I entered Kenyon College at eighteen as an aspiring poet with a strong interest in literary studies but encountered philosophy during my sophomore year. The departments at Kenyon and at the University of Michigan where I received my graduate training were largely analytic. I continue to regard myself as an analytic philosopher. The only text in my first course was Plato's Republic, however, and Plato remains my favorite philosopher. What initially struck me, and still does, is the way in which he weaves together closely reasoned pieces of analysis with myth, story, and symbol. Although Plato clearly believed that the latter are needed to express some truths or aspects of the truth, he employed them only after the resources of argument had been exhausted. This continues to seem to me the best way to do philosophy. I never lost the conviction that poetry, myth, symbol, and story can express truths and insights that can't be adequately expressed in other ways. Yet philosophy taught me that when poetry, story, symbol, and myth aren't tethered to reasoned argument and careful analysis, one runs the danger of falling prey to what Coleridge called “fancy” and distinguished from imagination. The latter, in his view, answered to something in reality. The former does not. These convictions have led to and shaped the understanding of the nature, role, and importance of religious argumentation expressed in this book.

Two facts are patently clear. As we saw in Chapter 2, philosophically sophisticated arguments are frequently deployed by adherents of the world's major religious traditions. What is equally clear, though, is that few of these arguments are universally persuasive. Not only do Christian arguments generally fail to convince Buddhists or atheists, for example, they often fail to convince many of their authors own co-religionists. In spite of the fact that both were Christians, Pelagius's arguments failed to convince Augustine, and Augustine's arguments failed to convince Pelagius. Chapter 2 argued that one of the main reasons for this is the person-relativity of most arguments. Two related sources of the person-relativity of religious proofs were explored in Chapters 3 and 4. The first is the embeddedness of the most interesting religious arguments in rich textual traditions. The second is that one's assessments of these arguments are deeply affected by one's possession of certain emotions, feelings, desires, and intuitions or by one's lack of them.

Rational beliefs can be grounded in perceptual experience, memory, testimony, rational intuitions, or inference. Other things being equal, beliefs grounded in any one of these ways are rational. For example, assuming that my memory is generally reliable and that I clearly remember having had toast for breakfast this morning, my belief that I had toast this morning is rational. The apparent self-evidence of “2 + 2 = 4,” or, “A whole is greater than its proper parts,” fully justifies my belief that 2 plus 2 is equal to 4 and that a whole is greater than its proper parts. A number of prominent philosophers of religion have argued that religious beliefs can be justified in a similar fashion. For example, William P. Alston, Richard Swinburne, Jerome Gellman, and I have argued that perceptual or quasiperceptual experiences of God occur and justify the religious beliefs of those who have them. Others such as Alvin Plantinga maintain that in certain circumstances, beliefs in God can be “properly basic.” That is, that like many of our memorial beliefs or beliefs in simple necessary truths, they can be fully justified without being grounded in further beliefs. Yet as Jonathan Edwards said in the mid-eighteenth century, “if we take reason strictly – not for the faculty of mental perception in general [which would include sense perception, memory, and rational intuition] but for ratiocination, or a power of inferring by arguments,” or reasoning, then “reason” refers to the faculty of rational inference and its exercises. The nature and proper role of inference and argument in religion is the subject of this book.

Until quite recently, philosophical studies of religious reasoning and argumentation have tended to focus almost exclusively on the validity of arguments for religious conclusions and the truth of their premises. This is not altogether surprising, since no invalid argument or argument with a false premise is a good argument. But even though truth and validity are necessary conditions of a good argument, they aren't sufficient, since an argument can meet both conditions and not be probative.

Arguments are constructed for various purposes, and these purposes must be taken into account when evaluating their success or failure. Furthermore, reasoning is always situated – it does not occur in a vacuum.

Christianity's critics have often accused it of mystery mongering. Hume, for example, says that “all popular theology, especially the scholastic, has a kind of appetite for absurdity and contradiction. If that theology went not beyond reason and common sense, her doctrines would appear too easy and familiar. Amazement must of necessity be raised: Mystery affected: Darkness and obscurity sought after: And a foundation of merit afforded to the devout votaries, who desire an opportunity of subduing their rebellious reason, by the belief of the most unintelligible sophisms.” And John Toland asserts that Christian theologians and priests have gone even further than the heirophants of the ancient mystery cults. The latter swore their initiates to secrecy but their mysteries were intelligible in themselves. Only Christians dared maintain that their doctrines were mysterious in a more radical sense, “that is, inconceivable in themselves, however clearly revealed.”

