PHILOSOPHY OF RELIGION

Analytical "philosophy of religion," still in its infancy in the 1960s, has developed markedly since then. Other approaches have certainly continued to play a part in philosophy of religion written in English, even more so in other languages. Process philosophy, for example, inspired by the thought of Alfred North Whitehead and exemplified in the ongoing work of Charles Hartshorne and others, has retained influence in philosophy of religion and in theology, probably more than in other areas of philosophy. Phenomenology, postmodernism, and other approaches characteristic of the European continent inspire important contributions to the subject. Indeed, there is often not a sharp line between different approaches. Continental writers such as Søren Kierkegaard figure extensively in undoubtedly analytical writing about religion, and analytical philosophy of religion makes such extensive use of medieval material as to be more or less continuous with neoscholastic treatments of the subject.

Although there had been a few earlier analytical essays about various religious issues, the main development of analytical philosophy of religion may be said to have begun in the 1950s with discussion of the "logical positivist" challenge to the cognitive significance of religious language. Most analytical philosophers then held, or were strongly tempted to hold, as an empiricist principle, that every (logically contingent) assertion, in order to have any cognitive meaning, must be verifiable or, more broadly, testable, in principle, by experience. It was charged, by Alfred Jules Ayer, Antony Flew, and others, that the affirmations of religious belief typically do not satisfy this criterion of meaning (A. Flew, R. M. Hare, and B. Mitchell in Brody 1974).

How, then, were the apparent truth claims of religions to be understood? Some were prepared, with Ayer, to treat major religious assertions as mere expressions of emotion, without any cognitive significance. Others sought ways of understanding such assertions as empirically verifiable in principle. John Hick (in Brody 1974) argued, for instance, that "eschatological verifiability," in a life after death, provides at least a partial solution to the problem. Still others, while granting that empirical testability is decisive for the meaning of typical factual assertions, sought to establish a different, and not merely emotive, type of meaning that could be ascribed to religious assertions. The most influential attempts of this type were inspired by the later writings of Ludwig Wittgenstein, particularly by his account of "language games" and their relation to forms of life.

The Wittgensteinian approach, as developed, for example, by Norman Malcolm (in Brody 1992) and D. Z. Phillips (1970), has generated very interesting studies of the relation of religious language to religious life. It is widely criticized, by some as giving inadequate weight to the apparent straightforwardly realistic intent of typical religious assertions, and by some as improperly shielding religious claims from rational criticism by relativizing them to religious language games. It remains, nevertheless, an important strand in contemporary discussion. Of all that has been done in analytical philosophy of religion, it is probably the discussion of religious language in general, and Wittgensteinian themes in particular, that have most interested professional theologians, perhaps because these themes have seemed more relevant than more metaphysical discussions to the work of interpretation and reinterpretation of traditions in which theologians are so much engaged.

Within analytical philosophy during the 1950s the verifiability criterion of meaning was already undergoing severe criticism and has since been virtually abandoned in anything like its original form. Many analytical philosophers continue to consider themselves empiricists and seek alternative ways of excluding claims that they regard as objectionably metaphysical. Many others, however, see the permanent contribution of analytical philosophy, not in a form of empiricism, or in any set of doctrines, but in a method, style, or discipline that can be applied to virtually all the historic issues of metaphysics and ethics and can be used in developing and espousing almost any of the classic philosophical doctrines.

The majority of work done in analytical philosophy of religion since the 1960s has been inspired by the later conception of analytical philosophy and has not focused on issues about religious language. It is characterized by metaphysical realism, taking the religious claims under discussion to be straightforwardly true or false. (For defense of this stance, see, e.g., Swinburne 1977, chaps. 2–6.) Some have suggested calling it philosophical theology rather than philosophy of religion, because the principal subject of most of it is God rather than human religious phenomena, though atheists as well as theists have certainly been important participants in the discussion. On this basis, mainly since 1960, a very substantial body of literature, dealing with most of the traditional issues of philosophical theology and some new ones too, has been created.

