From his initial book on God and Other Minds in 1967 to his monumental Warrant and Christian Belief in 2000, Alvin Plantinga has defended theism by lodging a circumstantial ad hominem objection against his nontheist opponents in which it is argued that they uphold epistemic standards for theistic belief that their own nontheistic beliefs fail to satisfy. It is widely assumed, as part of our Lockean legacy, that a belief that God exists can be epistemically rational, justified, or warranted only if it has adequate evidential support from beliefs that are either self-evident or evident to our senses. Without suitable argumentative support, theistic belief fails to measure up to proper epistemic standards and thereby violates our epistemic duties. Alvin Plantinga, who accepted this evidentialist assumption in his writings prior to the early 1980’s, mounts a vigorous attack on it in Warranted Christian Belief.[2] He had argued with considerable force in Warrant: the Current Debate and Warrant and Proper Function that what warrants a “basic belief,” a belief that is not based on or inferred from another belief, is that it results from the proper functioning of one’s cognitive faculties in the right kind of epistemic environment according to a design plan successfully aimed at truth.
Plantinga begins with basic beliefs that arise from our senses, memory, introspection, sympathy, and a priori reason, which comprise the “standard package” of cognitive faculties. He makes out a powerful case in these two earlier Warrant books that such beliefs are warranted when the faculty that produces them is functioning properly in the right sort of epistemic environment according to a design plan aimed at seeking truth. Someone who seems to see a tomato and then believes that there is a tomato out there has a warranted belief and moreover knows that there is a tomato out there. Her warrant for believing this, however, is subject to defeaters or overriders concerning something that is abnormal about her faculty of vision (she has cataracts) or the epistemic circumstances (she is in a factory that manufactures plastic tomatoes). For the sake of argument, this account of warrant will be accepted.
The next step in Plantinga’s argument is to show that it is possible that theistic, and in particular Christian, belief have warrant in an analogous way to that in which sensory and memory beliefs, etc., do. If theism is true, then God would want to reveal himself to created persons. Toward this end he implanted in them as part of their original cognitive equipment, along with the cognitive faculties in the standard package, a sensus divinitatis that would enable them to form true noninferential beliefs about God’s presence, nature and intentions upon having certain experiences, such as reading the Scriptures, hearing the choir sing, seeing a beautiful sunset, feeling guilt, and so on. Provided their sensus divinitatis is functioning properly on these occasions in accordance with its divinely determined design plan in the right sort of epistemic environment, their basic beliefs are warranted and constitute knowledge even if the subjects of the experiences are unable to offer any argument or justification for their beliefs. That they have such a noninferential warrant does not preclude them also have an evidentialist-based warrant: Plantinga is no fideist.
Plantinga also introduces a special supernatural process involving the internal instigation of the Holy Spirit by which one is directly caused by God, without any intervening worldly causes, to believe the great things of the Gospel concerning the incarnation, resurrection, salvation, and the like. Plantinga does not argue that these people are in fact warranted in their basic beliefs, only that it is possible that they are. To do the former would require giving evidence or arguments that God exists and has set things up the way in which Plantinga’s so-called Aquinas/Calvin model (A/C for short) says that he has.
This is a bare bones sketch of Plantinga’s so-called Reformed Epistemology and does not do justice to its rigor, depth and brilliance. Three objections will be made to it, which will be called the disanalogy problem, the relocation problem, and the immorality of exclusivism objection.
The analogy between the sensus divinitatis and the cognitive faculties of the standard package runs throughout Warranted Christian Belief and is central to the case that Plantinga constructs for it being possible to have a warranted basic belief in God.[3] The following quotations give some indication how Plantinga draws and uses the analogy.
