"Medieval Aristotelianism and the Case against Secondary Causation in Nature," pp. 74-118 in Thomas V. Morris, ed., Divine and Human Action: Essays in the Metaphysics of Theism (Ithaca, NY: Cornell University Press, 1988). 


Alfred J. Freddoso
University of Notre Dame

1. Introduction

Central to the western theistic understanding of divine providence is the conviction that God is the sovereign Lord of nature. He created the physical universe and continually conserves it in existence. What's more, He is always and everywhere active in it by His power. The operations of nature, be they minute or catastrophic, commonplace or unprecedented, are the work of His hands, and without His constant causal influence none of them would or could occur.

The Judaic Scriptures speak frequently and eloquently of the pervasiveness of God's causal activity. In a memorable rebuke to Job, Yahweh asks bitingly:

    Who has laid out a channel for the downpour and for the thunderstorm a path
    To bring rain to no man's land, the unpeopled wilderness; 
    To enrich the waste and desolate ground till the desert blooms with verdure?

    Has the rain a father; or who has begotten the drops of dew? 
    Out of whose womb comes the ice, and who gives the hoarfrost its birth in the skies? (Job 38:25-29) /75/

Nor is it only inanimate nature that Yahweh incessantly guides by His invisible hand: 
    Do you hunt the prey for the lioness or appease the hunger of her cubs, 
    While they crouch in their dens or lie in wait in the thicket? 
    Who puts wisdom in the heart, and gives the cock its understanding? 
    Who provides the nourishment for the ravens when their young ones cry out to God, and they rove abroad without food? (Job 38:39-41)
And in a cosmic hymn of praise near the end of the book of Psalms the entire created universe is enjoined to pay homage to the almighty creator whose decrees it obeys: 
    Praise Him, sun and moon; praise Him, all you shining stars. 
    Praise Him, you highest heavens, and you waters above the heavens. 
    Let them praise the name of the Lord, for He commanded and they were created;
    He established them forever and ever; He gave them a duty which shall not pass away.

    Praise the Lord from the earth, you sea monsters and all depths; 
    Fire and hail, snow and mist, storm winds that fulfill His word. 
    You mountains and all you hills, you fruit trees and all you cedars; 
    You wild beasts and all tame animals, you creeping things and you winged fowl. (Psalm 148: 3-10)

The Koran is no less insistent upon the ubiquitousness of God's causal activity:
    Allah is He Who raised up the heavens without any pillars that you can see. Then He settled Himself on the Throne, and constrained the sun and the moon to serve you; each planet pursues its course during an appointed term. He regulates it all and expounds the Signs, that you may have firm belief in the meeting with your Lord. He it is Who spread out the earth and made therein firmly fixed mountains and rivers, and of fruits of every kind He has made pairs. He causes the night to cover the day. In all this, verily, are Signs for a people who /76/ reflect. In the earth are diverse tracts adjoining one another, and vineyards, and corn-fields and date-palms, some growing from one root and others from separate roots, which are irrigated by the same water, and yet We make some of them excel others in the quality of their fruits. Therein also are Signs for a people who understand. (13:2-5) 
Even though this essay is concerned only with the role God plays in the production of effects in nature, it is also worth noting that the sacred writers do not shy away from counting even the free actions of rational creatures as products of God's ever-present causal influence: "O Lord, You mete out peace to us, for it is You who have accomplished all we have done" (Isaiah 26:12). In two passages destined to be cited often in the medieval Christian debates over divine causation, St. Paul emphatically reiterates this attribution of good deeds to God: "There are different works but the same God who accomplishes all of them in everyone" (1 Corinthians 12:6); and again, "It is not that we are entitled of ourselves to credit for anything. Our sole credit is from God" (2 Corinthians 3:5). As Paul puts it in summation, "He is not far from any of us, since it is in Him that we live, and move, and have our being" (Acts 17:27-28). 

Given such powerful foundational sentiments, it is little wonder that theistic philosophers have through the centuries felt impelled to address the metaphysical questions quite inevitably prompted by the belief that God is actively involved in the production of effects in nature.1 All have agreed that God is the primary or first cause of every natural phenomenon. But some have gone on, rather astonishingly, to make the additional claim that God is the only cause of such phenomena. In other words, they have denied that there is any such thing as genuine secondary (i.e., creaturely) causation in nature. In keeping with common usage I will call this position occasionalism, and, leaving aside for now various subtle and necessary qualifications, I will take as representative occasionalists al-Ghazali and Gabriel Biel from the middle ages and Nicholas Malebranche and George Berkeley from the modern era.2 /77/

Occasionalism is a remarkably devout theory of divine causation that (perhaps surprisingly, but, then again, perhaps not) managed to attract far wider and deeper support within the increasingly secularized atmosphere of pre-Kantian modern philosophy than it ever did within the unabashedly religious milieu of late medieval scholasticism. Nonetheless, we cannot fully appreciate thinkers like Malebranche and Berkeley unless we understand the fact that their views about the extent of God's causal role in nature, far from popping into existence ex nihilo in the 17th and 18th centuries, emerged instead from a long, even if never dominant, tradition in theistic philosophy--a tradition whose Christian spokesmen have always been eager to trace back to St. Augustine, a hero whom they regard as untainted by the corrupting influence of Aristotelian naturalism. 

My aim here is to take a first small step toward determining whether occasionalism can provide theists with a plausible and satisfying philosophy of nature, one that passes both philosophical and theological muster. Specifically, I explore the nature of and motivations for occasionalism, contrast it with the Aristotelianism its advocates have perennially rebelled against, and then confront it with what I take to be the strongest objections hurled at it by three prominent medieval Aristotelians--to wit, St. Thomas Aquinas and the 16th century Jesuits, Luis de Molina and Francisco Suarez. The Aristotelians do, to be sure, join with the occasionalists in (i) asserting that God is an immediate cause of every natural effect and in (ii) rejecting the claim that God's causal activity in nature is exhausted by His creating and conserving material substances and their causal powers.3 But in stark /78/ opposition to the occasionalists, they hold that each material or corporeal substance possesses and exercises its own proper causal powers. Such powers are not, they insist, supplanted or rendered otiose by God's causal activity in nature. Instead, God contributes to the ordinary course of nature only as a universal or general cause who cooperates with or concurs with secondary causes.4 True, as Malebranche was almost maliciously fond of pointing out in retrospect, the medieval scholastics were never able to reach an enduring consensus about the metaphysics of God's 'general concurrence' with secondary causes.5 But here I shall simply ignore their internal disputes and focus instead on their shared antipathy to occasionalism. 

In section 2 I formulate what I believe to be the essential thesis of occasionalism and then I will expound this thesis further in section 3, clarifying along the way the difference between genuine efficient or active causation and so-called occasional causation. Next, in section 4, I identify three conceivable versions of occasionalism (viz., the no-action, no-essence and no-nature theories) and show how each stems from a distinctive theological motive. I go on to argue in section 5 that while some of the Aristotelian objections to occasionalism are dialectically impotent against all three versions, there are other objections that succeed in pushing the occasionalist toward the no-nature theory, a theory characterized by a "radical" Berkeleyan metaphysics that does away utterly and completely with the Aristotelian conception of the natures of material /79/ substances. The main conclusion is, in fact, that anyone espousing occasionalism should accept the no-nature theory, since this is the one form of occasionalism that seems capable of withstanding the Aristotelian onslaught. Finally, in section 6 I briefly indicate for future discussion what I take to be the main strengths of the no-nature theory as well as its most serious problems.

2. The essence of occasionalism

All the thinkers to be discussed in this paper take causation in the most basic and proper sense to be a relation between substances on the one hand and states of affairs on the other.6 Typically, substances (agents) act upon other substances (patients) to bring about or actualize or produce states of affairs (effects).7 So both agents and patients may properly be said to contribute causally to the effects produced, and both acting and being acted upon may properly be thought of as modes of causal contribution. Moreover, since causal contributions involve the exercise of causal powers or, more generally, the actualization of causal tendencies and dispositions, we can also distinguish active from passive causal powers. The active causal powers of a substance delimit the range of effects that might be produced when that substance acts on suitably disposed patients; the passive causal powers of a substance delimit the range of effects that might be produced when that substance is acted upon by suitably positioned agents. (In what follows I will presuppose a basic understanding of this distinction between active and passive causal contribution, though I acknowledge that the distinction stands in need of further elaboration.)

In keeping with the above remarks, I will take as undefined the causal locution 'Substance S causally contributes to state of affairs p's obtaining at time t.' As I am using this primitive locution, it implies that p in fact obtains at t. However, taken by itself, it has nothing to say about what kinds of substances can be causal contributors. It does not, for example, require that causal contributors be capable of acting freely, or that they /80/ be endowed with sentient powers, or even that they be living. So the conceptual possibility is left open that some causal contributors, both active and passive, are non-free, non-intelligent, non-sentient, or non-living substances. 

Similarly, this undefined locution says nothing either about the time (if any) at which S makes its causal contribution or about the specific nature of that contribution. So, for instance, S might make its causal contribution to p long before t and not even exist at any time proximate to t. Again, S's causal contribution to p may be more or less direct, more or less closely connected with S's causal tendencies or (in the case of rational beings) intentions, and more or less determinative of the specific character of the effect in question.8 Finally, S's causal contribution might be either wholly active, wholly passive, or active in one respect and passive in another. (This last alternative occurs, for instance, when a substance brings about changes in itself, or when a substance is employed as an instrument or tool by some 'principal' agent.)

In what follows I will be interested mainly in active or efficient causation, and accordingly I will now define two further notions: 

    (D1) S is an active [or: efficient] cause of p at t =df
      (i) S causally contributes to p's obtaining at t, and

      (ii) S's causal contribution to p's obtaining at t is at least in part active.

    (D2) S is a strong active cause of p at t =df
      (i) S is an active cause of p at t, and

      (ii) no substance distinct from S is an active cause of p at t

The philosophers discussed all hold, first, that no substance other than God can be a strong active cause of any state of affairs and, second, that God is in fact a strong active cause of at least some states of affairs. Each of these tenets is a non-negotiable element of orthodox western theism. The first is based on sacred writings like those adduced above, /81/ writings that underscore the radical dependence creatures have on God for their being and causal efficacy (if any). The second is entailed by the doctrine of creation ex nihilo as well as by the doctrine, properly understood, that no created thing can remain in existence for any interval of time without being directly conserved by God throughout that interval.9 Given the rhetorical excesses of certain occasionalists (Malebranche comes immediately to mind), one might at first be tempted to characterize occasionalism as the thesis that for any state of affairs p and time t, if there is any substance at all that causally contributes to p's obtaining at t, then God is a strong active cause of p at t. This would make God the sole efficient cause of every effect produced in the created world, including all those involving rational creatures. 