Among the traditional topics the attributes of God received rather early analytical attention. (For general treatments see Swinburne 1977, chaps. 7–15; Kenny 1979; Wierenga 1989.) Analysis of the concept of God was easily seen as an appropriate subject for analytical philosophy, and issues about the attributes had been connected, since the Middle Ages, with problems about predication, an appealing point of entry into philosophical theology for those interested in the philosophy of language. According to some of the most influential medieval theologians, God is so different from creatures that positive attributes of creatures cannot in general be predicated of God univocally, that is, in the same sense in which they are predicated of creatures. How then can we predicate anything of God? Various Scholastic theologians developed various solutions, the best known being the theory of analogical predication of Thomas Aquinas. Analytical philosophers of religion have taken up the problem and some of the medieval views, along with more contemporary concerns—for instance, about the ascription of psychological predicates to a being who is supposed not to have a body. (Cf. Maimonides, Thomas Aquinas, and Alston in Brody 1992).

The two divine attributes that have received the most extensive analytical discussion are omniscience and eternity. The central issue about eternity is whether to understand it (as medieval and early modern theology generally did) as involving existence outside of time or rather as involving existence without beginning or end in time, as many contemporary thinkers have proposed. Critics of divine timelessness, such as Nelson Pike (1970) and Nicholas Wolterstorff (in Brody 1992), have questioned the compatibility of timelessness with God's consciousness or action or interaction with creatures. Eleanor Stump and Norman Kretzmann, however, have presented an influential defense of the traditional timeless conception (in Brody 1992), and the issue remains vigorously debated.

Omniscience and eternity are related topics, for one of the most discussed issues about God's knowledge concerns God's relation to time: Does God have complete knowledge of the future? In particular, does God know, infallibly and in every detail, how free creatures will use their freedom? Traditional theologies generally gave an emphatically affirmative answer to this question; but some modern philosophers and theologians have disagreed, arguing that the doctrine of total, infallible foreknowledge compromises the freedom of the creatures. The extensive analytical literature on this issue (e.g., in Fischer 1989) is continuous with older discussions, and opinion remains divided.

A related old debate, recently revived, concerns what has been called "middle knowledge": Does God know, completely and infallibly, what every actual and even merely possible free creature would freely do (or would have freely done) in every possible situation in which that creature could act freely? In the late sixteenth century, Luis de Molina, a Jesuit, proposed an ingenious theory of divine providence according to which God uses such subjunctive (and largely counterfactual) conditional knowledge to control the course of history without having to interfere metaphysically with the freedom of creatures. This theory of middle knowledge was widely embraced by Jesuits, but opposed by Dominicans, who argued that there cannot be such determinate conditional facts about everything that would be freely done by particular creatures in all possible circumstances. This historic controversy was introduced into current analytical discussion by Anthony Kenny (1979) and Robert Adams (1987), who have both defended the Dominican objection to middle knowledge; but the opposite position has been argued by a vigorous school of contemporary Molinists, including Alvin Plantinga (1974) and Alfred Freddoso (1988).

Regarding the relation of God to ethics, it was almost universally held in the 1960s that fundamental ethical principles must be independent of theology and that an acceptable theological account of the nature of ethical facts is impossible. Since then, however, it has come to be widely held by theists, and granted by many nontheists, that facts about God, if God exists, could play a central role in explaining the nature of ethics and that theistic philosophers should be expected to avail themselves of this possibility. The most discussed type of theological theory in this area is the divine-command theory of the nature of ethical obligation, or of right and wrong (Helm 1981). Several thinkers, such as Philip Quinn (1978), have tried to reformulate and explain the theory in such a way as to defend it against the traditional objections to it. Adams (1987) has proposed a form of the theory that rests on semantical assumptions very similar to those of some of the most influential contemporary exponents of metaethical naturalism but employs different (theistic) metaphysical assumptions.

The grounds proposed for belief or disbelief in the existence of God have naturally claimed at least as much analytical attention as the attributes of God. This is a subject so intensively discussed for centuries that one might have expected little novelty in the treatment of it. But in fact investigations have been rather innovative, and the state of debate has changed significantly since 1960. One striking change is that the traditional arguments for the existence of God, then widely dismissed, even by theologians, as hopelessly discredited, have many defenders at the turn of the twenty-first century.

This is connected with a more general phenomenon, which is that analytical philosophers, especially those inclined to construct and defend constructive metaphysical theories, demand less of arguments than has commonly been demanded in the past. Virtually no one thinks any one "theistic proof" conclusive; but if arguments must be either conclusive or worthless, there would be little useful reasoning about any of the most important philosophical issues. Theistic apologists are accordingly less apt to seek a single "knockdown" proof than to try to show that several traditional (and perhaps also novel) arguments have something of value to contribute to a "cumulative case" for theism, an approach exemplified by Richard Swinburne (1979). Extensive work has been done interpreting, developing, and criticizing all the main types of theistic arguments. Those that have probably received the most attention and development are the "ontological" and the "teleological" (to give them their Kantian names).

