Chapter 8: Overcoming Reductionism by John B. Cobb, Jr.

Existence and Actuality: Conversations with Charles Hartshorne
by John B. Cobb, Jr. and Franklin I. Gamwell (eds.)

Chapter 8: Overcoming Reductionism by John B. Cobb, Jr.

John B. Cobb, Jr. is Ingraham Professor of theology, School of Theology at Claremont, and Avery Professor of Religion at Claremont Graduate School.

Process philosophy has undertaken to overcome the reductionism of the Newtonian world view within a naturalistic context and guided by the actual course of scientific thought. The greatest single contribution to this work has been by Alfred North Whitehead. Yet there are limitations to his achievement as well. In The Nature of Physical Existence1 Ivor Leclerc has both acknowledged Whitehead’s crucial contribution to the philosophy of nature and criticized him for continuing reductionistic tendencies. The purpose of this paper is to set Charles Hartshorne’s contribution to the overcoming of reductionism against the background of Whitehead’s work.

Section I briefly clarifies what is meant by reductionism and its status in the current discussion, Section II presents Whitehead’s systematic refutation of reductionism. Section III describes and evaluates Leclerc’s critique of Whitehead. Section IV reviews Whitehead’s position in the light of Leclerc’s critique. Section V describes Hartshorne’s distinctive contribution.


The problem of reduction is often discussed today in terms of levels and their relations. Different sciences treat the world at different levels ranging, for example, from the level of subatomic particles, through molecules and cells, to the level of human behavior. These sciences have demonstrated that much can be learned about each level without referring to other levels. Hence a considerable degree of autonomy is possible for theory at each level without referring to other levels.

On the other hand, the study of one level is sometimes able to illumine what is taking place at a higher level. Chemistry clarifies biological phenomena, for example, and physics throws light on chemistry. A great deal of scientific advance has taken place by interpreting phenomena at one level in terms of events at a lower level.

Indeed, many scientists have believed that the complete explanation toward which science moves is necessarily reductive in this sense. If biological phenomena are fully understood, this will be through biochemical explanation. If all physical events could ever by fully explained, this would be in terms of the behavior of quanta. This is the view of thoroughgoing reductionism.

Historically, reductionism has been associated with materialism and determinism. The units of which the world is made up were thought to be material particles which remained unchanged in themselves as they moved about in space and formed the diverse configurations which made up the physical objects in our world. Some thinkers still picture the world in this way. They interpret the principle of indeterminism as indicating limits on the possibility of human knowledge of this world but not as indicating that the cosmology is incorrect. Even if the shift from deterministic to statistical laws with respect to these physical existents is accepted as reflecting an actual indeterminacy in their behavior, the reductionism is not greatly affected. The ideal remains that events at other levels are to be explained in terms of these statistical laws governing events at this basic level.

By no means all scientists today accept this ideal of reductionism. Many are impressed by the degree of autonomy characterizing phenomena at other levels. They do not deny that some levels have been reduced to others and that further reductions are possible, but they see that success in reduction has often been exaggerated. Rather than assume that reduction must ultimately be possible throughout science, they take it as an open question whether particular physical existents at higher levels are reducible to physical existents at lower levels.2

For the philosopher of nature this raises an interesting question. Since complex physical existents can be broken down into component physical existents, how can they have properties that the components lack? The tendency is to answer this question in terms of structure or architecture or system. When smaller existents are organized in a certain pattern, the properties of the patterned system will be different from the properties of the elements that constitute it.

There can hardly be any doubt that this is true, in some sense. Even a pile of rocks will have properties that none of the rocks individually possess -- instability for example -- but this is hardly a significant challenge to the reductionist goal. The properties of the pile, while not characterizing the individual rocks, can be derived from what is known of rocks and their behavior in external relation to other rocks. From the time of Democritus, reductionists have never questioned the importance of spatial arrangements.

The kind of structure which is now emphasized is quite different from this. It makes possible complex interactions among the structured elements so that the whole structure operates in relation to other structures in ways quite unlike the operations of the components apart from that structure. It is highly improbable that laws describing the behavior of the particles within this structure and information about spatial relations could ever explain the behavior of the structure as a whole.

This is remarkably like the old organismic view which held, rightly, that the whole is more than the sum of its parts. But like the earlier view it is weak in explanatory power. It is now evident that the properties of the whole are not found in the parts except as they are organized into that whole, and that for this reason the reductionistic program is not successful. The question remains, Why? Is there, as vitalists supposed, in addition to the mechanism some other principle at work? If so, What? If not, how does the arrangement of the parts produce properties which are not ultimately functions of the parts together with their spatial relations?


Much of Whitehead’s cosmology is a response to questions of this sort. This response is rich and complex. There are three major ways in which his philosophy opposes reductionism.