In this regard [with respect to basicality], the sensus divinitatis resembles perception, memory, and a priori belief. (175)
Due to one cause or another, the faculty itself [the sensus divinitatis] may be diseased and thus partly or wholly disabled. There is such a thing as cognitive disease; there is blindness, deafness, inability to tell right from wrong, insanity; and there are analogues of these conditions with respect to the operation of the sensus divinitatis. (184)
Failure to believe can be due to a sort of blindness or deafness, to improper function of the sensus divinitatis. (186)
And the first thing to see is that, on this model, faith is a belief-producing process or activity, like perception or memory. (256)
Faith, the whole process that produces them, is specifically designed by God himself to produce this very effect – just as vision, say, is designed by God to produce a certain kind of perceptual beliefs. (257)
Presumably one wouldn’t want to say that perceptual beliefs get warrant from experience only if there is a good (noncircular) argument from the existence of perceptual experience to the truth of perceptual beliefs; if not, however, what is the reason for saying it in the case of theistic or Christian belief? (329)
Why suppose that if God proposes to enable us to have knowledge of a certain sort, he must arrange things in such a way that we can see an argumentative connection between the experiences involved in the cognitive processes he selects and the truth of the beliefs these processes produce? That requirement is entirely gratuitous and also false, since it doesn’t hold for such splendid examples of sources of knowledge as perception, memory, and a priori intuition. (331)
The final two quotations make explicit Plantinga’s circumstantial ad hominem objection to his nontheist opponents – that they set higher epistemic standards for a justified theistic or Christian belief than they do for beliefs based upon perception and memory. It will be argued that they are not guilty of having a double standard, since Plantinga’s analogy fails badly in cognitively relevant respects, thereby permitting them to make invidious distinctions between the epistemic credentials of standard package and sensus divinitatis induced beliefs.
The heart of the analogy is that we can predicate of both types of experience the notion of being produced by a cognitive faculty that is “functioning properly,” as contrasted with one that suffers from a “disease,” “dysfunction,” “malfunction,” “pathology,” or “disorder.”[4] A dilemma argument can be constructed in regard to the predication of these terms. Either they are supposed to be predicated in the same sense of both theistic and standard package beliefs or they are not. On both alternatives Plantinga’s argument for the possibility of theistic and, in particular, Christian belief being warranted fares badly.
If Plantinga assumes that they are predicated in the same sense, he winds up with a false analogy. For there are agreed upon objective tests for a cognitive faculty in the standard package being in a state of dysfunction, malfunction, pathology, or disorder. But it is obvious that there are no agreed upon objective tests for a person’s sensus divinitatis suffering from a dysfunction, malfunction, pathology, or disorder. It will not do to charge this objection with resting on an unacceptable verificationist requirement and then have verificationism die from Plantinga’s favorite death of self refutation when it is required that it be applied to itself. The point of the objection is not that every type of cognitive experience must admit of a distinction between proper and improper functioning that measures up to verificationist standards, only that Plantinga’s analogically-based argument commits him to this being so for his sensus divinitatis since it is true of the cognitive faculties in the standard package.
In regard to basic religious beliefs that are internally instigated by the Holy Spirit, it is obvious that the notion of proper functioning could have no application to them since they are supernaturally caused directly by God. Such instigation, furthermore, is not a faculty but a process and thus cannot be said to have any function and therefore cannot be said to malfunction or be subject to a pathology; for there is no correct way for God to supernaturally cause worldly occurrences. Plantinga recognizes this difficulty: “A caveat: as Andrew Dole points out in ‘Cognitive Processes, Cognitive Faculties, and the Holy Spirit in Plantinga’s Warrant Series’ (as yet unpublished), it is not obvious that one can directly transfer necessary and sufficient conditions for warrant from beliefs produced by faculties to beliefs produced processes.” (257) Plantinga gives no response to this caveat, nor do I think one can be given.
Plantinga continually talks about the sensus divinitatis in natural law terms; but, whereas for Aristotelian natural law theorists questions concerning an individual’s nature and proper mode of functioning are to be answered, at least in part, by empirical inquiry, there is nothing analogous in regard to determining the nature and proper functioning of the sensus divinitatis or for what constitutes a proper way for the internal instigation of the Holy Spirit to occur.