In the history of the debate over secondary causation, Aristotelians have sometimes tried to saddle occasionalists with this alarmingly strong thesis and then to discredit them in the eyes of religious believers by pointing out that what follows is a manifestly unorthodox denial of human free choice. For instance, after attributing to Gabriel Biel and Peter D'Ailly the claim that secondary 'causes' bring about nothing at all, Molina continues: "Now if this view is understood to apply to all causes in general, even to the will and to free choice, as its authors seem to intend, then ... clearly it must be judged as an error from the point of view of the faith; for it completely destroys our freedom of choice and thus robs our works of every vestige of virtue and vice, of merit and demerit, of praise and blame, of reward and punishment."10 Suarez inveighs in like manner against those who reject the principle that creatures have and exercise active causal power: "What's more, there are other truths of the faith that cannot stand without this principle. The first and most important of these is the truth of free choice, which cannot consist in anything but a power and manner of acting ... Thus the Council of Trent (session 6, canon 4) condemns those who claim that created free choice 'does not act at all, and is purely passive'."11

However, as far as I have been able to tell, no important occasionalist has ever in fact intended to deny that there is such a thing as creaturely /82/ free choice or that such free choice involves a genuine active causal power to produce effects. (Just which effects is a matter I will take up in a moment.) Al-Ghazali and Berkeley, for instance, clearly believe that there are created spiritual substances with the active power of free choice. Even Malebranche, who wonders aloud whether free choice can properly be called a 'power', nonetheless recoils from the thought that our sinful deeds might be ascribable solely to God as an active cause.12 What's more, in the text cited by Molina, Biel is comparing the causal genesis of natural effects with the causal genesis of grace through the sacraments, and in such a context one would not antecedently expect to find any explicit discussion at all of free, as opposed to natural, causation.13 So in all fairness we should, it seems, define occasionalism in a way that prescinds from questions concerning God's causal influence on creaturely free choice.

All the philosophers we are dealing with here are libertarians; that is, they hold that a substance S is a free cause of p at t only if (i) S has a rational nature, i.e., is endowed with higher intellective and volitional capacities, (ii) S is an active cause of p at t, and (iii) S's causally contributing to p's obtaining at t does not itself obtain by a necessity of nature. This third condition must, of course, be spelled out in more detail, and I have attempted to do this elsewhere.14 But the intuition underlying it is clear enough for our present purposes: if an act is free, then it does not issue forth in a deterministic manner from the causal history of the world. 

Notice, by the way, that despite the tendency of many 16th century Thomists (the Banezians) to characterize God's causal influence on human action in very strong terms, these thinkers would nonetheless accept all three of the conditions on free causation just laid out. Their dispute with 'strong' libertarians (the Molinists) has to do with the intrinsic nature of God's contemporaneous causal contribution to free action rather than with the causal history of the world at the relevant moment. This issue, too, I have dealt with in more detail elsewhere.15 We are now ready to state the essential thesis of occasionalism: /83/

(OCC) For any state of affairs p and time t, if (i) there is any substance that causally contributes to p's obtaining at t and (ii) no created substance is a free cause of p at t, then God is a strong active cause of p at t

So, according to occasionalism, God is the sole efficient cause of every state of affairs that is brought about in "pure" nature, i.e., in that segment of the universe not subject to the causal influence of creatures who are acting freely. Interestingly, however, even this seemingly minor limitation on the extent of God's causal influence turns out to be less restrictive than one might at first expect. To see this clearly, and also to get a deeper understanding of the spirit behind occasionalism, we will have to take a brief but, I hope, illuminating excursion into the metaphysics of efficient causation.

3. Real and occasional causes

Occasionalists have been wont to draw a sharp distinction between genuine or real or proper causation on the one hand and so-called occasional or sine qua non causation on the other.16 Molina gives this concise summary of Biel's discussion of causation: 

    He is of the opinion that secondary causes bring about nothing at all, but that God by Himself alone produces all the effects in them and in their presence, so that fire does not produce heat and the sun does not give light, but instead it is God who produces these effects in them and in their presence. Hence ... he claims that secondary causes are not properly causes in the sense of having an influence on the effect; for it is only the First Cause which he affirms to be a cause in this sense, whereas secondary causes, he claims, should be called causes sine qua non, insofar as God has decided not to produce the effect except when they are present ... He also asserts with Peter D'Ailly that when God produces an effect in conjunction with a secondary cause, e.g., heat in conjunction with fire, He contributes no less than He would contribute were He to produce the same effect by Himself--in fact, He brings about more, since not only does He produce the heat with a concurrence just as great as if the fire were not present, but He also brings it about that the fire too is in its own way a cause of the heat.17 /84/
On Biel's account, occasional or sine qua non causation turns out to be little more than the mere counterfactual dependence of the 'effect' on the 'cause'. (In fact, to be perfectly accurate, Biel defines causation itself wholly in terms of counterfactual dependence, and then treats proper causation and sine qua non causation as two species of causation so defined--where it is only 'proper' causation that involves the actual derivation of the effect from the cause. I will pass over in silence this rather puzzling use of the term 'proper'.) More precisely, x is an occasional cause of y just in case (i) x is posited, (ii) if x is posited, then y is posited, (iii) if x were not posited, then y would not be posited in the same way, and (iv) it is not the case that y's being posited results from the exercise of any causal power on the part of x.18

So, for instance, if this fire is an occasional cause of the searing of this flesh, then it is not the case that the fire acts in such a way as to sear the flesh. Rather, the fire is an occasional cause of the searing of the flesh simply by virtue of the fact that the real or proper cause of the searing of the flesh, viz., God, would not have seared the flesh in the same way had the fire not been present. In general, then, an occasional cause is not such that the effect is derived from it or made to occur by it; nor does it exercise power or exert influence of any sort. Such conditions are satisfied only by those causes that are proper or real or genuine causes. (Notice, by the way, that classical occasionalists explicitly acknowledge God to be a real cause of natural effects; and so, unlike Hume and his progeny, occasionalists are not interested in reducing all causation to some sort of mere counterfactual dependence.)

It is clear, I trust, that the primitive causal locution I introduced above was meant to express real or proper causation rather than mere counterfactual dependence. That is to say, S causally contributes to p's obtaining only if p's obtaining derives (at least in part) from S's exercising or actualizing some causal power. Thus, S's presence in a set of circumstances in which p is made to obtain never by itself suffices for it to be the case that S causally contributes to p's obtaining--even if p's obtaining is counterfactually dependent on S's presence.

Now an occasional cause is not an instrumental cause, at least not if the notion of instrumental causality is explicated plausibly. An instrumental cause is a genuine causal contributor--more specifically, a genuine efficient or active cause.19 To be sure, the pen I am using to write /85/ this paper cannot on its own be an active cause of the paper's being written. Nonetheless, it does have active causal powers that are exercised under the right sort of conditions to produce an effect whose specific characteristics derive in part from the nature of those powers. When this particular pen is moved in the right way by the right sort of 'principal' agent, it is an instrumental cause that actively contributes to the paper's being written, and to its being written in black (say) rather than blue or red ink, and to its being written in ink of this consistency rather than of some other consistency, etc. By contrast, a merely occasional cause is such that there just is no direct natural connection between its causal properties (if any) and the specific character of the effect.20 If God alone properly causes the flesh to be seared in the presence of the fire, then there is no exercise of an active causal power on the part of the fire. Thus God is in no sense using the fire as an instrument to produce the searing of the flesh. Indeed, He could just as easily decide to refrigerate the flesh when it is brought close to the fire--and in that case, according to the occasionalists, the fire would be a 'cause' of the flesh's being cooled in exactly the same sense as that in which it is now a 'cause' of the flesh's being seared. There is thus no natural or intrinsic connection between the character of the effect (searing) and the nature of the fire or its powers. Molina is not slow to seize upon this point in order to demonstrate just how weak the notion of an occasional cause is: "Since God is just as able to make a given thing cold in the presence of fire and to make it hot in the presence of water as vice versa, fire could just as easily be a cause of cooling and water of heating as vice versa. Indeed, since God could create an angel or some other thing in the presence of a rock, a rock could be a cause of creation--which, even though Gabriel concedes it, is obviously as absurd as can be."21

Nor can occasional causes be either advising causes or disposing causes, as these notions are normally understood in Aristotelian theories of efficient causation. An advising cause of an effect E is, roughly, a rational agent who--by means of counsel, inducement, provocation, /86/ request, persuasion, threat, command, prohibition, etc.-- influences another agent to contribute freely to E. This, of course, is just the familiar sort of causal influence we ordinarily have on one another's free actions. As should be obvious, an advising cause is a genuine efficient or active cause. For instance, when I threaten to withhold my daughter Katie's allowance, I bring about in her a change of belief regarding whether her announced course of action is, all things considered, the most desirable. When she then chooses to act in an alternate way, I am said to be an advising cause of this piece of free behavior on her part. In Aristotelian jargon, I am a perfecting cause of Katie's new belief and an advising cause of the free behavior that issues from that belief.22

So no occasional cause is an advising cause, nor is any advising cause as such a mere occasional cause. In some cases this is perfectly obvious: a fire is not a rational agent and hence cannot, as it were, 'persuade' God to sear human flesh brought close to it. As we shall see in a moment, however, an occasionalist might be tempted to regard the free choices of rational creatures as, in effect, pieces of advice to God about how He ought to act in the realm of nature. 

A substance is a disposing cause with respect to an effect E when it produces in some patient a condition required in order for E to be brought about in the way that it is fact brought about. For instance, the farmer is typically a disposing cause of the corn's growing in the soil: by plowing, planting, fertilizing, etc., the farmer produces (i.e., brings to perfection or completion) in the relevant patients many of the conditions required for healthy cornstalks to be generated in the ordinary way. So, once again, as with the advising cause, the disposing cause of an effect E causally contributes to E by virtue of being a perfecting cause of certain other effects that are preliminary to E. Hence, it cannot be a mere occasional cause of its effects.

Now let us return to (OCC) and to the question of just which states of affairs created substances might, on the occasionalist view, be free causes of. Suppose that Stephen freely puts a kettle of water over a gas flame and that a few minutes later the water begins to boil. To many it will seem evident that Stephen is an active cause of the water's boiling at the time in question. They will reason as follows: True, /87/ Stephen could not by himself alone, without relying in some way on the active powers of other substances, have caused the water to boil. Nonetheless, he freely initiated a genuine causal sequence by putting the kettle over the flame. More specifically, by exercising his active causal powers he brought it about that other substances (e.g., the gas flame, the metal constituting the bottom of the kettle) exercised their active causal powers in the relevant way. It follows that Stephen was an active cause of the water's boiling. The only remaining question has to do with what sort of active cause he was. Some might argue that he was merely a disposing cause who put other substances (e.g., the gas flame) into a position to be perfecting causes of the water's boiling. Others might retort, less plausibly at first glance, that he himself was a perfecting principal cause who was using the other substances as his instruments. Yet whatever might be said about that issue, it is beyond dispute that Stephen caused the water to boil.23

An occasionalist, however, cannot endorse this line of reasoning. For, according to occasionalism, the boiling of the water does not in any way derive from the action of the fire or of any other material substance. To the contrary, God alone is a direct active cause of the water's boiling. An immediate consequence is that Stephen cannot be either a principal cause or a disposing cause of the boiling of the water in the ways just suggested. He cannot be a principal cause who uses the fire as his instrument, because the fire is an instrumental cause of the effect only if it is an active cause of the effect. But this is ruled out by occasionalism. Likewise, Stephen is a disposing cause in the manner described above only if the fire is a perfecting cause of the water's boiling. But this, once again, is ruled out by occasionalism. 

But perhaps we are being too hasty here. Perhaps we should rethink the whole issue before us in the way that the occasionalist theory ostensibly invites us to, viz., by substituting God for the relevant natural substances such as the fire and the metal pot. Can Stephen in that case be reasonably regarded as an active cause of the water's boiling? 