Hume's and Toland's explanations of this phenomenon differ. But whatever one thinks of their explanations, there is little doubt that the appeal to, and adoration of, mystery is a characteristic feature of much Christian thought and practice. The Pseudo-Dionysius, for example, begins his Mystical Theology by asking the Trinity to guide him to the “most exalted” and hidden secrets of scripture “which exceedeth light and more than exceedeth knowledge, where … the mysteries of heavenly truth lie hidden in the dazzling obscurity of the secret silence, outshining all brilliance with the intensity of their darkness,” and exhorts his disciple “to leave the senses and the activities of the intellect” and their objects behind, and “strain … toward him whom neither being nor understanding can contain.”

Nor are themes like these peculiar to Christian mystics and mystical theologians. They are commonplace in the church fathers and in a number of later Christian theologians. Two examples will suffice to illustrate my point. Consider first John Chrysostom. St. Paul said:

Anselm's Proslogion is not only cast in the form of a prayer, it is larded with passages from the Psalms and other portions of scripture. The two are connected. As Sr. Benedicta Ward notes, “when [Anselm] uses the language of the psalms, he is not quoting; he is speaking with the language of the scriptures.” Verses from the psalms “are woven together with Anselm's own thoughts, and are prayed spontaneously by him. This is the traditional monastic use of the psalter, where the words of the Hebrew psalms become the prayer of Christ.” As Alcuin wrote to Charlemagne, “if you come to the psalms with a serious mind, and look with the spirit of understanding, you will find there the word of the Lord incarnate, suffering, risen, and ascended.”

For Anselm and his fellow monks, reading, meditation, and prayer “were different aspects of the same thing, not separate exercises in their own right. Reading was an action of the whole person, by which the meaning of a text was absorbed, until it became a prayer. It was frequently compared to eating.” Thus Anselm's “Meditation on Human Redemption” opens with the words, “Taste the goodness of your Redeemer, burn with love for your Savior. Chew the honeycomb of his words, suck their flavor, which is more pleasing than honey, swallow their health-giving sweetness. Chew by thinking, suck by understanding, swallow by loving and rejoicing. Be happy in chewing, be grateful in sucking, delight in swallowing.” And indeed, “the text ‘O taste and see how gracious the Lord is’ was applied more often to the reading of the scriptures than to the Eucharist before the twelfth century.”

These views were by no means peculiar to Anselm. For example, Bernard's Thirty-Fifth Sermon on the Song of Songs says that the Bible is “the wine cellar of the Holy Spirit,” and calls it “a tasty matter for rumination that fattens his stomach, … and makes all his bones sing with praise…. And in the Vajracchedikasutra,” the Buddha says that “learning or teaching” one of its four-line verses “produces more merit than giving away whole world-systems filled with jewels.”

In short, the ingestion of religious texts was widely believed to be essential to spiritual development. These texts were not only transmitted by books or other physical instruments, however, but perhaps most often orally.

Until quite recently, philosophical studies of religious argumentation have almost exclusively focused on the validity of the arguments and the truth of their premises. This is not altogether surprising since no invalid argument or argument with a false premise is a good argument. But while truth and validity are necessary conditions of good arguments, they aren't sufficient since an argument can meet both conditions and not be probative.

Arguments are constructed for various purposes, and these purposes must be taken into account when evaluating their success or failure. Moreover, reasoning is always situated – it does not occur in a vacuum. Arguments are the products of men and women with various needs, hopes, fears, sensitivities, and proclivities, and with diverse individual histories, who are responding to highly specific problems and difficulties. They are not the expressions of a view from nowhere that abstracts from the existential specificity of the reasoner and/or the particularities of his or her concrete situation. Yet the latter frequently determines whether an argument is or is not probative in a particular situation or for a particular person.

