International Journal for Philosophy of Religion 32: 129‑147, 1991.

© 1992 Kluwer Academic Publishers. Printed in the Netherlands.

 

Articles

 

Together bound: A new look at God's discrete actions in history

 

FRANK G. KIRKPATRICK

Department of Religion, Trinity College, Hartford, CT 06106

 

A common thread running through many of the traditional attempts to reject or modify the use of agent language with respect to God is the conviction that causal law is exhaustive of all that happens in the world. Consequently specific, particular, historically unique divine actions cannot be understood except as violations of that law and thus as incoherent or unintelligible to reason. This tread runs virtually unbroken from the work of Rudolph Bultmann to the recent explorations by Maurice Wiles.1 If a divine action is to be understood at all it will have to be as something other than a discrete, particular action alongside other actions within a world in which all occurrences fall within the traditionally reductionistic explanatory nexus of cause and effect because ‘interference with’ or violation of that nexus from ‘outside’ is metaphysically absurd. Langdon Gilkey put it well for a generation of theologians when he said in an article nearly 30 years ago that the analogy of the mighty acts of God in history is "empty, ... void since the denial of the miraculous."2 For the Bible the language of action was naively univocal for both God and human agents God acted as we did though on a far grander scale: but for us, Bultmann argued, such univocity is dead.3 Few contemporary theologians expect to see a particular act of God ‘alongside’ or ‘among’ the other acts that constitute the history of the cosmos because causal law rules out any occurrences that do not conform completely to the finite cause‑effect paradigm developed by modern science. Gilkey challenged theologians to find an "an ontology of events specifying what God's relation to ordinary events is like" but did not really think such an ontology could be forthcoming precisely because he accepted, as did most 20th century theologians, the monolithic sway of causal law explanation. Contemporary theology simply accepted without question the essential premises of this neo-Kantian dualism: namely, that God (whatever God really was in Godself) was ‘in’ a radically other ontological context and could not act within our

 

130

 

ontological context without violation of the very principles of understanding by which our context is metaphysically represented.

 

The interesting theological question now is whether this ‘liberal’ view of action is in need of major revision in order to permit a conceptually adequate understanding of discrete divine acts in history that neither violate the explanatory power of causal law nor require a sacrifice of reason. Can we do more than simply preserve God's action through notions of ‘master acts’ at the beginning of history but not thereafter? Is there a univocal notion of action capable of explaining particular divine acts as well as human acts without reducing the explanation of acts to the level of pure causal law, thus eliminating the essential role of freedom in the initiation of any act?

 

The possibility of establishing a univocal basis for human and divine action gets little or no credence on at least two major grounds. The first is simply that the notion of an ontologically transcendent but historically active God violates neo‑Kantian scientific limits on human understanding. The second is that the notion of an ‘acting God’ seems too anemic when compared to the more robust concepts of ‘Being‑Itself,’ or the ‘Unconditioned,’ which Tillichian and transcendental theology believe to be ‘demanded’ by the inherent limits of the finite world, especially by the faculty of reason itself. Each of these two grounds, ironically, plays off against while contradicting the other. Either human understanding is unable to stake any claim to objective representation of the extra‑mental in the first place,4 or its very inability to move beyond the limitations of representation somehow ‘point’ to an ontological transcendence which is, because of those very limitations, unable to be conceptualized even while it must be affirmed.5 In either case, the notion of a particular, ontologically distinct being, capable of acting upon and within the cosmic order, and worthy of the kind of worship due only to a divine being, is summarily dismissed. Only process theology has consistently tried to make a metaphysically respectable case for the notion of God as a singular being existing and acting in relation to other beings. But process theology's notion of God runs into various difficulties, not least of which is its claim that God does not, in the usual sense of the word, act by doing something to something else.6 The process God ‘acts’ only by, as D.D. Williams put it, "being felt by his creatures ... [or simply] by being."7

 

The difficulty most theologians face in utilizing models for God drawn from the ‘finite’ ontological context is that they seem necessarily too limiting of God's freedom and power. If we are to use the personal agent model to understand God straightforwardly, it will be necessary to show that it does not limit either God's freedom of action or God's range and depth of power in any metaphysically or religiously meaningful way. Of

 

131

 

course, we are predicating the use of the agent model on the basis of a fundamental experience of relationship in which God is felt as together bound with us. Without this relational foundation, our knowledge of divine action would remain abstract and unconnected to the experience it is intended to reflect. As David Tracy has argued, the theological task must remain wedded both to common human experience and to the traditions of the Christian community and especially to its scriptures.8 This fact is important because it presupposes some limits from the very outset, the most important being that God has been experienced as an active, relational being. The question, therefore, is whether metaphysical reflection on that experience can provide an intelligible set of conceptual categories for understanding and articulating what it means for an active, relational divine being to act in the world. But ‘liberal theology's’ understanding of divine action must first be addressed.

