitive processes responsible for producing theistic belief are aimed at providing us with knowledge about Him (WCB 188–189). In discussing Freud's and Marx's criticisms of religious belief, along with his response to them that involves an articulation of the A/C model, Plantinga arrives at what he regards as an important conclusion: de jure concerns about theistic belief are not independent of de facto ones. More specifically, for Plantinga, epistemic concerns about theistic belief cannot be rightfully separated from ontological or metaphysical ones (WCB, 191).

Plantinga develops his generalized A/C model further to formulate a second model, the extended A/C model, ‘in which full-blooded Christian belief in all its particularity is justified, rational and warranted’ (WCB, 242). The extended A/C model consists of a three-tiered cognitive process involving the Bible, the internal instigation of the Holy Spirit, and faith (WCB, 243–244). The general idea behind this second model is that, when someone is made aware of divine truths contained in the Christian Scriptures, the Holy Spirit works on him or her to bring about faith. That is, the operation of the Holy Spirit causes such a person to embrace the ‘great things of the gospel’, which include specifically Christian beliefs about the Trinity, incarnation, atonement, etc. On the extended A/C model, as with its generic cousin, such specific Christian beliefs are generated in a basic way and can also possess warrant (WCB, 258).

Through Plantinga's discussion in WCB, we are given a sophisticated defence of his famous claim that belief in God is properly basic. As he puts it in his article ‘Reason and Belief in God’, one of the early classic pieces on RE, ‘it is entirely right, rational, reasonable, and proper to believe in God without any evidence or argument at all’ (Plantinga (1983), 17). In WCB, this point is extended to include Christian belief. If Christianity is true, then (based on the extended A/C model), Christian belief is warranted. In this case, Christians can have knowledge about God, the Trinity, incarnation, atonement, etc., without needing any evidence or arguments supporting the truth of Christian belief (WCB, 258).

Developing the Diderot Objection

In the preceding section, I provided an outline of Plantinga's RE and his attempt to provide a defence of the rational acceptability of Christian belief. Can an Imam just as well reason this way to defend the rational acceptability of Islamic belief, making, of course, the necessary substitutions and adjustments to account for the differences between Christianity and Islam? Plantinga seems to think so:

[C]ouldn't we find a model under which the beliefs in question have warrant, and such that, given the truth of those beliefs, there are no philosophical objections to the truth of the model? Well, probably something like that is true for the other theistic religions: Judaism, Islam, some forms of Hinduism, some forms of Buddhism, some forms of American Indian religion. Perhaps these religions are like Christianity in that they are subject to no de jure objections that are independent of de facto objections. (WCB, 350)

In their interesting comparative work on Plantinga's RE and other world religions, Erik Baldwin and Tyler McNabb consider how Plantinga-style RE may be used by members of non-Christian faiths. Looking specifically at Islam, they consider how Plantinga's standard and extended A/C models may be adopted by Muslims. Despite some reservations, Baldwin and McNabb present a detailed discussion based on the Qur'an, ahadith (reports of sayings and actions attributed to the Prophet Muhammad), and Islamic theological commentaries to show how Plantinga's RE can be plausibly applied in an Islamic context (Baldwin and McNabb (2019), ch. 10). Similarly, Jamie Turner (2021) offers a slightly different but more robust version of Islamic RE inspired by the medieval Islamic thinker Ibn Taymiyyah (d. 1328), which he maintains is wholly compatible with Plantinga's two A/C models. Since Plantinga himself effectively concedes that an Imam could indeed reason in a way resembling his thinking about the epistemology of religious belief, something that others have demonstrated in detail, I will not pursue a defence of the compatibility of his RE with Islam. Instead, I will focus on developing the Diderot Objection by showing how such compatibility constitutes a defeater for rationally held Christian belief for the reflective Plantingian Christian.

In WCB, Plantinga explains the basic idea behind defeaters as follows: ‘[Y]ou have a defeater for one of your beliefs B just if you acquire another belief D such that, given that you hold that belief, the rational response is to reject B (or hold it less firmly)’ (WCB, 366). Endorsing John Pollock's distinction between a rebutting defeater (roughly speaking, evidence that your belief is false) and an undercutting defeater (roughly speaking, evidence that undermines the reasonableness of your belief, a belief that may nonetheless be true), Plantinga cites an example from Pollock of the latter:

You enter a factory and see an assembly line on which there are a number of widgets, all of which look red. You form the belief that indeed they are red. Then along comes the shop superintendent, who informs you that the widgets are being irradiated by red and infrared light, a process that makes it possible to detect otherwise undetectable hairline cracks. You then have a defeater for your belief that the widget you are looking at is red. In this case, what you learn is not something incompatible with the defeated belief (you aren't told that this widget isn't red); what you learn, rather, is something that undercuts your grounds or reasons for thinking it red. (You realize that it would look red even if it weren't.) (WCB, 359)

He goes on to explain that defeaters of this kind are what he calls ‘rationality defeaters’; that is, ‘given belief in the defeating proposition, you can retain belief in the defeated proposition only at the cost of irrationality’ (WCB, 359). In Pollock's example, the person, let's call her Brigit, would be irrational to insist that, despite what the superintendent tells her, the widget she is looking at is red just because it appears red to her. After acquiring information from the superintendent about the widgets being irradiated by red and infrared light, information that (let us suppose) Brigit has no reason to discount, the widget's seeming red (no matter how powerful this seeming is) is insufficient for Brigit to rationally retain belief in its redness. In these circumstances, the kind of rationality being violated is what Plantinga calls ‘internal rationality’, a variant of proper function rationality. Such rationality, he explains, is ‘a matter of proper function of all belief-producing processes “downstream from experience”’’ (WCB, 110; emphasis mine). An internally rational person will form the right response to sensuous imagery and doxastic experience, have (sufficient) coherence among his or her beliefs and when needed draw the right inferences from them, make appropriate practical decisions in response to his or her beliefs, and, in general, exercise epistemic responsibility (WCB 110–111).

In discussing matters of epistemic responsibility, Plantinga construes internal rationality to include epistemic justification, which has as a central feature the idea of ‘being within one's epistemic rights, having flouted no epistemic duties or obligations’ (WCB 365). Internal rationality, then, has a ‘dual aspect’ as Plantinga puts it: (1) the proper functioning of cognitive faculties ‘downstream from experience’ and (2) the fulfilling of one's epistemic duties, such as considering how coherent the belief in question is with the rest of your beliefs, seeking potential defeaters, considering objections you've encountered, etc. (WCB, 255). For Brigit to insist that the widget is red just because it seems red, despite being aware of and believing what the superintendent says (an undercutting defeater), is for her to be internally irrational. She is violating her epistemic duties by ignoring the obvious implications of the defeater: the widget would look red even if it wasn't red. Note here that charging Brigit with internal irrationality does not call for denying the sensuous imagery and doxastic experience connected to her belief in the widget's redness. Nor does it involve calling into question what Plantinga calls ‘external rationality’, which requires the proper functioning of cognitive faculties with respect to the sensuous experience on which belief is based and the formation of the right kind of doxastic experience; as Plantinga sometimes describes it, external rationality is the proper functioning of cognitive faculties ‘upstream from experience’ (WCB, 246, 346). What further reflection on Pollock's widget example shows is that concerns about internal rationality cannot be fully met simply by appealing to the proper functioning of one's cognitive faculties at the external level and by the right formation of the experience resulting in belief, especially when one is faced with defeaters.

