This book addresses the question whether there is rational justification to believe that God, as conceived of by traditional Western theism, exists. There are contemporary fideists who hold that there is no need to justify belief by appeal to arguments, since the mere fact of belief is supposed to be self-justificatory. There is no inconsistency in offering both a fideistic and an argumentative support of belief, as do some contemporaries. My concern is only with the latter part of their justification. The question, then, is whether there are any good arguments either for or against believing that God exists.
I do not pretend to answer this question, since I completely ignore inductive arguments based on design, beauty, and lawlike regularity and simplicity for the existence of God, as well as those based on evil to show the improbability of his existence. A proper discussion of these arguments is the topic for a separate book of considerable length, since it would have to deal with the applicability of Bayesian models of probability to the aggregation of the premises of all the different inductive arguments for and against God's existence. One of the valuable lessons to be learned from Richard Swinburne's The Existence of God, which makes out such a Bayesian case for belief, is that the issues are exceedingly complex and need to be treated by those who are steeped in probability and confirmation theory, which eliminates me.
Our question whether there are any good arguments either for or against belief takes on special importance in the light of the startling resurgence of theism within philosophy during the past thirty or so years. What might surprise some is that the three leaders of this movement, William Alston, Alvin Plantinga, and Richard Swinburne, are themselves analytical philosophers. Some mistakenly see analytic philosophy as the natural enemy of theism, no doubt because certain movements in twentieth-century analytic philosophy, such as logical atomism, logical positivism, and some versions of ordinary language philosophy, developed theories of meaning that were employed to slay the dragon of theism by showing that it did not measure up to certain minimal standards of meaningfulness. But it is a mistake to identify analytic philosophy with these movements and their dogmas.
An ontological argument attempts to deduce the existence of God from an analysis of the conception of God, thereby showing that it is necessary that God exists. From the mere logical possibility that this conception of God is instantiated, it is supposed to follow that it is necessarily instantiated. It has been dogmatically pronounced by a long line of philosophers from Hume up through the logical positivists that an ontological argument cannot work, since existence can never be a logical consequence of an entity's essence. This dogma is called into question by what appear to be perfectly legitimate ontological arguments for certain types of abstract entities, that is, entities that could not logically be located in either space and/or time, such as numbers, properties, and propositions. God, as conceived of by the great medieval theists, is such an abstract entity, though differing from these abstracta in having a life (an illimitable one at that, which is supposed to be had all at once!) and also having a causally efficacious will that can timelessly bring about effects in the universe. Given God's status as an abstract entity, the question naturally arises whether the same style of deductive reasoning from premises knowable a priori to an existential conclusion, such as figures so prominently in mathematics in which existential questions are internal to the system, being decidable by deductions from axioms and definitions, might not also apply to the existence of God. It certainly would be a blatant begging of the question to deny that it can. It is no accident that the ontological argument is the darling of the mathematically inclined theistic philosophers in the Platonic tradition, those who think that the mathematical method of reasoning is the pathway to true knowledge.
This chapter will consider many different types of ontological arguments. The first is based on a weak version of the principle of sufficient reason according to which everything has a possible explanation and argues that a contradiction follows from the assumption that God does not exist consisting in there both being and not being a possible explanation for this negative state of affairs. The second version, of which there are several variants, is based on some premise that is supposed to formulate a necessary truth about abstract entities in general, from which, in conjunction with other necessary premises, God's existence is supposed to follow.
This atheological argument differs sharply from the ones considered in the preceding three chapters in two important respects. First, it has never been given before. Second, it fails to undermine in any way the traditional conception of God and therefore is not an occasion for theists to go back to the drawing board and redesign this conception.
Why, then, consider it? The reason is that it brings into clear relief an overlooked feature of this conception concerning God's “absolute actuality.” God, as creator, has a standpoint that is “outside” of or independent of possible worlds – the infinitely many different ways that things could be. God's existence or actuality is absolute, not being in any way relative to one of these worlds: He exists or is actual simpliciter. In his role as creator, God contemplates them and makes a unique creative decision as to which one alone shall be actual simpliciter. His choice is unique because, of all the creative choices he might have made, this choice alone is actual simpliciter, not just actual relative to some world. As a result of the absolute nature of God's creative choice, one world has a special sort of ontological honor bestowed upon it consisting in it alone having absolute actuality, that is, being the actual world simpliciter. Thus, the absoluteness of God's perspective as creator is propagated down the line via his unique creative choice to a particular world. And what is more, the universe – the largest spatiotemporal aggregate of objects and events – that is created by his unique actualization of one possible world itself has an absolute, unqualified existence. It exists simpliciter.