 

In 1968 Gordon Kaufman tried to provide an explanation of an act of God, for a theological consciousness under the sway of what it assumed was the scientific view of causal explanation. This view, which both Gilkey and Bultmann had earlier assumed, led Kaufman to proclaim that the notion of a God "who continuously performs deliberate acts in and upon his world ... has become very problematic for most moderns."9 It is, he proclaimed, "precisely by excluding reference to such a transcendent agent that we gain genuine knowledge of the order that obtains in nature..."10 In other words, Kaufman accepted without qualification the explanatory intimacy of causal law for the ‘natural’ order and the radical dualism which separates that order from the divine reality. His way around that dualism was to assert that the whole course of history, and not particular acts within it, is the ‘master’ act of God. This master act "is not a new event that suddenly and without adequate prior conditions rips inexplicably into the fabric of experience {emphasis added}, a notion consistent neither with itself nor with the regularity and order which experience must have if it is to be cognizable. Rather, God's act is viewed as the course of precisely that overarching order itself: it is God's master act that gives the world the structure it has and gives natural and historical processes their direction. Speaking of God's act in this sense in no way threatens the unity and order of the world as a whole."11

 

For Kaufman, God is not a ‘being’ or agent standing in relationship with the other agents in the world, but is rather the ‘wholly other’, the ‘limit’ of our ontological context. Kaufman's contribution was that the notion of God's transcendence can be understood as analogous to the transcendence of the human agent over his or her acts. Kaufman seemed to draw heavily on a residual Cartesianism12 in distinguishing between the ‘real’ or ‘true’ agent, hidden behind her acts, and the acts themselves that

 

132

 

are external and observable to others. As is the case with human agents, God can be known only when God acts and in and through those acts ‘reveals’ Godself.

 

But Kaufman's dualism remained unmodified throughout his development of the model of ‘interpersonal transcendence.’ As McLain has pointed out, for the ‘residual Cartesianist’ "the essence of the self consists of inaccessible, ‘private’ attitudes and feelings and states which are known to others only when disclosed."13 The dualist assumption necessarily voids any justified inferential link between what we see and what Kaufman claimed is the hidden reality ‘behind’ what we see. We have, in other words, no meaningful analysis or explication of the relationship he has labelled ‘behind’, or ‘transcendent of.’ In that case, the analogy of interpersonal relationship through action really gets us nowhere.

 

Like Kant, Kaufman winds up with a concept of God as an ultimate limit of experience and knowledge. As a limiting concept, it is not itself capable of being understood as we understand other things by means of the principles of knowledge within the world of experience. We only ‘know’ that God is not one of ‘those things.’ But just as Kant was unable to provide us with any concrete concept of God as a being who acts, so Kaufman preserved God's transcendence only at the price of rendering our knowledge of God completely equivocal. There remains no univocal relation between our knowledge of the rest of the world and our ‘knowledge’ of God. The unity of God with God's act is shattered, propelling God out beyond the ontological context in which human agents act and experience the actions of others.

 

The difficulties in Kaufman's position in at least one respect remain evident in the 1986 Bampton Lectures (God's Action in the World) by Maurice Wiles. Wiles is convinced that, in the words of Walter Kasper, “The God who no longer plays an active role in the world is in the final analysis a dead God.”14 However, Wiles is hard pressed to go beyond affirming that "the primary usage for the idea of divine action should be in relation to the world as a whole rather than to particular occurrences within it" and he quotes Kaufman approvingly in this context.15 The problem, as Wiles sees it, of postulating specific divine actions is that these would require us to "treat God as just one more causal agent alongside others in the world" [emphasis added].16 Presumably being just one more causal agent would negate those qualities about God which make God God, not human. Additionally, however, Wiles is troubled by the moral implications of regarding God as one agent among others, namely that if God can act, then why does God permit or fail to overcome evil? God's pattern of actions can only be read as capricious. Finally,

 

133

 

Wiles is concerned that if God acted in history, God's action would effectively negate human freedom, thereby undermining one of the most important divine values in the created order.

 

One of the best recent attempts to talk of God as an agent is Thomas F. Tracy's God, Action, and Embodiment.17 Tracy notes that at the heart of our understanding of agents is the ascription to them of ‘character traits’ based on our observation of their actions. If God, therefore, is to be described accurately as loving, just, and forgiving, there must be some intentional acts performed by God on the basis of which we ground these divine predicates. Tracy concludes that "we will be able to give an account of who God is on the basis of what he does ... [and] ... [i]nsofar as God's actions fall under the informal criteria that govern the use of terms like ‘loving,’ ‘just,’ and ‘wise,’ these character trait predicates can be applied to him just as ‘literally’ as they are applied to persons. None of the rules ... for the use of these predicates need be modified in order to ascribe them to a divine agent on the basis of his intentional actions."18 At the heart of the unmodified application of these predicates is the fact that God does things in order to realize God's intentions and that those things are done in and upon the very same world in which we do things. Tracy's argument links the dualistic halves which Cartesianism and Kantianism traditionally have kept separate and which have made it impossible for the model of agency‑action to transcend (what the dualist calls) the finite ontological context in order to be straightforwardly applicable to God.