There are further damaging disanalogies between Plantinga’s A/C experiences and those in the standard package. Whereas there is universal participation in the very same doxastic practices based on the experiences in the standard package, this is not so for sensus divinitatis based experiences. Plantinga has an explanation for this disanalogy based upon the serious damage that the sensus divinitatis suffered as a result of Original Sin, a damage that is repairable only by the supernatural intervention of the Holy Spirit. But to explain why there is this disanalogy does not explain it away.
Another disanalogy is that there is no standard package analogue to religious diversity. There is widespread disagreement among persons of different religions in regard to how they respond to reading the New Testament in that only some find themselves suddenly believing that a triune God exists who has atoned for our sins. In contrast, persons have pretty much the same doxastic responses to their standard package experiences.
Yet another disanalogy concerns the ways in which these two different belief-forming faculties can be shown to be reliable. In basic agreement with William Alston, Plantinga argues that neither perceptually-based beliefs nor sensus divinitatis-based ones can be shown to be reliable, that is, to produce true beliefs for the most part. The reason is that any argument for either’s reliability will suffer from
epistemic circularity, a malady from which an argument for the reliability of a faculty or source of belief suffers when one of its premises is such that my acceptance of that premise originates in the operation of the very faculty or source of belief in question. If you give an epistemically circular argument for the reliability of a faculty, then you rely on that very faculty for the truth of one of your premises. (119)
Pace both Alston and Plantinga, it can be shown that it is possible to give an argument for the reliability of our faculty of sense perception, but not for the sensus divinitatis faculty, that avoids epistemic circularity.
Lets begin with an attempt to show via an inductive argument that our perceptual experiences are reliable in general. Let n be the number of known perceptual experiences. For each of these n experiences there are agreed upon objective tests that can be appealed to determine whether it is veridical. These tests consist, in part, in agreement among observers, predicative success, and, most important, being caused in the right way, which involves normalcy in the observer and the medium. If it turns out that the vast majority of these n experiences satisfied these tests for veridicality, it can be inductively inferred that the vast majority of our perceptual experiences are veridical and thus that our perceptual experiences are reliable in general. There is no analogous inductive argument for the reliability of the sensus divinitatis faculty, since there are no agreed upon objective tests for determining when experiences induced by this faculty are veridical.[5]
Plantinga and Alston will immediately respond that my inductive argument for the reliability of our senses suffers from epistemic circularity. This is because the tests for veridicality that it makes use of are themselves based on perceptual experiences and thus can be known to be satisfied by some perceptual experience only if it is assumed that perceptual experience is reliable in general. It is true that we test the veridicality of a given perceptual experience by appeal to other perceptual experiences, but this does not involve epistemic circularity. The reason is that there is no need to assume that the latter are veridical, since what is in question with these tests is coherence among perceptual experiences. Thus the premises that make up my inductive argument can be known without assuming the conclusion – that perceptual experiences are reliable in general.
The mutivocalist horn of the dilemma fares no better than does the univocalist one. Plantinga now is to say that these terms are predicated with a different sense of the two kinds of basic beliefs. Herein Plantinga cannot make use of the results established by his two earlier Warrant books in his argument for the possibility of having a warranted basic belief that God exists. Thus, his basic A/C experiences can be said to admit of the dysfunction-proper function distinction just as do standard package beliefs, only the tests for the former will be radically different from those for the latter. Whereas the latter are based on empirical tests that have a grounding in what is vouchsafed by science, the latter will be based on criteria that are internal to the different religious doxastic practices and thus will vary across these practices, thereby posing the problem of religious diversity. And this is language-game fideism, a most vile doctrine and one which Plantinga’s ardent theological realism rightly rejects.[6]
Plantinga has made the bold claim that if theism is true, then something like his account of warranted basic theistic belief is true. And, on the basis of this conditional proposition, he infers, by the law of modus tollens, that a denial that basic theistic belief is warranted in pretty much the same way as his model depicts, also denies the truth of theism.[7] The de facto issue of the truth of theism, therefore, cannot be separated, as many have claimed, from the de jure issue of the warrant for theistic belief. If our objections to Plantinga’s model are justified, then the modus tollens refutation of theism fails, since the conditional proposition in question is false. If theistic and, more specifically, Christian beliefs are too be warranted, they must be warranted in some manner that is radically disanalogous to the way in which standard package beliefs are. This robs Plantinga of his circumstantial ad hominem response to his atheistic opponents.