1. The most fundamental basis for rejecting reductionism as adequate to explain the physical world is his doctrine of prehensions or internal relations.3 According to this doctrine the relations of one entity to others are constitutive of the entity in question. It is for this reason that the properties of a system cannot be derived from the properties of its constitutive parts, that is, from the properties possessed by these entities when outside the system. It is well known, for example, that a virus outside a cell lacks important properties it exhibits when in the cell. But in less dramatic ways this is true of atoms and molecules. Molecules exhibit properties that cannot be derived from the properties of the atoms constituting them when these atoms exist outside the molecular structure. If the properties of the molecule are derivative from those of the atoms, this can only be from the properties which the atoms have in this molecular structure.

The doctrine of internal relations has important implications for the scientific enterprise. That enterprise has been for the most part committed to analysis. The doctrine of internal relations explains why analysis is of limited, although real, value. Alongside the autonomous study of phenomena at each level and the reductive study of phenomena at one level in terms of those at a lower level, we need a study of phenomena at each level as they are shaped by phenomena at a higher or more inclusive level.4

The effect of the doctrine of internal relations on the understanding of the nature of the physical existent is radical. It destroys the notion of material substance and substitutes that of an event in which ~the many become one, and are increased by one."5 Since both relativity and quantum theory up until now have been couched in terms of substance, the theoretical effects of a shift to event thinking would be vast. These effects might provide the basis for the still needed synthesis of relativity and quantum theories. But that is not the project of this paper.

2. At least in the instance of the living cell, Whitehead believed that the constitutive entities are not all physical in the usual sense. There are entities or events also in the empty space. It is these events, and not the molecular structure of the cell, that account for the life of the cell. The molecules provide the necessary stability, but only the occurrences or "occasions" in empty space are capable of the novelty and spontaneity that are the distinguishing mark of life. Thus Whitehead posits a type of entity wholly neglected in ordinary reductionistic accounts.

3. The doctrine of internal relations and the assertion of occasions in the empty space of living cells still do not do justice to the reasons that complex entities cannot be explained in terms of the simpler ones of which they are composed. Whitehead saw that in the human instance we have immediate understanding of ourselves as unified experiences that cannot be identified with the experiences of the physical units making up our bodies. These human experiences are the dominant, presiding, or final percipient occasions which constitute us as "living persons." Similar occasions are to be found among other high-grade animals. Hence the full analysis of these animals requires not only the recognition that all the entities that make up their bodies are internally related to other such entities but also that there is present another set of entities of a much higher grade of experience which constitute the psyche, soul, or mind of the animal. These psychic experiences are internally related to the bodily ones, and the bodily ones are internally related to them.


In his renewal of emphasis upon the philosophy of nature, Ivor Leclerc has recognized the importance of the contribution of Whitehead. Indeed, much of Leclerc’s writing has been an explanation and defense of Whitehead’s project. However, Leclerc is an original thinker in his own right, and in The Nature of Physical Existence he criticizes Whitehead in important ways, especially for what he sees as a remaining tendency to reductionism. It will be useful to examine his criticisms, and to reexamine Whitehead in their light before presenting Hartshorne’s contribution to these central issues.

Leclerc recognizes that Whitehead tries to deal with the properties of larger and more complex entities in his doctrine of societies. But Leclerc believes there are limitations in Whitehead’s solution of this problem. He insists that such compounds of lower-level entities as atoms, molecules, and cells must be recognized as having individual existence in their own right, whereas he finds Whitehead attributing to them only social unity. Leclerc develops his own position through a double criticism of Whitehead.

Leclerc’s first criticism centers on the doctrine of causality. In Leclerc’s interpretation of Whitehead, activity or agency is located entirely in the perceiver. What is perceived functions as a passive datum for the new event or occasion. Leclerc interprets Whitehead’s position to be like that of Leibniz in presenting the cause-effect relation as one of perception. This implies that no force is exercised by the datum perceived. The agency is entirely in the perceiver.

Leclerc juxtaposes to this the proposal that acting be "conceived as an ‘acting on’ and thus as a ‘relating’."6 He believes that among the entities composing a compound "this reciprocal acting constitutes a tie or bond between them, this bond being the relation -- which exists only in the acting, and not as some tertium quid. The word ‘relation’ -- in this respect like the word ‘perception’ -- connotes both the act, the relating, and what the act achieves."7

Up to this point, I believe, Leclerc’s differences are not with Whitehead’s philosophy as developed in Process and Reality and Adventures of Ideas. They are with earlier writings in which the doctrine of causal efficacy was not developed. They are also with Whitehead as interpreted in some secondary writings, including Leclerc’s own otherwise valuable commentary in Whitehead’s Metaphysics: An Introductory Exposition. In that book he repeatedly cautions the reader not to take Whitehead’s language at face value, since Leclerc is sure that Whitehead could not mean what the language says. For example, whereas Whitehead consistently speaks of past occasions as actual, Leclerc says that Whitehead could not mean this. In Leclerc’s interpretation, "only concrescing, i.e., ‘acting’ entities are actual in the full, proper sense. The acting of antecedent actualities is completed; as such they are, in the strict sense, no longer ‘actual’."8 Whereas Whitehead speaks of past occasions as "functioning" in the self-creation of new occasions, Leclerc tells us that in Whitehead’s systematic position this "functioning" cannot imply agency.9 He adds: "This should be stressed because the contrary supposition might arise from Whitehead’s statement . . . that upon objectification an actual entity ‘acquires causation whereby it is a ground of obligation characterizing creativity’." 10