Presumably, to begin with, everyone will agree that Stephen cannot be a principal cause who is using God as his instrument, and so he cannot in that sense be a perfecting cause of the water's boiling. 

What's more, occasionalists still cannot grant that Stephen is even a /88/ disposing cause of the boiling of the water. Recall that a disposing cause of an effect E produces in some patient a condition required in order for E to be brought about in the way that it is fact brought about. That is, there must be some intrinsic or natural connection between the disposing cause's preparatory activity and E's being brought about in the way it is. But if occasionalism is true, then there is no such connection between any human action and any effect brought about in nature. Whenever God brings about E, His action in bringing about E, i.e., His causal contribution to E, is exactly the same whether or not there has been any antecedent activity on the part of finite free agents, and regardless of the nature of any such activity.24 Even if God has decided to act in such a way that He ordinarily brings about E only when finite free agents have first acted in certain ways, this implies only that the finite agents are serving as an occasion for God's producing E, not that they are disposing causes of E

Someone might suggest at this point that Stephen's freely putting the water over the flame renders him an advising cause who by his free action influences God (in a non-deterministic manner) to cause the water to boil.

However, this line of reasoning is also flawed, and in at least two distinct ways. First, all the occasionalists I know of would eschew even the faintest hint that any creature has any sort of causal power over God's actions. Nor is this attitude peculiar to occasionalists; philosophical theologians have traditionally held that the divine perfection absolutely excludes God's being acted upon as a causal patient by any creature. Second, it seems at least mildly outrageous to claim that all of our free actions might be signs to God of what we would like Him to do. Our freely offered petitionary prayers undoubtedly count as such. But can the same be said of, say, Stephen's putting a kettle of water on a stove under ordinary circumstances? Clearly not. 

Notice that these arguments are perfectly general. So even though in setting up the example I seemed to be conceding that a finite free agent might cause a change in some other substance (e.g., that Stephen /89/ might move the kettle from one place to another), the same considerations will apply to this alleged effect of Stephen's as well. In that case, however, the occasionalist still has not found a way to accommodate the common view that free creatures often make genuine causal contributions to states of affairs that go beyond their own free acts of will and simple bodily movements.

In fact, the occasionalist may not even be entitled to claim that we are proper or real causes of our own bodily movements. For in eliciting (to use the scholastic term) acts of will we might turn out to be nothing more than occasional causes of simple bodily movements. Malebranche enthusiastically embraces this conclusion, since he sees in occasionalism a theoretically satisfying account of how mind and body are interrelated. Mind and body do not causally interact with one another, but this, he avers, is merely a special instance of the more general truth that no creature, be it corporeal or spiritual, causally acts upon any other creature. Instead, it is God alone who coordinates created effects in general and who guarantees the customary concatenation of mental and bodily events in particular.25

Berkeley, by contrast, has deep misgivings about the claim that free creatures are active causes only of their own acts of will. In the Philosophical Commentaries he asserts that "we move our legs our selves," and that, contrary to what Malebranche thinks, it is not just God who moves our legs on the occasion of our willing that our legs move.26 And in the Three Dialogues, when Hylas charges in effect that (OCC) makes God Himself responsible for the most heinous sins, Philonous replies in part: "I have nowhere said that God is the only agent who produces all the motions in bodies. It is true, I have denied there are any other agents beside spirits: but this is very consistent with allowing to thinking rational beings, in the production of motions, the use of limited powers, ultimately indeed derived from God, but immediately under the direction of their own wills, which is sufficient to entitle them to all the guilt of their actions."27

But this reply immediately suggests a difficulty. Suppose that I move my leg and that at the same time the shoe on my foot moves along with it. Can Berkeley consistently grant that I have moved my shoe as /90/ well as my leg? We have already seen enough to know that on the occasionalist view my leg cannot count as an instrument by means of which I move the shoe. The leg, after all, is a corporeal substance and hence cannot be 'put into action' by a principal agent. The close correlation between the leg's movement and the shoe's movement is entirely God's doing. Nor, as we have seen, can I plausibly be regarded as a disposing or advising cause of the shoe's moving, and this by virtue of the fact that I will to move my leg. It seems to follow from what Berkeley says that I can move my leg and even my foot, but not my shoe!

Though this is by no means the last word, I believe that Malebranche has the better of the argument here. Be that as it may, my purpose so far in the paper has merely been to define occasionalism with a modicum of rigor and to show that it must place severe limitations on the causal influence of free creatures in order to remain faithful to its root conviction that God is a strong active cause of effects in nature. In short, according to occasionalism free creatures causally contribute at most to their own basic actions, and it may even be that those basic actions include only their own acts of will. 

Another question raised by (OCC) is this: Do occasionalists mean to exclude the exercise of passive as well as active causal power on the part of creatures? There seems to be no unanimity on this point. Malebranche, for instance, attributes to matter the three essential properties of impenetrability, infinite divisibility and magnitude, each of which seems to delimit in some (perhaps fairly minimal) way the sorts of effects that can be produced when a material body is acted upon.28 So, for instance, if bodies A and B are essentially impenetrable, then God cannot cause B to pass through A (though He could, of course, bring it about that when B comes into contact with A, instead of being deflected it immediately comes to occupy the very same place it would have occupied had it passed through A). By contrast, Berkeley and (in at least one place) al-Ghazali seem to look upon passive powers with as much suspicion as active ones, since both kinds of power would impose restrictions on the manner in which God acts in nature.29 What's more, /91/ it is arguable that a corporeal substance's having passive causal powers entails its also having some active causal powers, at least to the extent that its passive powers render it apt to be used as an instrumental efficient cause. For instance, even as minimal a property as impenetrability is such that a body which has it can be used by a suitably situated agent (God, at least) as an instrument to deflect other bodies. I will not press this point any further here, but will return to it below.

4. Three brands of occasionalism 

Given that (OCC) captures the essence of occasionalism, there are at least three interestingly different theories about God and nature that share (OCC). I do not claim that any philosopher has ever actually held either of the first two theories. Rather, I am employing them as "useful fictions" designed to help us (i) isolate three distinct motives for occasionalism and (ii) reconstruct in an informative way the dialectic by which the Aristotelian arguments push all occasionalists in the direction of a Berkeleyan philosophy of nature.

    4.1 The No-action Theory 
The first, which I will call the no-action theory, combines (OCC) with a robust essentialist account of material substance. The essentialist element of the theory (at least as I will be using the term 'essentialist' here) is encapsulated by the following two theses:30 
    (E1) It is metaphysically necessary that for any material substance x, there is a (lowest-level) natural kind K such that x essentially instantiates K

    (E2) For every (lowest-level) natural kind K, there is a non-empty set P of active and passive causal powers such that it is metaphysically necessary that a substance instantiates K if and only if it instantiates every member of P

So on the no-action theory, non-free creatures never exercise any active causal power, even though each one has such power essentially. As we will see in section 5, something like the no-action theory serves /92/ as the apparent target of many of the Aristotelian anti-occasionalist arguments, despite the fact that in all likelihood no occasionalist has ever actually held this theory. Still, I want to ask what sort of argument a theistic philosopher might be able to give for the no-action theory.

There is an alternative and perhaps more revealing way of stating my objective here. As will become clear in a moment, the best-known arguments for occasionalism involve a denial of essentialism with respect to material substances, as defined by (E1) and (E2). But it seems perfectly possible to assert (OCC) or something very much like it without even addressing the issue of whether essentialism is true. So even if the no-action theory seems extremely implausible, the following question is still worth considering: Is there any philosophically interesting argument for (OCC) that does not involve a denial of either (E1) or (E2)? 

Ironically, it is Suarez, rather than any occasionalist, who gives the most penetrating answer to this question. After introducing the thesis that "created things do nothing, but God brings about everything in their presence," he goes on to comment:

    I do not see a foundation for this position that carries any weight. Yet its principal foundation seems to have been that to whatever extent efficient causality is attributed to the creature, to that extent the divine power of the creator is diminished; for either God does everything, or He does not do everything; the latter detracts from the divine efficacy, and for this reason we will show below that it is false and erroneous, since it implies that something exists without depending on God. But if God does everything, then I ask again whether He does it immediately and by a power that is sufficient, or only mediately and by a power that is not sufficient. The latter detracts from the divine perfection. But if the former is true, then any other efficient causation is superfluous, since one sufficient and efficacious cause is enough to produce the effect.31
What we have here is a challenge to those who, like the medieval Christian Aristotelians, claim that there is a via media between occasionalism and the theory on which God contributes to natural effects only mediately, i.e., by creating and conserving material substances and their powers. For if God is an immediate active cause of every effect brought about in the realm of pure nature, then non-free creatures are immediate active causes of natural effects only if some such effects /93/ come immediately from both God and creatures. But, the argument goes, it is impossible to give a coherent and theologically orthodox account of how an effect might be brought about directly or immediately by both God and a creature--i.e., an account that does not render one of those alleged causal contributions wholly redundant. But God's contribution is on all accounts non-redundant. Therefore, God is the sole active cause of every effect in nature. 

This argument is certainly much stronger than Suarez's introductory remark would lead one to believe. How, after all, can God be thought to cooperate or concur with creatures without compromising His perfection in general or His omnipotence in particular? Despite Suarez's evident confidence in the truth of his own position, the scholastics could reach no agreement on this matter--a point that, as I mentioned above, Malebranche was especially delighted to emphasize. So one reason for embracing (OCC) is the conviction that the only philosophically respectable alternative to it is a view according to which God is not an immediate cause of natural effects.32

    4.2 The No-essence Theory 
The second brand of occasionalism is what I shall call the no-essence theory. According to this theory, even if material substances have active and passive causal powers, they do not have such powers essentially or by nature. More precisely, the no-essence theory accepts (OCC) but rejects both (E1) and (E2) in favor of the anti-essentialist thesis that no material substance has any of its causal powers essentially.33 

The chief motivation for the no-essence theory is the conviction that if material substances had their causal powers essentially, then there would be constraints on God's power that are both philosophically repugnant and, given the authenticity of certain miracle stories from the sacred texts, theologically untenable. Suppose, for instance, that this human flesh (say, the flesh of Shadrach) is essentially or by nature /94/ such that if it is exposed to extreme heat in the absence of impediments (e.g., asbestos clothing), it is incinerated; and suppose further that this fire (say, the fire of Nebuchadnezzar's furnace) is essentially or by nature such that in the absence of impediments it automatically, as it were, incinerates unprotected human flesh brought into suitable proximity to it. What follows, the occasionalists claim, is that according to essentialism it is metaphysically necessary for this flesh to be incinerated when brought into contact with this fire in the absence of natural impediments--a claim confirmed and, indeed, insisted upon by such contemporary essentialists as Rom Harre and Edward Madden.34 Any alleged human flesh not incinerated under such conditions would not be real human flesh; by the same token, any supposedly raging fire that did not incinerate unprotected human flesh brought near it would not be real fire. Or so, at least, essentialists seem bound to hold. And yet, the occasionalists retort, the sacred writings tell of God's miraculously sparing (real) human beings thrown unprotected into (real) raging infernos.35 That is, according to the sacred writings it is possible for real human flesh to be exposed to real raging fire and yet not be incinerated. The Bible and the Koran line up on one side, Aristotle and his heathen friends on the other; one must choose between the faith and the philosophers. 