As I've noted, Plantinga himself concedes that the full scope of proper function rationality includes other important matters connected to the fulfilment of one's epistemic obligations. These have a decidedly ‘internalist’ aspect, as they involve a person's own beliefs and experiences (e.g. matters of doxastic coherence) which he or she has access to upon reflection (for a further discussion of this crucial point, see Sudduth (1999), esp. 170–171). As most people familiar with Plantinga's work in religious epistemology are aware, the internalist–externalist debate in general epistemology looms in the background in considerations of his RE. While I cannot embark on any sort of substantial discussion or assessment of that debate here, I will nevertheless make what I think are a few important observations that are relevant to my discussion of Plantinga's RE.

To begin with, notice that, although Plantinga initially appears to abstain from adopting internalism in crafting his RE, he ends up tacitly accepting something like it by including a no-defeater condition in his account of how internal rationality can obtain under his A/C models:

Suppose my experience is of the sort that goes with the testimony of the Holy Spirit . . . so that the great things of the gospel seem powerfully plausible and compelling to me: then (given that I have no undefeated defeaters for these propositions) there will be nothing dysfunctional or contrary to proper function in accepting the beliefs in question. Indeed, given those experiences, it would be dysfunctional not to form them. (WCB, 255; emphasis mine)

Plantinga's externalism, then, is in a sense moderated by internalism (or quasi-internalism), at least when it comes to rationality defeaters. This point is a critical one and should be borne in mind lest one thinks that my upcoming presentation of the Diderot Objection, which I will articulate discursively, is inappropriately approaching Plantinga's RE from an internalist position tout court. It is not. Instead, I will present this objection to Plantinga's RE within the (primarily externalist) epistemological framework that Plantinga himself lays out in WCB, one that factors in potential rationality defeaters.

Insofar as defending RE is concerned, I think that Plantinga is right to include a no-defeater condition in his A/C models if only to avoid having his defence of the epistemic propriety of religious belief collapse into irrationalist fideism. As I noted in my discussion of the widget case, for Brigit to maintain a naive ‘steadfastness’ and continue to insist that the widget she is looking at is indeed red, despite being presented with information by the superintendent that constitutes an undercutting defeater, is for her to be (internally) irrational. There are several similar cases where Plantinga's RE cannot, without bringing in the constraints of internal rationality, account for a subject's irrationality in accepting a particular belief despite it being true and the result of properly functioning cognitive faculties. Here is one such case, devised by Laurence BonJour (a variant of his well-known example of Norman the clairvoyant that was originally presented as a challenge to reliabilism):

[S]uppose that a certain person, call him Boris, was indeed designed by God and that deep within his brain or psyche, God has implanted a very narrow and specialized module designed to guarantee that Boris will have a true belief about some monumentally important matter. To be specific, suppose that this module is so constructed that at some appropriate temporal interval before the mundane world comes to an end with the Second Coming, Boris will be caused to believe with maximal firmness and conviction that this is about to occur. We may suppose that the belief is accompanied by no distinctive phenomenology, beyond the strong impetus to belief itself. Now the time has come, and Boris finds himself believing that the world will soon end, and believing it as firmly as he believes that 2 + 2 = 4 or that he is a human being. (BonJour (2002), 254–255)

As BonJour goes on to explain, it is clearly irrational for Boris to accept and act on this belief given that he (i.e. from his perspective) has no rational basis for thinking that the end of the world is imminent; indeed, on the contrary, Boris has several reasons for being suspicious of this belief (ibid., 255). For Boris to maintain steadfastness in his conviction given these circumstances is irrational, even if his belief that the world is about to end is the result of properly functioning cognitive faculties (i.e. is externally rational) and the world is, in fact, about to end. Since all the externalist conditions for knowledge are, ex hypothesi, present in the case of Boris, the irrationality of his steadfastness can only be accounted for by appealing to internalist conditions on proper function rationality (downstream from experience).

It is in the kind of context created by the cases of Brigit and Boris that I will be presenting the Diderot Objection to Plantinga's RE. In doing so, I will grant Plantinga's primarily externalist framework within his RE (as presented in his A/C models) and limit the scope of my objection to its internalist aspect that is connected to proper function rationality. I will also assume, for argument's sake, that Christianity (Plantinga's understanding of it, at any rate) is true. Finally, I will assume that Peter, from whose perspective I shall articulate the Diderot Objection, is a reflective Christian of the Plantingian kind (i.e. Peter holds Christian belief and thinks that he is warranted in doing so through Plantinga's RE) and that his Christian belief is, in fact, the result of properly functioning cognitive faculties (i.e. is externally rational). Now, to the objection.

Let p stand for any proposition that is part of fundamental Islamic doctrine but is incompatible with fundamental Christian doctrine. In detailing the Diderot Objection, I will let p stand for the proposition ‘Jesus is merely a human prophet’ although several other alternatives can be used (e.g. ‘God has no son’, ‘God is not a Trinity of Persons’, etc.). Take, now, Peter's belief that Jesus is God. Since Peter holds this belief and sees that it is incompatible with p, he believes the following proposition:

1. (1) p is false.

Since Peter is a reflective Christian of the Plantingian kind, he holds that his Christian belief is warranted through Plantinga's standard and extended A/C models. According to Plantinga's theory of warrant, which is central to both models, false beliefs are (probably) not warranted (WCB, 186–188).⁴ Peter understands this point and sees that it also applies to Islamic versions of the standard and extended A/C models. False beliefs are not warranted in these models either. Let us also suppose that Khadijah is, like Peter, a reflective Plantingian who holds that her Islamic belief is warranted through analogues of Plantinga's standard and extended A/C models. In encountering and engaging with a reflective Plantingian Muslim like Khadijah who believes that p, Peter sees that the following must be true:

1. (2) If p is false, then Khadijah's belief that p is not warranted.

From (1) and (2), Peter validly infers and believes (3):

1. (3) Khadijah's belief that p is not warranted.

Although Peter holds that Khadijah's belief that p is not warranted, he nevertheless affirms the following:

1. (4) Khadijah believes that p is warranted.

It is important to note that (3) and (4) are not inconsistent. (3) is a statement about Khadijah's belief that p lacking warrant, whereas (4) is a statement about what Khadijah herself believes about her belief that p being warranted (i.e. it is about a ‘metabelief’: a belief about a belief). Clearly, Khadijah and many other Muslims have metabeliefs about the epistemic status of their Islamic beliefs. From Peter's perspective, (3) and (4) are both true. To see this, consider the following scenario. Suppose that Peter spends some time with Khadijah discussing Christian and Islamic beliefs, in particular enquiring about her belief that p. She explains to him that, many years ago, she read through the Bible and the Qur'an carefully, and that she also prayed to God to show her the truth about p. A week or so following this, Khadijah tells Peter, she was overwhelmed with a firm sense of assurance that God answered her prayers since p became powerfully plausible and compelling to her. By God's grace and guidance, says Khadijah to Peter, she knows that p is true.