My point is that, at the time Leclerc wrote his interpretation of Whitehead, he understood Whitehead’s systematic position to be that past actual entities function in the present concrescence only as passive perceived data, not as truly causally efficacious. His interpretation of Whitehead did not change as he came later to recognize that this is not an adequate doctrine of cause. The results are odd. Whereas Whitehead repeatedly stressed that it is the distinctive task of his philosophy to show how one actual entity is constitutively present in other actual entities, Leclerc criticizes Whitehead for picturing past entities as remaining passive data for perception. Whereas Whitehead understands an actual entity to be constituted by the creativity which he defines as the many becoming one and being increased by one, Leclerc criticizes him for thinking that the many merely characterize a creativity which then leaves them behind as it constitutes a new entity.

Although it is unfortunate that Leclerc has in this way introduced confusion into the discussion,11 it is fortunate that we have his strong and independent corroboration of the importance of a doctrine of real causal relationality bonding the entities of the world together. Whitehead would certainly agree with Leclerc that actual entities exercise true agency on other actual entities, and that this "acting constitutes a tie or bond between them, this bond being the relation -- which exists only in the acting and not as some tertium quid."12

Although Leclerc’s criticism of Whitehead’s doctrine of causality is based on misinterpretation of the texts, Leclerc goes on to draw antireductionistic conclusions beyond those of Whitehead summarized in the preceding section. The passage just cited continues as follows: "This means that by virtue of the mutual activity of relating, there exists a form or character common to the entities acting. This form or character is not one inhering in each entity separately and individually -- in which case the character would be a mere class name -- but is a character of a relation."13

These sentences are not entirely clear. They could be interpreted as saying nothing different from what Whitehead asserts, namely, that in the case of a society the form or character common to the entities is derived by all subsequent ones in their prehension of earlier ones. But Leclerc elsewhere makes clear that he means more than this. "Specifically the question here is: which are to be identified as ‘acting’ entities in the primary sense -- as opposed to being only derivatively acting, that is, by virtue of the acting of the constituents of the entity in question? For example, are we to conceive chemical atoms as active in the primary sense while molecules are active only derivatively? But atoms, in current theory, are themselves composites; does this then imply that they too are derivatively active, it being their constituents, electrons, protons, etc., which are primarily active? But some at least of these constituents, e.g., protons, would themselves seem to be composite, so that by the logic of this argument the truly active entities must be identified with the ultimate constituents, those which are not themselves composite."14

Leclerc finds that the evidence of modern science counts against such a reduction. "The action of one compound ‘atom’ on another is not, by the scientific evidence, an aggregate action of the electrons and protons individually on each other."15 Molecules and cells also function as units of action, hence as real individual physical existents or substances.

Leclerc believes that when the subordinate entities in a compound act upon one another in a fully reciprocal way, so that they share in one another’s constitution, the whole attains a unity that is not attributable to its parts. This whole he calls the dominant monad. "There is a most important difference between the dominant monad and the other constituents of the ‘corporeal substance’. The dominant monad will not be simply one monad among others, differing from them only in having a greater internal complexity -- having consciousness, for example; the dominant monad will in a certain sense be inclusive. . . . What has emerged in this new position is the conception of a monad or substance which is not a constituent of a compound but itself a compound."16

This does not mean for Leclerc that the constituent parts lose their identity in the compounds. "The actual unity of necessity transcends the constituents individually, and since it is the unity which is actual, in relation to it the constituents must necessarily be potential; this is necessarily entailed in the relationship. In themselves, however (that is, not in reference to the actual unity) they are actual substances, and must be in themselves actual in order to be able to be constituents. Furthermore, they must be in themselves actual in order to be able to act; the actualized unity is not due solely to the acting of the dominant or unifying monad, for the latter acting to be an ‘acting on’ requires a responsive acting in the others."’17


Leclerc pictures the world as composed of a hierarchy of individuals each of which is compounded of smaller individuals until the ultimate stratum is reached. A similar vision can be found in Whitehead, especially in Science and the Modern World. For example, after asking whether a molecule in a human body is affected by the decisions of the dominant human occasion, he says that it would be more consonant with his philosophy to say that the direct effects would be negligible, but that the indirect effects would be important. "We should expect transmission. In this way the modification of total pattern would transmit itself by means of a series of modifications of a descending series of parts, so that finally the modification of the cell changes its aspect in the molecule -- or in some subtler entity."18

This passage clearly suggests that the bodily organs prehend the dominant occasion, the cells prehend the organs, and molecules prehend the cells. From the point of view established in Process and Reality, if prehensions are attributed to organs and cells and molecules, then there are actual occasions at these successive levels. Furthermore, in Process and Reality, there are several passages that imply that there are molecular prehensions19 and others that indicate that there can be occasions at successive levels, for example, electronic occasions and still more ultimate ones.20 Hence it is misleading for Leclerc to contrast his view so sharply with Whitehead’s on this topic. Something like Leclerc’s doctrine can be derived from Whitehead’s. Also, Leclerc’s critique ignores other features of Whitehead’s way of overcoming reductionism which were discussed in Section II.