This anti-essentialist argument should sound familiar, since it is epitomized by the famous philosophical bromide that there is no necessary connection between cause and effect. That is, roughly, there are no non-trivial conceptual limits on what sorts of causal transformations are possible. Thus al-Ghazali:

    In our view the connection between what are believed to be the cause and the effect is not necessary. Take any two things. This is not That; nor can That be This. The affirmation of the one does not imply /95/ the affirmation of the other; nor does its denial imply the denial of the other. The existence of the one is not necessitated by the existence of the other; nor its non-existence by the non-existence of the other. Take for instance any two things, such as the quenching of thirst and drinking; satisfaction of hunger and eating; burning and contact with fire; light and the rise of the Sun; death and the severance of the head from the trunk; healing and the use of medicine; the loosening of the bowels and the use of a purgative ... They are connected as the result of the Decree of God (holy be His name), which preceded their existence. If one follows the other, it is because He has created them in that fashion, not because the connection in itself is necessary and indissoluble ... 
    Let us consider only one example--viz., the burning of a piece of cotton at the time of its contact with fire. We admit the possibility of a contact between the two which will not result in burning ... [The philosophers] reject this possibility.36 
This, of course, is similar to the billiard-ball argument which Hume found in Malebranche and whose non-theological counterpart he proceeded to make famous in the English-speaking philosophical world. Notice, there is no little irony involved in the fact that Ghazali's argument should have been taken over by such a sceptic about miracles as Hume, given that the argument was originally prompted by the belief that God had miraculously saved His faithful servants from being incinerated! 

But what exactly is wrong with essentialism? After all, even an essentialist can allow that God is able, say, to change the flesh of Shadrach into stone for the duration of his sojourn in the furnace, or to put an invisible heat-resistant shield between him and the fire.

Occasionalists are wholly unwilling to settle for this concession. God, they will insist, does not have to act in opposition to his creatures from without, as it were; He does not have to vie with them in order to exercise control over them. Rather, He controls them from within as their sovereign creator and governor. They are beholden to His word: He can command the fire not to burn the flesh even while it remains fire; He can suspend the flesh's susceptibility to being incinerated without changing it into a different substance or altering its natural kind. Only those whose piety has been corrupted by the philosophers will disagree. al-Ghazali puts it this way: 

    [The philosophers say:] Whenever we suppose fire with all its qualities, and suppose two similar pieces of cotton which are exposed to fire in /96/ the same way, how can we conceive that one of them should burn, and the other should not? There is no alternative for the other piece. (From this idea, they come to disbelieve the story that when Abraham was thrown into fire, burning did not happen, although fire continued to be fire. They assert that this cannot happen, unless fire should be devoid of heat (which would put an end to its being fire), or unless Abraham's person or body should turn into a stone or something else which might resist the influence of fire. And, they add, neither this nor that is possible.)37
On al-Ghazali's view, then, material substances may have active (and passive) causal powers as long as God has full control over these powers "from the inside," i.e., as long as no creature has any such power essentially or with metaphysical necessity. So I will take the no-essence theory to consist of (i) (OCC), (ii) the anti-essentialist thesis that no material substance has any causal power essentially, and (iii) the claim that every material substance has active causal powers. According to this theory, when God keeps the fire from burning Shadrach, He does not have to act against the fire's natural inclination or tendency to consume Shadrach. Instead, He can merely ensure that the fire lacks this tendency, yet without destroying it or changing it into another kind of thing. 

Interestingly, there is some debate about whether al-Ghazali himself is an occasionalist, i.e., whether he accepts (OCC). As a matter of fact, in problem 17 of The Incoherence of the Philosophers he sets forth two distinct theories. His clear intent, it seems to me, is to find a religiously acceptable alternative to essentialism, since he is convinced that essentialism rules out certain of the miracles recorded by the sacred writers.38 He finds two such alternatives. The first consists of (OCC) and the anti-essentialist thesis; the second includes the anti-essentialist thesis along with the dual claim that (i) every material substance has active and passive causal powers and that (ii) every material substance exercises active and passive causal powers. So it is only the first of Ghazali's theories that counts as a version of occasionalism, and the argument against calling him an occasionalist must ultimately be based /97/ on the fact that he seems to retreat from the first to the second theory in the face of objections. 

Now whatever position al-Ghazali himself may finally have settled upon, the no-essence theory, as I have formulated it, is a hybrid constructed from the two theories he actually propounded. While the no-essence theory is once again not especially plausible, its very conceivability shows that one can accommodate the anti-essentialist motivations for occasionalism without thereby giving up the claim that created substances have causal powers. As such, the no-essence theory provides a possible refuge from the 'prevention or preemption' objection that will be directed at the no-action theory in section 5. 

    4.3 The No-nature Theory 
Finally, the third brand of occasionalism is what I dub the no-nature theory, and although Malebranche fudges a bit on whether material substances have passive causal powers, he and Berkeley are the champions here, with Biel playing the role of an extremely sympathetic fellow-traveller. In short, this theory consists of (i) (OCC), (ii) the anti-essentialist thesis, and (iii) the claim that no material substance has any active or passive causal power at all.39 The motivation for this theory is not any abstract metaphysical qualm about how one and the same effect might be immediately produced by both God and a creature; nor is it any relatively narrow worry about essentialism's ruling out a certain fairly small class of miracles. It is, instead, the sweeping and startling conviction that the attribution of any power at all (especially any active power) to any corporeal substance is not only unnecessary but blasphemous, not only philosophically confused but downright idolatrous. Malebranche puts the point as follows:
    If we consider attentively our idea of cause or of power to act, we cannot doubt that this idea represents something divine ... We therefore admit something divine in all the bodies around us when we posit forms, faculties, qualities, virtues, or real beings capable of producing certain effects through the force of their nature; and thus we insensibly adopt the opinion of the pagans because of our respect for their philosophy. It is true that faith corrects; but perhaps it can be said in this connection that if the heart is Christian, the mind is basically pagan.40 /98/

    To render God all the respect due Him, it is not enough to adore Him as the sovereign power and to fear Him more than His creatures; we must also fear and adore Him in all His creatures. All our reverence must be directed toward Him, for honor and glory are due only Him ... Thus, the philosophy that teaches us that the efficacy of secondary causes is a fiction of the mind, that Aristotle's, and certain other philosophers', nature is a chimera, that only God is strong and powerful enough not only to act in our soul but also to give the least motion to matter, this philosophy, I say, agrees perfectly with religion, the end of which is to join us to God in the closest way. 
    We ordinarily love only things capable of doing us some good: this philosophy therefore authorizes only the love of God, and absolutely condemns the love of eveything else. We should fear only what can do us some evil: this philosophy therefore sanctions only the fear of God and absolutely condemns all others.41

And in the concluding sections of the Principles, Berkeley continues the attack on Aristotelian natures: 
    If by nature is meant some being distinct from God, as well as from the laws of nature and things perceived by sense, I must confess that word is to me an empty sound, without any intelligible meaning annexed to it. Nature in this acceptation is a vain chimera introduced by those heathens, who had no just notions of the omnipresence and infinite perfection of God. But it is more unaccountable, that it should be received among Christians professing belief in the Holy Scriptures, which constantly ascribe those effects to the immediate hand of God, that heathen philosophers are wont to impute to nature ... Fain would we suppose Him at a great distance off, and substitute some blind unthinking deputy in his stead, though (if we may believe Saint Paul) He be not far from every one of us.42
Such memorable passages confirm Charles McCracken's perceptive observation:

The chief goal of the philosophical labours of both Malebranche and Berkeley was the same: to recall Christian philosophy to a recognition of the total and immediate dependence of all things on God. To both, belief in nature, if by that term be meant a realm of entities that produce effects by their own power, is the hallmark of the pagan, and the antithesis of the Christian, view of the world.43

As our discussion of the no-essence theory revealed, the Aristotelian /99/ account of substance or individual nature has two distinguishable elements: (i) the bare attribution of causal powers to material substances and (ii) the further thesis that each material substance has its basic causal powers essentially. The no-essence theory rejects one half of this conception of nature, its essentialism. But for a no-nature theorist the mere rejection of essentialism is not an adequate response to the danger posed by "heathen" naturalism, a naturalism that in effect attempts to replace the action of a provident God with that of natures conceived of as intrinsic causal principles. One must instead eradicate every vestige of the natures championed by Aristotle and the other pagan philosophers, and this entails nothing less than the complete repudiation of causal power for corporeal substances. What we have here, then, is the most extreme form of occasionalism and, as I will now argue, the only form of occasionalism worth defending.

5. The Aristotelian objections 

The medieval Aristotelians exhibit little patience with those who espouse occasionalism or theories closely resembling it. (Occasionalism is often treated along with theories that limit the active causal contribution of material substances to the production of merely accidental, as opposed to substantial, changes.)44 In one place St. Thomas goes so far as to call occasionalism stupid, a breach of decorum for which he is roundly applauded by Molina: "Everyone rejects this position, and St. Thomas justifiably calls it stupid. For what could be more stupid than to deny what is obvious from experience and sense perception? But it is evident to the senses that secondary causes elicit and exercise their own operations."45

Yet, for all their bravado, the Aristotelians were only partly successful in constructing a reasoned case against occasionalism. Many of their philosophical and theological objections, whatever their 'objective' merits, seem in retrospect to be at least dialectically ineffective, especially to those of us who have had the opportunity to read Malebranche and Berkeley. 

Take, for instance, the battery of objections that appeal to sensory evidence. Molina points out that such propositions as The sun giveslight /100/ and Fire gives heat, which, he claims, our sensory experience shows to be evidently true, would be false if material substances were not active causes of anything.46 We might add other, perhaps more complex, examples such as The wind is blowing the leaves around or The acid is making the litmus paper turn red or The oak tree is sprouting leaves, but the point is basically the same: If material substances are not active causes, then much of what we ordinarily say in making perceptual reports is literally false.

Along similar but not entirely identical lines, St. Thomas argues that, given the truth of occasionalism, evidently true propositions about sensings themselves, e.g., I sense the heat of this fire, would be false. For even if in the presence of fire God produces in me a sensation of heat, He cannot, according to occasionalists, bring it about that it is the fire's heat (read: the heat produced by this fire) that I sense.47

In yet another vein, St. Thomas argues that if occasionalism were true, then the diversity of effects in nature would not result from a diversity of causes. Rather, the (real) cause in every instance would be the same, viz., God Himself, though the effects would be diverse. But, St. Thomas retorts, "this appears false to the senses. For it is only heating, and not cooling, that results from putting something near a hot object; nor does the generation of anything but a human being result from the semen of a human being."48

So in at least three different ways occasionalism runs afoul of common opinions that emanate from and are grounded in ordinary sense experience. But, contends Molina, "what experience attests to should not be denied in the absence of a compelling reason; but not only is there no compelling reason, there is not even a plausible reason that might recommend the claim that created things do not truly exercise the actions which experience teaches originate from these same causes."49

The first-stage occasionalist response to such arguments is to claim that while we perceive the various and diverse effects that occur in nature, we do not perceive that these effects derive from or are brought about by created causes. At most what is obvious is that these effects are produced in the presence of such-and-such corporeal substances. Yet, /101/ as we saw in our discussion of occasional causation, this sort of presence is in itself a strong enough relation to ground the counterfactual dependencies that Aquinas' diversity argument points to. We need make no appeal to the alleged action of material substances. 