 

The legacy these attempts to defend a notion of God as agent leaves us includes a requirement that we show how human freedom to act is compatible with but transcends the limitations inherent in causal explanations of bodily operations. If we can show that causal law (which traditional liberal theology assumes describes all physical, and even mental, operation within the ‘natural’ or ‘finite’ universe bounded by space‑time), is compatible with the operation of free agency, then we will have established the basis, in human action, for understanding divine action as compatible with causal law. At the heart of this compatibility is the claim that free agency transcends and supervenes on the causal realm of nature with respect to its initiation of free acts, and utilizes it as an infrastructure for the execution of those acts without violating the causal law which appropriately describes the workings of that infrastructure in the absence of intentionality.19

 

We can begin our analysis by claiming that the fundamental unit to be understood is the agent. This may seem trivial but it is not. It has been a hallmark of science that the fundamental units of reality are its least complex, indivisible elements. Hilde Hein notes that traditional scientific thinking has "been indoctrinated with the geometric principle that compli-

 

134

 

cated things are built out of simple ones, and that fruitful analysis is the result of careful separation of the simple elements which together produce a complex synthesis."20 This conviction has driven science to identify such elemental entities as the atom, the quark, or the meson, as the ‘building blocks’ of the universe. If one follows this line of thinking, anything more complex than one such elemental unit will not be truly basic and will be capable in principle of being broken down into more basic constituent parts. From this point of view an agent is not a fundamental unit, nor is the agent's action. They are considered to be complex arrangements of more elemental units and the explanation of an agent's action is therefore necessarily reducible to these more basic, less complex entities. Explanation will be in terms of atomistic causes which only appear in the guise of free acts when they enter highly complex relations with other atomic elements. But the real causes are themselves non-intentional and subject only to the strict laws of cause and effect in the mental/physical realm. This kind of explanation is inherently reductionistic (and thus deterministic) in that it eliminates a unique role for freely formed intentions which originate actions and which are not themselves entirely the effects of prior, non‑intentional causes.21

 

If we are to preserve the freedom of the agent, we need to question the assumption of reductionism. But it is not adequate simply to assert the ‘two angles’ view which holds that reductionistic determinism is real from one point of view and freedom from another. This way of dealing with the problem simply provides for obscurity, confusion, and, ultimately, epistemologically incoherent dualism. Rather, what we need is a way to accept the genuine observations which underlie reductionistic determinism but to place them within a more inclusive but still basic context (i.e., the agent) which includes freedom but resists reduction into more fundamental units which omit such freedom. Getting at that more inclusive context is aided by the development in modern biology of what is called the ‘systems’ or hierarchy approach.

 

This notion holds that as reality becomes more complex, the higher level complex entities include but go beyond (transcend) the lower levels. These higher level entities are themselves basic units, not reducible to their lower level parts precisely because they hold those ‘parts’ together at a higher level of integration. A higher level being, such as an animal, is a hierarchically organized unity, holding together a series of levels of reality such as the physiological and the organic, but transcending them in the sense that the ‘laws’ which describe the operation of the animal ‘as a whole’ cannot be accounted for solely by reference to the laws operative at the lower levels. The entity as a whole is subject to principles of explanation unique to itself (though dependent upon and not in contradiction to

 

135

 

the principles explaining the lower levels which it integrates), and is not reducible through analysis to principles that apply exhaustively at the lower levels. The higher level entity, for example, can be self‑directing in a way that is simply not possible for the lower levels and its capacity for self‑direction cannot be explained simply by ‘adding up’ the principles applicable to the lower levels. As Irving Laszlo has put it, "Organisms of all kinds are built by the integration of systems into superordinate systems, and these again into still higher level systems, until we encounter the organism as a whole. Each subsystem finds constraints imposed on its behavior by the higher system, with the result that the total organism's functional behavior dominates the behavior of all its parts, through successive hierarchically organized steps."22 (Note that it is the higher levels which put constraints on the lower, not, as in reductionism, the other way around.)

 

Although much of systems philosophy developed in order to clarify the work of the biological sciences, it can be applied to an understanding of agents because it provides a basis for challenging the dominance of the reductionistic assumption. In particular, it opens up the possibility of considering the agent as a ‘systems’ unit, one which is hierarchically organized but, as a unit, not analysable reductively into its constituent parts without eliminating some of the essential traits which characterize its higher level unity. By starting with a "fresh set of categories" provided by systems philosophy and the philosophy of action, we can, I believe, cut through liberal theology's subservience to an older Newtonian scientific view which virtually eliminated any serious discussion of discrete divine action precisely because it eliminated any possibility for free human action.

 

Edward Pols has contributed to that ‘fresh set of categories’ which can be extended to an understanding of God as an Agent capable of performing specific acts within the cosmic order (though Pols himself does not explicitly provide that extension of his categories to God). Pols' work complements the systems approach and allows us to break through the neo‑Kantian/transcendental Gordian knot which has traditionally been drawn so tightly as to forbid any opening to divine action.23

 

Pols argues that the ‘system’ of the agent is a unifying hierarchy which both distinguishes the agent as a unique unity and reconciles and integrates that uniqueness with the ‘lower’ levels of nature normally explained by the reductionistic causal laws of physical science. If Pols' analysis of human action is correct, then the explanation of both human and divine agency will have the necessary metaphysical room for freedom and power of action, irreducibility to lower levels of reality, transcendence

 

136

 

over those levels, and integral but non‑dualistic relation to them. Thus, a basis for understanding God's relationship with other beings will have been created which neither reduces God to a less‑than‑personal level of reality, nor limits God's freedom and power of action over an infinite range of reality, nor requires an apparent ‘higher’ conception of God as something other than ‘a’ personal/agential being. At the same time, however, it establishes an ontological link between God's being and the being of other entities, a link which undergirds our straightforward knowledge of God derived from our experience of being together bound with God and reflects the necessary metaphysical foundation for articulating that experience.

