James W. Felt, "Why Possible Worlds Aren't"

From The Review of Metaphysics, 50 (September 1996), 63-77. “Agents in this world determine the events in this world, together with their definite ways of being . . . . But where are the agents that define all the other ways things could have been, all the hypothesized possible worlds?”

Here is an uncommon argument for doubting the existence of possible worlds. It calls into question the whole spectrum of supposed possible worlds, from Lewis’s radically plural real worlds to the world-stories of Adams and Plantinga. More than that, it challenges the tacit presuppositions of most of those who have attacked such views. Yet despite its strangeness I cannot but think this unorthodox position is correct. I shall furnish metaphysical reasons for thinking that it is, and then proceed to show why the possible worlds taken for granted in contemporary discussions are based on a misconception, a metaphysical mistake that we would be better off without.

I rest this unusual claim on the ground of a metaphysics that is at odds with the metaphysical viewpoint implied in the theories of possible worlds. I suggest a different and, I think, superior way of conceiving the world, experience, and what we mean by possibility.

We are faced here with the deliberate choice of philosophic first principles, not with demonstrations proceeding from commonly acknowledged or even tacitly presupposed principles. So there can be no knockdown argument for such a choice, only reasons for thinking it better than its alternatives. Lewis himself has noted that “one man’s reason is another man’s reductio,”¹ and, after arguing against “ersatz” substitutes for his thoroughgoing modal realism, he advises: “Join the genuine modal realists; or foresake genuine and ersatz worlds alike.”² On the view that I shall propose, Lewis’s modal realism turns out to be itself a reductio rather than a reason, and so I must set it aside with all its variants, ersatz or otherwise.

Philosophic beginnings are chosen, but their consequences are not. As Gilson put it: “Philosophers are free to lay down their own sets of principles, but once this is done, they no longer think as they wish-they think as they can.”³ These first principles are chosen not only for their initial plausibility, but most importantly because they appear to make more intelligible sense of experience than do their opposites. Consonance with experience is the final criterion for accepting or rejecting any philosophic standpoint. As Whitehead mentioned, it has been said that systems of philosophy are never refuted, they are only abandoned, either by reason of the mutual incoherence of their principles or because they are inadequate to account for experience as we find it.⁴

I begin this consideration, then, by recommending the plausibility of three metaphysical principles that suggest themselves, both initially and in their consequences, as characterizing our immediate, ongoing, changing experience.

Three Principles of Becoming and Being.⁵ These principles are meant to express, at least partially, the character of human experience viewed in terms of its changing patterns over time. They link the dynamism of activity with its own formal patterns; they link temporal actuality with possibility.

PRINCIPLE (A): Past actuality, whether immediate or remote, is definite, exact, unambiguous.

Lady Macbeth observed that what’s done cannot be undone. But also, what’s done, being done, has its own definite character. Though knowledge of the past fades, including knowledge of one’s own past self, this past is not in itself ambiguous. We have to cope in the present with what has in fact been decided, by us and by others, in the past. The pattern of that past is settled, now and always. And how does this settled pattern come about?

PRINCIPLE (B): Present actuality involves a process of determination whereby from the indefiniteness of what might be (the range of possibility), there is created the definiteness of what actually is (actuality).⁶

Take the writing of a philosophic essay. It is a process by which more general initial thoughts take on the definition of exact formulation, a process by which the vagueness of what might be written takes on the definiteness of what is written. A more vivid analogy is that of a dune buggy on a beach. The margin in which you can drive is perhaps already determined by the surf on one side and sheer palisades on the other, so that these two factors literally determine a range of possibility for your driving. However, within that width you can drive (steer) as you please, and it is just your activity of driving that determines the track that you make in the sand. The definiteness, the pattern of that track, is created precisely by your act of driving.

PRINCIPLE (C): Only actual events create this definiteness of settled actuality within the given range of possibility.