The fallaciousness of any ontological argument and the contingency of all real existence had become such commonplaces, especially among empiricists, that it had a certain "shock value" when Norman Malcolm in 1960 published a defense of an ontological argument (reprinted in Brody 1992). Malcolm claimed to find in Anselm's Proslogion, besides the famous argument of its second chapter, a second ontological argument in which it is not existence but necessary existence that figures as a perfection. Malcolm also held that necessary existence cannot be excluded from theology on general philosophical grounds. Whether a statement expresses a necessary truth, he argued, depends on the language game in which it figures; and a religious language game can treat the existence of God as a necessary truth. These two features of Malcolm's article foreshadow the main tendencies in the development of ontological arguments since then: (1) attention to more modal versions of the argument and (2) the attempt to rehabilitate the idea of necessary existence.

Ontological argument studies have been greatly influenced by the dramatic development of modal logic, which was gathering momentum in the 1960s and burst into the center of American philosophical consciousness in the 1970s. In 1962 Hartshorne published a modal proof of the existence of God relying only on the premises that God's existence must be necessary if it is actual and that God's existence is at least possible. Subsequent discussion has established that this proof, and related proofs from slightly slenderer assumptions, are valid in the system of modal logic (S5) most widely thought to be appropriate for the context. David Lewis (in Brody 1974) and Plantinga (1974) have given the argument a form that takes account of developments in modal predicate logic as well as modal propositional logic (or in de re as well as de dicto modality). The argument is still of limited value for proving the existence of God, because those who would otherwise doubt the conclusion are likely to doubt the possibility premise, given the rest of the argument. But the modal development of the argument is helpful in structuring discussion of questions about necessary existence.

In the 1950s it was the opinion of almost all analytical philosophers that the existence of a real being, such as God (as distinct from merely abstract objects, such as numbers), cannot be necessary in the strongest, "logical" sense. This opinion has come to be widely doubted, however, and the traditional view that God should be conceived as an absolutely necessary being has regained a following. (For contrasting views see Adams 1987, chaps. 13–14, and Swinburne 1977, chaps. 13–14.) Several factors have contributed to this change. The identification of necessity with analyticity, on which the rejection of necessary existence was commonly based, is under attack. W. V. O. Quine's influential doubts about the adequacy of the notion of analyticity led Quine himself to skepticism about necessity. But others, influenced in some cases by an interest in necessity de re, have been inspired to seek a more robustly metaphysical conception of necessity. Since a conception of the latter sort was generally held by the great philosophers of the Middle Ages and the seventeenth century, a growing and more sympathetic understanding of those periods of the history of philosophy has also tended to undermine the most dismissive attitudes toward the idea of necessary existence.

The most popular argument for the existence of God in the eighteenth century was the teleological or design argument, usually in a pre-Darwinian form drawing its evidence largely from biological adaptations. This type of argument was discredited both by the devastating critique it received in David Hume's Dialogues concerning Natural Religion and by the development of an alternative explanation of the biological phenomena in terms of natural selection. A major rehabilitation of the design argument has been undertaken by Swinburne (1979). Instead of the biological evidence, he takes as his principal evidence the most pervasive, highest-level regularities in the universe. Since they constitute the most fundamental laws of nature, to which all scientific explanations appeal, he argues, there cannot be any scientific explanation of them. There may therefore be no viable alternative to a theological explanation for them, if they are to be explained at all. Deploying the apparatus of Bayesian probability theory, and responding to Hume's objections, Swinburne tries to establish that a theological explanation is indeed more plausible than no explanation at all. Swinburne's argument depends at some points on controversial metaphysical theses and has inspired an extended atheistic response by J. L. Mackie (1982); but the teleological argument has at least been shown to have much more philosophical life in it than had been thought.

The leading argument for atheism, aside from the various critiques of theistic arguments, has long been the argument from evil. The evils that occur in the world are incompatible, it is argued, with the existence of an omnipotent, omniscient, perfectly good God. In the earlier years of analytical philosophy of religion this was usually a charge of demonstrable, logical incompatibility; and attempts to provide theists with a "solution" to the "problem of evil" concentrated accordingly on trying to show the possibility of a perfect deity having permitted the evils. Borrowing a Leibnizian idea, for instance, Pike argued that for all we know, this might be the best of all possible worlds (in Adams and Adams 1990). Plantinga (1974) developed a much-discussed version of the traditional "free will defense," arguing that even if there are possible worlds containing less evil, and as much moral good, as the actual world, an omnipotent God may have been unable to create them because it may be that creatures (whether humans or angels) would not have freely done what they would have to do freely in order for one of those worlds to be actual. The adequacy of such theistic responses to the "logical" form of the argument from evil has been keenly debated, but it has probably become the predominant view that the argument does not afford much hope of a tight, demonstrative proof of atheism.