Now, Peter's interpretation of this interaction (given what he believes) must include something like the following construal of her experience and testimony: for Khadijah, it strongly appears that p even though p is false (i.e. it strongly appears that Jesus is merely a human prophet even though he isn't). Khadijah's strong religious seeming that resulted in her belief that p, unfortunately, did not save her from ‘epistemic misfortune’, thinks Peter. He continues to think this despite recognizing Khadijah as a reflective Plantingian like himself, who can use Plantinga's RE in an Islamic context to give an account of her religious beliefs, including her belief that p. Insofar as a Plantingian Muslim like Khadijah is concerned, then, Peter sees that the following is also true:

1. (5) Khadijah can use Islamic versions of the standard and extended A/C models to give an explanation for why her belief that p is warranted.⁵

Making the next premise a plausible one (or as plausible as possible) requires filling out in a bit more detail the kind of reflective Plantingian Muslim that Khadijah is. Let us further stipulate that Khadijah's reflectiveness constitutes a part of her being what Robert McKim calls a ‘person of integrity’. Here is how he describes this category:

By ‘people of integrity’ I mean people who, at least in the ideal case, know a great deal, avoid exaggeration, admit ignorance when appropriate, have an interest in the truth, and are intelligent, serious, sincere, decent, sensible, reflective, and so on . . . Such people hold the relevant beliefs in all sincerity and endeavor to live in accordance with them. (McKim (2016), 230)

As McKim suggests, to be a person of integrity might be an ideal to aspire towards. Even if Peter thinks that no one is actually a person (i.e. a model) of integrity, he may nonetheless think that Khadijah's progress towards this ideal is comparable to his (e.g. he sees that both he and Khadijah know quite a bit about religion, have an interest in the truth, are reflective individuals, etc.). Given this, Peter further holds that this proposition is true:

1. (6) If Khadijah believes that p is warranted and Khadijah can use Islamic versions of the standard and extended A/C models to give an explanation for why her belief that p is warranted, then Khadijah is Peter's Plantingian religious peer (i.e. Khadijah is, like Peter, a person of integrity who believes that her religious beliefs are warranted in the Plantingian way).

To be clear, by a ‘Plantingian religious peer’ I simply mean that Khadijah is a person of integrity like Peter who deploys a Plantingian way of thinking about her religious beliefs. Obviously, since Peter does not think that Khadijah's belief that p is true or warranted, he does not think that she is his Plantingian religious peer in that sense.⁶ From (4), (5), and (6), Peter validly concludes that (7) is true:

1. (7) Khadijah is Peter's Plantingian religious peer.

Finally, from (3) and (7), Peter arrives at (8):

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

This conclusion, I maintain, constitutes for Peter, upon reflection, an undercutting (internal) rationality defeater for his belief. Consider again his belief that Jesus is God. Peter holds that this belief is true and warranted in light of Plantinga's extended A/C model. He reflects on how his own conviction about Christ's divinity was formed in the kind of way that Plantinga describes when presenting this model:

We read Scripture, or something presenting scriptural teaching, or hear the gospel preached, or are told of it by parents, or encounter a scriptural teaching as the conclusion of an argument (or conceivably even as an object of ridicule), or in some other way encounter a proclamation of the Word. What is said simply seems right; it seems compelling; one finds oneself saying, ‘Yes, that's right, that's the truth of the matter; this is indeed the word of the Lord.’ I read, ‘God was in Christ, reconciling the world to himself’; I come to think: ‘Right; that's true; God really was in Christ, reconciling the world to himself!’ (WCB, 250)

On the extended A/C model, recall, Christian conviction does not arise through argument but rather manifests itself as a basic belief. More specifically, as Plantinga explains:

[I]n each case there is presentation or proposal of central Christian teaching and, by way of response, the phenomenon of being convinced, coming to see, forming of a conviction. There is the reading or hearing, and then there is the belief or conviction that what one reads or hears is true and a teaching of the Lord. (WCB, 251)

Plantinga's point here can be accepted by Plantingian Muslims like Khadijah, who espouse Islamic versions of the standard and extended A/C models. Suppose that Peter learns from Khadijah how her belief that p was formed. She explains to Peter:

We read Scripture, or something presenting scriptural teaching, or hear the Qur'an preached, or are told of it by parents, or encounter a scriptural teaching as the conclusion of an argument (or conceivably even as an object of ridicule), or in some other way encounter a proclamation of the Word. What is said simply seems right; it seems compelling; one finds oneself saying, ‘Yes, that's right, that's the truth of the matter; this is indeed the word of the Lord.’ I read, ‘Jesus is merely a human prophet’; I come to think: ‘Right; that's true; Jesus really is merely a human prophet!’

Khadijah provides further clarification:

[I]n each case there is presentation or proposal of central Islamic teaching and, by way of response, the phenomenon of being convinced, coming to see, forming of a conviction. There is the reading or hearing, and then there is the belief or conviction that what one reads or hears is true and a teaching of the Lord.

Despite Khadijah sharing all this with him, Peter maintains her belief that p is false and unwarranted. Khadijah, he thinks, does not know that Jesus is merely a human prophet. Of course, Khadijah herself is convinced she knows (i.e. is warranted in believing) that p and that this is through the Islamic versions of the standard and extended A/C models; her belief that p simply seems right and compelling. But all of this is perfectly consistent with the falsity of her belief that p, thinks Peter. This observation should serve to reinforce the point that being able to think about his belief that Jesus is God in a parallel way does not, by itself, secure its warrant. Simply thinking that Christian belief is true and rational is not the same as the belief actually being true and rational, obviously. Indeed, this point is part of Peter's assessment of Khadijah's situation in accounting for why her belief that p lacks warrant. Yes, Khadijah may think that her belief that p is warranted because she thinks that Islamic belief is true (i.e. that the Islamic versions of the standard and extended A/C models are true). But, notes Peter, this alone does not secure p's warrant for Khadijah, a proposition that he holds to be false. No matter how convinced Khadijah is in her belief that p or in her ability to proffer a rich theological account explaining how her belief that p arises through the work and guidance of God Himself, muses Peter, Khadijah does not know that p. Yet, if Peter sees this, sees that Khadijah's belief that p lacks warrant despite her being a Plantingian religious peer, how can he be sure his belief that Jesus is God is warranted? Pressing this concern does not require one to excise Plantinga's RE from its externalist home and inappropriately place it in foreign internalist territory. Rather, it is to raise a concern about internal rationality that has now become relevant and salient given Peter's acceptance of (8).

As explained earlier, I am not, in presenting the Diderot Objection, attempting to shoehorn Plantinga's RE into a general internalist framework, one that incorporates a KK or JJ thesis/principle (‘If S knows/is justified in believing that p, then S knows/is justified in believing that S knows/is justified in believing that p’). As we saw earlier in the example of Boris, circumstances may arise where properly functioning cognitive faculties that bring about a true belief are insufficient to allay concerns about internal rationality. Thus, even if Peter's Christian belief is true and the result of properly functioning cognitive faculties, as I have granted in formulating the Diderot Objection, his thinking about the rational acceptability of his belief that Jesus is God is challenged by his belief in (8), an undercutting rationality defeater:

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

What (8) offers to Peter is a concrete example of how a person may hold an unwarranted religious belief despite appearing to have a religious profile very similar to his, one that includes powerful religious seemings and the ability to provide an account of them in a Plantingian manner. The force behind (8) can easily be amplified by Peter noting that there are numerous reflective Plantingian Muslims like Khadijah.