This theistic view of actuality and existence as absolute or non-world relative is challenged by David Lewis's extreme version of modal realism according to which “every way that a world could possibly be is a way that some world is.” Logical space is a plenitude of isolated physical worlds, each being the actualization of some way in which a world could be, that bear neither spatiotemporal nor causal relations to each other. This means that no world is “absolutely actual” or “actual simpliciter.”
The existence of evil poses both a logical and empirical problem for theism, forming the basis of both a deductive argument that attempts to deduce a contradiction from the existence of both God and evil and an inductive argument that contends that it is improbable that God is the creator of a world that contains the amount and kinds of evils found in this one. Since the deductive argument tries to prove that there is no possible world that contains both God and evil, all the theist need do to rebut it is to show that it is logically possible that God and evil coexist, and this can best, maybe only, be done by actually describing a possible world in which both exist. In such a world God will have a good reason or morally exonerating excuse for permitting evil. Such a response constitutes a defense. To rebut the inductive argument, the theist must construct a theodicy, which is a defense plus an empirical argument for the actual existence of the possible world articulated in the defense.
Often a defense is confounded with a theodicy, resulting in the bogus demand that the theist produce a theodicy as a rebuttal of the deductive argument. H. J. McCloskey, as an example, first espouses the deductive argument. “Evil is a problem for the theist,” he says, “in that a contradiction is involved in the fact of evil on the one hand, and the belief in the omnipotence and perfection of God on the other.” He shows a surprisingly short memory span because later in his essay he says of a certain defense that “it does not in itself provide a justification for the evil in the universe. It shows simply that the evil which occurs might have a justification.” And when he discusses the free will defense (hereafter FWD), he says that it “can at best show that moral evil may have a justification, and not that it has a justification.” These admissions are inconsistent with his earlier claim that theism is inconsistent in holding that both God and evil exist. He seems to be requiring of an adequate defense that it also succeed as a theodicy, which belies a serious conceptual confusion on his part. It is desirable, of course, that a successful defense have the makings of a plausible theodicy, but this is not logically required.
One of the fondest memories for most of us is the time we nailed our parents with the regress of why-questions, the crowning point of which was when we triumphantly asked, “Yeah, so who made God?” At this point we were told to go play marbles in traffic. Little did our parents realize that the cosmological argument could have served as Jim Dandy to the rescue; for it demonstrates that our initial explanatory demand must ultimately terminate with a self-explaining explainer, in which a self-explaining being is one whose existence is entailed by its nature or essence, that is, one for whom there is a successful ontological argument.
But if there is a successful ontological argument, as the cosmological argument supposedly proves, why screw around with the cosmological argument? Why not go right for the jugular and give this ontological argument, thereby having an end to the matter of whether God exists? This was one of Kant's objections to the cosmological argument. This charge of redundancy misses the mark, since there could be a successful ontological argument even if we were unable to give it. Thus, we could know on the basis of the cosmological argument that there is a successful ontological argument and thereby that God exists, but not be smart enough to give it.
The most telling objection that can be lodged against the cosmological argument is that it is impossible for such a being to exist, thereby showing that this argument's conclusion is necessarily false. Any argument for an impossible conclusion, to say the least, leaves something to be desired, whether in the way of validity or soundness. This will be the ground on which every version of the cosmological argument will be attacked. The points made in the Chapter 6 criticisms of ontological arguments will form the basis of my polemic.
Before getting down to a detailed analysis of different versions of the cosmological argument, it will be helpful to describe their generic nature. A cosmological argument is made up of three components: (a) a contingent existential fact, (b) an explanatory argument, and (c) a version of the principle of sufficient reason (hereafter PSR). It begins by demanding an explanation for the existential fact in (a) or some fact that it entails in conjunction with certain other premises or assumptions.