Furthermore, it is unlikely that any amount of psychological evidence will be sufficient to settle the question of whether, as common usage suggests, we actually perceive the wind blowing the leaves or the fire heating the kettle of water.50 But, the occasionalists urge, even if it should turn out that almost all our perceptual reports are indeed literally false, we must acknowledge that our linguistic practices have a myriad of different functions, the overwhelming majority of which do not require the literal truth of our perceptual statements. Of all the occasionalists, it is Berkeley who emphasizes this point most forcefully: "The communicating of ideas marked by words is not the chief and only end of language, as is commonly supposed. There are other ends, as the raising of some passion, the exciting to, or deterring from an action, the putting the mind in some particular disposition; to which the former is in many cases barely subservient, and sometimes entirely omitted, when these can be obtained without it, as I think does not infrequently happen in the familiar use of language."51

In fact, to insist on literal truth outside of strict scientific or philosophical contexts will often impede ordinary communication. In replying to the objection that a philosopher who denied that fire heats or that water cools would be deservedly laughed at, Berkeley concedes the point in these celebrated lines: "I answer, he would so; in such things we ought to think with the learned, and speak with the vulgar. They who to demonstration are convinced of the truth of the Copernican system, do nevertheless say the sun rises, the sun sets, or comes to the meridian: and if they affected a contrary style in common talk, it would without doubt appear ridiculous. A little reflection on what is here said will make it manifest, that the common use of language would receive no manner of alteration or disturbance from the admission of our tenets."52

As an outspoken rationalist, Malebranche is considerably less reluctant than Berkeley to scorn common opinion and practice. He openly /102/ acknowledges (and laments) that, largely as an effect of original sin, naturalism has become deeply embedded in ordinary thought and language: "We are led by an almost natural prejudice not to think of God with respect to natural effects, and to attribute power and efficacy to natural causes; ordinarily only miracles make us think of God, and sensible impression initiates our view of secondary causes. Philosophers hold this view because, they say, their senses convince them of it; this is their strongest argument. In the end this view is held by all those who follow the judgment of their senses. Now, language is formed on this prejudice, and we say as commonly that fire has the power to burn as we call gold and silver our good."53

Occasionalists, then, are not without systematic resources when countering Aristotelian objections that appeal to the evidence of the senses. For even if those who deny that fire heats and water cools are indeed epistemically flawed in some objective sense, they are nonetheless able to articulate a prima facie coherent metaphysical framework from within which arguments of the sort in question will fail to appear compelling. 

Even less convincing is the Aristotelian argument charging that occasionalism engenders a sort of causal anarchism. Occasionalism would of course be wholly implausible if it entailed that in any given circumstances C, no one causal transformation is more likely or less likely to occur in C than any other. In that case, to cite al-Ghazali's unforgettable example, you might have good reason to wonder whether the man standing before you now was one of the fruits sold at the market a few hours ago. But in fact occasionalism entails only that such bizarre causal transformations are metaphysically possible, not that they are likely or probable:

    If you could prove that in regard to things which 'can exist' there cannot be created for man a knowledge that they 'do not exist,' then these absurdities would be inescapable ... God has created for us the knowledge that He would not do these things, although they are possible. We never asserted that they are necessary. They are only possible--i.e., they may, or may not, happen. It is only when something possible is repeated over and over again (so as to form the Norm), that its pursuance of a uniform course in accordance with the Norm in the past is indelibly impressed upon our minds.54 /103/
Indeed, it is just God's steadfast adherence to certain arbitrarily chosen "Norms" that serves as the occasionalist surrogate for Aristotelian natures. That is to say, constant divine intentions provide the stability and regularity in the universe that Aristotelians attribute to the natures of corporeal substances. So instead of invoking causal dispositions, tendencies and inclinations rooted in the natures of corporeal substances, the occasionalist appeals to God's abiding intention to act in certain fixed ways. Where the Aristotelian will claim that, say, fire has an active causal disposition to heat bodies brought near it, the occasionalist will claim instead that in bringing about natural effects God resolutely follows the (defeasible) rule according to which He heats bodies brought near fire. And where the Aristotelian appeals to the regularity guaranteed by the natures of corporeal substances in order to justify the scientific practice of making inductive inferences about what would occur in counterfactual situations, the occasionalist instead grounds these same inductive practices in the firmness of God's causal intentions. (Notice in passing that this gives occasionalists a great theoretical advantage over their positivist cousins, who generate "the problem of induction" by their agnosticism about the existence and/or nature of the 'real' causes of natural phenomena.) In short, every important naturalist concept (nature, law, disposition, power, etc.) has an occasionalist analogue. There need be no substantive disagreement between occasionalists and Aristotelians about what will or would occur under such-and-such circumstances; there will be disagreement at most about what might possibly occur. And even here theistic Aristotelians can claim no advantage. They, too, must (or at least do) admit that God can prevent the (real) fire from incinerating the (real) human flesh of Shadrach; they, too, must concede that God has the power to transform (or better: transubstantiate) a fruit into a man.55

Two types of theological argument used by the Aristotelians are also ineffective. First, the Scriptural and Patristic testimony cited by the Christian Aristotelians in favor of secondary causation in nature is scanty and, more to the point, counterbalanced by other texts that /104/ seem to support occasionalism.56 Nor does any Aristotelian I know of make a serious effort to explain the meaning of the latter texts or to argue on hermeneutical grounds that they should be given less weight than texts that militate against occasionalism. Perhaps the Aristotelians simply take it for granted that what is evident to the senses or deeply entrenched in common opinion constitutes a nearly indefeasible constraint on how Sacred Scripture and the Fathers are to be interpreted. At any rate, the most imaginative and ingenious attempt to formulate a decisive argument on this front comes not from an Aristotelian but from Malebranche, who makes an admirable attempt to transform the counterintuitive character of the texts supporting occasionalism into a reliable sign of its truth.57

Second, both Aquinas and Molina argue that God's almightiness is honored more by the view that natural substances possess and exercise their own causal powers than by occasionalism. Thus Molina: "One extols God's power more by claiming that He can effect the operations of all things both by Himself and through the powers He confers on secondary causes than by claiming that He alone can effect them.58 Aquinas makes the same point negatively: it detracts from God's power and perfection to deny that He communicates (or is even able to communicate) to creatures the perfection of exercising their own causal power.59

But this objection is clearly inconclusive. Occasionalism, after all, assigns to God a much more intensive, if not extensive, causal role in nature than does the concurrentist view held by the Aristotelians. For according to concurrentism, God's active causal contribution to ordinary natural effects is 'general' and thus is not by itself sufficient to determine the specific nature of those effects; this is instead a function of the 'particular' secondary causes involved.60 So occasionalists, too, can lay claim to a legitimate sense in which their position celebrates God's power to a higher degree than does that of their opponents. 

In summary, then, many of the standard Aristotelian objections seem /105/ impotent against a philosophically sophisticated occasionalist. Indeed, it may appear at this point that the most an Aristotelian can hope for against any version of occasionalism is a mere standoff. It may appear that there just are no potent objections whose premises have any purchase on convinced occasionalists. This appearance, however, is misleading. 

Let's look at the situation a little more closely. The occasionalist must hold either (i) that corporeal substances have no causal powers at all or (ii) that, having such powers, they never in fact contribute causally to any effects. That is, an occasionalist must espouse either the no-nature theory or else a 'weaker' no-action or no-essence theory. Let's begin by exploring this second option. How might an Aristotelian respond to it? Consider the following chain of reasoning, which I will dub the prevention or preemption objection: 

Suppose for the sake of argument that the no-action theory is true. Now imagine that an ordinary piece of wood falls into a raging fire at a particular time T and that it is completely consumed shortly thereafter. Assume that no creature is a free cause of this effect. (As we saw in section 3, occasionalists seem forced to grant this assumption in any case.) God, then, is the sole active cause of the wood's being consumed. Nonetheless, both the fire and the wood have active and passive causal powers, and these by nature. Now assume that the fire's active causal powers are by and large what we usually take them to be. (If they are not, then alter the example so that the effect is one that the fire has the active power to produce.) In that case the fire has at T a natural inclination to consume any piece of ordinary wood thrown into it. Of course, since the no-action theory is true, we know that this inclination is never realized. But why not? There are just two possible answers. The first is that God acts on the fire in such a way as to prevent it from exercising its power to consume the wood; the second is that, without doing anything to the fire, God preempts its action by causing the wood to be consumed before the fire can act on it. In either case, in order to be a strong active cause of the wood's being consumed, God has to ensure that the fire will not exercise its proper causal power. 

To put it more generally, the no-action theory entails that in order to produce all natural effects by Himself, God must continually prevent or preempt the causal activity of created material substances. He must, as it were, constantly ward off his own creatures and the powers He has given them in order to play His rightful causal role in the ordinary course of nature--an awkward consequence, to be sure, and, embarrassingly, /106/ one that flouts the occasionalist's own deepest convictions. For, as we saw in our discussion of the fiery furnace, occasionalists deem it blasphemous even to suggest that God must struggle with his creatures or in any way oppose them from without in order to govern them and exercise control over them. Yet the no-action theory carries with it just this suggestion that creator and creature tend toward working at crosspurposes. 

At this juncture the occasionalist might fleetingly contemplate a retreat to the no-essence theory. For if the root problem with the no-action theory is that it attributes to corporeal substances certain essential causal powers, tendencies and dispositions that God must actively oppose 'from without', then perhaps this problem can be solved simply by denying that such causal properties are essential to the substances in question. We can see this more clearly as follows: 

To prevent or preempt the action of a corporeal substance is to oppose it or frustrate it in some way; otherwise its natural tendencies will be actualized. But if no powers or tendencies are had essentially by any corporeal substance, then God does not have to act against His creatures 'from without'; He can instead simply regulate their causal tendencies 'from within' to suit His purposes. For example, if He brings it about that the fire simply lacks a causal tendency to consume pieces of wood thrown into it, then He need not prevent or preempt the fire's action. Accordingly, He can be a strong active cause of the wood's being consumed in the presence of the fire without at the same time having to act against or thwart the fire. So, in general, if the no-essence theory is true, then God is able to arrange things in such a way that He never has to frustrate in any way the causal tendencies of any corporeal substance. And there will be just the sort of harmony between creator and creature that one would antecedently expect.

Perhaps this line of reasoning provides the no-essence theorist with an escape from the 'prevention or preemption' objection. (I am not certain that it does, but will not linger over the point here.) Nevertheless, two further arguments are no less damaging to the no-essence theory than to the no-action theory. In fact, the sheer obviousness and power of these arguments may explain why the 'prevention or preemption' objection, despite its considerable strength, does not appear in the Aristotelian sources I have cited. 