 

The heart of Pols' argument is that the agent is a ‘basic’ unified whole who utilizes in the execution of his or her freely chosen intentions sub-wholes or ‘infrastructures’ which are, taken by themselves in isolation from their integration into the unity of the agent, quite capable of being understood in non‑intentional causal‑law ways. Ultimately, the explanation of an agent's action must include the agent's own intention, and not simply be an analysis of the parts of the biological or physical infrastructure through which the act was carried out. An explanation that "comes back to [the agent] himself is final, satisfactory, and not to be set aside by any other. This means that it is also not to be set aside by scientific explanations that avoid the category of action entirely, and are couched instead in terms of entities, processes, functions, states, or events related under the laws of nature."24

 

For Pols, each act in a hierarchy of acts has an ‘ontological authenticity’.25 It is a basic unity which can only be explained by reference to itself. It is the work of a rational agent, the exercise of "the most fundamental and concrete sense of power accessible to our intelligence".26 It is important to note that Pols suggests that any exercise of power is power exercised by an agent, an entity, a being. This suggestion implies that if we want a notion of God as able to exercise power, then that notion must be one of an agent, not of something more abstract or ‘robust’ like Being-itself, or the ‘Unconditioned’ Horizon of being.

 

Action is, as Pols puts it, opaque to complete explanation by any other mode of explanation than the agent‑in‑act:

 

whatever functions in explanation in a way not reducible to or replaceable by other explanatory categories is an explanatory ultimate, and what is thus ultimate can not itself be explained ... [S]omething in the development and activity of the agent should be inexplicable except by calling attention to the fact that the agent indeed did thus and so ... [T]here must remain some features of the agent‑in‑act that are inex-

 

137

 

plicable in the sense that no explanatory technique can further illuminate what is expressed in the statement that the agent indeed acted in just that way. Otherwise agency is inauthentic in the sense that it is not what it purports to be.27

 

This means that metaphysically one has reached the north pole of explanatory capacity when one has reached the act of an agent, or, more precisely, when one has reached the initiation of an act through the intention of its agent.

 

It simply will not do to try to tease apart ‘segments’ of the act or its agent in order to discover the ‘underlying’ explanation for it. If we try to do this, the act becomes inherently mysterious. But Pols is suggesting that the unity of the act is the ultimate explanation of what the act unifies. To tease apart that unity by reductive analysis is to lose the act, and the agent whose act it is, in the process. We get off on the wrong foot if we start by asking "what causes the agent's intention?" because that question already presupposes the explanatory intimacy of a reductionistic cause‑effect explanation which assumes that something must have ‘caused’ the intention. What Pols is suggesting is that explanation in terms of the unity of the agent and his or her act is an explanation sui generis. It is not piggybacked onto or fitted into a more basic, more comprehensive causal explanation. In fact, Pols argues, causal explanation only takes on its real significance within the more inclusive, comprehensive explanation in terms of agent‑act. Explanation in terms of causes is abstracted from explanation in terms of intentions. It is explanation through deliberate omission of what deploys causal chains.

 

The act of an agent is ‘originative’. It initiates and unifies (holds together as an irreducible unit) a series of subacts and an infrastructure through which the act manifests itself. This series of subacts and infrastructure "includes all the functions and processes that take place within the persistent physical structure of the body..."28 and includes the firing of neural cells, the contraction of muscles, the movement of bones, the circulation of blood, and all the other biological and physiological functions that are necessary to the existence of the biological body of the agent. This entire infrastructure can be understood in itself in a thorough, precise, and reductionistic way through the explanatory categories of the natural sciences without any reference to the unity it acquires in and through the agent's action provided that this understanding abstracts it theoretically from the more inclusive context of action in which it serves as the infrastructure of the act.

 

When the agent acts, however, her act overarches and unifies this

 

138

 

infrastructure hierarchically without separating it into a discrete series of independent atomic parts which only seem to be held together artificially. The originative act remains throughout "an absolutely seamless unity." 29 "[H]owever fine‑grained your analysis of the infrastructure may be, extending even to the details of its mediation between the world outside an agent and the agent himself, you will not find there the full story of the act itself."30 The act, in other words, cannot be divided up into parts (except for purposes of abstraction and theory only). Its unity extends throughout the whole temporal sequence necessary for the carrying out of the act. (And Pols argues that different acts have different temporal reach. My act of writing this text may extend over a year or more and include the sub‑act of typing one letter on my keyboard which may take less than 1 second.) Each act may include subacts, subwholes, or a multiplicity of complex infrastructural mechanisms, but the act itself is a seamless whole and not to be understood solely by picking apart its constitutive elements. It must be seen as a whole to be understood as a whole because the act is what it is precisely through its unifying of a multiplicity of elements in its infrastructure. Pols indicates the richness of this notion of the act's capacity to unify its infrastructure by listing a series of synonyms conveying its meaning: the power of the originative act "orders, disposes, uses, deploys, shapes, binds together, [and] wields the multiplicity of the spatiotemporal elements that make up its infrastructure."31 The intention is not to be understood as simply the first temporal stage of the act. Rather, as Pols insists, the intention ramifies and unifies the whole of the act (which takes both time and space). The intention is present throughout the whole of the act. The act of the agent is "everywhere and at every time in the manifold of an infrastructure understood in other terms."32