In order for arguments such as those discussed in Chapter 1 to be good ones, it is not enough that they be valid and noncircular and have true premises. For as George Mavrodes pointed out, proofs are “person-relative” – a valid, noncircular argument with true premises can prove something to Mary without proving it to John. To take a trivial example, if Mary knows that all the premises of a valid, noncircular argument are true and John does not, the argument can extend Mary's knowledge without extending John's. The proof's person-relativity in this case depends on the fact that Mary knows something that John does not. Other sources of person-relativity are less obvious, however.

The Purposes of Religious Arguments

Arguments are constructed for various purposes, for example, and these have a bearing on their success. Consider Udayana's, Madhva's, and Ramanuja's contrasting attitudes toward proofs for the existence of God. All three were theists. Udayana was an adherent of the Nyaya-Vaisesika school, which identified God with Shiva. Madhva and Ramanuja were Vedantists who identified God with Vishnu. All three accepted the authority of the Vedas.

Philosophy's Quarrel with Rhetoric

Western philosophy has tended to drive a wedge between rational discourse and persuasion. Rational discourse is the domain of philosophy. Persuasion, on the other hand, is the domain of rhetoric, and rhetoric is epistemically and morally suspect. Thus, Plato argues that rhetoric isn't a species of knowledge, but rather a mere “knack” or “technique” (empeiria). The rhetorician has mastered the devices and stratagems that enable him to speak persuasively but lacks a theoretical understanding of their nature, of the psychological and social mechanisms that ensure that some techniques will be effective and others not. Furthermore, the rhetorician's aim is not to produce knowledge or true opinion but to convince his audience of the truth of his assertions, whether they are in fact true or not. And because the rhetorician's aim is persuasion, its mastery involves a command of those devices that make speakers persuasive. These include sound arguments. But they also include plausible but specious proofs, ad hominem attacks, appeals to one's hearers’ baser emotions and prejudices, the creation of a favorable personal impression whether it is warranted or not, verbal style and ornament, and so on. The rhetorician's means of persuasion are thus both rational and nonrational. And this is morally problematic. Genuine arts aim at the good of their subject matter. The aim of medicine, for example, is to produce health. By contrast, rhetoric is not concerned with the spiritual and moral well-being of its potential audiences but only with persuasion. The rhetorician qua rhetorician is indifferent to whether he produces knowledge or true opinion, on the one hand, or false opinion, on the other, and to whether he persuades us by employing reason and appealing to our nobler sentiments or convinces us by using specious arguments and pandering to our baser desires. The philosopher wishes to benefit the souls of her hearers and so employs rational means to produce knowledge or true belief. The rhetorician as such is not concerned with what would benefit us.

But of course, matters are not so simple. Plato himself concedes (in the Phaedrus) that the first defect could be remedied. The second cannot. But the force of his complaint is mitigated by at least three things. First, Plato himself is a master of rhetoric as he must surely have recognized.

This chapter examines an influential proof of God's existence, attempts to defend the coherence of the concept of omnipotence, arguments for and against the personhood of ultimate reality, and competing accounts of the relative importance of grace and free will. While this chapter will bring out the strengths and weaknesses of the arguments it examines, its primary purpose is to provide the reader with historically important instances of religious reasoning which he or she can refer to when reading later chapters.

Samuel Clarke's Cosmological Argument for the Existence of God

Proceeding from the assumption that anything that exists must have a sufficient reason for its existence, Samuel Clarke (1675–1729) argued that the existence of contingent beings can only be explained by postulating the existence of a self-existent being, that is, a being that is both essentially causeless and self-explanatory or intrinsically intelligible (although not necessarily intelligible to us). If a being is self-existent, however, it is also logically necessary.

We can formulate Clarke's argument as follows:

1. 1. If something exists, it is either self-existent (and hence self-explanatory) or some other being causes it to exist.

2. 2. A contingent being isn't self-existent. Therefore,

3. 3. A contingent being is caused to exist by some other being. (From 1 and 2.)

4. Contingent beings are usually caused by other contingent beings. Samuel Clarke's existence, for example, was caused by his parents, their existence was caused by their parents, and so on. Yet what about contingent beings as a whole?

5. 4. Either the series of contingent beings has a first member (a contingent being that isn't caused by another contingent being) or it doesn't (the series is beginningless).

6. 5. If the series has a first member, then a self-existent being exists and causes it. (From 3. Since the first member is contingent, another being causes it. By hypothesis, the first member isn't caused by a contingent being. Hence, it is caused by a self-existent being.)