First of all, the no-essence theory, like the no-action theory before it, entails that each corporeal substance has causal powers and tendencies, and yet that no such causal power or tendency is ever actualized. But this /107/ consequence is objectionable in itself, regardless of whether God has to oppose or resist His creatures from without. For what possible reason could a perfectly wise and provident God have for endowing creatures with causal powers, none of which they will ever have the opportunity to exercise? Such powers, hidden from view, would be wholly superfluous, utterly lacking in point, serving no conceivable purpose in the divine providential scheme. St. Thomas expresses this point in two slightly different ways: 

    Reason shows that there is nothing in natural things which does not serve some purpose. But if natural things did not do anything, then the forms and causal powers they are endowed with would have no purpose--just as, if a knife did not cut anything, then it would have its sharpness to no avail.61 

    It is contrary to the notion of wisdom that there should be anything without a purpose in the works of one who is wise. But if created things in no way acted to produce effects, but instead God alone did everything directly, then these other things would be employed by Him to no purpose in the production of effects. Therefore, the position in question is incompatible with the divine wisdom.62

This objection seems to be very powerful, especially--but not only--if we make the fairly modest assumption that God could have created a world that appears very much like ours without endowing any corporeal creature with active causal powers. After all, if He could have done so, why would He choose instead to create a counterpart universe containing wholly unactualized causal powers? It would be as if He had created many tools (recall the knife) which He then proceeded to ignore in bringing about by Himself alone the very effects that the tools were intended to be instrumental in producing. One need not subscribe to a strong principle of sufficient reason in order to appreciate the point of this argument. 

The second objection is a rather subtle one that has both strategic and tactical significance. If corporeal substances cause nothing, then we can have no scientific knowledge. For the chief aim of natural science is to lay bare the natures of the material substances that are the real causes of spatiotemporal effects, and the method of science is to /108/ reason from effects to powers and from powers to natures. So if corporeal substances do not act, then scientific reasoning will not help us come to a knowledge of the natures of material things. In short, if occasionalism in any of its forms is true, then natural science is impossible. Thus St. Thomas: 

    If effects are not produced by the action of created things, but only by the action of God, it is impossible for the power of any created cause to be manifested through its effects. For an effect does not manifest the power of the cause except by virtue of the action which, proceeding from the power, terminates in the effect. But the nature of a cause is not known through the effect except insofar as its power, which flows from the nature, is known through the effect. Therefore, if created things did not act to produce effects, it would follow that no nature of any created thing could ever be known through an effect. And we would be deprived of all the knowledge of natural science, since in natural science demonstrations are derived mainly from the effect.63
This argument is strategically important because it demonstrates how the debate between Aristotelians and occasionalists is readily transformed into a debate over the character and aims of natural science. I will come back to this in section 6. But here the argument can be used tactically to show that those who accept (OCC) can have no justification for embracing either the no-action theory or the no-essence theory. For if corporeal substances never produce any effects, then, barring divine revelation, none of us will ever know just which powers any given corporeal substance has.64 Worse yet, scientific reasoning will not even help us establish the general claim that corporeal substances have causal power. So those who believe that material substances never act will, given their own theory, have no justification for attributing to those substances any active causal power at all. Even from within the occasionalist perspective, then, both the no-action theory and the no-essence theory suffer from grave deficiencies. The Aristotelian arguments seem to have rendered a move toward the no-nature theory not only appropriate but mandatory. 

But at this point in our dialogue there is likely to be a breakdown in communications--one that helps explain why the most impressive of /109/ the Aristotelian objections to occasionalism seem question-begging when directed at the no-nature theory. The problem, simply stated, is that the no-nature theory would never occur to a medieval Aristotelian as a coherent possibility worthy of being refuted. To Aquinas, Molina, Suarez or any other robust Aristotelian, denying active causal power to an entity amounts to nothing less than denying that entity the status of being a substance. As St. Thomas says in the opening chapter of De Ente et Essentia (and here he is simply echoing the Aristotle of Physics II), things that exist in the most privileged sense, i.e., the primary beings or substances, are just those things that have natures.65 And from these natures, in turn, flow the intrinsic causal powers, tendencies and dispositions that are definitive of the various natural kinds and that are thus possessed by the individual instances of those natural kinds. So it is metaphysically necessary that, say, tomato plants tend to produce tomatoes and that fire tends to incinerate human flesh brought into close proximity to it. What's more, it is just these natures and natural kinds, revealed to some extent and in an imprecise way through ordinary sense experience, that are the objects of systematic scientific inquiry. What, after all, could be more obvious than the contention that in order to have scientific knowledge about a tomato plant, one must acquire a systematic understanding of its causal tendencies and dispositions--more specifically, its principles of generation, development and degeneration, the conditions under which it flourishes and the conditions under which its development is impeded, etc.? Unlike the medievals, we now know that such inquiry will steer us in directions hitherto undreamt of--toward postulational biochemistry and evolutionary biology, for instance. But the fundamental guiding idea is the same: the scientific enterprise is aimed at discovering the natures of material substances and providing genuine causal explanations of spatio-temporal phenomena in terms of those natures. To repudiate such natures and the causal powers that flow from them is tantamount to denying that there are any material entities at all and a fortiori to denying that scientific knowledge is possible. To the medieval Aristotelians (and perhaps to others among us as well) such denials seem utterly preposterous. And so they are, as long we take it for granted that corporeal entities must be substances, i.e., beings with natures. 

And thus of course we come to Berkeley, whose antinaturalism /110/ knows no bounds. Since corporeal things have no active or passive causal powers, there just are no material substances of the sort that Aristotelian (or, for that matter, Lockean) naturalists have conceived there to be.66 Only spirits have natures that serve as intrinsic sources of causal power, and hence only spirits are primary beings or substances. There are, to be sure, bodies--those very bodies whose sensible characteristics we continually perceive. But if a mark of substance is the possession of causal power, then there simply are no material or corporeal substances.67

Notice, by the way, that Berkeley and the theistic Aristotelians do not disagree about whether corporeal things can exist without being cognized or 'perceived'. The Aristotelians, after all, do not claim that the ontological independence (or per se existence) of a corporeal substance consists in its ability to exist without being known by any mind. For, according to their theological beliefs, there can be no substance that is not known by the divine intellect. Rather, the disagreement is precisely about whether corporeal things have natures, i.e., about whether they are substances or first beings endowed with causal power. 

But perhaps we have moved too quickly here. Perhaps we have overlooked a possible alternative. Berkeley's position is what I shall call the full-fledged no-nature theory. On this view corporeal things in no way delimit or determine the effects they are involved in. They have neither active nor passive causal power and hence are not independent beings or substances. Instead, they are "ideas", i.e., mere modes or modifications of spiritual substances. But might an occasionalist not hold, as Malebranche seems to, that even though material substances lack all active causal powers, they nonetheless have passive causal powers? In this way the occasionalist would be able to affirm both (OCC) and the metaphysical independence of corporeal things. I shall call this the modified no-nature theory.

Such a theory would in effect be putting asunder what Aristotle had joined together, viz., structured or formed matter on the one hand and /111/ active causal power on the other. God would supply all the active causal power in nature, while material substances would receive and channel God's causal influence as patients. To be sure, these substances would be wholly inert, but they would at least be "thicker" than on the Berkeleyan view. 

Notice, by the way, that the same arguments that led us from the no-action theory to the no-essence theory will dictate that the passive powers in question be non-essential. al-Ghazali, for instance, insists that God could bring it about that of two exactly similar pieces of cotton exposed to equally intense heat in exactly similar circumstances, one is consumed by the heat and the other is not. But if the second piece's susceptibility to being consumed in those circumstances were essential to it, then God would have to thwart that susceptibility in order to keep the cotton from being consumed.68 So the friend of the modified no-nature theory should presumably hold that any passive causal powers a corporeal thing has are not essential to it. 

How plausible is the modified no-nature theory from within the occasionalist perspective? My suspicion is that the very notion of a substance endowed with passive but not active causal powers is incoherent. At the very least, as I noted above, the fundamental passive powers associated with material substances in particular (e.g., impenetrability) seem clearly to render the things that have them apt to be used (at least by God) as instrumental efficient causes. But if this is so, then no occasionalist can consistently accept the modified no-nature theory. Since this theory seems to have been Malebranche's, it follows that the arguments presented here support Berkeley against Malebranche, and the full-fledged no-nature theory against the modified no-nature theory. 

Any tenacious and epistemically conscientious friend of occasionalism will, I believe, be ultimately led by the Aristotelian objections to embrace the no-nature theory--and the full-fledged version at that. This is what I set out to prove, and I believe that my cumulative argument is a strong one. Yet since the Aristotelian objections examined so far do not appear to undermine the no-nature theory itself, the next step is to evaluate the no-nature theory on philosophical and theological grounds. I cannot do this in any depth here, but in the concluding section I will briefly indicate what I take the strengths and weaknesses of the no-nature theory to be. /112/

6. Prospects for the no-nature theory 

One of the no-nature theory's chief strengths is, I believe, its ability to provide a very clear and intellectually satisfying account of how the natural sciences are to be integrated into a theistic vision of the universe. Theistic metaphysics seems bound to resist the quintessentially modern dogma that the natural sciences are autonomous in the sense that (i) no alleged source of revelation (e.g., Scripture or Tradition) can rightfully serve in any way to correct or modify their claims, and that (ii) they in no way point beyond themselves to a more ultimate sort of explanation (whether in terms of efficient or final causes) for natural effects. The no-nature theory has the resources to curb these pretensions in a much more direct and simple (some would say simplistic) way than any 'theistic naturalism' can. 

Both Berkeley and Malebranche understand (OCC) to be a mortal enemy of the brand of scientific realism that engenders these claims to autonomy. This may not be apparent at first, since Malebranche clearly accepts the existence of unobservable entities with reference to which the fundamental laws of motion are to be formulated--and this makes him a scientific realist in at least one standard sense. However, we must distinguish here between between what I shall call entity realism and what might appropriately be termed explanatory realism. For while Malebranche accepts on scientific grounds the existence of unobservable theoretical entities, the postulation of these entities does not on his view enable us to grasp the causal powers or natures of material substances or to provide genuine causal explanations of observable effects--as opposed to "explanations" in terms of occasional or what he sometimes calls natural causes. Malebranche and Berkeley are both explanatory anti-realists. That is, they hold that the purpose of natural science is not to discover the real causes of natural phenomena or the natures of those causes, but to discover and systematize regularities and correlations among the Real Cause's observable effects in nature, and on this basis to make accurate predictions. First Malebranche and then Berkeley: 

    The study of nature is false and vain in every way when true causes are sought in it other than the volitions of the Almighty, or the general laws according to which He constantly acts.69 /113/

    If therefore we consider the difference there is betwixt natural philosophers and other men, with regard to their knowledge of the phenomena, we shall find it consists, not in an exacter knowledge of the efficient cause that produces them, for that can be no other than the will of a spirit, but only in a greater largeness of comprehension, whereby analogies, harmonies, and agreements are discovered in the works of nature, and the particular effects explained, that is, reduced to general rules, which rules grounded on the analogy, and uniformness observed in the production of natural effects, are most agreeable, and sought after by the mind.70