Nevertheless, there are differences between Leclerc’s position and the one that is most fully articulated in Process and Reality. As Whitehead developed his doctrine in that book, he seems to have conceived all actual occasions as minuscule in size. This is true not only of physical ones but also of those in the empty spaces of the cells and those constituting the animal psyche. Despite the indications, just noted, that there are occasions at various levels, such as subelectronic, electronic, and molecular, his most detailed analysis indicates that the larger units are societies of mutually external smaller entities and do not have the unity of actual occasions. The animal does have a series of unifying, dominant occasions which jointly constitute the psyche, but even these can exist only in the interstitial spaces in the brain.

The most interesting case, and the one to which Whitehead gave most attention, is the living cell. For Leclerc the cell is a dominant monad which is a compound of all its subordinate elements. Whitehead’s analysis was discussed in Section II where the important concept of occasions in empty space was discussed. The question now, however, is that of the unity of the cell. Does the cell have the unity of an occasion so that the cell as cell prehends and acts? Or are all the prehensions and actions of the cells the prehensions and actions of its constituent parts, either those constituting the molecules or the occasions in empty space?

In the analysis of the cell Whitehead introduced the distinction between the nonsocial nexus of living occasions in the empty space of the cell and societies of occasions (e.g., molecules) constituting its physical components. In this connection he considered the possibility that the living occasions in the cell might constitute a single personally ordered society rather than a nonsocial nexus.21 That would mean that at any moment there would be a single living cellular occasion internally related to the world rather than a great multiplicity of minuscule occasions to which severally the internal relations of the cell must be attributed.

His reason for rejecting this theory is interesting. He had in view at that point in his writing two types of occasions: those socially organized and those not organized into societies. Social order was constituted for him by deriving a common character from antecedent events. In short, social order involved repetition of form. In a nonsocial nexus, on the other hand, the unit events could be spontaneous. They need not derive their character from antecedent events. Given these choices, social order explains the mechanical character of relationships. To explain life one must turn to occasions in nonsocial nexus. Hence to see the cell as governed by a single personally ordered society would deny the life and spontaneity which was the reason for turning to the occasions in the empty spaces.

There are obvious problems with this position. Whitehead at that point was forced to explain the order in the cell in terms of its molecular structure, to which spontaneity was denied, and to explain the life of the cell in terms of the events in its empty space, which he depicted as radically unordered. It is hard to think that this combination can account for the type of order and the type of spontaneity actually exemplified in a cell.

Whitehead recognized this problem, and even while writing Process and Reality he advanced beyond this point. He developed an understanding of another type of entity which combines order and life.22 He did so because he recognized that there is something like social order in the human soul and that this is combined with life and freedom. If successive human experiences constituted simply a nonsocial nexus, they would lack significant connection through time. But for them to constitute a society, as Whitehead had conceived of societies up to that time, each would have to repeat identical characteristics derived from antecedent members. Experience would have to be either chaotic or endlessly repetitive. In fact it is neither.

Whitehead saw that the problem had arisen because he had described societies in terms of the derivation of a common character from antecedent members. That meant that the more social an occasion is, the more it is conformal to its past. But another pattern is possible. The main requirement for social order is derivation of its characteristics from some aspects of the past members of the society. This derivation can be from their originative elements rather than from those that they have derived from their antecedents. In Whitehead’s terms, an occasion may prehend the conceptual feelings of antecedent occasion, not only the physical feelings of those occasions. In this way spontaneity in one moment can be transmitted to the next as the base from which its own spontaneity arises. Whitehead called this quite different ordering of occasions a "living person."

In Process and Reality Whitehead did not employ this new concept in conceptualizing the cell. Indeed, he explicitly rejected such application.23 But in Adventures of Ideas his comments on life suggest that he may have changed his mind. There he stressed that it is the coordination of spontaneities that constitutes life.24 His discussion is more reminiscent of what he said in Process and Reality about a living person than of what he said about a nonsocial nexus.

Even if Whitehead had clearly affirmed the presence of a cellular "living person" and also of molecular and atomic ones, there would remain a difference with Leclerc. The locus of the cellular living person would not have been the cell as a whole but the empty space within the cell. Even in the case of the human living person the locus is not the brain as a whole. Instead, Whitehead posits that the "route of presiding occasions probably wanders from part to part of the brain."25 Whitehead seems to have thought that all occasions, including dominant or presiding ones, must occupy spatio-temporal regions external to the regions of all other occasions. In this sense, even dominant occasions do not include other occasions in the spatial sense but lie alongside of them.