Now, it is true that Peter's Christian belief is ‘epistemically superior’ to Khadijah's Islamic belief insofar as external rationality is concerned, given my concession that his belief is true and the result of properly functioning cognitive faculties (whereas this is not the case for Khadijah). But this ‘epistemic superiority’, resting as it does on purely externalist and inaccessible conditions, fails to offer any support to Peter's perspective given the reflective dialectical context that he finds himself in given (8). For, given (8), the issue for Peter now, an issue concerning internal rationality, is deciding whether he should (a) continue to accept his Christian belief as true and warranted or (b) allow that reasonable doubts can be raised about the rationality of his Christian belief, since he recognizes a perspective (namely, Khadijah's) from which his, Peter's, Christian belief can be seen in the same light as he sees Khadijah's belief that p (i.e. false and unwarranted despite seeming to be otherwise).

Using (8) as an analogous reference point, Peter should see there is good reason for him to think that it would appear to him that Jesus is God and, in general, that Christian belief is true and warranted, even if Christianity is false. After all, as Peter accepts, it appears to Khadijah that Jesus is merely a human prophet and, in general, that Islamic belief is true and warranted, even though (as Peter himself thinks) Islam is false. Given this, Peter cannot, without begging the question, reasonably fall back on his religious seeming to opt for (a) over (b); by accepting (8), the justificatory status of this seeming is now under question. True, even if Peter accepts (8), he may continue to have a strong seeming that Jesus is God, whereas he has no such seeming for the claim that Jesus is merely a human prophet. This persistent internal asymmetry between these two beliefs is, however, of no help to him. As Johnathan Matheson puts it, ‘[t]he mere persistence of a seeming does not indicate that any justification coming from that seeming has not been defeated’ (Matheson (2015), 48).

One obvious way for Peter to annul or at least mitigate the threat posed by (8) is to argue that Christian belief is true (and consequently, he is warranted in accepting that Jesus is God through the A/C model and its Christian extension), whereas Khadijah has no good argument that Islamic belief is true (and consequently, she is not warranted in accepting that p through the A/C model and its Islamic extension). Unfortunately, in what may strike many as the anticlimax of his project in WCB, Plantinga maintains that he cannot show Christianity to be true and that the prospects of demonstrating its truth are unlikely (WCB, 270). He states that a Reformed Epistemologist such as himself does not claim as part of his philosophical position that Christian beliefs do, in fact, have warrant; nor does he try to argue that Christianity is true (WCB, 347). Whether Christianity is true, says Plantinga, is the ‘really important question’ but it is beyond the competence of philosophy to establish its truth (WCB, 499). So, what then is Peter to do? It seems compelling to him that Jesus is God, and he notes that it also seems compelling to Khadijah (and others like her), his Plantingian peer(s), that p is true.

Independent evidence in support of his seeming would allow Peter to ward off the threat posed by (8). As Richard Swinburne explains in his review of Plantinga's work:

Whether various sets of evidence (some public, some private) make it probable that Christian beliefs are true is the question that Plantinga does not discuss. A positive answer – say, that Christian beliefs are probably true on the evidence available to all – would have enabled him to tell us not merely that if they are true, Christian beliefs have warrant; but that (probably) they have warrant. And he would have had a message of reason – which in my view the Christian religion usually claimed that it can provide – for the weak believers and for those outside the field. (Swinburne (2001), 208)

Let us consider Swinburne's suggestion for a moment with reference to Pollock's example of the widgets and my extension of it using Brigit. For Brigit to be told by the shop superintendent that the widgets are being irradiated by red and infrared light constitutes an undercutting rationality defeater of her belief that the widgets are red (based on their seeming red). If, however, Brigit has evidence (public or private) independent of her seeming (e.g. remembering seeing the widgets earlier outside the factory in broad daylight and observing that they were red; or, reading pamphlets in the factory from the manufacturer stating that the widgets are indeed red), then the superintendent's information no longer constitutes a defeater for Brigit's belief that the widgets are red. As Plantinga sometimes puts it, such evidence would constitute a ‘defeater-defeater’ (WCB, 368–369). In his response to Swinburne, Plantinga contends that appealing to public evidence to show Christian belief is true is not a viable option. Commenting on Swinburne's extended probabilistic argument of natural theology in favour of Christianity, which Plantinga takes to be the best on offer, he writes:

[T]his argument (and other arguments like it) does not succeed in showing that Christian belief is very probable with respect to public evidence. In fact, these arguments don't even show that Christian belief is more probable than not with respect to that evidence; they show, at most, that such belief is not wholly improbable with respect to it. Accordingly, if I'm right, the best arguments for the public rationality of Christian belief are not particularly successful – at any rate they don't show that Christian belief is likely with respect to public evidence. (Plantinga (2001), 219–220)

Now, one might disagree with Plantinga's assessment of the public evidence that supports the truth of Christian belief. Indeed, a number of Christians like Swinburne do.⁷ Be that as it may, if Peter was forced to rely on public evidence in order to deal with the threat posed by (8), this would go against the ethos of Plantinga's RE. It would result in a collapse of RE into the kind of ‘evidentialism’ Plantinga rejects, namely, the claim that religious belief is rationally acceptable only if there are good arguments for it (WCB, 82). As William Lane Craig writes in observing the relationship between Plantinga's earlier version of RE and defeaters:

Plantinga seems on the verge of falling into a sort of crypto-evidentialism. For in considering whether a person who holds to belief in God as properly basic may be open to argument, Plantinga appears to allow that belief in God so held may be overcome by argument, so that the theist in order to be rational may have to abandon his belief in God . . . But then [theistic belief] will have to be surrounded by an enormous and elaborately constructed citadel, bristling with defensive armaments to ward off the enemy. In such a case, one wonders how much has been gained by making belief in God properly basic. (Craig (2015))

In defending the version of RE presented in WCB, however, Plantinga eschews public evidence in considering matters of religious (Christian) knowledge. In his response to Swinburne, he maintains that ‘[p]robability with respect to public evidence . . . is neither necessary nor sufficient for warranted Christian belief’ (Plantinga (2001), 221). If this is the case, then it wouldn't matter what Peter's (or anyone's) assessment of the public evidence supporting Christianity and Islam is, when considering whether Christian belief is warranted. Even if Peter arrived at the conclusion that the public evidence supporting Islam is extremely strong while the public evidence for Christianity is extremely weak, it would not (according to Plantinga's RE) threaten any warrant his belief that Jesus is God enjoys. Indeed, Plantinga himself makes a similar point when he discusses contemporary historical biblical criticism in WCB. Even when its alleged results go counter to Christian belief, they do not serve as a defeater for it (WCB, 358).

