Leslie Armour teaches philosophy at The University of Ottawa, Faculty of Arts, P.O. Box 450 STN A, Ottawa, Ontario, Canada KIN 6N5. His most recent book is Infini Rien: Pascal’s Wager and the Human Paradox. His Being and Idea: Developments of Some Themes in Hegel and Spinoza appeared in Germany in 1992. He is currently working on a study of the concept of idea.
The following article appeared in Process Studies, pp. 203-218, Vol. 21, Number 4, Winter, 1992. Process Studies is published quarterly by the Center for Process Studies, 1325 N. College Ave., Claremont, CA 91711. Used by permission. This material was prepared for Religion Online by Ted and Winnie Brock.
According to the author, a conceptual scheme is necessary and appropriate in Whitehead’s interpretation of experience and this is his most important claim.
(The Canadian symposium on Whitehead’s Metaphysics was held May 25, 1992, at the annual meeting of the Canadian Philosophical Association, Prince Edward Island, Canada.)
In Process & Reality, Whitehead defines the task of “Speculative Philosophy” as “the endeavor to frame a coherent, logical, necessary system of general ideas in terms of which every element of our experience can be interpreted” (PR 3/4).l Whitehead’s suggestion is that a conceptual scheme can be necessary and appropriate to the interpretation of experience, not merely a set of tautologies consistent with all experience. This, I think, is his most important claim, for it is this which enables him to frame an ontology which includes all that there is.
Yet any such relation of necessity and contingency seems to present itself as a dilemma: The claim that there must be some necessary truths arises chiefly from the association of necessity and constancy. If experiences are to be explained, something must remain constant throughout the sequence to be explained. If there are no necessities in reality, it is not certain that anything will be or has been constant. Yet Whitehead was convinced that our experience is Heraclitean: The most basic and important category is that of events. In such a universe, nothing stays the same. Every actuality seems contingent.
The conflict between flux and explanation is perhaps the oldest element in western metaphysics. One can read Plato as struggling to reconcile Parmenides. who seemed to have discovered the logical conditions for intelligibility, and Heraclitus, who grasped the truth about experience. Not unnaturally, Whitehead saw all of subsequent philosophy as a series of struggles with Plato’s problems and insights. And he insisted that the abandonment of this quest was a major disaster:
The claim of science that it can produce an understanding of its procedures within the limits of its own categories, or that those categories themselves are understandable without reference to their status within the widest categories under exploration by the speculative Reason — that claim is entirely unfounded. Insofar as philosophers have failed, scientists do not know what they are talking about when they pursue their own methods, and insofar as philosophers have succeeded, to that extent scientists can attain an understanding of science. (FR 58-59)
If we are to understand Whitehead and get to the roots of his problem, we must begin at a level more general than the one on which he opens the discussion in Process and Reality. We need to tackle the relation between the necessary and the contingent, because every attempt to describe the empirical world entails the use of propositions which have logical forms and have a structure which must include variables, some of whose values are states of the world to be described.
Though, in Modes of Thought,3 Whitehead expresses doubts about the utility of some kinds of formal logic for metaphysics, he insists that it contains a vital clue as to how this relation can be expressed:
Before we finally dismiss deductive logic, it is well to note the function of the ‘variable’ in logical reason. . . . The variable, though undetermined, sustains its identity throughout the arguments. . . . Thus the variable is an ingenious combination of the vagueness of ‘any’ with the definiteness of a particular indication.”
But we must be clear that it cannot be the case that all those variables are such that, when values are assigned to them, the result is a necessary truth. For unless error is impossible, it has to be a contingent fact that some of the propositions we take to be true are true. And if error is impossible, it does not matter what we say. If it does not matter what we say, we can assert both P and not-P from which everything follows. In that case there is no point in doing anything, including metaphysics.
To get even this far, we have to think about the logical forms within which our assertions are made or within which whatever is true or false can he based. Because we are talking about Whitehead here I shall follow him and call these “propositions” and “propositional forms” (see especially PR 184/280-81). For the point that Whitehead wanted to make is more general than any assertion or statement, and it runs beyond the limits of consciousness. It has to do with the forms which are available for the accommodation of the kind of order of which making sense is the principle sub-species of which we are aware.