The Relocation Problem
Philip Quinn has argued that for intellectually sophisticated adults in our culture there are defeaters for Plantinga-type basic beliefs in God consisting in the prevalence of apparently unjustified evils and certain naturalistic explanations for religious beliefs. While Plantinga reject’s Quinn’s claim that there are defeaters for his basic beliefs in God, he agrees with him to this extent: There are at least potential defeaters that Quinn’s “intellectually sophisticated adults in our culture” must neutralize if their basic belief in God is to count as rationally justified. And because they feel challenged by them, they wonder whether they really are warranted in having their A/C-based beliefs and cannot rest in these beliefs until that they have defeated the defeaters. Plantinga himself recognizes the need to defeat these potential defeaters, since, over and over again, he says that the theist’s basic belief in God is internally rational only if they have done the best they can to think long and hard about all the issues surrounding their belief, which includes defeating these defeaters.
Internal rationality…requires…that you have done your best or anyway well enough with respect to the formation of the belief in question. You have considered how it fits in with your other beliefs, engaged in the requisite seeking for defeaters, considered the objections that you have encountered, compared notes with the right people, and so on. Clearly, on the model [the A/C one]…someone who accepts the Christian beliefs in question can easily meet these conditions….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. 255
Thus it turns out that for Quinn’s scientifically sophisticated adults in our culture to have warrant for their basic belief in God they must have arguments against the potential defeaters. Does this requirement clash with Plantinga’s repeated claim that a basic belief in God can be warranted even if the believer is unable to give any argument for her belief? Has Plantinga merely relocated the point at which the basic believer must appeal to arguments or evidential support, thereby creating the relocation problem? Plantinga’s response to this problem is that his basic believer need not give any argument or evidence for the existence of God and for his having set things up the way his A/C model says he has in order to defeat the potential defeaters. Plantinga’s attempts to defeat Quinn’s defeaters based upon apparently unjustified evil and certain naturalistic explanations of religious belief seem, at first blush, to be a good case in point. The problem of apparently unjustified evil can be neutralized by showing that there is too big an epistemic gulf between us and God for us to be able to fathom God’s reasons for permitting evil. And the challenges posed by those naturalistic explanations that are defeaters for religious belief, such as those of Marx and Freud, can be defeated by showing that they presuppose the falsity of theism and moreover have little evidential support in their own right. Thus, it appears that Plantinga’s relocation of the point at which his basic believers must be prepared to give arguments in support of their belief in God does not violate his claim that they can be warranted in their beliefs even if they cannot give any arguments or evidence for God’s existence and for his having set things up the way his A/C model says he has.
Plantinga’s epistemic gap solution to the problem of evil would escape the relocation problem if it were acceptable. But it isn’t. His solution has the unwanted consequence that we should not be in a better position to determine that the goods of the world attest to God’s existence rather than being necessary for the realization of an outweighing evil by some malevolent deity. Epistemic gap theology, if consistently applied, also has the consequence that we cannot ascertain what God’s purposes are and thus cannot determine how we should conduct our religious life. The proper answer to Plantinga’s rhetorical question, “Given that God does have a reason for permitting these evils, why think we would be the first to know?”, is that we should be the first to know, at least for those evils that directly affect us. For in a communal love relationship, each party should give the other some reason why she allows evils to befall the other.