This argument is a dramatic case in point of the thesis of Chapter 1 concerning the positive role of the atheological arguments; for Saint Augustine's famed theory of the relation between God's eternality and the temporality of the created world, which was accepted by all of the subsequent great medieval theists and became official church doctrine, was developed in response to it. Furthermore, the account of time that falls out of his theory has gained widespread acceptance down to the present day. It will be shown that this account of time itself forms the basis of an atheological argument. Had its import been understood by his Christian followers, they would have tarred and feathered rather than sainted him. The notion of God's eternality then will be subject to critical scrutiny and will be shown to render him a nonperson and thereby not religiously available.
In Book 11 of his Confessions, in which he develops this theory of time and eternity, Augustine begins with this Manichaean formulation of the creation–immutability argument:
Lo are they not full of their old leaven, who say to us, “What was God doing before He made heaven and earth?” “For if (say they) He were unemployed and wrought not, why does He not also henceforth, and for ever, as He did heretofore? For did any new motion arise in God, and a new will to make a creature, which He had never before made, how then would that be a true eternity, where there ariseth a will, which was not?”
The following is an explicit mounting of this argument, with 1–3 comprising the initial set:
1. The world came into existence at some past time, say t 0 ;
2. The world exists at some time if and only if God wills that it does;
3. God is immutable;
4. At t 0 God wills that the world exist [from 1 and 2];
5. There are times earlier than t 0 , say t -1, at which the world does not exist [from 1];
6. At t -1 God does not will that the world exist [from 2 and 5];
7. God has a property at one time that he lacks at another [from 4 and 6];
8. God is not immutable [from 7]; and
9. God is immutable, and God is not immutable [from 3 and 8].
In this chapter a broad overview will be given of atheological arguments, without any attempt to determine whether they ultimately succeed. The aim is to give us a feeling for these arguments and an appreciation for their positive role in bringing about important conceptual reforms in the way in which God's nature has been conceived of over time.
The aim of an atheological argument is to reveal a logical inconsistency in the theist's concept of God. Accordingly, it begins with an initial set of propositions each of which is accepted by the theist as a necessary conceptual truth. For example, from the initial set containing the sole premise that necessarily God is omnipotent, an attempt is made to deduce a contradiction, namely, that there is some task that an omnipotent being cannot do. Or, from the initial set containing the conceptual truth that God is benevolent in the sense of always choosing the best alternative, it is deduced that God both does and does not actualize some possible world, with appeal being made to the additional premise that necessarily there is no best of all possible worlds. In both of these atheological arguments, the initial set contains only propositions that the theist takes to be necessary truths.
It would seem that my definition of an “atheological argument” as a deduction of a contradiction from the theist's concept of God, with appeal to only necessarily true additional premises, is unduly restrictive; for some of the most important attempts to demonstrate the logical inconsistency of theism begin with an initial set of propositions that contain at least one proposition that is taken by the theist to be only contingently true, for instance, that there exists evil, created free persons, and so on. For example, the deductive argument from evil, in one of its versions, begins with an initial set containing two propositions:
1. There exists an omnipotent and benevolent God; and
2. Evil exists,
from which it is deduced that evil does not exist, with appeal being made to some extra premises that articulate conceptual truths about omnipotence and benevolence. Herein the initial set is not comprised solely of conceptual truths, the theist taking 2 to be only contingently true, and also 1, if God's existence is taken to be contingent.
This argument's initial set is composed of these three propositions:
1. God essentially knows all and believes only true propositions;
2. God essentially does not change from one time to another in respect to any nonrelational property; and
3. There are true temporal indexical propositions (to be called “A-propositions”) to the effect that certain events and/or times are now past, present, or future.
Theism's commitment to 1 and 2 results from its conceiving God to be essentially both omniscient and immutable; and it is committed to 3 because, as pointed out in the previous chapter, theism (at least as represented by Judaism and Christianity) is a historically rooted creed that accords not only reality but significance to the sort of A-facts over which 3 quantifies, such as that the world came into being in the past,has reached a certain stage in its spiritual development at present,and awaits a future day of judgment.
The qualification on God's omniscience, that he believes only true propositions, which will be dropped in the future for the sake of brevity, is due to our not wanting to allow an omniscient being to go around having any false beliefs, even if he does know every true proposition. Notice that an omniscient being must not only know of every true proposition that it is true but must also know every true proposition. As we shall see later in this chapter, these are not the same, since one could know that some proposition is true without knowing it.