This aspect of Plantinga's RE in WCB can be supported on the same theological grounds that Plantinga cites for thinking that a loving God would not require us to know Him through arguments. As Craig suggests, a loving God would also not require our knowledge about Him to be contingent on our ability to refute objections (i.e. defeaters) presented against theistic or Christian belief. In circumstances where the believer is faced with potential defeaters of such belief, the non-propositional warrant it has can be safeguarded by the warrant itself constituting an intrinsic defeater-defeater (Craig (2015)).⁸

Reflective Plantingian Muslims like Khadijah can be sympathetic to the theological impetus for accepting this ‘intrinsic defeater-defeater’ proposal, seeing that it can be found among Muslim thinkers. For instance, in his spiritual autobiography, The Rescuer from Error, the Islamic polymath Abu Hamid Al-Ghazali (d. 1111) gives an account of his quest for religious certainty that touches on several themes raised in my discussion of Plantinga's RE. In The Rescuer, Al-Ghazali informs us that he sought ‘certain knowledge’ about religious and indeed all matters (Khalidi (2005), 61), something very similar to Calvin's ‘firm and certain knowledge’ that Plantinga discusses and endorses as the basis of Christian faith (WCB, 248). In explaining the nature of knowledge thus construed, Al-Ghazali introduces an example that resonates with the idea of an intrinsic defeater-defeater:

It became apparent to me that certain knowledge is that in which what is known is laid bare in such a way as to leave no room for doubt, and is unaccompanied by the possibility of error or illusion, to the point that the mind cannot even conceive it. Rather, what is secure from error should be so closely associated with certainty, that if someone tried to show that it was false by turning stone into gold or a stick into a snake, for example, that would not make it doubtful or refute it. Thus, if I came to know that ten is greater than three, and someone said to me: ‘No, three is greater than ten, in proof of which I will turn this stick into a serpent’, then went on to do so in plain view, I would not as a result of that come to doubt what I was cognizant of. The only outcome would be wonderment at how he is able to perform such a feat. As to doubt concerning what I know, there is none. (Khalidi (2005), 61; emphasis mine)

Al-Ghazali goes on to explain that, in his search for this kind of knowledge, he found all his beliefs were open to doubt and it was impossible for him to overcome this doubt through discursive reasoning (ibid., 63). He narrates that he was eventually cured of his scepticism and acquired epistemic certainty, not through any proofs or arguments but by God Himself:

This disease [of extreme skepticism] defied all cure and lasted for almost two months, during which I embraced the sophistical creed in actual fact, though not in speech nor expression. Eventually, God cured me of this disease and my mind was restored to health and balance. The rational necessary beliefs were once again accepted and trusted, both securely and certainly. This did not come about by composing a proof or by an arrangement of words, but rather by a light that God Almighty cast into my breast, which is the key to the greater part of cognizance. Whoever supposes that enlightenment depends upon explicit proofs has narrowed the expanse of God's mercy. (ibid.; emphasis mine)

Notice how Al-Ghazali here appeals to God's mercy in a way that is similar to Plantinga's appeal to God's love. Both Al-Ghazali and Plantinga offer a theological basis from which they derive the proper basicality and epistemic certainty of specific beliefs. In The Rescuer, Al-Ghazali, drawing from the Sufi tradition in Islam, presents a model of Islamic faith where proper immersion in Islamic practice yields religious (Islamic) knowledge that is indisputable. Such knowledge is neither based on nor can be defeated by rational argument (ibid., 85–86). In addition to the notions of proper basicality and epistemic certainty, Al-Ghazali's Sufi understanding of faith includes the idea that the non-propositional warrant enjoyed by Islamic belief can serve as an intrinsic defeater-defeater.

To return now to the issue at hand, here is the problem. Peter holds that he is warranted in his belief that Jesus is God. He is challenged in this by also believing (8), which constitutes an undercutting rationality defeater for his belief:

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

Peter, qua Plantingian Christian, cannot respond to (8) by appealing to public evidence for the truth of Christian belief, as we have seen.⁹ Might (8), a defeater, itself be defeated by Peter's appeal to the sheer compelling nature of the seeming that accompanies his belief that Jesus is God? That is, might the strength of this seeming serve as an intrinsic defeater-defeater for (8)? In WCB, Plantinga describes the essential phenomenology accompanying the epistemic process of his extended A/C model as involving a ‘belief or conviction that what one reads or hears is true and a teaching of the Lord’ (WCB, 251). Surely, this sort of phenomenology, a belief or conviction, can be had by a reflective Plantingian Muslim like Khadijah; indeed, Khadijah may aver that the phenomenology of her conviction about the truth of p exceeds that of Peter's in intensity. Furthermore, Khadijah may submit to Peter that the compelling seeming behind her conviction that p serves as an intrinsic defeater-defeater of any potential defeaters of p to which he may draw her attention. After all, she may say, Allah – The Most Gracious, Most Merciful – would never put a believer in a position where he or she would be rationally required to abandon Islamic belief. What is more plausible, Khadijah might go on to reason, is this. When faced with potential defeaters of p, Allah will strengthen the Divine Light cast into her heart by which she believes the truth of p more intensely.¹⁰ As I have indicated, such an interpretation of the notion of an intrinsic defeater-defeater is not ad hoc in the Islamic context, given that it may be found in the writings of Muslim thinkers like Al-Ghazali.¹¹

To the extent that Khadijah can appeal to a phenomenology comparable to Peter's that accompanies her conviction that p, alongside a comparable Islamic understanding of an intrinsic defeater-defeater of any potential defeaters of p, she is still Peter's Plantingian religious peer. These additional features of Khadijah's being such a peer are, however, compatible with her belief that p being false and unwarranted, as Peter recognizes and continues to hold. Even if Khadijah is absolutely convinced in her belief that p and can provide an account of how Allah would safeguard her conviction that p against any and all defeaters of p, Khadijah is, thinks Peter, unfortunately persisting in holding a false and unwarranted belief. What we have in all this, then, is an additional detailing and clarification of (8):

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

(8), then, remains a defeater for Peter and the rationality of his belief that Jesus is God, even if Peter finds himself strongly convinced of Christ's divinity and is also able to offer a Christian account of how the work of the Holy Spirit can constitute an intrinsic defeater-defeater of any potential defeaters brought against his belief. As I argued earlier, by accepting (8), the justificatory status of Peter's seeming is under question. It is hard to see why the compelling nature of Peter's seeming should tip the scales in his favour given the epistemic stand-off with Khadijah, especially if he thinks that Khadijah's seeming can be just as powerful as his own, if not more.

In the remainder of this section, I will consider one last strategy Peter might deploy in responding to (8), along the lines advocated by Andrew Moon (2021). Moon argues that a believer whose religious belief is prima facie justified and warranted can legitimately deflect potential defeaters of his or her belief by resorting to epistemically circular reasoning. Moon lays the foundation of his case by adopting Michael Bergmann's framework in thinking about epistemic circularity, which rests on several critical points. These include Bergmann's distinction between epistemically circular beliefs and epistemically circular arguments. With respect to belief, Moon explains, ‘S's belief that <some belief source X is reliable> is epistemically circular if S used X to come to believe that <X is reliable>’ (Moon (2021), 792). And, with an argument, ‘S's argument for the conclusion that <X is reliable> is epistemically circular if S used X to come to believe in one of the premises of that argument’ (ibid.). Then there is Bergmann's distinction between ‘malignant circularity’ and ‘benign circularity’. The former kind of circularity disqualifies a belief from being justified; the latter kind does not disqualify it as such (ibid.). According to Moon (again, following Bergmann), the circularity in using a deliverance of X to support the belief that X is reliable is malignant when the subject is or should be seriously questioning X's trustworthiness (ibid.). Moon then connects this point to defeaters, maintaining that S should seriously question X's trustworthiness if S has a defeater for thinking that X is reliable. In this case, using a deliverance of X to support belief in X's trustworthiness will result in malignant circularity (ibid., 793).