7. 6. If the series of contingent beings doesn't have a first member (and is therefore beginningless), a self-existent being exists and causes the whole series. (From 3. Since the existence of the series is contingent, another being causes it.

8. Since the cause of the series of contingent beings isn't part of the series, it isn't itself contingent.)

9. 7. If contingent beings exist, a self-existent being exists and causes them. (From 4, 5, and 6.)

In An Essay concerning Human Understanding, John Locke defined reason as “the discovery of the certainty or probability of such propositions or truths, which the mind arrives at by deduction made from such ideas, which he has got by the use of its natural faculties; viz. by sensation or reflection” (4.18.2). Rational belief is proportionate to the strength of the evidence at one's disposal. “The mind, if it will proceed rationally, ought to examine all the grounds of probability, and see how they make more or less for or against any proposition, before it assents to or dissents from it; and upon a due balancing of the whole, reject or receive it, with more or less firm assent, proportionably to the preponderancy of the greater ground of probability on one side or the other” (4.15.5). What is true of beliefs in general is true of religious beliefs. They are rational only if they are (1) properly basic (by being immediately grounded in the mind's intuitive awareness of its own ideas, for example), (2) inferred from those ideas by sound deductive or inductive reasoning, or (3) the content of a revelation whose credentials are certified by beliefs meeting the first or second condition. Although modern intellectuals may doubt whether religious beliefs meet these standards, Locke did not. God's existence can be demonstrated, and the evidence at our disposal makes it probable that the Bible is God's revelation,

Locke's view is sometimes called evidentialism. Evidentialism has much to be said for it. Like scientific beliefs, religious beliefs appear to be “evidence essential” – they are rationally held only if one is entitled to believe that someone in one's intellectual community has good evidence for them. (I am entitled to believe there are quarks, for example, but only because I have good reason to believe that physicists have evidence for them.) Religious beliefs are also controversial. Responsible inquirers have called them into question, and some doubts about them are not unreasonable. Under these conditions, one can appropriately be asked what entitles one to believe. The only entitlement that is likely to convince others of one's right to believe is good evidence. But finally (and most important), Christians, at least, have often assumed that there is good evidence for their position and that those who examine it without prejudice will be persuaded by it.

The anti-Pelagian arguments of Augustine discussed in Chapter 1 rely heavily on appeals to Christian scriptures. They are not the only proofs discussed in that chapter that do so, however. Peter Geach thinks that being omnipotent (being able to do all things) should be distinguished from being almighty (having power over all things), and claims that although the former is arguably incoherent, being almighty must be ascribed to God to ensure that the powers of darkness will ultimately be defeated. He concedes, though, that while a belief in God's ultimate triumph is essential to traditional Christian belief, its truth can be established only by an appeal to revelation – not by philosophical argumentation. Again, one of Ramanuja's objections to Advaita Vedanta is that their denial of the reality of distinctions undercuts Advaita's appeal to the Vedas, which both he and the Advaitins believe to be an inerrant revealed text.

Chapter 3 argued that religious reading inflects the construction and appreciation of religious arguments in the major religious traditions. For their influence to be epistemically benign, however, the relevant texts must be truthful. Now many if not most of the texts in which the religious intellectuals of these traditions are immersed are thought to be revealed texts. If they really are revealed, then presumably they are truthful. Yet are the alleged revelations authentic?

The question of revelation comes up in Chapter 5 as well. The unstated but implicitly assumed presuppositions of the religious rhetorician's enthymatic arguments are often allegedly revealed truths. So if the rhetorician's arguments are to carry their full weight, the revelations must be authentic.

Do the appeals to reason in these cases therefore have a limit? Or can the appeal to revelation itself be rational?

Vedantin and Christian Views of Revelation

Orthodox Hindus regard the Vedas as revealed texts. Vedanta believes that the Vedas are eternal. “Two kinds of eternity are distinguished…. A thing is kutastha nityata [immutably eternal] if is unchanged forever. A thing is pravaharupa nityata [mutably eternal]” if it exists for ever, changing incessantly but always according to the same pattern. (If the movements of our solar system were everlasting, they would therefore be said to be mutably eternal.) Shankara appears to have thought that the Vedas are mutably eternal, produced anew at the beginning of each world cycle according to the same unchanging pattern.