In summary, although Whitehead developed categories for understanding the unity of a society as more than the commonality of its tiniest parts, he applied these categories explicitly only to the higher animals. Even here his account indicates that he thought of the dominant monads as spatially minuscule in extension rather than as spatially inclusive of the occasions at lower levels. Hence, despite Whitehead’s extremely important contributions to overcoming reductionism, Leclerc is not wrong in detecting a continuing tendency to a particular type of reductionism. He affirms complex societies, and societies of societies, in some of which there are dominant occasions or monads. But the dominant monads are pictured as members of the societies rather than as compound individuals.


Leclerc published his theory in 1972. He seems to have been oblivious to the fact that Charles Hartshorne had published an almost identical theory in 1936, even using much of the same terminology, in "The Compound Individual." The chief difference is that, whereas Leclerc developed his views through criticism of Whitehead, Hartshorne thought of himself as interpreting and developing ideas partly derived from Whitehead.

Hartshorne focuses on atoms and cells as clear instances of individuals.26 He notes that "all individuals apparent to the senses are compounded of numerous much smaller individuals."27 Such compounds are distinguished from composites in the same way Leclerc distinguishes compounds from aggregates. The constituents of the aggregates lack the degree of mutual immanence characteristic of those in compounds. Like Leclerc, Hartshorne, in this essay, uses "substance" as the equivalent of individual.

Hartshorne, like Leclerc, holds that where there is a true compound individual there is a "dominating unit."28 For him, as for Leclerc, this includes the constituent entities without reducing their own substantial identity.

To point out the unrecognized similarity of Leclerc’s "new" proposals of 1972 and the position espoused by Hartshorne at least since 1936 is, on the one hand, to claim originality for Hartshorne and, on the other, to indicate that the independent and thorough work of Leclerc adds to the credibility of Hartshorne’s speculations. It might also suggest that future work should build on Leclerc’s more recent and better-documented formulations. The remainder of this essay argues that this would be a mistake -- that Hartshorne’s formulations have more far-reaching applicability and greater power of illumination.

Leclerc develops his theory in direct consideration of the objects of scientific inquiry. Like Whitehead prior to Science and the Modern World, he brackets out the human knower from the world he studies. Hence his formulations are adapted to a world of objects and their relations. He does not exclude certain kinds of subjectivity from this world, indeed he insists on activity in a sense that goes beyond observed phenomena. But he does not wrestle with the question of where an instance of such activity can be humanly observed. His work is, as he states, exclusively in the field of the philosophy of nature.

Whitehead had recognized that our one direct access to a unitary entity is in our own immediate experience. Much of his analysis of the dynamic structure of every unitary entity is based on phenomenological analysis of his own experience. He speculated that to be an actual entity at all was to have a dynamic structure analogous to that of human experience. Yet much of his thinking about the natural world was done before he made this move.

For Hartshorne this approach was eminently congenial. In his case it did not have to achieve reconciliation with another, more objective, study of nature that had occupied him for many years. On the contrary, its application in cosmology as in metaphysics was direct and unproblematic. Accordingly, Hartshorne’s whole cosmology reflects the model of human experience more directly than does Whitehead’s.

From the point of view of cosmology this difference is not entirely in Hartshorne’s favor. Cosmology has never been Hartshorne’s chief interest, and difficult questions may be too easily answered when one is not immersed, as Whitehead was, in the relevant science and mathematics. Nevertheless, some modifications of Whitehead by Hartshorne, even in the area of cosmology, lead to differences of doctrine which seem to be marked advances.

Since Hartshorne begins with human experience, the first cosmological question that arises for him does not have to do with atoms and molecules but with the relation of human experience to the body, especially to the central nervous system and the brain. Approached in this way the most natural answer is that human experience is the unified subjective concomitant of the complex pattern of physical events that constitute the body or some portion thereof. At different times Hartshorne has written of this as the body as a whole, the central nervous system, and the brain. It is my own judgment that the evidence favors the view that human experience occurs in a more or less extended portion of the brain rather than in the brain as a whole.29 But discussion of these important topics belongs to another paper. Here the brain will be taken as the physical correlate of human experience, and the flow of human experience will be termed the psyche. The traditional mind-body problem will be discussed as the psyche-brain problem. In these terms, then, for Hartshorne the psyche is located where the brain is located, not in some tiny point within it. The numerous, more limited, brain events occur in portions of the same region in which the unified human experience is taking place.

Hartshorne and Whitehead both assume that evolutionary evidence implies full continuity between human beings and other animals. Hence, where similar physiological structures and behavior occur, both posit that there are unified animal experiences analogous to unified human ones. Neither attempts to identify the exact point at which such experiences emerged in evolution. Both suppose that they are lacking in plants. That means that the internal relations of the plant to its environment are to be found in the parts of the plant as these are discovered in analysis. Both also suppose that the smallest units of reality have a character which resembles human experience at least in that it involves internal relations. But beyond this they diverge.

