Charles Hartshorne and the Ontological Argument
by Eugene H. Peters
Eugene H. Peters (Ph.D., University of Chicago) is professor of philosophy at Hiram College, Hiram, Ohio 44234, and the author of The Creative Advance and Hartshorne and Neoclassical Metaphysics. Eugene H. Peters died of leukemia on May 13,1983, at the age of 54. The following article appeared in Process Studies, pp. 11-20, Vol. 14, Number 1, Spring, 1984. Process Studies is published quarterly by the Center for Process Studies, 1325 N. College Ave., Claremont, CA 91711. Used by permission. This material was prepared for Religion Online by Ted and Winnie Brock.
Charles Hartshorne believes that the ontological argument forces on us the disjunction, either "God exists" is true necessarily or "God exists" is false necessarily. This, he holds, is simply a logical extension of what Anselm discovered, for, as Anselm saw, God cannot be understood in the imperfect mode of contingency, whether as existing or not. Hartshorne thinks that the possibility of "God exists" may be asserted as an intuitive postulate. If so, then the second of the disjuncts -- that "God exists" is false necessarily -- is false (and indeed could not be true). Hence, "God exists" is true necessarily.
Yet, as R. L. Purtill has pointed out, a precisely parallel argument holds if we conjoin with the original disjunction the premise that "God does not exist" is possible.1 For then the first of the disjuncts -- that "God exists" is true necessarily -- is false (and must he so). "God exists is then false necessarily.
Moreover, if to show that "God exists" is possible one must show that some actual set of circumstances requires Godís existence, then, says Purtill, one would, in adducing such a set of circumstances, "have proved Godís existence without the ontological argument."2 Likewise, if by showing that some actual set of circumstances is incompatible with Godís existence, one shows that "God exists" is not possible, then, in adducing such a set of circumstances, one would have "proved Godís non-existence without the ontological disproof."3
Perhaps the defender of the doctrine, "God exists" is possible, need only conceive a set of circumstances to require Deity, while the opponent of the doctrine need only conceive a set of circumstances incompatible with divine existence, e.g., a completely evil universe. Without settling the question whether an actual or only a conceivable set of circumstances need be adduced, Purtill says he inclines to favor the first option -- as in the cosmological and design arguments. But then the possibility of God would be based on a proof itself sufficient to show that God exists (granting that the proof succeeds). "Thus, if the ontological argument is sound it is superfluous."4
In his provocative essay, Purtill has certainly thrown down the gauntlet to the Anselmian. Not only has he focused attention on the crucial postulate of the ontological argument, the postulate of logical possibility, he has advanced the thesis that a justification of that postulate would already include the conclusion of the argument and thereby render it superfluous.
At the outset it should be said for purposes of clarity that the Anselmian is not trying to establish that God is one possibility among others, that is, that God is a potential being. Rather, he is concerned to show that God is not inconceivable or (putting it another way) that the concept of God hangs together in the understanding and breaks down neither logically nor ontologically. Perhaps to accomplish that task fully would require a full-blown doctrine of God and the world. In any event, it is not simply an intuition (an "intuitive postulate") that perfection is not impossible -- except to one sufficiently enlightened by examination of the conceptual problem of God. My contention is that such an examination can provide the assurance one needs to assert that God is conceivable.
That assurance, if it can be achieved, must be internal to the doctrine of God itself and need not be sought in and through arguments such as the cosmological and design arguments. Doubtless these arguments have a bearing on the ontological argument, and I will later suggest what it may be. My point here, however, is that to ask how God is possible is (to put it in broadest terms) to ask how an individual being may be so conceived as to function universally in the universe, each thing in the universe in turn functioning with respect to that individual being. To be sure, the functioning must be specified. But what I am concerned to stress is the utterly unique role the individual being God, in functioning universally, is understood to fill. Hartshorne, in explaining how we may conceive that role, simultaneously provides the intuition that a divine being is possible.