There has therefore been increasing interest in probabilistic arguments from evil, as presented, for example, by William Rowe (in Adams and Adams 1990), whose thesis is that evils show theism to be implausible, or at least constitute evidence against theism, which might contribute to a cumulative case for atheism. Theistic responses to this type of argument must address issues of plausibility and not merely of possibility. Some have been methodological, attempting to show that the relevant probabilities cannot be determined, or that the explanatory structure of the situation keeps the evils from being even relevant evidence (e.g., Stephen Wykstra in Adams and Adams 1990). Others have tried to give plausible accounts of why evils might have been necessary for greater goods. One widely debated hypothesis, developed in different ways by Hick (in Adams and Adams 1990) and Swinburne (1979), for instance, is that evils, and possibilities of evil, play an essential part in making the world a context for the moral and spiritual development of free creatures.

All such explanations of why God would permit great evils have seemed to some morally or religiously objectionable. Among theists who take this view, Marilyn Adams has argued that we should accept that we simply do not know why God has permitted horrendous evils but that within a religion that affirms, as Christianity does, God's love for individuals who suffer them, it is important to have a coherent account of how God may be seen as redeeming them (Adams and Adams 1990). She points to traditional religious ideas of suffering shared with God or with Christ as suggesting how horrendous evils might be "defeated" by forming an organic whole with incommensurably great religious goods.

One of the more dramatic developments of the period under review is the development of a defense of the rationality of theism that professes not to be based on arguments or evidence. Plantinga maintains that belief in the existence of God can be "properly basic," a basic belief being one that is not inferentially based on any other belief (Plantinga and Wolterstorff 1983). It has been held by many that some beliefs (formed, perhaps, in sensation or memory) do not need inferential support from other beliefs for their justification. Plantinga argues that more beliefs than some have supposed are reasonably held without being based on the evidence of other beliefs and that there is no compelling reason to deny that some religious beliefs have this basic status. He suggests that religious beliefs not based on "evidence" constituted by other beliefs may nonetheless be based on other sorts of "grounds," which might be found, for example, in religious experience. Plantinga's view (which he has dubbed "Reformed epistemology") has been keenly debated. One of the most discussed issues is whether it allows an adequate basis for distinguishing between rational and irrational religious beliefs. (For a moderately critical view see R. Audi in Audi and Wainwright 1986).

A related but importantly different view has been developed by William Alston (1991). Religious experience has been a major subject of discussion in philosophy of religion (e.g., W. James, W. T. Stace, and C. B. Martin in Brody 1974; Wainwright, 1981), as it has been in modern theology. Not all of the discussion has been epistemological or focused on the justification of belief. Pike (1992), for instance, has written about the phenomenology of mysticism, arguing, against the older theory of Stace, that there are mystical experiences of theistic as well as nontheistic content. Alston's approach is thoroughly epistemological, however, and he focuses on the experience of more ordinary religious believers rather than of those adepts typically singled out as "mystics."

Relying on carefully discussed analogies with sense perception, Alston argues that in some circumstances experiences as of God addressing, or being present to, a person can reasonably be regarded as perceptions of God. His argument is placed in the context of a "doxastic practice" conception of the justification of beliefs. He argues that we are able to form and justify beliefs only in socially established practices in which we have learned to be responsive to such factors as experiential cues and communal traditions as well as to beliefs that we hold. In Alston's view we have no choice but to rely on socially established doxastic practices, and it is presumptively rational to do so, even though we typically have little or no independent evidence of the reliability of the practice. He argues that this presumption of rationality applies also to religious doxastic practices that are socially established, and in particular to practices in which participants have learned to form beliefs of having perceived God in various ways. Alston offers vigorous rebuttals of several major objections to basing religious beliefs on religious experience. In his opinion the most serious problem for his view, which he treats at some length, is that posed by the existence of diverse religious traditions whose well-established doxastic practices lead them to form apparently conflicting beliefs on the basis of their religious experience.