Whitehead’s account showed why complex organisms cannot be reductively explained by the activities of their component parts when these are studied apart from the structures of the organism. It gave additional reasons why the cell cannot be reduced to its molecular components. But when dominant occasions in animals and the occasions in empty space in cells are included in the animal and cellular societies, then the activities of the societies can be explained by the activities of the components when they are components of those societies. In societies other than animals and cells these component parts are finally only the subatomic entities.

Hartshorne proposes that wherever the evidence counts against reduction, we posit that the entities in question are real individual agents, internally related to the environment. The case of the living cell may be taken up again. If the activity of the cell suggests that it is taking account of its environment as a cell, then we should attribute internal relations and unified activity to the cell as a cell. That means that we should view the cell by analogy with our own experience and see the relation of the cell to its constitutive parts on the analogy of the relation of our own experience to our brains. There is no need to locate the cellular experience at some point within the cell, associating it exclusively either with the molecules or with the empty space. The cellular experience is the subjective unification of the whole complex of events. That means also that the cell as cell interacts with other cells in the larger organism.

The structure of the human brain gives rise to unified subjective experience. It does not appear that anything analogous occurs by virtue of the structure of a rock or even of a tree. The question as to whether the structure of a cell is more like that of a brain or of a tree is a factual one. If, as seems to be the case, unicellular organisms are responsive to their environment in their unity, then the evidence supports the attribution to them of internal relations to that environment. In this case the cell constitutes an important level not reducible to lower ones for the same reason that animal life constitutes a level not reducible to physiology.

If this is so, then the internal relations of the tree to its environment can be located in its cells. Does that mean that the level of the tree can be reduced to the level of the cell? No. What happens in the growth and death of the tree can, indeed, be best studied at the cellular level, but the cells in the tree behave as they do because of the structure of the tree. One cannot explain the tree by a study of the cells apart from the tree.

The question of which levels have the radical autonomy introduced by their own internal relations and which are irreducible only by virtue of the internal relations and actions characterizing their parts is always a factual one. Hartshorne attributes internal relations and action to many levels of organization -- for example, to atoms and molecules. Clearly these are for him factual questions to be settled by such evidence as we can acquire. It is this freedom from prejudice against the attribution of internal relations and action at whatever level the evidence suggests them that is most promising for guiding further investigation. It is the use of the psyche-brain analogy that provides a clarity to Hartshorne’s treatment lacking in Leclerc’s and shows that the paradigm that solves an important problem in philosophy of nature also solves a crucial problem in philosophical anthropology.

In employing a single paradigm to solve a variety of problems Hartshorne follows Whitehead’s direction. Whitehead taught that all reality should be understood through a common categoreal scheme. Hartshorne’s doctrine of compound individuals goes beyond Whitehead but conforms to his spirit and program. It has also helped Hartshorne to formulate a doctrine of God and the world that goes beyond Whitehead’s at a point at which Whitehead’s doctrine strains his categoreal scheme to the limit.

That God is different from everything else is to be expected. If God were not profoundly different, God would not be God. On this point there is no disagreement between Whitehead and Hartshorne. But both are also convinced that to view God as an exception not only to the general character of creaturely beings but to metaphysical principles as well is to render talk of God finally nonsensical. They declare that God is not an exception to the categories but instead their chief exemplification.

Hartshorne takes an additional step. He has already used the psyche-brain analogy to understand the world of physical existents. He proposes that we use it also with respect to God.

Such a move was not available to Whitehead. For him the soul at any moment is located at some one point in the brain. To think of God as similarly located at one point in the cosmos would be absurd. It would not help to think of God as flitting around in the interstices. Accordingly Whitehead opted for thinking of God as unlike everything else in that God is nonextended spatially.

When Whitehead declared God to be without spatial extension, he had the primordial nature chiefly in view. The primordial nature is God’s mentality, and in Whitehead’s scheme mentality is not, of its own nature, spatio-temporal in character. Hence no problem arose. But Whitehead went on to affirm that God has a physical pole as well which feels all the feelings of the world. This seems to introduce something like spatial extensiveness into God. In the world it is physical relations that are in their very nature extensive. Whitehead does not explain how the physical feeling of the world in God leaves God free from the extensiveness of that which God feels. It is also not clear how God has this immediate feeling of all feelings without being spatially proximate to the feelings felt.

Answers to these questions are possible within Whitehead’s framework, and speculations about God are notoriously difficult. But where a simpler and more satisfying answer is available, obscurity and complexity appear unnecessary. Hartshorne offers the simpler and more satisfying answer.

Hartshorne conjectures that God is related to the universe as we are related to our brains. We are not our brains, but what happens in our brains is immediately related to our experience. We are spatially immediately present to every event in our brains. We are coextensive with the sum of these events. But we are numerically different from the sum of these events, and our experience is not exhausted by them. For example, we enjoy consciousness, whereas probably none of the brain events are thus favored. Further, our decisions affect the events in the brain just as these events affect us.