The definition of deity, as unsurpassably great, identifies one and only one being, though the definition contains no empirical elements. As Hartshorne puts it, God is the one individual definable a priori.5 Thus, in conceiving deity -- though the conception is a bare abstract thought -- we possess the principle of individuation for God; we describe no mere class of beings. The perfection of that individual being consists precisely in its all-comprehensive functioning. And this in turn may be understood by speaking of God as the ground of all possibility.6 "The basic referent of Ďpossible,í" says Hartshorne, "is to the divine capacity to create and enjoy creatures"7 That capacity, being coextensive with possibility itself, furnishes God a status transcending that of any of the unimaginably many alternative possibilities, making God the individual being to which there is no alternative. Godís inclusiveness of the possible is paralleled by the possession of whatever actualities are realized. To speak of Godís modal coextensiveness, as Hartshorne does, is the abstract, non-dynamic way of referring to Godís unsurpassable capacity to make alternatives possible and to produce (without necessarily determining) actual creatures.
I submit that this conception itself, when adequately elaborated, shows the possibility of God. For unless the conception contains hidden contradictions, it shows that deity would exist no matter what else did or might exist, no matter what else did not or might not exist. In Hartshorneís words, "the existence of God is not a possibility competing with other possibilities . . . [Godís] role is not competitive."8 To say that nothing can rival God in greatness is simply to say that any other existent does (or would) compete with other possibilities and hence is (or would be) inferior to deity. It is Godís embracing all possible and actual things which at once defines Godís uniqueness as an individual being, explains the noncompetitiveness of thyme existence, and explicates its unsurpassibility. God is possible because it is possible for possibilities to have a ground, namely, an individual who functions as their presupposed source. Or, to state the point another way, God is possible because by existing no entity can contravene the very condition or presupposition of its being.
We would smell a paradox in the declaration: "I canít disclose whether I am capable of uttering an English sentence." Similarly, we may find it paradoxical to ask Purtillís question whether some set of circumstances is incompatible with -- or, alternatively, requires -- Godís existence. For God is conceived as the ground or condition of any and every set of circumstances. In his "rejoinder to Purtill," Hartshorne aptly declares: "From the unsurpassable power of God to adapt to circumstances I deduce that nothing could conflict with his existence." 9 He adds: "From the unsurpassable power of God to produce effects I deduce that . . . anything could show his existence."10 At a minimum, it is possible that God so exist as to conflict with no circumstance whatever but, rather, so exist as to furnish its very possibility.
It should be evident that Hartshorneís understanding of God represents a departure from that of Anselm and other classical theologians. As George L. Goodwin sees it, "the key move in the Hartshornean re-conception of God is precisely the introduction of possibility into the Godhead."11 Possibilities do not exist in a vacuum. They belong (or subsist) within the context of Godís being, since God is their indispensable source or principle. Hence, God is not actually complete and perfected, the exhaustion of all possible values. Rather, God is modally complete. This means that God includes whatever is in fact actual, and can and will possess any predicate or compossible set of predicates that becomes actualized.12
What is missing in the classical view of God is the logical (and modal) distinction between property and instance, between the abstract concept and its concrete embodiment. God was, in the tradition, said to be Godís perfection. If that be true, then God "is evidently the mere content of an abstract definition."13 And being merely abstract -- albeit necessary -- God falls short of actuality or concreteness. Hartshorne escapes the paradox of abstractness, to which J. N. Findlay has called attention. Hartshorne distinguishes the existence of deity -- that Godís identity is instantiated in actual states -- from the actual states which exhibit the divine identity. It is Godís unlimited capacity to possess successive states, each an inclusive appropriation of its corresponding world state, that makes sense of modal completeness or co-extensiveness. Yet the view is unmistakably dipolar, for the perfection of God is that abstractly conceivable capacity of God to have and enjoy creatures, a capacity unsurpassable because limited by no actual or possible thing.