 

The act is a basic unit33 (not to be further reduced) just because it is the unification of less basic elements which, apart from their ordering and deployment within the act, are nonoriginative and dependent upon a more inclusive unifying power. That inclusive unifying power is, ultimately, the agent who stands to the act as the act to its infrastructure. The category of the act "is more fundamental than that of action, although obviously inextricably mixed up with it."34 And the agent is "an entity with an ontological status more fundamental than the acts themselves."35

 

One implication of this view is that agents ‘transcend’ the causal order which they employ. Drawing upon ‘systems’ or ‘hierarchic’ analysis in the work of Paul A. Weiss, Pols suggests that the agent presides over the hierarchy of lower levels of reality (e.g., organism, organ, cell, organelle, and molecule) through which the act is carried out. The agent is the ‘real causality’ through which the causality of the lower levels is initiated and

 

139

 

unified.36 The agent is the entity through whom subacts originate (and in this sense the agent ‘transcends’ his or her acts).37

 

But this transcendence is not dualistically conceived. The agent is never cut off from his or her acts. "He becomes in his acts: but his becoming is the unifying power in each of his acts and the unification of all them... An act wells up out of an entity ... [and] ... over against the potential multiplicity of his acts his unity is absolute."38 The causal ‘necessity’ which may prevail at lower levels of the agent's infrastructure does not obtain exhaustively at the level of the agent taken as a unity: the agent's acts are not predetermined by the causal laws to which the agent is asymmetrically related and which the agent utilizes in carrying out his or her intentions. In this way, both reductionism and dualism are avoided.

 

            The brilliance of Pols' analysis is that it provides us with a concept of agent and act which integrates without dualism the authenticity of free (intentional) action (the supervening and superordinate power of the originative and unifying act) and the reality of causal law (the subordinate power of the derivative and unified infrastructure) through which the intention is enacted. By forcing us to think from the point of view of the basic unity of the agent, Pols leads us to understand the agent as the ontologically fundamental reality from which we can derive, by a process of abstraction (eliminating for the moment consideration of its ultimate source and unifying power), the meaning and application of causal law. The agent needs the causal order to carry out his or her intentions. In this sense the agent is dependent upon causal law and cannot act in such a way as to disregard its reality. Nevertheless, the origination and ultimate explanation of the agent's acts which utilize causal law transcends and cannot be reduced to causal description alone.

 

            Now, one extremely important consequence follows from this analysis. Every time an agent acts intentionally she ‘interferes’ with the natural constants which would have remained the same following the ‘natural’ sequence of events had she not interfered. An interference in this sense means simply an occurrence which would not have been had the agent not acted. Deterministic prediction of what the ‘normal’ course of events would have been if only non‑agential natural factors prevailed is upset if agents interfere with that course of events by deciding to enact particular intentions at particular times and places. From the point of view of causal law, an ‘action’ is an ‘interference’ in the (otherwise) closed causal nexus. An interference would be a ‘violation’ of causal law if it contradicted what causal laws have determined as causally possible (e.g., water turning into wine simply by verbal command). All acts are interferences with causal law but not necessarily violations of it. (We will take up the issue of violation below.) However, interference always means utilization of the

 

140

 

infrastructures described by causal law. An act which does not overturn or contradict the natural constants, nevertheless subordinates them and unifies them. The interference question, therefore, is a problem only on the assumption that causal law is explanatorily monolithic and exhaustive ‑ an assumption which we have every reason to reject.

 

How does Pols' analysis apply to a concept of God as Agent without reducing God to the level of human agents? William P. Alston has recently used the phrase ‘partial univocity’ to describe the kind of attribution to God of the categories of personal agency which we can legitimately make.39 While differing with some of the things Alston thinks requires univocal attribution to be only partial (e.g., he believes God has no body and I am not sure God cannot have a body), I think the notion of partial univocity is a good one because it makes the crucial assumption that if God is to be intelligible at all (and the experiential fact of being together bound with God in some kind of real relationship, requires some degree of intelligibility), God must share at least partially in what I have called our ontological context. This partial univocity provides us with what Alston calls "a hard literal core to our talk about divine action."40 As Alston says, "unless our understanding of divine purpose, intention, and will" have some partial univocity with what we mean when we attribute these to human persons, "we would, justifiably, doubt that the divine states in question deserve to be called ‘purpose,’ ‘intention,’ and so on."41 For all persons within the biblical tradition, it is mandatory to retain the concepts of divine purpose and intention because they are the core of the Biblical description of God.

 

If the world described by causal law is the world as it is ‘provided nothing interferes,’ and if interference is what an agent does when she acts by employing that ‘level’ of the world as it is hierarchically structured ‘under’ her initiating action, then all action at least partially transcends the closed world of causal law and interferes in it. Thus no charge of interference can be leveled against divine action simply because all actions are interference with an infrastructure in which a course of events would have been otherwise had God (like other agents) not acted.