This doctrine does not differ greatly from Whitehead’s. It may even be that he in fact adopted something very like it, for in Modes of Thought he identified God with the whole in a way that is highly congenial to this doctrine. In traditional terms this is the doctrine that God is everywhere instead of the doctrine that God is nowhere. Both conceptually and religiously it seems superior.

All analogies have limits. The limits of this one appear quickly. We have no conscious awareness of the events in our brains, even though they largely determine the content of our consciousness. Our consciousness is directed to the external world. God has no external environment. God’s consciousness is of those events that, for God, are analogous to the brain events for us. That would indeed be a very different mode of consciousness! But this model does not pose the radical puzzle of nonextensive experience of extensiveness.


Hartshorne has understood himself more as a metaphysician than as a cosmologist. His concern is more with the question of what anything must be to be at all than with determining which entities in the universe have which characteristics. On the whole he has accepted and adopted Whitehead’s cosmology. Nevertheless, much in his thought is distinctively his own, and this is true of some of his cosmological ideas. In this paper I have argued that his wide extension of a paradigm derived from reflection about the relation of the psyche to the brain has proved fruitful in carrying forward a basically Whiteheadian way of overcoming reductionism.


I. Ivor Leclerc, The Nature of Physical Existence (London: George Allen and Unwin, 1972).

2. For a recent mainstream discussion of levels and the possibility of reduction, see John Cowperthwaite Graves, The Conceptual Foundations of Contemporary Relativity Theory (Cambridge, Mass.: MIT Press, 1971). Graves writes on page 20: "I will assume, along with Carnap, Nagel, and most other philosophers of science, that a theoretical reduction of one level A to another level B requires the following two things: (1) Each term that appears in A should be definable in terms of the language of B. At best such definitions should be explicit, but if this fails one can use something like a system of ‘reduction sentences.’ Putting the same point in more ontological terms, each entity in A should be fully characterized in terms of the properties and relations of the entities in B. This would constitute a reduction of the material aspects of A into the language of B. (2) Once this translation has been carried out, we can express the laws of A in terms of the language of B. But at this stage these A-laws will appear only as unjustified assertions within B. To complete the reduction, we must also show that the A-laws as expressed in the language of B, can indeed be derived from the basic laws of B by some deductive procedure. Success here would constitute a reduction of the formal aspects of A to those of B.’’ While not denying that in specific cases this reduction can be carried out, Graves sees no evidence of its universal success.

3. Alfred North Whitehead, Science and the Modern World (New York: Macmillan, 1925), p. 152.

4. In Science and the Modern World, pp. 115-16, Whitehead wrote: "The concrete enduring entities are organisms, so that the plan of the whole influences the very characters of the various subordinate organisms which enter into it. In the case of an animal, the mental states enter into the plan of the total organism and thus modify the plans of the successive subordinate organisms until the ultimate smallest organisms, such as electrons, are reached. Thus an electron within a living body is different from an electron outside it, by reason of the plan of the body. The electron blindly runs either within or without the body; but it runs within the body in accordance with its character within the body; that is to say, in accordance with the general plan of the body, and this plan includes the mental state. But the principle of modification is perfectly general throughout nature, and represents no property peculiar to living bodies."

In Process and Reality, corrected edition, ed. David Ray Griffin and Donald W. Sherburne (New York: The Free Press, 1978), p. 100, Whitehead wrote: "The first stage of systematic investigation must always be identification of analogies between occasions within the society and occasions without it. The second stage is constituted by the more subtle procedure of noting the differences between behavior within and without the society, differences of behavior exhibited by occasions which also have close analogies to each other. The history of science is marked by the vehement, dogmatic denial of such differences, until they are found out.’’

5. Process and Reality, p. 21.

6. The Nature of Physical Existence, p. 309.

7. Ibid., pp. 309-10.

8. Ivor Leclerc, Whitehead’s Metaphysics. An Introductory Exposition (London: George Allen and Unwin, 1958), p. 101.

9. Ibid., p. 110.

10. Ibid.

11. In his presentation at the First International Whitehead Symposium at Bonn, August 25-28, 1981, Leclerc repudiated his earlier interpretation and criticism of Whitehead on this point, recognizing that Whitehead seriously intended to affirm the causal efficacy of past occasions.

12. The Nature of Physical Existence, p. 309.

13. Ibid., p. 310.

14. Ivor Leclerc, ‘‘Some Main Philosophical Issues Involved in Contemporary Scientific Thought," in Mind in Nature, ed. John B. Cobb, Jr., and David Ray Griffin (Washington: University Press of America, 1977), pp. 103-4.