If, as Goodwin says, Hartshorneís key move is to introduce possibility into the Godhead, then of course time is introduced as well. For possibility, according to Hartshorne, is essentially futurity -- the determinability of some actual state of affairs, its Inn -- finished, yet-to-be-completed aspects. I am not referring here to what Hartshorne calls "pure" or eternal possibilities, which he says are wholly rational and presuppose no choice, decision, or selection.14 It is in this latter sense that Codís existence can be said to be a possibility. But in speaking of possible divine states we refer to what is possible for God relative to time and place. It is possible for God to enjoy tomorrowís values, and to suffer its disvalues, only when and as these become actual facts.
To explicate the manner of Godís functioning with respect to all actual and possible entities, Hartshorne employs the psychical categories -- volition knowledge, love, experience, valuation, etc. These are not for him symbolic usages, nor are they merely analogical. They are true of God literally. Thus, for example, Godís omniscience is the clear and complete cognition of all things just as they are. God knows the past in all its variety and detail, the present in its creative becoming, and the future as that set of possible (and probable) states furnished by the temporal advance to date.
Now my argument is that Hartshorneís neoclassical way of conceiving Cod furnishes the postulate of logical possibility for the ontological argument. For one thing, it escapes the logical difficulties which afflict the classical actus purus doctrine. But its positive strength is (1) to provide an understanding of Godís identity as an individual through abstractions alone ("This does not imply that God is a merely abstract entity, but only that what makes God God and no other individual is abstract"15), and (2) more specifically, to identify God as the sole individual with strictly universal functions, defined with relation to actuality as such and with respect to possibility as such. Hence, God as an existing being can be seen to conflict with nothing whatever, God being the foundation (the ground or source) of all things and the preserver of all through the unsurpassable power to include all omnisciently.
The intuition into Godís unique universal role and utter noncompetitiveness yield is also the intuition that no existent could either cause Godís existence or prevent it. Such independence of any possible condition identifies a pure possibility -- what is possible in a special sense because nothing conceivable threatens or rivals it.
At this point, we must face a possible objection. For without contending that Hartshorneís doctrine of God is incoherent or that something might, if it were actual, put an end to Godís existence, in short, without contending that God is impossible, one might claim simply that a world without God makes sense. The claim appears modest enough -- merely the assertion that all beings may be and perhaps are imperfect in some fashion.
My view is that the other arguments for God -- those other than the ontological argument -- do not directly provide the postulate of logical possibility. But they do function to protect that possibility from alleged alternatives -- such as the one Just suggested, namely, that it happens or might happen to be the case that all beings fall short of perfection. Accordingly, the ontological argument has an independent status; it is not superfluous. Yet its postulate that divine existence is possible can and must be defended against specious claims and proposals that a world without God is as possible as one with God.
Consider the above-mentioned objection. It might be put as follows: "God does not exist, but not because some unfortunate accident incompatible with divine existence has brought about Godís nonexistence or prevented Godís existence. No, the world is just a Godless world in which all beings are surpassable." In effect, the objector holds that either a Godfull world or a Godless world is possible, though only the latter obtains.
It seems to me that the other arguments can and do connect appropriately with such an objection. For consider, to begin with, the fact that no one could ever know the world to be Godless. I am assuming of course that the objector means to insist (among other things) that in the Godless world no individual possesses omniscient awareness. Let us examine this hypothesis. How might we establish it? Even if we surveyed each part of the universe and found them all cognitively deficient, we would not have succeeded in showing that God does not exist. For God is no part of the universe. "Nothing less than the all-inclusive reality, or at least the universal cause of reality, is even a candidate for the status Ďdivine.í " 16 But, in any event, the notion that beings imperfect in knowledge could inspect all parts of the universe is implausible at best. So if one holds that all individuals are or might be imperfect, none having omniscience, one is committed to a position which is strictly unknowable. "For how can any incomplete survey of reality constitute evidence that there is not a complete survey?"17
Of course, it might be argued that an unknown and unknowable truth -- say, the truth about all things and their imperfections -- is nonetheless a truth, since it is true by dint of its relation to reality. But note, we now have the hypothesis that, though no knower is free of ignorance and confusion, there is still a pure and comprehensive truth about reality. It is, we might say, a perfect truth -- one such as a perfect knower would possess were such a knower to exist. But we do not assess beliefs and ideas by comparing them with bare, unexperienced reality. We compare them with various aspects of our experience: expectations are examined in light of subsequent experiences, concepts are evaluated by their adequacy to perceptions, etc. So what sense can we find in the supposition that truth is independent of experiencing or knowing? Indeed, is not reality just "what is or would be experienced and known by an experience free from doubt, unclarity, inconsistency, and unanswered questions"18?