 

If one is looking for ‘room’ for divine acts, one will find it in exactly the same way as one finds room for human acts, namely as the transcendent source and unifying power through which the causal infrastructure is employed or deployed in the enactment of the agent's intentions. Transcendent in this context means simply presiding over levels of reality less inclusive than the employer of those levels. God can be conceived as the Agent who, through God's creative originative act, unifies reality as a whole and gives it the enduring order which it possesses and which we, as human agents, depend upon when we act in less

 

141

 

extensive, creative ways. If God is the creator and sustainer of the total infrastructure (nature) within which all actions take place, it hardly seems a serious objection to point out that God is ‘interfering’ with what God causes to exist, especially given that all agents interfere with the natural order whenever they act. This means that there need be nothing ‘unique’ about the relation of specific divine actions to the infrastructures of the world that is not also ‘unique’ about the relation of human acts to their infrastructures. The link between divine action and the world is identical in this respect to the link between human action and the world. God's acts unify and ramify throughout a complex of infrastructural mechanisms to achieve their intended purposes. The typology of God's relation to those acts would be the same as a human agent's relation to her acts.

 

From the point of view of causal law per se, the origination of an act is always mysterious in the sense of being beyond exhaustive causal explanation. But it is not mysterious in the sense of being an affront to reason if we take the hierarchy or systems approach to explanation.42 Thus the origination of something that is a divine act is neither more nor less mysterious, in this respect, than the origination of something that is a human act. Simply by virtue of being an act, it transcends causal reduction and explanation and requires its own category of explanation, but one which is not unintelligibly cut off from the causal order.

 

But a divine act may be mysterious in other ways without thereby falling into the incoherence or unintelligibility of dualism. It is clear that most human acts require an empirical infrastructure to be enacted. But it is not clear that a divine Agent needs to utilize the same mechanisms in the same way to achieve divine intentions. It may well be the case that the complete ‘how’ of divine action is hidden from human understanding. Determining the ‘how’ of any action presupposes the uncovering of the mechanism linking intention to the deployed causal structures. Since the mechanism itself falls within a causal scheme, it is necessarily opaque to a full explanation of action which must refer to a utilization of that mechanism from beyond itself. The mechanisms and operative principles by which the divine Agent deploys the infrastructures of the world may therefore be even more inaccessible to human investigation than those which are employed in human action, especially given the infinite, and infinitely complex, range and power of divine action. The mystery of divine action, however, does not nullify the literalness of understanding God as an Agent. Rather it exemplifies it.

 

            There still remains the question of whether God, in addition to interfering in the causal order, also can violate, contradict, or bypass it in accomplishing God’s intentions.  (There are other questions, such as how to understand God’s power over, presence to, and creation of the cosmos, not

 

142

 

to mention God's relation to evil, into which we cannot enter here but which will be decisively altered by the notion of God as Agent who performs specific actions.)

 

If God is an Agent, then God can bring about a causally‑compatible natural occurrence (e.g., rain falling) by utilizing the causal infrastructures (the condensation of moisture in the atmosphere from clouds) which make rain possible. God can ‘intervene’ in the course of events which deterministic natural conditions would, without the agent's intervention, have caused to happen in accord solely with the laws of nature (just as a human agent can intervene in the natural sequence of events in order, say, to speed up the blinking of his eye which otherwise would happen according to a predictable natural pattern for biological/physiological causes). God may initiate a causal chain of events resulting in the falling of rain. The initiation of that chain could reasonably be assumed to be a basic action since God would not first have to cause something else to happen in order to get the chain moving. The events within the chain need not, of course, be basic divine actions provided only that the initiation of the chain is. Rarely does God, of course, according to those records which speak of divine action, perform basic acts for their own sake. Normally God does one thing (basically) in order that other things (causal chains) get set in motion in order to achieve a longer range purpose. (For example, God causes it to rain in order to punish the inhabitants of the earth in order to encourage the survivors to worship God more wholeheartedly, etc.) God could, of course, simply implant the desire to worship God directly in their hearts and minds. But this direct implantation (a basic act) would seem inconsistent with the Biblical witness that God will not coerce human worship and prefers to have it occur voluntarily.

 

There is no need to deny that a discrete divine act utilizes (and in this sense depends upon) natural infrastructures and their causal order. But if this is all that is meant by a divine act (the free, non‑deterministic initiation of causal chains), then divine action does not violate the causal order. When, then, would divine action be such a violation? Would it require God's doing something without utilizing a causal infrastructure or initiating causal chains? Can God bring it about that water begins falling from a cloudless sky absent all the causal conditions we normally assume must be present for rain to fall? If we say that God cannot cause it to rain if clouds are not present, are we not saying that God needs the infrastructures of the physical world to accomplish God's purposes there even if those purposes do not require a cause other than God's intention to begin the sequences through which they are realized? The indispensability of infrastructures even for divine actions is not alleviated by trying to think of all divine actions as basic. Even basic or originative acts require an infrastructure for

 