Principle (C) is roughly the converse of Principle (B). (B) asserts that actual events always exemplify, because they produce, definiteness, an exact pattern of actuality. (C), on the other hand, asserts that such definiteness requires actual events for its own creation. Of itself the range of possibility leaves undetermined the particular patterns that may be defined within it.

This principle embodies the central point of controversy around which this essay pivots. It is derived from reflection on our experience of the creation of the patterns of ongoing temporal events. Thus the pattern of tracks in the sand is created precisely, and only, by the driving, just as the exact way in which you will walk through the next doorway will be created precisely, and only, by your act of walking through it. The act, by its very nature, creates its own pattern. Better, you by your act of walking create the pattern.⁷

Corollary (1): In temporal events at least, the definiteness or pattern of the event does not temporally precede the event itself.

This is what Henri Bergson meant by saying that the possible does not precede the real:

As reality is created as something unforeseeable and new [because undefined), its image [its pattern of definiteness) is reflected [mentally] behind it into the indefinite past; thus it finds that it has from all time been possible, but it is at this precise moment [of temporal creation) that it begins to have been always possible, and that is why . . . its possibility, which does not precede its reality, will have preceded it [for the mind] once the reality has appeared. The possible is therefore the mirage of the present in the past.⁸

This viewpoint is so unusual and so critical to the argument of this essay that it bears more illustration. Thus, the pattern of the actual track in the sand had no existence in the world prior to the driving that made the track. However, when we look at the track already made, we naturally suppose that that pattern must have been possible even before it was made—that it must have constituted a particular possibility preceding the making of the track. We suppose that this particular possibility was only one of an infinity of such possibilities awaiting realization by the driving. Yet we make this supposition only by means of mental rear-view mirrors. Only after the track is made do we have a pattern to refer to and then to project mentally into the past as having been antecedently possible.

It is almost impossible to exaggerate the seductiveness of this natural tendency to relocate the present back into the past. Yet we deceive ourselves when we suppose that the pattern in fact preexisted the driving.

This becomes more evident if we consider the production of a piece of literature or of music. Mozart created the pattern of notes which is his 40th symphony. Once created, this definite pattern can be thought of as having always been possible—otherwise how could it have been created at all? But only with the actual creative activity that went into the composition of the symphony does there exist a pattern, a single “possibility,” to talk about at all. Prior to its conception by Mozart it was not, of course, impossible—that is, there were no intrinsic impediments to its creation—yet there were no definite forms at all, to be called possible or even impossible. There was literally nothing specific to talk about or think about.⁹

To think otherwise is to suppose that the exact pattern of notes which is the 40th, existed as such, perfectly defined, in some ghostly limbo of possibilities (one is reminded here of Quine’s “slum of possibles”),¹⁰ awaiting the infusion of actuality by Mozart. On such a view there would never be artistic creation or novelty, only selection from among pre given definite possibilities.

To return finally to the dune buggy: according to the common way of thinking of the possible and the real, there is an infinity of possible tracks lying ahead on the beach, and, by your driving, you select one for realization. This way of thinking implicitly likens the beach to a railroad yard, where the locomotive enters upon one of several tracks already laid down in advance. In fact, there are, however, no preexistent possible tracks in the sand just because there is nothing to define them ahead of time. Yet a single track does get created by the actual driving. Only after the driving, however, is there a particular track one can talk about or suppose, by hindsight, to have been possible prior to its creation.

In the wording of Corollary (1) I added the qualification, “in temporal events at least.” By this I meant that the form or pattern of an actual event (its way of being or becoming) has no existence in time prior to the event itself. Prior to the event its own form simply is not, and (as Parmenides or even Wittgenstein would say) cannot even be spoken of. This I consider an ineluctable conclusion if one takes seriously that novelty does arise in the world. More than that: I now propose to remove the above qualification, and to state, more broadly:

Corollary (2): No definiteness of form obtains at all, temporally or atemporally, except in virtue of the activity of a real agent.