There is, however, another (and perhaps more convincing) attack on the contention that a Godless world is possible. For if a Godless world is one without an omniscient knower, this, if true, is a wholly negative fact. We assume that the objector does not hold that some set of conditions or accidents has put God out of business or kept God from coming to be in the first place. Rather, the objector takes it to be conceivable that all beings just are, simply are, imperfect.
Since it is assumed that no fact in the Godless world is such that, were it nonexistent, God would exist, the absence of God is a sheer absence, with no positive bearing or significance. Most negations -- like ĎĎno fire in the room" -- imply some positive factual alternative -- like "every part of the room below ninety degrees centigrade" or "air in the room clear and fresh." Indeed, such facts, falling under observation, give us ground for denying states of affairs which are incompatible with them.
But if God is just "not there," and yet all else is the same -- if a world where no individual is perfect is a conceivable world -- we would lack any positive evidence for the divine absence. This absence would in principle be unknowable, since nothing one might observe would or could constitute the meaning of that absence. But what no one would or could conceivably experience does not represent a conceivable state of affairs.
Hartshorne contends that all fact must have its positive aspect, merely or exclusively negative facts being impossible. This "axiom of positivity" he asserts to be a truism, yet one with immense consequences.19 One such consequence is that "Ďthe non-existence of Godí could have no possible positive meaning, and . . . this suffices to render it logically null and void." 20
Now, the supposition that God might fail to exist, that this is just as possible as Godís existing, is not an empirical supposition. Rather it is simply an attempt to identify a priori something conceivable. My contention, to repeat, is that against such a supposition arguments other than the ontological argument are effective. Thus they may protect the postulate of logical possibility (so essential to the ontological argument) by discounting an alleged correlative or complementary possibility of nonexistence. They do not then directly establish that "God exists" is possible. That, as I have argued, is an implication of the concept of God itself. What those arguments provide the ontological argument is an a priori repudiation of the conjunct "and also possibly God fails to exist, all things being imperfect."
I have not, in fact, employed the lineaments of the classical cosmological argument in seeking an a priori negation of that conjunct. Bather I have had recourse to what Hartshorne calls the epistemic or idealistic argument.21 My two theses have been (1) that in a world of imperfect beings the truth that no omniscient knower exists would simply be unknowable -- which cannot possibly be so -- and (2) that in such a world no fact could furnish evidence for Godís nonexistence, hence Godís nonexistence would be a purely negative fact, and so nonsense. I believe that various arguments can be used against that a priori hypothesis of a world without deity, depending on how in particular (with which emphases and details) one undertakes to describe that world.
Where does the famous problem of evil fit into the discussion? Arguments from evil take us beyond the position of our objector who claimed only that a Godless world is also possible. For those arguments aim to show that God is either improbable or downright impossible. They represent a direct assault on the possibility postulate of the ontological argument. It is the concept of God which is at stake and which, in and of itself, must either conquer or be conquered by the alleged incompatibility of evil with Godís existence.
If God is conceived as all-powerful, as in classical theology, and this attribute is taken to mean that all decisions are Godís -- that in truth God is the only agent -- then the theist must surely deny either the existence of God or the existence of evil. In fact, the very conceivability of evil in any possible world would, as Hartshorne declares, wreck theism if divine power and goodness could not coexist with evil. For in that case God would have to be conceived as contingent even if God were to exist. Since evil is conceivable, and contingency of existence is surpassal.1e, theism is absurd unless Godís existence is compatible with the existence of evil.22
But of course Godís existence is not compatible with the existence of evil if God exercises a sheer monopoly in decision-making. So if God exercises such a monopoly, theism is absurd. Yet is not divine power capable of excluding all evil and guaranteeing universal harmony? Neoclassical theism answers that to be a creature at all is to be a creator on the nondivine level.23 So if God has creatures, their self-determinations will always add definiteness to the range of potentiality furnished by their causal conditions. Of course, if every creature is to some extent self-determining, then there is always risk of conflict and suffering. Since, for Hartshorne, nothing -- from angels to man to particles -- is simply determined by deity, risk of evil is unavoidable.