143

 

their complete enaction since only the initiation of the act by the agent's simply doing it without doing something else first is what is covered by the notion of a basic act. God's act can be basic (in the sense that God can simply decide to have it rain) without denying the necessity of its utilization of an infrastructure (clouds and other atmospheric conditions). To claim that God can act without any infrastructure may be to imply what is literally inconceivable.43 Rain‑from‑clouds simply is part of the infrastructure which God deploys when God chooses to make it rain whenever and wherever God chooses. God's choice may, of course, require an extraordinary condensation, speeding up, or manipulation of the causal processes far in excess of anything human agents can do, but God's purposes will still be accomplished by means of these processes and not by bypassing them entirely. To someone born paralyzed the act of moving one's limbs may seem miraculous. To persons unable to control the forces of nature, the ability to do so at will may seem equally so. But no divine act need be understood as in strict violation of the laws of the nature since all such laws are ultimately under the control of the agent who can utilize them all for his own purposes. Therefore there is nothing God cannot do in the world through those infrastructures. Even turning water into wine may not be a violation of causal law so much as it would be a manipulation of subatomic causal processes, the result of which would be the molecular and atomic change from water into wine. It might appear miraculous to us because we have no ability to exploit the subatomic levels of nature as a divine agent can. Nevertheless God's act in this regard is not a violation so much but an infinitely powerful utilization of those levels. I'm not at all sure what would be gained by claiming that God is bypassing the processes of nature entirely if the end result is that God can do with the causal processes whatever God wants. Violating (as distinct from deploying) their causal patterns is simply not necessary for God to accomplish the divine purposes. As we become freed from the presumption that all action is merely a sub‑species of causal law, we become open to the enormously complex and unpredictable ways in which action can occur (especially given divine power) within the world. These ways are so radically different from the limitations of reductionistic causal explanation as to open up room for what have traditionally been called miracles but now can be seen simply as God's way of deploying the infinitely complex infrastructures under God's control.

 

There are exceedingly difficult questions (and these are the ones to which Wiles has pointed us) to be dealt with regarding how we can ever determine which occurrences that take place in the world are to be ascribed to the specific intentions of God, i.e., are specific acts of God. But the ontological possibility of such acts requires us only to accept the

 

144

 

coherence of the notion of God as an Agent standing in the same relation to God's acts as human agents stand to their acts. If we can perform specific historical acts for which we are the primary agents, then surely God, as Agent, can perform specific historical acts for which God is the primary Agent.

 

The difference between God's capacity and ours, of course, would therefore have to reside in the range, power, and moral integrity of what God does. It would not have to require the traditional dualist notions of divine timelessness and spacelessness, though the arguments against those notions cannot be entered into here. What I have tried to show is that there is a ‘hard, literal’ core to a metaphysically credible notion of God as personal Agent and that talk about divine action is intelligible even in the ‘post‑modern’ world. I have not touched on the more religiously sig­nificant questions, such as whether there is any evidence (and if so, what counts as such) for actual divine action in the world. Or, if there is some evidence, what its religious import is for the lives of believers. The answers to these questions are clearly religiously crucial. All I have tried to do is to show that there is a metaphysically sound basis for regarding them as intelligible and rationally meaningful and not to be dismissed because modern people ‘cannot accept’ their underlying premise that there is a God who acts in the world.

 

Notes

 

1.    Two of the more recent attempts to deal with the issue of divine action have been Peter C. Hodgson, God in History (Nashville: Abingdon Press, 1989) and Maurice Wiles, God's Action in the World (London: SCM Press, 1986). Both works reflect the contemporary theological consensus that however divine action is construed, it must not entail specific divine acts in history or nature because such a notion is not intellectually convincing (Wiles, p. 11) because history "admits of no suprahistorical interruptions" (Hodgson, p. 197). Both Wiles and Hodgson (in somewhat different ways), like Gordon Kaufman before them, accept only the creation of the totality of the universe as the one and only divine act. To accept the idea that God might perform specific, particular acts, is to conceive God as interventionist, as one being alongside others, whose acts would then be subject to the charge that they are arbitrary, manipulating, infrequent, and unresponsive to the enormity of evil.

 

2.    Langdon Gilkey, "Cosmology, Ontology, and the Travail of Biblical Language," Journal of Religion 4l (1961):200.

 

3.    Rudolph Bultmann, Jesus Christ and Mythology (New York: Charles Scribner's Sons, 1958), esp. the chapter "The Meaning of God as Acting," pp. 60-85.

 

4.    This is the position taken by such philosophers as Richard Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979) and

 

145

 

by deconstructionists generally. See especially Mark C. Taylor, Erring (Chicago: University of

Chicago Press, 1984) and Thomas J.J. Altizer et al., Deconstruction and Theology (New York: Crossroad, 1982).

 

5.    This is the position by some representatives of transcendental Thomism. See, e.g., Richard Viladesau, Answering for Faith (New York: Paulist Press, c. 1987), pp. 39-42.

 

6.    Alan White, "Introduction," The Philosophy of Action, ed. Alan White (London: Oxford University Press, 1968) p. 2.

 

7.    Daniel Day Williams, "How Does God Act?: An Essay in Whitehead's Metaphysics," in Process and Divinity, ed. William L. Reese and Eugene Freeman (LaSalle, IL: Open Court, 1964), pp. 170‑171. This view is not far from Maurice Wiles' claim that God acts only through the creation of the structures of reality as such and Hodgson's claim that God acts "by being efficaciously present as who and what God is ... by being the normative shape, the paradigm" of a transfigurative praxis. (Hodgson, op. cit., p. 205).