In order to neutralize the problem of evil, Plantinga’s basic believer must construct a theodicy. Herein evidence must be given that the envisioned scenario in which God has an excuse for permitting the known evils actually obtains. But this alone doesn’t create the relocation problem, since supposedly these excusing conditions can be discovered without having to establish that God exists. For example, it can be empirically determined that an evil proved to be necessary for the realization of an outweighing good. The problem is that for many known evils this cannot be done. At best a defence can be given for God’s permitting them in which it is shown that it is logically possible that God have a morally exonerating excuse for permitting them. But this will not serve to defeat these defeaters for scientifically sophisticated in our culture.
Given our inability to show that these evils in fact are in fact justified, it seems clear that the only way a theodicy can succeed is if in our background knowledge we have good arguments for the existence of God. The reason is that just how strong the evidence for a theodicy must be will depend upon whether we have in our background knowledge, and, in particular, whether it contains good arguments for the existence of God. This is especially evident with the “greater good” theodicy wherein there is a need to appeal to an a heavenly after-life compensation. And herein the relocation problem breaks out, for it turns out that Plantinga’s basic does need evidence, after all, for the existence of God. Having Plantinga type A/C based beliefs in one’s background knowledge will not do for lowering the epistemic bar for a theodicy, since, as argued previously, such experiences could not be warranted in the way Plantinga would have them be.
Some A/C based beliefs, especially those caused by the internal instigation of the Holy Spirit, are exclusivist in the sense that they have the consequence that all religions other than Christianity are mistaken in some of their fundamental creedal beliefs, especially in respect to whether God is triune and that only those who believe this will find salvation. Plantinga attempts to protect these exclusivist Christian beliefs from the charge of being self-servingly arbitrary, egotistical, conceited, and arrogant. But these are mere personality flaws, and the real objection should be that it is immoral to have religious exclusivist beliefs. I will confine my objection to exclusivist beliefs that get promulgated, which is not much of a restriction, since such beliefs invariably find public expression. The immoral consequence of exclusivist religious beliefs will be the ultimate defeater of exclusivist beliefs based on A/C type experiences, especially those that involve the instigation of the Holy Spirit.
My moral objection to exclusivist religious beliefs borrows from William James’s will-to-believe doctrine. Let us grant for the sake of argument that epistemic reasons trump pragmatic and moral reasons. But when epistemic reasons are of no avail or cannot settle the issue, one’s belief should be based on pragmatic or moral reasons, or, at any rate, one is morally permitted to believe in such a case upon insufficient evidence. Since belief cannot be produced at will, my concern is with getting oneself to have a belief. Plantinga does not see how it can be shown that the epistemic reasons for exclusivist Christian beliefs are superior to those for the exclusivist beliefs of other religions, and thus clears the way for pragmatic or moral reasons to decide the issue of what one should believe, that is, try to get oneself to believe. The following is a will-to-believe type argument for not believing in a exclusivist religious proposition that one promulgates.
A Will-to-Believe Defeater Argument
1. For any proposition, p, if one cannot show that p is epistemically warranted and the consequences of believing p are worse than they would have been if one were not to have believed p, then one morally ought not to believe p. premise
2. A/C-based exclusivist religious beliefs cannot be shown to be epistemically warranted. premise
3. The consequences of believing A/C-based exclusivist religious propositions are worse than they would have been if one were not to have believed them. premise
4. One morally ought not to be believe A/C-based exclusivist propositions. From 1, 2, and 3 by modus ponens and universal instantiation
5. There is a defeater for A/C-based exclusivist religious propositions. From 4
This hardly is a decisive argument. Its premises call for further clarification and support.