15. The Nature of Physical Existence, p. 311.

16. Ibid., p. 303.

17. Ibid., p. 305.

18. Science and the Modern World, p. 215.

19. Process and Reality, p.323.

20. Ibid., p. 91.

21. Ibid., p. 104.

22. Ibid., pp. 106-7.

23. Ibid., p. 107.

24. A. N. Whitehead, Adventures of Ideas (New York: Macmillan, 1932), p. 262.

25. Process and Reality, p. 109.

26. Charles Hartshorne, "The Compound Individual," in Philosophical Essays for Alfred North Whitehead, ed. F. S. C. Northrop (New York: Russell and Russell, 1936), p. 193.

27. Ibid., p. 194.

28. Ibid., p. 215.

29. Hartshorne’s formulations are sometimes quite similar to this, e.g., a man’s consciousness is everywhere in some limited area in his nervous system, rather than localized in a point." Charles Hartshorne, A Natural Theology for Our Time (LaSalle, Ill.: Open Court, 1967), p. 95.

Response by Charles Hartshorne

Cobb’s account of Whitehead and Leclerc I find highly illuminating. His account of me is entirely acceptable. The issues between Leclerc and Whitehead have always seemed to me rather difficult to grasp. So far as I understand them, they are as Cobb says.

It is true that I am more a metaphysician than a cosmologist (student of the special structures of this cosmic epoch). I know the history of philosophy far better than I do current science, except some small parts of the latter having to do with sensation and animal behavior. I do know with some intimacy what it is to he an empirical scientist, but only as dealing with some very special ranges of phenomena.

Cobb does not mention the fact that my application of the mind-body analogy to the idea of God is Platonic. I refer to the idea of the World Soul in the Timacus. As a Harvard student of philosophy I chose Plato and Spinoza as my special topics in the history of philosophy; and I have always been something of a Platonist; though never much of a Spinozist. I have been strongly influenced by the challenge Spinoza issues to us all to take seriously the question of the modal structure of reality. We should learn from him to choose among: complete necessitarianism or denial of any contingency at all; a wholesale contingency of the system of nondivine things, between the parts of which there is no further contingency; piecemeal contingency among the parts of the system as well as of the system as a whole. Spinoza took the first of these options. But his reasons are unconvincing, once one sees that his definition of God as absolutely infinite substance is either meaningless or contradictory, since there are incompossible yet positive possibilities. As Whitehead saw, and so many have not seen, "all actuality is finite," only possibilities can be absolutely infinite. Definiteness is "the soul of actuality," and definiteness means this but not that, or that but not this, where this and that are alike possible, genuinely conceivable. One cannot prove anything by assuming the logical coherence of the classical idea of an ens realissimum or unsurpassable actuality, for this coherence is in no way known or knowable. In addition the sheer denial of contingency violates the principle of contrast. "Everything is necessary" deprives ‘necessity’ of any distinctive meaning.

Granting contingency, which alternative is more reasonable: within the world there is no contingency, all is tightly interlocked, there are no open possibilities for choice or decision; or there are such inner-worldly open possibilities? That there is this world, not some other, is without ultimate necessity, and so it exists by free choice or mere chance, yet within the world there is no chance or free choice at all -- is this a reasonable view? Theistically it amounts to giving God and God alone effective freedom to decide; for the rest there is just the content of the divine decision. In that case how could we, who have no libertarian freedom, conceive the freedom we attribute to God? Linguistically (linguistic analysts please note) this is an incoherent position. If our apparent freedom to select among truly open possibilities is not genuine, neither can we know what such freedom would be in God.

If all is to be conceived by analogy with our human nature, then either Spinoza is right and the eternal, immutable essence of the cosmic soul necessitates everything in the cosmic body, and there is no chance, randomness, or genuinely open alternatives either within the world or as between this and other possible worlds; or there is freedom both in our decisions and in God’s. Either only the real is possible, or, in both the human and the cosmic mind-body relation, there is contingency. Contingency is not an irrational idea; deductive reason is subsidiary to the ultimate rationality, which, as Whitehead says, is the wise, free creation of novel orders. Causal explanation is misunderstood if taken as the effort to show necessities. Rather it is the effort to show possibilities and impossibilities, so that we avoid wasting our energies attempting the latter and make reasonable choices among the former.

Spinoza does give one a sense of the unity of the world and of the world with God, even though he exaggerated this unity, or conceived it too simply. But his predecessors mostly underestimated the unity in both respects. Above all they attempted to combine belief in divine knowledge devoid of inner contingency or change with the assertion, rejected by Aristotle for clear and logically cogent reasons, that the world known by this knowledge is contingent. Aristotle avoided the contradiction by denying divine knowledge of the contingent aspects of the world, Spinoza, by denying all contingency. There remains the admission of contingency both in the world and in God as knowing that world. Socinus took this remaining option and Spinoza noted the fact. Too bad that so few others did note it. They might have learned something. For, unlike Spinoza. they were unwilling to deny all contingency.

I do not comment in detail on Cobb’s paper. He is as clear as I can be about the things he discusses. It is gratifying to be so well understood as I feel I have been by the four former students participating in this volume, the other three being Peters, Alston, and Ogden.