For philosophy of religion as for contemporary theology, the problem of conflicting truth claims of different religions is, if not a new issue, one that is coming into increasing prominence. Hick (1989) has done much to draw attention to it. He argues that it is not plausible to suppose that one traditional form of religious experience is veridical while others are not, and he tries to articulate a way in which many apparently conflicting forms could all be at bottom veridical, proposing to regard them as apprehending different "phenomenal" manifestations of a single "noumenal" transcendent "reality." Not that Hick thinks all religious beliefs equally acceptable; the main criterion he proposes for the value of religious traditions and belief systems is their fruitfulness in producing morally and spiritually recognizable saints, people notably advanced in a transformation from self-centeredness to Reality-centeredness. Among the issues in the vigorous debate about Hick's view are the adequacy of the conceptual apparatus he borrows from Immanuel Kant and whether it is compatible (as he means it to be) with a fundamentally realist and cognitivist conception of religious belief.

See also Atheism; Ayer, Alfred Jules; Bayes, Bayes' Theorem, Bayesian Approach to Philosophy of Science; Empiricism; Epistemology, Religious; Evil, The Problem of; God, Concepts of; Hare, Richard M.; Hume, David; James, William; Kant, Immanuel; Kierkegaard, Søren Aabye; Mackie, John Leslie; Malcolm, Norman; Modal Logic; Mysticism, Nature and Assessment of; Ontological Argument for the Existence of God; Phenomenology; Postmodernism; Quine, Willard Van Orman; Religious Experience; Religious Experience, Argument for the Existence of God; Religious Pluralism; Stace, Walter Terence; Teleological Argument for the Existence of God; Theism, Arguments For and Against; Thomas Aquinas, St.; Whitehead, Alfred North; Wittgenstein, Ludwig Josef Johann.

Bibliography

Adams, M. M., and R. M. Adams, eds. The Problem of Evil. New York: Oxford University Press, 1990.

Adams, R. M. The Virtue of Faith and Other Essays in Philosophical Theology. New York: Oxford University Press, 1987.

Alston, W. P. Perceiving God. Ithaca, NY: Cornell University Press, 1991.

Audi, R., and W. J. Wainwright, eds. Rationality, Religious Belief, and Moral Commitment: New Essays in the Philosophy of Religion. Ithaca, NY: Cornell University Press, 1986.

Brody, B. A., ed. Readings in the Philosophy of Religion: An Analytic Approach. Englewood Cliffs, NJ: Prentice-Hall, 1974; 2nd ed., 1992. Somewhat different selections, both comprehensive and both excellent, in the two editions.

Fischer, J. M., ed. God, Foreknowledge, and Freedom. Stanford, CA: Stanford University Press 1989.

Hartshorne, C. The Logic of Perfection and Other Essays in Neoclassical Metaphysics. La Salle, IL: Open Court, 1962.

Helm, P., ed. Divine Commands and Morality. Oxford: Oxford University Press, 1981.

Hick, J. An Interpretation of Religion. New Haven, CT: Yale University Press, 1989.

Kenny, A. The God of the Philosophers. Oxford: Clarendon Press, 1979.

Mackie, J. L. The Miracle of Theism. Oxford: Clarendon Press, 1982.

Molina, L. de. On Divine Foreknowledge (Part IV of the Concordia ). Translated and with an introduction by A. J. Freddoso. Ithaca, NY: Cornell University Press, 1988.

Phillips, D. Z. Faith and Philosophical Enquiry. New York: Schocken, 1970.

Pike, N. God and Timelessness. New York: Schocken, 1970.

Pike, N. Mystic Union. Ithaca, NY: Cornell University Press, 1992.

Plantinga, A. The Nature of Necessity. Oxford: Clarendon Press, 1974.

Plantinga, A., and N. Wolterstorff, eds. Faith and Rationality. Notre Dame, IN: University of Notre Dame Press, 1983.

Quinn, P. L. Divine Commands and Moral Requirements. Oxford: Clarendon Press, 1978.

Swinburne, R. The Coherence of Theism. Oxford: Clarendon Press, 1977.

Swinburne, R. The Existence of God. Oxford: Clarendon Press, 1979; 2nd ed., 1991.

Wainwright, W. J. Mysticism. Madison: University of Wisconsin Press, 1981.

Wierenga, E. R. The Nature of God. Ithaca, NY: Cornell University Press, 1989.

Robert M. Adams (1996)