This, doubtless, is my most controversial claim. Yet given what I have contended in justification of Corollary (1), it seems intuitively plausible to make an analogous assertion concerning patterns of definiteness (“possibles”) conceived just in themselves, apart from any emergence in the temporal world. So conceived they are atemporal, not of themselves involved in time. ¹¹

Some examples of such pure patterns of possibility would be Plato’s Forms, the possibilia of the medievals, and Whitehead’s “eternal objects.”¹² However, if one accepts that a particular pattern of definiteness is realized (in the literal sense: becomes part of a res) in temporal events only as the result of the defining activity of an agent, it seems to me reasonable to think that definite patterns of pure possibility, apart from any temporal realization, also require the activity of a defining agent. This seems more plausible than supposing that such distinct, definite patterns simply exist, in their very distinctness, within the general range of possibility.

I have noted that with regard to first principles, such as the above, there can be no knockdown argument one way or another. Rather, we are at a level at which we must in effect play philosophic cops and robbers. “Bang, I got you!”“No, you missed me!” As the saying goes, you pays your money and you takes your choice.

Why possible worlds aren’t. Let us now apply the above principles to current notions of possible worlds, and—because of its fundamentality and lucidity—first of all and chiefly to the possible worlds of David Lewis. How does he conceive possible worlds, and why does he think they exist? Here is his well-known reply:

I believe that there are possible worlds other than the one we happen to inhabit. If an argument is wanted, it is this. It is uncontroversially true that things might be otherwise than they are I believe, and so do you, that things could have been different in countless ways. But what does this mean? Ordinary language permits the paraphrase: there are many ways things could have been besides the way they actually are. On the face of it, this sentence is an existential quantification. It says that there exist many entities of a certain description, to wit “ways things could have been.” I believe permissible paraphrases of what I believe; taking the paraphrase at its face value, I therefore believe in the existence of entities that might be called “ways things could have been.” I prefer to call them “possible worlds.”¹³

Twelve years later, looking back at the controversies his view stirred up, Lewis took pains to insist on the existential import of what he is claiming:

I must insist that my modal realism is simply the thesis that there are other worlds, and individuals inhabiting those worlds; and that these are of a certain nature, and suited to play certain theoretical roles. It is an existential claim, not unlike the claim I would be making if I said that there were Loch Ness monsters, or Red moles in the CIA, or counterexamples to Fermat’s conjecture, or seraphim. It is not a thesis about our semantic competence, or about the nature of truth, or about bivalence, or about the limits of our knowledge. For me, the question is of the existence of objects-not the objectivity of a subject matter.¹⁴

That is what Lewis claims; he does not pretend to be able to prove it, but rather to offer plausible reasons why it is so. In On the Plurality of Worlds he only argues that modal realism is useful, whereas in the above argument from Counterfactuals he had made a stronger claim. However, if the above principles are right, Lewis’s argument runs right off the rails at its second step. That is where he supposes that saying that “things might be otherwise than they are” is equivalent to saying that “there are other ways things could have been.” The first statement, however, reflects how the present state of affairs arose within a general range of possibility set by its antecedents, while the second statement posits the existence of a multiplicity of distinct, particular possibilities, none of which has been actualized by the course of events.

These are not two different ways of saying the same thing; they are two different statements altogether. To say that the track in the sand could have had a different shape does not entitle one to say that there exist other definite shapes that it could have had. What agency has produced the definiteness of these supposed “other shapes” whereby we may talk of them? And what agency has produced the definiteness of the multiplicity of “other ways things could have been,” Lewis’s possible worlds? Agents in this world determine the events in this world, together with their definite ways of being, just as you, in your act of walking, will determine the way in which you next walk through a doorway. But where are the agents that define all the other ways things could have been, all the hypothesized possible worlds? Lewis tells us that he is making an existential claim about these possible worlds. In his view there is the same existential definition to these worlds as there is to the one we happen to inhabit, in which exactly these and these events obtain and others do not. What can possibly be the agent or agents that give their respective definitions to Lewis’s plurality of possible worlds?