So the aim of the natural sciences is merely to give us theories that are, to use Bas Van Fraassen's apt expression, empirically adequate. Their aim is emphatically not to find the real (as opposed to natural or occasional) causes of natural phenomena or to provide a true picture of the causal mechanisms at work in the created universe. Indeed, according to the no-nature theory, there are no such causal mechanisms waiting to be discovered--though this is not to deny that there are extremely complicated "analogies, harmonies and agreements" that might suggest such mechanisms to the "pagan mind". When we reach the inevitable and perfectly legitimate questions about real causes, natural science exhibits its inherent explanatory limitations and points toward metaphysics and theology, to which it is subordinated in the hierarchy of disciplines. (This is one major point at which occasionalism breaks with Humean empiricism.) In Malebranche's words: 
    I grant that recourse to God or the universal cause should not be had when the explanation of particular effects is sought ... In a word, we must give, if we can, the natural and particular cause of the effects in question. But since the action of these causes consists only in the motor force activating them, and since this motor force is but the will of God, they must not be said to have in themselves any force or power to produce any effects. And when in our reasoning we have come at last to a general effect whose cause is sought, we also philosophize badly if we imagine any other cause of it than the general cause. We must not feign a certain nature, a primum mobile, a universal soul, or any such chimera of which we have no clear and distinct idea; this would be to reason like a pagan philosopher. For example, when we ask how it is that there are bodies in motion, or that agitated air communicates its motion to water, or rather how it is that bodies push one another, then, since motion and its communication is a general effect on which all /114/ others depend, it is necessary, I do not say in order to be a Christian but to be a philosopher, to have recourse to God, who is the universal cause, because His will is the motor force of bodies and also produces the communication of their motion.71 
The no-nature theory, then, deflates the pretensions of the natural sciences to be autonomous and self-sufficient disciplines that can provide a complete understanding of natural phenomena. If the no-nature theory is true, then the study of nature must inevitably point beyond itself to a more inclusive "wisdom" that investigates the Real Cause of natural phenomena--a wisdom that willingly assigns a prominent place to the study of nature and yet at the same time directs it from without. It is against this background that we must understand the conclusions Berkeley draws from his discussion of the natural sciences in the Principles:
    After what has been premised, I think we may lay down the following conclusions. First, it is plain philosophers amuse themselves in vain, when they inquire for any natural efficient cause, distinct from a mind or spirit. Secondly, considering the whole creation is the workmanship of a wise and good agent, it should seem to become philosophers, to employ their thoughts (contrary to what some hold) about the final causes of things ... Thirdly, from what has been premised no reason can be drawn, why the history of nature should not still be studied, and observations and experiments made, which, that they are of use to mankind, and enable us to draw any general conclusions, is not the result of any immutable habitudes, or relations between things themselves, but only of God's goodness and kindness to men in the administration of the world ... Fourthly, by a diligent observation of the phenomena within our view, we may discover the general laws of nature, and from them deduce the other phenomena, I do not say demonstrate; for all deductions of that kind depend on a supposition that the Author of Nature always operates uniformly, and in a constant observance of those rules we take for principles: which we cannot evidently know.72
This mix of theism with explanatory scientific anti-realism is very powerful and in many ways very attractive, combining as it does a fitting modesty (even humility) regarding the aims and claims of reason with an optimism about the ability of reason guided by faith to attain wisdom. Thus are avoided the prideful presumption of autonomy for reason on the one hand and an equally destructive pessimism about /115/ reason's ability to attain ultimate truth on the other--all of which is extremely desirable from a theistic perspective. What's more, as anyone familiar with the current philosophical scene will realize, no-nature theorists can ride piggyback, as it were, on recent penetrating defenses of scientific anti-realism.73 This adds considerable philosophical weight to the already impressive theological case for the no-nature theory. 

Although I myself do not subscribe to the no-nature theory, I am forced to admit that the sort of 'theistic naturalism' I favor has much to learn from and emulate (if it can) in the way occasionalism deals with and situates the natural sciences. The ever-present danger for attempts, like that of the medieval scholastics, to incorporate naturalism into a theistic vision of the world is just that, as Malebranche and Berkeley never tire of reminding us, secondary causes tend to usurp rather than to complement the role of the First Cause. In short, it is ever to be feared that the tail will end up wagging the dog. 

Yet while, from a theistic perspective, the no-nature theory provides a much more impressive and satisfying philosophy of nature than most contemporary theistic intellectuals have cared to admit, this theory is not without apparent weaknesses, and these will have to be explored in detail before we can make a final assessment. Three particular problem areas present themselves. 

The first is ontological. Just what is the ontology of corporeal things on the no-nature theory? Is Berkeleyan idealism mandatory for the no-nature theorist? If so, can it be made plausible? If it is not mandatory, what are the alternatives? Can bodies have the sort of metaphysical independence characteristic of substances without having any active causal power at all? How so? Might they simply be "bundles" of independently existing sensible qualities? And just what sort of ontology would that entail? (We might also wonder in passing whether such a theory could give a plausible account of the difference between living and non-living things, or of the difference between finite spirits closely associated with bodies, e.g., human beings, and finite spirits not so attached to bodies, e.g., angels.)74 

The second problem comes from natural theology. The no-nature theory seems to implicate God too deeply in the causation of physical /116/ evil. Of course, every theistic metaphysics must address questions about why God permits evil. But the no-nature theory has a special problem. Aristotelian theists use individual natures as a causal buffer between God and evil. For they claim that God is only a general cause of the effects of secondary causes, and so they can argue with some plausibility that the defectiveness of evil states of affairs (whether they be moral evils or physical evils) is traceable solely to the causal contribution of the secondary or creaturely causes. The no-nature theory, by contrast, does away with natures, and holds that God is the sole active cause of every state of affairs in nature. So if any such states of affairs are evil or defective, this defectiveness can be traced only to God's causal contribution. But this seems to make God the doer--and not just the permitter--of evil. 

In response to this problem Berkeley develops two different lines of argument. The first is to treat the laws of nature (i.e., the rules that God follows) as surrogate buffers between God and evil. The argument is that God's following these general rules is an intrinsic good that outweighs whatever particular evils result from the process. But this clearly will not do, since even when He is following rules, God is still the sole cause of the resultant states of affairs. A more promising reply, though one that stands in need of careful articulation and development, is simply to deny that there is any such thing as natural or physical evil for corporeal things: 

    Our prospects are too narrow: we take, for instance, the idea of some one particular pain into our thoughts, and account it evil; whereas if we enlarge our view, so as to comprehend the various ends, connections, and dependencies of things, on what occasions and in what proportions we are affected with pain and pleasure, the nature of human freedom, and the design with which we are put into the world; we shall be forced to acknowledge that those particular things, which considered in themselves appear to be evil, have the nature of good, when considered as linked with the whole system of beings.75
This argument may be stronger than it first appears. Since an Aristotelian nature defines normal and/or paradigmatic existence for members of a given natural kind, it follows that doing away with corporeal natures is tantamount to doing away with the objective /117/ grounding for the notion of evil (i.e., defectiveness with respect to the norm or paradigm) as applied to bodies. So whereas on the Aristotelian view God merely permits, say, the intrinsic evil of a tree's becoming diseased but does not Himself produce the disease as a particular cause, on the no-nature theory the tree's being diseased is no more an intrinsic evil and no less an intrinsic good than its being healthy. So God can in good conscience, as it were, directly cause the tree to be diseased. In short, given the no-nature theory, there is no independent standard that measures physical or natural good and evil. (Whether this line of reasoning can convincingly be extended to human pain is a moot question, of course.) In any case, the ultimate fate of the no-nature theory depends in large measure on whether or not this argument can be adequately elaborated and defended. 

Finally, the acceptability of the no-nature theory must be assessed in the light of revealed doctrines, and the results of this assessment will vary according to denominational allegiance. The 'higher' the church, or so it seems, the less acceptable the no-nature theory. For instance, it is fairly clear (though not beyond debate) that no account of substance consistent with the no-nature theory can provide a satisfactory metaphysics for the Catholic doctrine of the Real Presence of Christ in the Eucharist. The no-nature theory may also be incompatible with an orthodox understanding of the doctrine that the sacraments are causes of grace. There are undoubtedly other such examples lurking in the wings. 

I hope to discuss all these issues at length in another place. For now I will be satisfied to have shown that occasionalism in general and the no-nature theory in particular are eminently worthy of serious study by today's theistic philosophers. Many contemporary religious thinkers, unduly beholden to Kant and his successors, evidently refuse to take doctrines concerning divine causation as metaphysical truths and seem almost embarrassed by the thought that God might be intimately involved in the production of natural effects--it is as if Berkeley's worst nightmare had come true. Indeed, some even make the astonishing and astonishingly naive claim that the natural sciences have themselves established the impossibility of any causal contribution to natural effects on the part of a transcendent agent. (One cannot help but wonder where this sort of information about the deliverances of modern science comes from.) In stark contrast, traditional philosophical theologians almost unanimously regarded as too weak even the strong-sounding /118/ (to modern ears) claim that God's causal contribution to natural effects consists precisely in His creating and conserving material substances and their causal powers. In this paper I have tried to show by example how we can learn much from an examination of this tradition and its various elements.76 


1. On the inevitability of such metaphysical questions being raised about revealed doctrines and in defense (contrary to much modern theology) of the legitimacy of raising them, the best popular piece I know of is John Courtney Murray, S.J., The Problem of God (New Haven: Yale University Press, 1964), pp. 31-76. 

2. I will be citing the following editions of the works of these four philosophers: (a) al-Ghazali: al-Ghazali's Tahafut al-falasifah: The Incoherence of the Philosophers, trans. Sabih Ahmad Kamali (Lahore: Pakistan Philosophical Conference, 1963), problem 17, "Refutation of Their Belief in the Impossibility of a Departure from the Natural Course of Events," pp. 185-196; (b) Biel: Collectorium circa quattuor libros sententiarum 4, pt. 1, eds. Wilfridus Urbeck and Udo Hoffman (Tuebingen: J.C.B. Mohr, 1975), q. 1, "Utrum sacramenta legis novae sint causae effectivae gratiae," pp. 1-36, esp. 14-18 and 27-36; (c) Malebranche: Nicolas Malebranche: The Search After Truth and Elucidations of the Search After Truth, trans. Thomas M. Lennon and Paul J. Olscamp (Columbus: Ohio State University Press, 1980), esp. The Search After Truth, b. 6, p. 2, chaps. 2 and 3, pp. 440-452, and Elucidations, Elucidation 15, pp. 657-683. (Note that page references for both The Search after Truth and the Elucidations will be to this one volume.) (d) Berkeley: Three Dialogues between Hylas and Philonous, ed. Robert M. Adams (Indianapolis: Hackett Publishing Co, 1979), and A Treatise Concerning the Principles of Human Knowledge, ed. Kenneth Winkler (Indianapolis: Hackett Publishing Co., 1982). 

3. It cannot be emphasized enough that the position being rejected here (viz., that God's action in the world is exhausted by creation and conservation) is regarded as too weak by almost all medieval Aristotelians as well as by the occasionalists. (Its main medieval spokesman was William Durandus, an early 14th century Dominican bishop who, as far as I know, is the only theologian whose name is explicitly associated with this doctrine by 16th century writers.) In stressing the importance of this historical point in another context, I injudiciously claimed that the scholastics regarded the position in question "as in effect a form of deism." Peter Van Inwagen has persuaded me that this use of the term 'deism' is not only historically and philosophically inaccurate but unfortunately inflammatory. So I hereby recant. I would still insist, however, that the near unanimity of the tradition on this point constrains any contemporary theistic philosopher from simply assuming without argument that Durandus' position is theologically orthodox. 