The reason for the restriction of God's immutability to nonrelational properties, in which a nonrelational property is one that an individual could possess even if there were not to exist any other nonuniversal particulars, save for spaces and times, is to escape such “Cambridge” changes in God as being thought about by Jones at one time but not at another, herein the change being in Jones rather than in God. How the argument proceeds from here depends on whether it has as its target the timelessly eternal God of St. Augustine or the omnitemporally eternal God of the Bible, that is, a God that endures throughout a beginningless and endless time and logically could not begin or cease to exist.
So far only epistemological or truth-directed reasons have been considered for believing that God exists. It is now time to consider pragmatic reasons based on the desirable consequences of belief. There is the big question, which will not be considered, of whether individuals, as well as society at large, are better off having religious faith. While religion and the faith it inspires has a mixed track record, having been among the causes of man's greatest and basest deeds, we have no way of determining whether things overall would have gone, or will go, better with it than without it. My interest, however, is not with questions of empirical fact but with the conditional question “If overall there were to be desirable consequences of believing that God exists, would that justify, that is, constitute a sufficient reason for, one's believing?”
The desirable consequences can take different forms and thereby serve as the basis for different types of pragmatic argument. They can be prudentially desirable if they satisfy the needs, wants, or interests of the believer. Or they might be desirable on moral grounds either by enabling us to engage in the practice of morality, that is, following moral rules even when doing so sacrifices our own self-interest, or making the believer and/or her society morally better. Kant's moral argument for believing is an example of the former: The demands of morality to follow the categorical imperative make sense only if we assume that there exists a God who will bring it about that those who do so are rewarded with happiness, something that rarely happens in this life. Because this justification of faith takes us so deeply into the nature of morality, it will not be possible to consider it in this book, nor am I up to tackling it. Given that it is rational or reasonable to promote one's own interests as well as those of morality, both sorts of consequences can serve as a justification or sufficient reason for believing. Pragmatic arguments from prudence will be discussed first, then those from morality.
Pragmatic arguments from prudence
The most influential of the prudential-type pragmatic arguments for faith is Pascal's “wager” in his Pensées.
Throughout history, in almost every society, persons have claimed to experience God or some ultimate reality in the same sort of immediate way that they experience physical objects through their senses. Their experiences have been quite diverse, running the gamut from extrovertive and introvertive mystical experiences of some underlying unity into which the experient is totally or partially incorporated to numinous experiences and direct experiences of God's presence as a very powerful, loving, nonhuman person who shows concern for the well-being of the experient. Different descriptions are given of the apparent object of their experiences – God, one's personal purusha, Brahman, the eternal one, the undifferentiated unity, and the like. In spite of these different characterizations, the subjects agree in taking the object of their experience to be the ultimate reality, the really real, and, most important, are convinced that man's greatest possible good consists in achieving a direct, experiential contact or union with this reality. For these reasons, I shall call all these experiences “religious experiences,” even though they are not taken to be such by everyone who has them.
I shall duck the thorny issue of whether there is a common phenomenological content to all or most of these experiences, at least the unitive ones, that supposedly gets interpreted differently by mystics in diverse cultures, as well as the larger problem of whether the phenomenological–interpretative and observation–theoretical language distinctions are viable. While I am in basic agreement with the twentieth-century “mystical ecumenicalists” – Underhill, Otto, Stace, Suzuki, Merton, and N. Smart – who accept these dualisms and find a common phenomenological content, it will not be necessary for me to defend their views against their contemporary critics, such as R. C. Zaehner and Steven Katz. I will assume that these distinctions are viable relative to our epistemic circumstances – the manner in which we experience the world and process information. Our perception of something as a chair involves a minimal amount of theoretical interpretation, if any, but our perception of a track in a Wilson cloud chamber as a movement of an electron involves a highly theoretically based inference. Similarly, our experience of an undifferentiated unity or a personal, loving presence is relatively noninferential as contrasted with our perception of an evil as a test set for us by God or of a sunset as an expression of God's love.
Email your librarian or administrator to recommend adding this to your organisation's collection.