Following his endorsement of Bergmann's account of epistemic circularity, Moon extends it into the religious domain. He does so by introducing Plantinga's distinction (made after Plantinga's work in WCB) between a ‘defeater-defeater’ and a ‘defeater-deflector’. A defeater-defeater ‘nullifies the defeating power of something that is already a defeater’, whereas a defeater-deflector ‘prevents something from being a defeater in the first place’ (ibid.). Now suppose we have a hypothetical Christian believer, ‘Hannah’ as Moon calls her, for whom Christian belief is prima facie justified and warranted (without defeaters). Let ‘Christian belief*’ denote the proposition that ‘Hannah's Christian belief was formed reliably’. According to Moon, Hannah can reasonably deflect defeaters that purport to undermine the rationality of her belief that her Christian belief was reliably formed. She can do this by arguing for Christian belief* using the following argument, which I will call the ‘Deflector Argument’ or DA for short:

1. (I) Christian belief is true.

2. (II) If Christian belief is true, then Christian belief* is probably true.

3. (III) Christian belief* is probably true. (ibid., 793–794).

If we grant the assumption that Hannah's Christian belief is prima facie justified and warranted (without defeaters), says Moon, then we can say that Hannah's belief in (I) is at least initially justified. Hannah's belief in (II) can also be justified given Plantinga's argument that connects the truth of Christian belief to its probable warrant, alongside her observation that her Christian belief was formed in a way that is not unusual for Christian believers (hearing the preaching of the gospel at church and then forming a religious conviction in response). It is reasonable for Hannah to think that the circumstances in which she formed her Christian belief are probably those in which it would be reliably formed through the work of the Holy Spirit (ibid., 794). From (I) and (II), Hannah validly concludes (III).

Hannah's use of the DA involves epistemic circularity, as Moon makes clear:

[Hannah] formed Christian belief on the basis of some process: P. Then, using Christianity as a premise, she inferred claims about the Holy Spirit's active role in the production of her belief, which led her to believe that P is probably reliable. So, a doxastic deliverance of P played a role in justifying her belief in the probable reliability of P. (ibid., 795)

Nevertheless, avers Moon, the epistemic circularity involved in Hannah's use of the DA is not malignant. Using a deliverance of X to support the belief that X is reliable is malignant, recall, when a subject is or should be seriously questioning X's trustworthiness. Moon explains why these conditions are not met in Hannah's case:

We can simply stipulate that Hannah is not seriously questioning or doubting Christian belief*. Should she be seriously questioning or doubting it? The only plausible reason to think that is if she has some good reason to: a defeater. But in this case, Hannah can use her modus ponens argument to prevent herself from ever gaining a defeater in the first place. (ibid.)

Moon takes Christianity as an ‘epistemically self-promoting proposition’ for someone like Hannah. By justifiedly believing it, she gains good evidence that her belief is reliably formed (ibid.). This is because the truth of Christianity has implications for how the Holy Spirit reliably guides believers like Hannah to form their Christian beliefs (ibid., 797). Christianity is also an ‘epistemically others-demoting proposition’, as its truth has additional implications for how some other people's disbelief in it is, because of sin, unreliably formed (ibid., 802–803). From her Christian vantage point, argues Moon, Hannah can deflect potential defeaters from religious disagreement by using both the epistemically promoting and demoting aspects of Christian belief. Hannah's reasoning, in schematic form, has this general structure comprised of three conditions:

Suppose S believes p and T believes ~ p.

Condition 1: S has prima facie justification to believe p and has no defeaters (independent of the proposed defeater from disagreement).

Condition 2: S is justified in believing that if p, S's belief that p is probably reliably formed.

Condition 3: S is justified in believing that if p, then T's belief that ~ p is formed unreliably. (ibid., 803)

This, in summary, is Moon's case for thinking that some religious believers can use epistemically circular reasoning to ward off potential defeaters, including defeaters that stem from religious disagreement. I will defer a complete evaluation of it for some other time. For now, let me offer a few critical remarks that I think are sufficient to show why it fails to offer Peter a way out of his epistemic stand-off with Khadijah.

The first thing to note is that Moon's strategy, even if viable, applies only to believers who do not currently have defeaters for their religious belief or any other reason that should cause them to seriously question the trustworthiness of the processes that form it. Hannah, as Moon stipulates, does not have defeaters for thinking that her Christian belief was reliably produced; she is not and should not (given this stipulation) be seriously questioning it. By contrast, Peter does have defeaters for his Christian belief, as I have explained in my scenario; he is and should be seriously questioning it. Following Quinn, I've cast Peter as an ‘intellectually sophisticated adult theist’. As Quinn explains, this sort of theist does indeed have defeaters for his or her theistic belief and requires defences against them (Quinn (1985), 484). The Diderot Objection, as I have been arguing, constitutes one such defeater. Because of this, Moon's strategy is nugatory in the dialectical context involving Peter's reflective concerns about his Christian belief given the Diderot Objection.

Let me set aside this problem, however, and assume for the sake of argument that Peter is in the same position as Hannah. Suppose we grant that Peter's Christian belief is prima facie justified and warranted (without defeaters). Can Moon's strategy now offer Peter a reasonable way to resolve his stand-off with Khadijah in his favour? I think not. Here's why. The general structure of Hannah's reasoning, set out by Moon as resting on Conditions 1–3, can indeed be met by a reflective Muslim like Khadijah (see e.g. Turner (2022), 135–139). Upon recognizing this feature of his epistemic stand-off with Khadijah, what is Peter to do? Moon offers two options for attempting to resolve the stand-off. One option is for Peter and Khadijah to consider arguments for the relevant de facto claims. Peter may offer Khadijah a historical argument for the resurrection of Jesus. Khadijah, on the other hand, may offer an argument that Christianity entails some logical inconsistency. The idea here is that a successful de facto objection to Christianity or Islam can overturn the prima facie justification (specified in Condition 1) held by either Peter or Khadijah (Moon (2021), 806). A second, non-rational option is for Peter and Khadijah to consider alternatives to rational arguments, such as prayer. Peter may pray that Khadijah sees the truth about Christianity. Khadijah may do the same, praying for Peter to see the truth about Islam (ibid.).

If these strategies are unsuccessful, Peter and Khadijah, as Moon notes, ‘might see that they are in a standoff and wonder, how do I know that I'm the one who's in the right? To answer that, they might think they are just as likely to be reliable as the other and so be moved toward agnosticism’ (ibid.). Despite conceding that there is an ‘intuitive pull’ to say that, in the situation being considered, Peter has a defeater for his Christian belief, Moon resists the inference to this conclusion. Given a stand-off, people in Peter's and Khadijah's situation, he writes,

can easily answer the question that they are in the right. They have a justified belief that their belief was formed reliably! And they have a justified belief that the other belief was formed unreliably. That's how they can justifiedly believe that they are in the right and not the other. They might grant that, from the other person's perspective, they can all make similar moves as them, but they have justification for thinking that the other person's moves are unreliable. So, there really isn't a good reason to think that either side should have to withhold their religious belief. I take this to be an instance of reasonable religious disagreement. (ibid.)

In his discussion, Moon doesn't clarify whether this is an instance of reasonable disagreement after both sides have engaged in sustained discussion and debate about the truth of religious claims. If this is something they should do, given an initial stand-off, then Moon's assessment here about the reasonableness of the religious disagreement seems plausible. One can think of several relatively non-controversial cases where two individuals are equally reasonable in ‘agreeing to disagree’ after protracted discussion and debate (e.g. regarding certain political matters). Each may walk away after the exchange, reasonably thinking that he or she is in the right while the other one isn't. This way of understanding Moon's position, however, has a problematic consequence.