Modal realists such as Lewis, however, would doubtless challenge the presuppositions inherent in that question. They would say that I am presupposing a distinction between the agents of this world and their hypothetical (“counterpart”) agents in other possible worlds; that I am attributing an ontic priority to the former; that I am making an undemonstrable distinction between this real world and the other worlds posited as possible, such that only agents of this (real) world are really agents, in contrast to so-called but unreal agents of possible worlds.¹⁵

Now I willingly grant the accuracy of this charge but deny that it is a fault. Cops and robbers. I think, though I cannot prove it, that a possible or hypothetical or counterpart agent, simply is not an agent.¹⁶ I think, though I cannot prove it, that there is an ontological difference as ways of being between possibility (I would rather say potentiality) and actuality. Lewis, on the other hand, writes: “Nor does this [actual] world differ from the others [other possible worlds] in its manner of existing. I do not have the slightest idea what a difference in manner of existing is supposed to be.”¹⁷

On the view here proposed, to be potentially is a real way of being that differs essentially from being actually.¹⁸ But potentiality necessarily inheres in what is actual. Thus the range of what is possible is determined by what is in fact actual, just as the range for driving the dune buggy is set by the actuality of the cliffs and the surf. Relations of possibility and necessity presuppose and depend on what is actual.

Furthermore, potentiality as such cannot consist in a set, even an infinite set, of distinct, definite possibilities. It is rather the inherent capacity within given actuality for any such definite possibilities to be realized by the activity of appropriate agents. Potentiality as capacity differs ontologically from any set of forms of definiteness that it makes possible. Hence to confuse possibility, as potentiality, with a set of hypothesized possible patterns is to make a category mistake. Furthermore, it leaves unanswered and unanswerable how one is to account for the definite distinctness of the hypothesized possibilities.

There is a remarkably apposite passage in Aquinas in which St. Thomas not only makes roughly the same point, but practically puts modern theories of possible worlds into that perspective. In a digression in his commentary on Aristotle’s treatise On Interpretation he writes:

There are various opinions about possibility and necessity, about can be and must be. Some people, like Diodorus, distinguish these by reference to what happens, saying that can’t be never happens, must be always happens, and can be sometimes happens, and sometimes doesn’t. The Stoics, however, distinguished them by reference to external obstruction, saying that the truth of must be can’t be obstructed, of can’t be is always obstructed, and of can be can sometimes be obstructed and sometimes not. Neither way of distinguishing them seems adequate. The first way gets things back to front: it is not that things must be because they always happen, but that they always happen because they must be, and similarly for the other definitions. And the second way appeals to something extrinsic and so to speak inessential: it is not that things must be because nothing obstructs them, but that nothing can obstruct them because they must be.¹⁹

By extrapolation, Thomas would say that it is getting things backwards to say that something is necessary because it is found in all possible worlds (if there are any), rather than to say that some particular thing is found in all possible worlds because it is necessary. Thomas evidently takes possibility and necessity to reside primitively within the relations of what is actual. Class inclusion is derivative from those relations, not definitive of them.

Since I am persuaded by the above principles, I must conclude that there can be no possible worlds in the sense required by Lewis’s modal realism, for there are no agents to produce the definiteness of their patterns. Possible worlds, in his sense, cannot be existential objects as he supposes; they can at most be his own conceptions of possible worlds.²⁰

What of other current conceptions of possible worlds—what Lewis calls “ersatz” possible worlds? It will suffice here to examine one such conception, for all of them suffer from the same defect. Lewis is right in maintaining that genuine and ersatz possible worlds fall together, if they fall at all.

Let us consider in a general way the world-story or world book interpretation of possible worlds suggested respectively by Robert Merrihew Adams²¹ and Alvin Plantinga. ²² Instead of thinking of a possible world as another object like the real world, as Lewis does, one can think of it as a set of propositions forming a story or book describing a world. If a proposition can denote a state of affairs, then suppose there exists a maximal, consistent set of propositions describing a totality of states of affairs. It is maximal in the sense that no further proposition can be added to it without either duplicating or contradicting some proposition already contained in it. One such world story will, furthermore, be distinctive, in that all its propositions will in fact be true. That is, it alone will describe the real world.