How does this doctrine fit with the view that God is unsurpassable power? To begin with, omnipotence -- as the making of all decisions by one agent -- is a pseudo-idea. This follows from the fact that a creatureís concrete decisions cannot be made by another individual or by any situation or law. Theological determinism, Hartshorne points out, divides all decision-making into (1) Godís, which is supposed to be utterly free (and inexplicable) and (2) creaturesí, which is supposed to be utterly unfree (and wholly explicable). To take the term decision from its context among creatures and extend it to God is then to leap an absolute gulf.24
Moreover, it is not weakness in God that God does not enforce universal harmony. It is an implication of Godís having creatures, and of their having self-determination, that there is risk of discord. Yet God sets limits to the freedom of action for creatures, namely, the laws of nature, which no creature decides. In this sense God guarantees limits to the potential chaos of the universe. What is often overlooked or denied is that God has receptive or passive power to take the contributions of the creatures and to decide what use to make of them in Godís ongoing life. Godís power would indeed be deficient were God incapable of receiving the determinations of the creatures. Of course, this requires that God be a self-surpassing being, not unmoved or immutable. It also requires that the sufferings of creatures, not merely their joys, be internalized by God, who thereby suffers with the creatures -- a departure from the classical view of divine bliss.
The heart of Hartshorneís doctrine is that God does not manipulate or set concrete events at all. What God decrees is what sort of risks of evil shall accompany what sort of opportunities for good. For risk is inseparable from freedom, as is opportunity. The laws of nature maximize the chances for good, while minimizing those for evil. "The chances of evil are subordinate to the chances for good, so that good is the primary overarching probability."25 Note that even if each creature chooses a good course for itself, achieving internal harmony, it is partly luck whether that course harmonizes with, or clashes with, the courses chosen by other creatures. In short, a good can conflict with other goods.
What restricts freedom is always other exercises of freedom that is, decisions made by oneself and others (including atoms and cells) in the past. "Even the laws of nature are divine decisions made long ago, and freely made."26 Past exercises of freedom, including Godís in establishing the order of nature, is what is meant by causality.
I have, I hope, presented the main features of Hartshorne s view of the problem of evil. What should be clear is that the basic issue is the concept of God. For where there is confusion or inconsistency in that concept, there is correlative difficulty regarding what it means to be a creature.27 If a creature could escape all evil, and indeed even its possibility, and escape it because Godís perfect power and goodness would prevent anything undesirable from occurring, then the very conceivability of evil would render theism absurd. Hence, the postulate of logical possibility, which, as I have argued, rests on the cogency of the idea of God, must by virtue of that very idea reveal the problem of evil to involve a misunderstanding. Thus, it is not here germane -- at least, not directly germane -- to recur to the various other arguments for Godís existence. What is germane is the clarification of the idea of God. There is reason to believe that the conceivability of evil, perhaps especially natural evil but also moral evil, impugns the classical doctrine of divine power and thereby shows that doctrine to be defective. Hartshorneís neoclassical view is that decision-making cannot be monopolized by deity (hence the classical doctrine is intrinsically impossible). For the only way God could have creatures or a world is to set creatures free, within limits -- or, better, set limits on the inherent freedom of creatures -- and thereby risk evil. So the problem of evil, it turns out, is based on the misapprehension that God could monopolize decision-making and still have creatures. Once it is seen that evil is to be explained by the presence of many free creatures in the world, and by their social relations, that misapprehension becomes evident to us. The classical doctrine of omnipotence and the problem of evil are each aspects of the misapprehension, and accordingly both are precluded by the neoclassical vision of God.