If Peter, given his stand-off with Khadijah, is required to engage with arguments for the truth of Christian belief in order for his belief to be ultimately rational, then his posture will be reliant on the sort of ‘crypto-evidentialism’ that concerned Craig (see my earlier discussion). It will, moreover, render Moon's defeater-deflector strategy useless for him. Suppose that Khadijah presents Peter with a defeater for his Christian belief by offering a de facto objection to Christianity. If he is required to defeat (not merely deflect) this objection, then he cannot permissibly use the truth of Christianity to defeat this defeater, as Moon acknowledges (ibid., 795).

On the other hand, if Moon means to say that Peter and Khadijah's disagreement, given the stand-off, just is a case of reasonable disagreement despite the absence of any serious discussion or debate about the truth of religious claims, this will strike many (including myself) as implausible. Consider, then, the second horn of the emerging dilemma for Peter. If Peter, given his stand-off with Khadijah, is not required to engage with arguments for the truth of Christian belief in order for his belief to be ultimately rational, then his posture seems unable to respond to concerns that his defeater-deflector strategy is meretricious and dogmatic (and especially so if, unlike Peter, Khadijah insists that one ought to discuss and debate to at least attempt a resolution to the stand-off). Moon is aware that these concerns may be raised against his position:

According to [this objection], my strategy makes it impossible for Hannah's Christian belief to ever get defeated since she can always flat-footedly appeal to the Holy Spirit any time a defeater looms. Hannah is thereby rationally sealed from open-mindedness about the possible falsity of her belief. This indicates that the strategy is problematic. (ibid., 798)

In response to this objection, Moon suggests that Hannah can still gain defeaters by considering de facto objections to Christian belief (ibid.). But this seems to be effectively giving the nod to a no-defeater requirement by saying that, in order to avoid this horn of the dilemma (dogmatism), one must accept the other one (crypto-evidentialism). Given its lack of a clear resolution to the dilemma I've presented, Moon's defeater-deflector strategy seems to be of no avail to Peter in warding off the threat posed by (8):

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

Reflective Plantingian Christians who agree with the line of reasoning that I have presented in (1)–(8), my discursive articulation of the Diderot Objection, have an undercutting rationality defeater for their Christian belief. In the next section, I will assess Plantinga's attempt to resist this sort of objection in his discussion of defeaters and religious pluralism in WCB.

Ad Plantinga: Defending the Diderot Objection

In WCB, although Plantinga does not deal specifically with the Diderot Objection, he nevertheless tackles a more abstract formulation of the problem in considering ‘whether a knowledge of the facts of [religious] pluralism constitutes a defeater for Christian belief’ (WCB, 438). Suppose that, in my example of the two reflective Plantingian believers Peter and Khadijah, Peter continues to believe that Jesus is God and thinks that this belief is warranted for him while also believing (8) is true (which, recall, he arrived at following a process of reflection):

1. (8) Khadijah's belief that p is not warranted, and yet Khadijah is Peter's Plantingian religious peer.

Peter should then be classified as an exclusivist, as his situation satisfies the three essential elements of exclusivism as stipulated by Plantinga: (1) he holds the tenets of one religion (Christianity) as true and rejects propositions incompatible with those tenets as false (e.g. p, which Khadijah believes is true and warranted for her); (2) he is fully aware of other faiths (e.g. Islam) and has noted that ‘the adherents of other religions sometimes appear to display great intelligence, moral excellence, and spiritual insight’ (WCB, 440); and, (3) he does not have a rationally convincing argument for the belief with respect to which he is an exclusivist (e.g. his belief that Jesus is God) that can be counted on to convince those who disagree with him (WCB, 440–441). Put in the context of how Plantinga frames exclusivism, the important question then is this: can Peter rationally continue to be an exclusivist given his belief in (8)?

Prior to his attempt to defend an affirmative answer to this question, Plantinga offers a generous concession by allowing the case under consideration to be one where the two believers who disagree – Peter and Khadijah, in my example – have similar sorts of internal markers for their respective convictions. More specifically, he grants that there can be an internal epistemic parity between the two, which for each believer includes propositional evidence for the belief and the phenomenology connected to it (WCB, 451–452). Nevertheless – and here is Plantinga's defence – someone like Peter can insist on epistemic privilege on his part. How so? In this way:

[T]he believer in question doesn't really think the beliefs in question are on a relevant epistemic par. She may agree that she and those who dissent are equally convinced of the truth of their belief, and even that they are internally on a par, that the internally available markers are similar, or relevantly similar. Still, she must think that there is an important epistemic difference: she thinks that somehow the other person has made a mistake, or has a blind spot, or hasn't been wholly attentive, or hasn't received some grace she has, or is blinded by ambition or pride or mother love or something else; she must think that she has access to a source of warranted belief the other lacks. If the believer concedes that she doesn't have any special source of knowledge or true belief with respect to Christian belief – no sensus divinitatis, no internal instigation of the Holy Spirit, no teaching by a church inspired and protected from error by the Holy Spirit, nothing not available to those who disagree with her – then, perhaps, she can properly be charged with an arbitrary egoism, and then, perhaps, she will have a defeater for her Christian belief. But why should she concede these things? (WCB, 453)

This defence rests on two interrelated components. The first is an appeal to what McKim calls ‘discrediting mechanisms’, which are

techniques that are used to discredit or explain away the views of others, typically by imputing a defect of some sort to those who hold them. The defects that are imputed to others are of different sorts. The familiar ones include, for instance, an inability to see beyond class interests, lack of imagination, carelessness, intellectual cowardice, intellectual conformity, wishful thinking, stubbornness, and sin. (McKim (2001), 135–136)

The second component involves Plantinga's appeal to Divine Guidance (specifically the testimony of the Holy Spirit) that constitutes a special source of knowledge (WCB, 454). These two components are interrelated because, in Plantinga's view, it is primarily the defect of sin and its cognitive consequences that create a significant epistemic divide between believer and unbeliever. Due to the ‘noetic effects of sin’, the unbeliever is unable to recognize important truths about God (WCB, 213–218).

Might Plantinga's epistemic privilege defence offer a plausible way for Peter to nullify the force of (8) as a potential defeater and thereby allow him to rationally affirm his exclusivist stance? I don't think so. To begin with, the threat posed by (8) need not be based on thinking that Plantinga's A/C models are false, as Plantinga is concerned to point out. Rather, the worry for Peter involves securing a rational basis for his confidence that he is warranted in his belief, whereas Khadijah is not, even though there is internal epistemic parity between himself and her. If Khadijah can have internal markers like Peter's for her conviction, markers that are compatible with her religious delusion (according to Peter) about p being true and warranted, why should Peter hold that he is not similarly deluded or some such thing?