It would be pointless here to review the alleged advantages of world-story possible worlds over those of Lewis, for if the above principles are sound it is evident that world-stories are as metaphysically specious as are his. For world-stories are presumed to enjoy definiteness of and on their own. The multiple—even infinite—possible world-stories are simply taken as given in all their detail. Only on that supposition does it make sense to distinguish one of them from another and to allege that one has the distinction of being wholly true. Yet no agent can be assigned that is capable of having made up the stories—at least no agent short of God.²³ Certainly no human mind could invent the complete story of a universe.

Once again, then, I must conclude that there are in fact no existing possible worlds for us to deal with, either genuine or ersatz, inasmuch as no agent is available to have created their alleged definiteness.

What, however, shall we say about Lewis’s contention that a good reason for thinking there are possible worlds is that it is useful to do so? Here is how he puts it:

Why believe in a plurality of worlds?—Because the hypothesis is serviceable, and that is a reason to think that it is true. The familiar analysis of necessity as truth at all possible worlds was only the beginning. In the last two decades, philosophers have offered a great many more analyses that make reference to possible worlds, or to possible individuals that inhabit possible worlds. I find that record most impressive. I think it is clear that talk of possibilia has clarified questions in many parts of the philosophy of logic, of mind, of language, and of science—not to mention metaphysics itself.²⁴

Lewis spells out a number of such instances, and it is not possible here to examine them individually. I submit, however, that in each case Lewis fails to show that it is useful to think of possible worlds as ontically existing, as modal realism asserts, rather than that certain possibilities, or even vaguely defined possible worlds, are simply conceived by us. Conceptions suffice in every case; there is no need for autonomously existing possible worlds.

Conclusion: Possible Worlds and Platonism. The Platonic method of philosophizing has been described by Pegis as “the method of modeling the properties of existing beings on the abstractions of the human intellect. In other words, it is the method of thinking that being takes its characteristics as being from what it reveals of itself in the state of being thought.”²⁵ Pegis intended, and so do I, to distinguish Platonism from the contrasting viewpoint of Aristotle or the adaptation made of it by Aquinas, and I doubt that one could find a more striking example of doing philosophy in the Platonic manner than the contemporary possible-worlds ontologies, especially the modal realism of David Lewis. Plato and Aristotle embodied the two fundamentally different ways of regarding the world and of doing philosophy that characterize thinkers in the whole course of Western philosophy. It is therefore illuminating to set theories of possible worlds into this thematic and historical context. I close this essay, then, by pointing to the Platonic strain that is paramount in the ontologies of possible worlds.

First of all, there is in Lewis’s thought an identification of definiteness with actuality, and to make this identification is precisely to adopt the Platonic viewpoint. The identification underlies his indexical theory of what it means to be actual, so that the inhabitants of other worlds, who are just as definite (and apparently just as active) as we, are supposed to have as much right to call their worlds actual as we do ours. For Lewis, all possible worlds are as real as our own, differing only in their spatio-temporal locus; they do not differ in their manner of existing.

Thus Lewis has in effect accepted the conclusion that Hartshorne reached, that belief in possible worlds erases the distinction between the possible and the actual. As Hartshorne put it: “One must admit that a possible world is as definite and complex as the corresponding actual one. This, I hold, reduces the distinction between possible and actual to nullity. Value is in definiteness, and definiteness is the ‘soul of actuality.’ Were possibility equally definite it would be redundant to actualize it.”²⁶

Again, in Lewis’s argument for the existence of possible worlds, the conviction that things could be otherwise becomes the conviction that there are other ways things could have been. “Otherwise” has become “other ways,” and that is exactly a Platonic move. Because we can conceive of other ways, those other ways are assumed to be constitutive of extramental reality.