 

8.   David Tracy, Blessed Rage for Order (New York: The Seabury Press, 1975), Ch. 3.

 

9.   Gordon Kaufman, “On the Meaning of ‘Act of God’”, Harvard Theological Review 61 (1968):175.

 

10.  Ibid., p. 176.

 

11.  Ibid., p. 192.

 

12.  See F. Michael McLain, "On Theological Models," Harvard Theological Review 62 (1969):155‑187 for a good critique of Kaufman's argument as Cartesian.

 

13.  Ibid., p. 163.

 

14.  Wiles, op. cit., p. 2.

 

15.  Ibid., p. 28. He refers to Kaufman's claim that "it is the whole course of history ... that should be conceived as God's act in the primary sense." Interestingly, Wiles is even more consistent than Kaufman in refusing to go beyond the notion of a single divine master‑act. Kaufman occasionally refers to God performing ‘subacts’ within God's overarching master act. But Wiles seems suspicious of reintroducing the causal problem if one permits God to perform a subact but not a discrete historical act.

 

16.  Ibid., p. 56.

 

17.  Thomas F. Tracy, God, Action, and Embodiment (Grand Rapids: William B. Eerdmans, 1984).

 

18.  Ibid., p. 152.

 

19.  I am not at this point entering into the debate within the philosophy of action over whether an intention is simply a particular type of ‘cause’ and thus whether all acts are caused. I want to defend a stronger view of freedom as that which transcends (is not limited by) the kind of causal explanations which have traditionally attributed the initiation of human action to factors which are not under the full and intentional control of the agent.

 

20. Hilde Hein, On the Nature and Origin of Life (New York: McGraw‑Hill, 1971), pp. 172‑173.

 

21.  Donald Davidson, for example, holds that causation is strictly a relation between events, not between an agent and an event which follows from the agent's act. See "Causal Relations," Journal of Philosophy 64 (1967):691‑703.

 

22.  Ervin Laszlo, Introduction to Systems Philosophy: Toward a New Paradigm

 

146

 

of Contemporary Thought (New York: Gordon and Breach, Science Publishers, 1972), p. 97.

 

23.  Pols' analysis of the action of an agent has many points in common with that of Richard Taylor and Roderick Chisholm who have argued for an ‘agent causation’ theory. All three hold that some causal sequences have their origin in the agents themselves (and not just in their volitions, desires, neural movements, etc.). This view is traditionally rejected on the grounds that it leaves direct agent causation an essentially mysterious phenomenon. (See Lawrence H. Davis, Theory of Action (Englewood Cliffs, NJ: Prentice‑Hall, 1979) (Prentice‑Hall Foundations of Philosophy Series), p. 11.) Pols' analysis wants to challenge the claim that explanation in terms of agents is inherently mysterious by utilizing, as Taylor and Chisholm do not, the systems or hierarchy notion.

 

24.  Edward Pols, Meditation on a Prisoner: Towards Understanding Action and Mind (Carbondale and Edwardsville: Southern Illinois University Press, 1975), p. 11.

 

25.  Edward Pols, "The Ontology of the Rational Agent," Review of Metaphysics 43.4 (June 1980):690.

 

26.  Edward Pols, "Power and Agency," International Philosophical Quarterly 11.3 (September 1971):295.

 

27.  Edward Pols, The Acts of Our Being (Amherst: The University of Massachusetts Press, 1982), pp. 36‑38.

 

28.  Pols, Meditation on a Prisoner, p. 103.

 

29.  Ibid., p. 99.

 

30.  Edward Pols, "Human Agents as Actual Beings," Process Studies 8.2 (1978):111.

 

31.  Pols, Meditation on a Prisoner, p. 105.

 

32.  Ibid., p. 110.

 

33.  But not necessarily a basic act since an act in Pols' sense can include a whole series of subacts some of which the agent does not perform directly without having to do something else. My typing this article can be an act but not a basic act since it involves a whole series of intermediary occurrences which take place between my intention and the finished product on the computer-printed page. Nevertheless Pols does suggest that his notion of an ‘originative’ act is similar to Danto's notion of a basic action in that neither is caused in a strict cause‑effect sense by prior acts. (See Meditation on a Prisoner, p. 95‑96.) The exact relation between acts and subacts in Pols is not entirely clear but its basic point is all that I want to establish for the purposes of my argument.

 

34.  Ibid., p. 72.

 

35.  Ibid., p. 309.

 

36.  Ibid., p. 65. This notion of the agent's ‘causality’ of its actions links Pols rather explicitly with the point of view adopted by Richard Taylor and Roderick Chisholm.

 

37.  Ibid., p. 310.

 

38.  Ibid., p. 315.

 

39.  William P. Alston, "Divine and Human Action," in Divine and Human Action: Essays in the Metaphysics of Theism, ed. Thomas V. Morris (Ithaca and London: Cornell University Press, 1988), p. 258.

 

40.  Ibid., p. 280.

 

147

 

41.  Ibid., p. 275.

 

42.   And thus complement the agent‑causation theories of Taylor and Chisholm.

 

43.  This fact may have serious consequences for our ability to ‘think’ the notion of divine creation ex nihilo of all that is besides God. Even if such an ultimate act is beyond our conceptual reach (since it transcends all infrastructures), the more directly relevant historical actions of God remain accessible both in understanding and in experience.