To circumvent this worry, Plantinga submits that someone like Peter can appeal to external factors to account for the difference between himself and Khadijah such that she is not his epistemic peer. Or, to put it another way, Peter can maintain that, insofar as internal epistemic parity is concerned, Khadijah is his Plantingian religious peer. When it comes to external epistemic parity, however, she is not. To account for this difference, Peter can appeal to discrediting mechanisms. This, I think, is the critical element in Plantinga's epistemic privilege defence on which it succeeds or fails. How reasonable, then, is it for Peter to use such discrediting mechanisms in defending his exclusivist perspective? Plantinga does not really answer this question in WCB apart from a brief mention of it in the abstract. While I can't consider here all the various concrete ways in which Plantinga's move can and has been applied to Muslims and Islam, let's consider some samples to get a real feel for how it may be done and to gauge its plausibility. I propose that we reflect on some instances in Aquinas and Calvin, who do not shy away from telling us what they really think about Muslims. In his Summa contra Gentiles, Aquinas ([1259–1265] 1955–1957) asserts that the message of the Prophet Muhammad was believed only by carnal, ignorant, brutish, and subdued men (1.6.4). On this point, Calvin seems to concur, extending Aquinas' assessment of the early Muslim community to the Turks (Muslims) of his time:

Mahomet [Muhammad] has reported himself to be the party that should bring the full revelation – over and besides the Gospel. And by means thereof, they [the Islamic Turks] have utterly become brute beasts . . . At this day, we see that those poor beasts busy their heads about as doltish and unsensible things as any can be. But it is the just vengeance of God, Who has given them over to a willful stubborn mind! (as quoted by Lee (2000), 7)

Putting aside the question of how reasonable it was for Aquinas and Calvin to have such views about Muslims during their time, it should be clear to us today that they constitute obviously implausible discrediting mechanisms of all Muslim belief. By ‘us’, I mean not just individuals who are impartial in the debates between Christians and Muslims but Christians as well. In the Second Vatican Council's statement on Islam and Muslims, for instance, we find this:

The Church has also a high regard for the Muslims. They worship God, who is one, living and subsistent, merciful and almighty, the Creator of heaven and earth . . ., who has spoken to men. They strive to submit themselves without reserve to the hidden decrees of God, just as Abraham submitted himself to God's plan, to whose faith Muslims eagerly link their own. (Second Vatican Council, Nostra Aetate 3, 1965; emphasis mine)

Quite the contrast with Aquinas and Calvin. But why do most of us, including many Christians, regard the Vatican Council's statement on Muslims as eminently more plausible than what Aquinas and Calvin had to say about them? Part of the reason, I submit, is because of our increased awareness of and contact with members of different world religions, including Islam. It is precisely this experience that was a key motivating factor for John Hick in thinking about and constructing his model of religious pluralism. In his autobiography, Hick discusses this experience with respect to his interactions with different religious communities in Birmingham, UK. Although there were obvious external differences among them, he observed something that was, in his view, shared by people in these communities:

[A]t a deeper level it seemed evident to me that essentially the same thing was going on in all these different places of worship, namely men and women were coming together under the auspices of some ancient, highly developed tradition which enables them to open their minds and hearts ‘upwards’ towards a higher divine reality which makes a claim on the living of their lives. (Hick (2002), 160)

One need not be a Hickian pluralist to appreciate and perhaps even accept Hick's insight here. My suspicion is that it is at least partly based on such an insight that Plantinga allows his conception of the exclusivist to have the feature of noting great intelligence, moral excellence, and spiritual insight among members of non-Christian religions. But if Peter is this kind of an exclusivist, it seems difficult for him to defend the reasonableness of deploying some of the discrediting mechanisms suggested by Plantinga. Consider, for example, how incongruent, jarring, and ad hoc formulating this description of Khadijah (or some approximation of it) may seem to him:

Khadijah is my Plantingian peer, just as convinced of her belief that Jesus is merely a human prophet as I am that he is God, has the same kind of evidence for her belief as I have for mine, seems to be a very intelligent, morally excellent, spiritually insightful woman but . . . her being entrenched in the basic and aboriginal sin of pride is blocking her from receiving the grace of the Holy Spirit!¹²

Plantinga himself never offers such concrete indictments of Muslim belief, to be fair. Still, such a dismissal of Khadijah is in perfect harmony with his account of epistemic disparity between believer and non-believer. It is perhaps for this reason that a Plantingian like Craig has no reservations in saying that Muslims (among other non-Christians) are ‘kicking against the Holy Spirit and his witness to the truth of Christian theism’ (Craig (2014)). The difficulty facing Peter in attaching himself to such a dismissal of Khadijah is that the grounds on which he affirms her laudable qualities are, generally speaking, the same grounds that detract from the discrediting mechanism being utilized (Peter's contact and interaction with Khadijah). If the defect of pride significantly permeated Khadijah's character, why regard her as someone who is morally excellent and spiritually insightful? Conversely, if Peter does regard Khadijah as morally excellent and spiritually insightful, why think that she is mired in pride?

One last point on this matter. As part of his overall consideration of Plantinga's epistemic privilege defence, someone in Peter's position may find it helpful to reflect on how plausible he would find a tu quoque retort from someone like Khadijah. Suppose that Khadijah takes a page out of the writings of Ibn Qayyim al-Jawziyya (d. 1350) and affirms that, despite being her Plantingian peer and internal epistemic parity existing between their beliefs, Peter (and Christians like him) unfortunately suffer from ignorance, envy, pride, fear, a perverse fascination with the irrational, and remain in their religion out of sinfulness (Hoover (2010)). She goes on to explain that, by God's Grace, she has managed to avoid these pitfalls and enjoys warrant for the full panoply of Islamic belief, including her belief that Jesus is merely a human prophet.

Since Plantinga's appeal to discrediting mechanisms does not, at least as it stands, deal adequately with such hard problems facing his exclusivism, this essential aspect of his epistemic privilege defence, and ipso facto the defence itself, fails. I conclude that (8) remains intact as an undercutting rationality defeater for Peter's belief that Jesus is God and that the Diderot Objection to Plantinga's RE stands.

Conclusion

In this article, I have shown how a specific form of the problem of religious diversity, what I have called the ‘Diderot Objection’, presents a difficulty for reflective Christians who endorse Plantinga's RE. If indeed reflective Plantingian Muslims can, in good faith, deploy Plantinga's RE in an Islamic context, this results in a defeater for Christian belief for such Christians who are aware of this. The attempt to reject this defeater by claiming that reflective Muslims are not peers with reflective Christians on religious matters is not a plausible one, especially when considered in concrete circumstances involving specific Muslim individuals. I will finish with a few final remarks.

If, as I maintain, the Diderot Objection to Plantinga's RE succeeds, then we have a successful de jure objection to Christian belief that is independent of a de facto one; this is because there is nothing in my formulation and presentation of this objection that requires thinking Christianity is false. The Diderot Objection raises worries about the epistemic, not alethic, status of Christian belief. While I believe this objection raises significant problems for Plantinga's RE, I do not think it entails that Plantinga's project collapses entirely or that it has no use in defences of religious belief. On the contrary, I hold that the concerns raised in my discussion can be overcome by a reasonable extension of RE in the direction of a moderate fideism (quite distinct from the irrationalist fideism Plantinga rightly rejects).¹³ Finally, I want to make it clear that, by pressing the Diderot Objection against Plantinga's RE, I am not claiming that Islamic belief and an Islamic version of Plantinga's RE are epistemically superior to Christian belief and Plantinga's Christian version of RE. In my view, a Christian version of the Diderot Objection applies with identical force to any Islamic RE that parallels Plantinga's.