In this regard, we may note the affinity between modal realism and the quintessentially Platonic “ontological” argument for the existence of God. When I conceive “that than which no greater can be conceived,” I must ipso facto conceive “a necessarily existing being.” Therefore such a being exists. This, however, slips in the tacit assumption that what I have conceived in my mind has an extramental counterpart, and that is 24 carat Platonism. Similarly Lewis, because he conceives other ways of being, including other possible worlds, habitually supposes that these other ways of being extramentally exist.

Again, potentiality (or range of possibility inherent within given actuality) becomes, for Lewis, possibilities, discrete, definite, conceivable patterns of being. Aristotle’s dynamis, as a manner of existing, has been replaced by an assemblage of conceivable patterns.

Lewis has passed from potentiality to patterns. But this is retrogression beyond Aristotle to Parmenides. It renders impossible Aristotle’s solution to Parmenides’ dilemma of becoming, for it denies any intrinsic distinction as ways of existing between being actually and being potentially.

The same pattern of thinking is apparent in the way Lewis relates causality to counterfactuals. He conceives that the way the letters appearing in his book causally depend upon the keystrokes of his word processor is just that if he had touched different keys, different characters would have appeared. “That is how the letters depend causally upon the keystrokes.”²⁷ Thus the causal power of actuality has passed into counterfactual patterns of possibility. The attempt to describe causal relationships has left him lost in the “woulds” of counterfactuals.

And they are dark woulds indeed, since they do not in fact succeed in describing causal power. They substitute instead a pattern of conceived hypothetical possibilities. These possibilities are not only purely conjectural, they even presuppose the causal power that they are meant to define.²⁸

Confiding to Dorothy why everything in Emerald City was green, Oz explained: “When you wear green spectacles, why of course everything you see looks green to you.”²⁹ Everyone, I suppose, wears philosophic glasses of one prescription or another, and Lewis has chosen to see everything through the lenses of possible worlds. He admits that this entails a vast expenditure of entitative suppositions, but nevertheless insists that the price is right.

The argument of this essay implies that the price, far from being right, is simply exorbitant. Possible worlds, in the currently accepted senses, are the monstrous issue of a metaphysical mistake; they are unnecessary for realizing the advantages Lewis seeks; and investing in them promotes rampant metaphysical inflation at a time when the credibility of philosophers’ currency is already badly undermined.

¹ David Lewis, On the Plurality of Worlds (hereafter, OPW) (New York: Basil Blackwell, 1986),207.

³ Etienne Gilson, The Unity of Philosophical Experience (Westminster, Md: Christian Classics 1982), 302-3.

⁴ Alfred North Whitehead, Process and Reality (hereafter, PR) (New York: Free Press, 1978), 6.

⁵ I have proposed these principles elsewhere, especially in Making Sense of Your Freedom: Philosophy for the Perplexed (Ithaca: Cornell University Press, 1994),56, and in a complementary essay, “Impossible Worlds,” International Philoso-phical Quarterly 23 (1983): 251-65. They amount to a reformulation of a fundamental conceptuality of Henri Bergson.

⁶ Here I am using “actual” and “actuality” in a broader sense than does Lewis, but the difference is irrelevant to this principle.

⁷ Principle (C) bears an obvious resemblance to Whitehead’s “ontological principle” as described in his eighteenth Category of Explanation: “That every condition to which the process of becoming conforms in any particular instance has its reason either in the character of some actual entity in the actual world of that concrescence, or in the character of the subject which is in the process of concrescence. This category of explanation is termed the ‘ontological principle.’ It could also be termed the ‘principle of efficient, and final, causation’. This ontological principle means that actual entities are the only reasons; so that to search for a reason is to search for one or more actual entities”; Whitehead, PR, 24.

⁸ Henri Bergson, “The Possible and the Real,” in The Creative Mind (Totowa, N.J.: Littlefield, Adams, 1975)101.

⁹ If it is true that a Mozart can conceive in his mind an entire symphony, and perhaps all at once, then I would grant that that conception constitutes the definiteness of the symphony even prior to its being written down or performed. The point is that the form of definiteness had no existence prior to Mozart’s creative activity, whether that was purely mental or also physical.