I will make no attempt to defend premise 1, since I argue for it at great length in Chapter 4 of my book, The Divided Self of William James (Cambridge: Cambridge University Press, 1999). It is important to realize that epistemic warrant is a genus of which there are different species, two of which are evidential warrant and another basic warrant.[8] Premise 2 is highly controversial, since there are those, such as Swinburne, who argue that exclusivist Christian beliefs are epistemically warranted in the evidential sense. Plantinga, however, is not among them. Premise 3 does not admit of any straightforward verification. In the first place it is very difficult, if not impossible, to total up the goods and evils that have actually resulted from exclusivist religious beliefs. Contrasted with many of the evils that have resulted from religious exclusivism, such as holy wars, persecution, intolerance, and bigotry, are many goods. It is my gut intuition that the actual evils outweigh the goods. But the issue becomes even more cloudy when we consider what would have happened in the counter-factual situation in which there are no exclusivist religious beliefs. It is my gut intuition that things would have gone better in that counter-factual situation than they did in the actual world. But gut intuitions do not make for a decisive argument. Nevertheless, I think this Will-to-Believe Argument is a potential defeater that should worry any scientifically sophisticated adult in our culture who has a basic exclusivist religious belief. Plainly, this defeater needs to be defeated.
It must be stressed that my Will-to-Believe Defeater Argument has a limited target, being directed against only exclusivist religious beliefs that get promulgated. It would appear that, for Plantinga, these beliefs are based on the internal instigation of the Holy Spirit, the sensus divinitatus being the source of the generic religious beliefs. I would not want to argue for premise 3 when it is generalized to all religious beliefs, including nonexclusivist ones.
3’. The consequences of believing A/C-based religious propositions are worse than they would have been if one were not to have believed them.
Thus my moral defeater argument does not apply to those nonexclusivist religious beliefs based upon the sensus divinitatus, but only to the exclusivist religious beliefs based on the internal instigation of the Holy Spirit.
[1] This paper grows out of my participation in an Author Meets Critics session on Plantinga’s Warranted Christian Belief at the Pacific Division Meetings of the American Philosophical Association. His critical response to my comments clearly brought out some of the blunders and confusions in my comments. Needless to say, he is responsible for any blunders and confusions that remain in this paper. For if he were not to have written Warranted Christian Belief, these blunders and confusions would not have occurred.
[2] Oxford: Oxford University Press, 2000. All references to this book will be included in the body of the paper in parentheses.
[3] See especially pages 69, 70, 92, 97-8, 100-1, 106, 110, 132, 134, 146, 148, 174-5, 178-9, 181, 184, 186, 191, 244-6, 256-7, 262-4, 286, 300, 303, 305, 328-31, 334, 343, 372, 479, 485.
[4] See above quotations from pages 184 and 186. See also 100-1, 110, and 184.
[5] For further discussion of this see my following publications: On the Nature and Existence of God (Cambridge: Cambridge University Press, 1991), Chapter 8; “Why Alston’s Mystical Doxastic Practice Is Subjective,” Philosophy and Phenomenological Research, LIV (1994); “The Overall Argument of Alston’s Perceiving God,” Religious Studies, 30 (1994); “Swinburne on Religious Experience,” in Reason and the Christian Religion, ed. by Alan Padgett (Oxford: Oxford University Press, 1994); Critical Study of Keith Yandell’s The Epistemology of Religious Experience, Faith and Philosophy, 12 1995).
[6] See Plantinga’s trenchant objections to fideism in his “Reason and Belief in God,” in Faith and Rationality, edited by A. Plantinga and N. Wolterstorff (Notre Dame: University of Notre Dame Press, 1983), pp. 87-91.
[7] See especially pages 169, 188-191, 201, and 285.
[8] In my presentation at the aforementioned Author Meets Critics session I committed a horrible fallacy of equivocation in which I sloshed back and forth between the generic sense and one of its species -- the evidential one. I am deeply indebted to Plantinga for having pointed this out to me.