The ontological argument properly focuses attention on the postulate of logical possibility and hence on the doctrine of God itself. That doctrine must carry within its conceptualities the warrant for the postulate. Should it be claimed that any set of circumstances or its possibility is incompatible with the existence of God, the burden then falls on the advocate of theism to show that the claim is not only false but confused. For it is impossible that a conceivable set of circumstances be such as to conflict with the existence of that individual who is presupposed by every actual and every possible state of affairs. Once the theist has clarified in a doctrine of God the universal functioning and hence the inalienability of the divine individual, the ontological argument proceeds of its own strength.
On the other hand, should it be assumed that there is a matching possibility opposed to the possibility of divine existence, the other arguments for God can be called on to exclude that assumption. Those arguments may proceed directly to the conclusion that a necessary being exists. But this in no way weakens the ontological inference from the concept of deity to necessary existence. In short, the ontological argument is by no means superfluous, since it does not rest on the other arguments to guarantee the postulate of logical possibility, and gains support from them only insofar as that postulate is protected by those arguments against options they show to be specious.
1B. L. Purtill, "Ontological Modalities," The Review of Metaphysics 21 (December, 1967), 303.
2Ibid., p. 306.
4Ibid., p. 307.
5See Charles Hartshorne, The Divine Relativity; (New Haven: Yale University Press, 1948), p, 31.
6See Charles Hartshorne, "What Did Anselm Discover?" The Many-faced Argument (New York: Macmillan Co., 1967), p. 324. See also Charles Hartshorne, "Rationale of the Ontological Proof" Theology Today 20 (July, 1963), 281, and Charles Hartshorne, "The Formal Validity and Real Significance of the Ontological Argument," The Philosophical Review 53 (May, 1944), 232, 240.
7Hartshorne, "What Did Anselm Discover?" p. 324.
8Ibid., p. 330.
9Charles Hartshorne, "Rejoinder to Purtill," The Review of Metaphysics 21 (December, 1967), 309.
11George L. Goodwin, The Ontological Argument of Charles Hartshorne (Missoula, Montana: Scholars Press, 1.978). p. 55. Goodwinís italics.
12See Charles Hartshorne, Anselm Ďs Discovery (La Salle, Illinois; Open Court Publishing Co., 1965), p. 123. Note the qualification added in Hartshorneís parenthetical comment, that "God does not have the anger of the angry man as the man has it . . ." Ibid.
13Charles Hartshorne, The Logic of Perfection (La Salle, Illinois: Open Court Publishing Co., 1962), p. 105.
14See Hartshorne, Anselmís Discovery, p. 185.
Charles Hartshorne, Creative Synthesis and Philosophic Method (La Salle, Illinois: Open Court Publishing Co., 1970), p. 246.
16Charles Hartshorne, "Is the Denial of Existence Ever Contradictory?" The Journal of Philosophy 63 (February, 1966), 88.
17Charles Hartshorne, "Royceís Mistake -- and Achievement," The Journal of Philosophy 53 (February, 1956), 127.
18Charles Hartshorne, "Royce and the Collapse of Idealism," Revue Internationale de Philosophie 23, 79-80 (1967, Fasc. 1-2), 51.
19See Charles Hartshorne, ĎNegative Facts and the Analogical Inference to ĎOther Mind,ííí Dr. S. Dadhakrishnan Souvenir Volume, ed. J. P. Atreya (Moradabad, India: Darshana International, 1964), p. 151.
20Ibid., p. 149.
21Hartshorne, Creative Synthesis, p. 286.
22Ibid., p. 292.
23See ibid., p. 293.
24See Charles Hartshorne, "A New Look at the Problem of Evil," Current Philosophical Issues: Essays in Honor of Curt John Ducasse (Springfield, Illinois: Charles C. Thomas Publications, 1966), p. 204.
25Ibid., p. 210.
26Ibid., p. 211.
27See ibid., p. 202.