4. Discussions of God's general or universal causation can be found in St. Thomas Aquinas, Summa Contra Gentiles III, chaps.66-70 and De Potentia Dei, Q. 3, A. 7; Luis de Molina, Liberi Arbitrii cum Gratiae Donis, Divina Praescientia, Providentia, Praedestinatione et Reprobatione Concordia, [hereafter: Concordia], ed. Johann Rabeneck, S.J. (Ona and Madrid: Soc. Edit. Sapientia, 1953), p. 2, "De concursu Dei generali," Q.14, A.13, disp.25-35, pp. 159-222; and Francisco Suarez, Disputationes metaphysicae, ed. C. Berton (Paris, 1866; reprinted in two volumes at Hildesheim: Georg Olms Verlagsbuchhandlung, 1965), vol.1, disp. 22, "De Prima Causa et alia eius actione, quae est cooperatio, seu concursus, cum causis secundis," pp. 802-843. Since those works of Aquinas to be cited here are generally known and widely available, I will follow the ordinary custom of citing the relevant texts without alluding to any modern editions.

5. Malebranche, Elucidations, pp. 676-680. 

6. The states of affairs in question typically involve states of the substances that are acted upon. Whether all such states must be thought of as involving accidents inhering in these substances is a moot ontological question that would receive different answers from different medieval scholastics. 

7. An atypical case would be that of creation ex nihilo, wherein there is no patient acted upon. 

8. Below I will have occasion to allude to the distinction between general (universal) causes and particular causes. Roughly, a substance is a general cause with respect to an effect E when its causal influence has had to be channeled (or, better, specified) toward E by some further cause. For instance, the sun is a general cause of this calf's being born. For the heat of the sun is a causal factor in the generation of the calf, but one that is channeled toward the production of a calf (as opposed to, say, a duckling) by further causes (in this case, a cow and bull). As can be seen, the same cause might be more general or less general with respect to different effects. 

9. Conservation must be taken as involving the conferral of existence on the thing in question and on every proper part of it. Only then, I believe, can one argue persuasively that conservation is God's perogative. 

10. Molina, Concordia, p. 160. 

11. Suarez, Disputationes metaphysicae, disp. 18, "De causa proxima efficiente eiusque causalitate, et omnibus quae ad causandum requirit," p. 597. 

12. Malebranche, Elucidations, p. 669. 

13. See Biel, Collectorium, pp. 15-18 and 31-34.

14. See my "The Necessity of Nature," Midwest Studies in Philosophy 11 (1986), pp. 215-242. 

15. See sec. 2.9 of the introduction to Luis de Molina, On Divine Foreknowledge (Part IV of the Concordia), Translated, and Introduction and Notes by Alfred J. Freddoso (Ithaca: Cornell University Press, 1988). 

16. On this distinction, see al-Ghazali, Incoherence of the Philosophers, pp. 185-187; Biel, Collectorium, pp. 15-18; Malebranche, Search after Truth, pp. 448-450; and Berkeley, Principles, pp. 51-53. 

17. Molina, Concordia, pp. 159-160. 

18. Biel, Collectorium, pp. 14-15. 

19. For an account of the different types of efficient causes, see Aquinas' Expositio in VIII Libros Physicorum Aristotelis, b. 2, lec. 5 (194b16-195a27). 

20. The notion of a 'direct' connection strong enough to ground causal derivation needs further honing here. Suarez, for example, has an objector propose the following sort of connection between fire and heating: heat inheres in the fire and this is why, even though the fire does not act, God heats (rather than cools) things in the presence of fire. Obviously, any Aristotelian will find this connection still too 'indirect' for true causal derivation. But the example suggests that articulating necessary and sufficient conditions for such derivation might not be very easy. See Suarez, Disputationes metaphysicae, disp. 18, p. 595.

21. Molina, Concordia, p. 161. For a reply, see pp. 102-103 below.

22. Suppose that Katie persists in her original intention and acts against my advice? In that case I am not a disposing cause of her behavior--indeed, she acted as she did despite my threats. Still, I am a perfecting cause of various psychological states in her, e.g., her belief that I disapprove of the behavior in question. 

23. I raise this second possibility, viz., that Stephen is a principal cause who uses the fire, pot, etc, as his instruments, because this is very much the image after which Aquinas models God's relationship to secondary causes. Berkeley, by the way, has some sharp criticisms of the idea that God might use instruments. See Three Dialogues, pp. 53-54. 

24. This follows from the fact that according to occasionalism God is a total particular cause of every effect in nature. The concurrentists, by way of contrast, hold that there are two ways for God to contribute causally to a natural effect: (a) by Himself as a particular cause, in which case the nature of the effect is derived from God's causal contribution alone; or (b), more commonly, as a general cause (indeed, the most general cause) cooperating with secondary causes, in which case the nature of the effect is derived from the secondary causes. 

25. Malebranche, Search after Truth, pp. 448-450 and Elucidations, pp. 669-671. 

26. Berkeley, Philosophical Commentaries, entry 548, in The Works of George Berkeley Bishop of Cloyne, eds. A.A. Luce and T.E. Jessop, vol.1 (Edinburgh: Thomas Nelson, 1948). 

27. Berkeley, Three Dialogues, p. 70. 

28. Malebranche, Search after Truth, b. 3, p. 2, chap. 8, pp. 243-247. 

29. Thus Berkeley in the person of Philonous: "I only ask whether the order and regularity observable in the series of our ideas, or the course of nature, be not sufficiently accounted for by the wisdom and power of God; and whether it does not derogate from those attributes, to suppose He is influenced, directed, or put in mind, when and what He is to act, by any unthinking substance" (Three Dialogues, p. 55). See also al-Ghazali, Incoherence of the Philosophers, pp. 187-188.

30. For an explanation of the notion of a lowest-level species or natural kind, see Michael J. Loux, "The Concept of a Kind," Philosophical Studies 30 (1976), 53-61. 

31. Suarez, Disputationes metaphysicae, disp. 18, p. 593. 

32. As I pointed out in n. 3, the claim that God's non-miraculous activity in nature is limited to creation and conservation is almost universally found wanting in the theistic metaphysical tradition. For a concise and engaging of just why this is so, see Molina, Concordia, pp. 161-163. I hope to develop Molina's arguments in more detail elsewhere.

33. Strictly speaking, an anti-essentialist could deny just one of (E1) and (E2) or even just their conjunction. However, to simplify the argument, I will assume that both (E1) and (E2) are being denied. As far as I can see, the differences involved have no bearing on my discussion of occasionalism. 

34. See R. Harre and E.H. Madden, Causal Powers (Totowa, NJ: Rowman and Littlefield, 1975), pp. 44-49. 

35. Specifically, the references are to chapter 3 of the Book of Daniel and to Koran 21:69-70 and 37:98. The miracle of the fiery furnance is not the only such miracle cited in this connection. Aquinas, for instance, alludes to God's keeping the waters of the Jordan from flowing downstream (Joshua 3:15-16), and the virginal conception of Christ. The important factor is that the created substances involved in this sort of miracle at least appear to have a natural inclination toward an effect contrary to the miraculous effect. Occasionalists will claim that this appearance is mere appearance, while concurrentists will aver that God accomplishes this sort of miracle by omission, i.e., by withholding His general concurrence from the ordinary course of nature--even while contrary tendencies remain in the created causes. 

36. al-Ghazali, Incoherence of the Philosophers, p. 185. 

37. Ibid., p. 188. 

38. The theistic Aristotelians will, of course, demur, insisting instead that one of the prerequistes for the action of a created cause is God's general causal influence. If God withholds this concurrence, there will (miraculously) be no action, even though the substances themselves retain their natures and their essential causal powers. Occasionalists respond by asking for a detailed metaphysical account of God's general concurrence. See my "The Necessity of Nature," 234-236. 

39. This is what I will later call the full-fledged no-nature theory. What I will dub the modified no-nature theory allows creatures to have passive causal powers. 

40. Malebranche, Search After Truth, p. 446.

41. Malebranche, Elucidations, p. 681. 

42. Berkeley, Principles, p. 89. 

43. Charles J. McCracken, Malebranche and British Philosophy (Oxford: Oxford University Press, 1983), p. 211. 

44. See, e.g., Aquinas, Summa Contra Gentiles III, chap. 69. 

45. Molina, Concordia, p. 160. 

46. Ibid., p. 161. 

47. Aquinas, De Potentia Dei, Q. 3, A. 7, resp.

48. Aquinas, Summa Contra Gentiles III, chap. 69. 

49. Molina, Concordia, p. 161. 

50. See Harre and Madden, Causal Powers, pp. 49-67 for an interesting discussion of the perception of causation. 

51. Berkeley, Principles, p. 19. 

52. Ibid., pp. 44-45. 

53. Malebranche, Elucidations, pp. 673-674.

54. al-Ghazali, Incoherence of the Philosophers, p. 189. 

55. The theistic Aristotelians differ from the occasionalists in claiming that the admittedly possible bizarre transformations mentioned here have a wholly different sort of causal progeny from normal causal transformations and are accomplished by God despite natural tendencies to the contrary on the part of creatures. The occasionalists rule out any such tendencies in the creatures themselves and in effect "relocate" them in the divine intellect and will. 

56. The longest and most illuminating discussion of this matter is found in Malebranche's Elucidations, pp. 672-685.

57. See Malebranche, Elucidations, pp. 672-685, esp. 672. I hope to discuss this section of Elucidation 15 at some length elsewhere. 

58. Molina, Concordia, p. 161. 

59. Aquinas, Summa Contra Gentiles III, chap. 69. 

60. Remember that nature of the effect is derived from its particular causes. See n. 8 above. 

61. Aquinas, De Potentia Dei, Q. 3, A. 7, resp.

62. Aquinas, Summa Contra Gentiles III, chap. 69. 

63. Ibid.

64. In fact, if the no-essence theory were true, we would be in a better position to know which powers particular substances lack, viz., powers sufficient to bring about the effect that God has in fact brought about. For recall that the no-essence theory is meant to rule prevention and preemption. 

65. Aquinas, De Ente et Essentia, chap. 1: "Nature seems to mean the thing's essence as ordered to its proper activity, for nothing is without its proper activity." 

66. It is the lack of causal power that, in Berkeley's metaphysics, is prior to and grounds the status, had by each corporeal thing, of being an idea. 

67. The connection between being and power was noticed at least as early as Plato. Thus the Athenian stranger: "I suggest that anything has real being that is so constituted as to possess any sort of power either to affect anything else or to be affected, in however small a degree, by the most insignificant agent, though it be only once. I am proposing as a mark to distinguish real things that they are nothing but power" (Sophist, 247d-e). 

68. al-Ghazali, Incoherence of the Philosophers, p. 190. 

69. Malebranche, Elucidations, p. 662. 

70. Berkeley, Principles, p. 68. 

71. Malebranche, Elucidations, p. 662. 

72. Berkeley, Principles, p. 69. 

73. The best-known, and justly so, of these defenses is Bas Van Fraassen's The Scientific Image (Oxford: Oxford University Press, 1980). 

74. The first of these questions is pressed by Suarez in Disputationes metaphysicae, disp. 18, 595. Malebranche tries to respond at Elucidations, pp. 661-662. 

75. Berkeley, Principles, p. 91.

76. A very distant ancestor of this paper was delivered in 1983 to a gathering of the Society for Medieval and Renaissance Philosophy, while a version much more like the present one was given in April 1987 to the annual Notre Dame-Calvin College Philosophy Colloquium. I am especially grateful to Calvin Normore and Philip Quinn for the penetrating questions they posed on those occasions.