¹⁰ W. V. O. Quine, From a Logical Point of View (Cambridge: Harvard University Press, 1953), 4.

¹¹ This is not to say that they are omni-temporal, as if they existed at all times, but rather that, precisely as patterns, they contain no reference to temporal realization at all.

¹² In a manner analogous to that of this essay, Lewis S. Ford has called attention to the difficulty of reconciling Whitehead’s theory of eternal objects with Whitehead’s own “ontological principle.” Ford elaborates this difficulty and suggests an emendation of Whitehead’s metaphysics to circumvent it. See Ford’s essay, “The Creation of ‘Eternal Objects,’” The Modern Schoolman 71 (1994): 191-222.

¹³ David Lewis, Counterfactuals (Cambridge: Harvard University Press, 1973),84.

¹⁵ Here I have avoided using the word “actual,” to which Lewis has attributed a special meaning. As the reader probably knows, Lewis takes “actual” to be an indexical term, like “here” or “now”—that is, as bearing an implicit reference to the world of the speaker. Thus “actual” refers exclusively to events taking place in the speaker’s own world. This allows Lewis to deny any special priority to the world that we inhabit since, as he says, inhabitants of other possible worlds have as much right to refer to their own worlds as actual as we do to our own; see Lewis, Counterfactuals, 85-6.

¹⁶ I have sketched some features of an appropriate theory of agency in chapter 7 of my Making Sense of Your Freedom.

¹⁸ Here and hereafter I decline to restrict the word “actual” to Lewis’s indexical sense.

¹⁹ Thomas Aquinas, “Commentary on Aristotle’s De Interpretatione,” 1.9 (18b26-19a22), in Aquinas: Selected Philosophical Writings, trans. Timothy McDermott (Oxford: Oxford University Press, 1993), 277.

²⁰ Nicholas Rescher comes close to this view when he insists that the human mind plays an essential role in establishing possibilities. He writes: “It is my central thesis that by the very nature of hypothetical possibilities they cannot exist as such, but must be thought of: They must be hypothesized, or imagined, or assumed, or something of this sort”; Nicholas Rescher, “The Ontology of the Possible,” in The Possible and the Actual, ed. Michael Loux (Ithaca; Cornell University Press, 1979), 167. Yet he does not, I think, go far enough, for he goes on to say that such a state of affairs need not actually be conceived but must be conceivable; see ibid., 169. This appears to assume an intrinsic definiteness of these possible objects of the mind apart from their being actually conceived. If so, it violates the above principles since it renders inexplicable the assumed definite-ness of these unconceived but conceivable states of affairs.

²¹ See Robert Merrihew Adams, “Theories of Actuality” Nous 8 (1974): 211-31.

²² Alvin Plantinga, The Nature of Necessity (Oxford: Clarendon Press, 1974),45. ,

²³ If one chose to introduce God for this purpose, as Leibniz seems to have done, it would make God not only an agent but the sole agent. If it is God who from eternity decides the definiteness of all actual events, then there is no deciding left over for temporal agents. However, that is just to deny, it seems to me, that there are any temporal agents at all. The above principles presuppose creative, decision-making activity on the part of temporal agents.

²⁵ Anton C. Pegis, ed., Introduction to Saint Thomas Aquinas (New York: Modem Library, 1948), xviii.

²⁶ Charles Hartshorne, Anselm’s Discovery (LaSalle, Ill.: Open Court, 1965), 189-90.

²⁸ Compare in this regard the attempt sometimes made to define personal freedom in terms of counterfactuals. Thus it is sometimes alleged that to say “Mary acted freely” is just to say “If Mary had considered different motives, she would have acted differently.” Counterfactuals again, but to what purpose? Freedom is a characteristic of Mary’s act itself, not of a set of hypothetical actions that Mary might have performed. To suppose the latter would be to make a category mistake.

²⁹ L. Frank Baum, The Wizard of Oz (Indianapolis: Bobbs-Merrill, 1903), 152.