These propositional forms must necessarily remain constant through a given assertion. As to what propositions are, more will emerge as I go along, but notice, for the moment, that “proposition” is to be taken just as the name for a certain kind of order which is involved in talking sense and not as the name for some entity, both occult and unnecessary which Professor Quine castigates as belonging to the class of empty verbalisms (LPV 108-09; Q 57).
In practice these propositions and their forms must remain constant through a given piece of theorizing or a given attempt to frame a picture of the world or for any possible expression, conscious or otherwise. For, if they do not, we cannot test our descriptions for consistency and, failing consistency, our picture once again, admits everything whatsoever.
There is an obvious element of necessity in this, but we must move very slowly in setting out just what “necessity” means. Professor Quine’s objection to many of its usual senses is not merely an allergic itch. It has good enough rational foundations. (These are most clearly stated in “Two Dogmas of Empiricism,” LPV 20-46.)
We might begin, though, by taking “logical necessity” to mean nothing more than whatever it is without which we cannot claim to go on talking sense about the world. This would allow us to take such necessities as either merely belonging to our discourse (even if they are crucial to it), or as built into the structure of the world without, in either case, taking them lightly. But it is part of Whitehead’s argument, as we shall see, that what is crucial to our discourse must have relations to wider structures in the world, if it is, really, to be intelligible at all.
At any rate, since P and not-P entail everything, the basic need for an element of logical necessity in this restricted sense is to make the description of the world selective. For any true statement or proposition must be selective in what it asserts. A consistent system is one for which not every theorem holds. One might think that this is a parochial view, one bound to the sorts of logics fancied by Frege, Russell, Quine and their successors.
But it holds, too, for logics like those of Hegel. Certain propositions which might be thought to be contradictory are regarded by Hegel as dialectical oppositions capable of resolution in a special process whereby their distinctness is preserved in a new unity. But these oppositions are always expressed in concepts which have both a common property and a distinct scope. Being and nothing collapse into one another. But it is neither true that they literally and fully exclude one another nor that they are identical. They have a uniting property, the property of excluding all determinateness. But they exclude determinateness in different ways. “Being” represents what there is by a single predicate which allows no further differentiation. “Nothing” indifferently excludes all differentiation. And so questions can be asked about how they can be seen together. What it is for two or more logical entities to have properties in common which are not shared with some third entity or for logical entities to be distinguishable from one another is, at least, for them to have descriptions which do not include everything.
But there is more to the issue than this. If we are to have a metaphysic, in the sense of a picture of what makes up any reality whatsoever, we must have some propositions which are true in any possible world. For this it is not enough to have propositions which one takes to be necessary for pragmatic reasons. The pragmatic reasons one has might hold in one world or in the face of one set of experiences, but not in another.
It is sometimes said that the basic propositions of any acceptable logic must be true in any possible world. This must surely be true of the principle of non-contradiction. But logics involve much more than this. Logics, in order to test inferences, must make use of propositional forms which are constructed so that there is some way of determining how propositions can be related to one another. Such propositional forms invariably limit the kinds of values which their variables can take. Thus the commonest of our logics, the logics developed by philosophers like Russell and Quine — one of the foundational works of which Whitehead helped to write — are logics of discrete states. That is to say they mark out entities, classes and their members or whatever, which are perfectly distinct from one another and are wholly determinate. Furthermore, they are so organized that propositions describing such states are imagined to have only two possible truth values. But it is by no means clear that the world in which we live consists of such stales. Indeed it is at least for this reason that Whitehead came to doubt the use of such logics in metaphysical construction (MT 105-8).
A logic which permits one to do metaphysics must either reflect or be neutral to the world which is described. It must not, that is, force us to assert that things or events have properties which they do not have. This is a metaphysical or ontological principle as well as a logical one. The possible worlds are all the worlds there are and can be. A world with a description that did not meet the conditions of the principle of non-contradiction, would be a world that could not exist. For it would contain everything.
Logics have another property which we should keep in mind. They are intended, as William Kneale said in the history he wrote with his wife Martha, to be valid for all possible subject matters.2 If they were not — if, for instance, we needed two or three logics, each for a different subject matter — we could not, in principle, tell how the subject matters were related. If we could not tell how they were related, we would have to express this new discovery in a logic which was adequate to all of them.
Thus, we could not tell whether any logic was better than another unless we could find some way of determining which propositions were true in every possible world, and therefore which were universally necessary truths. Such propositions are often said to be analytic or tautological, but it is because he does not see how one can make the analytic-synthetic distinction anything more than arbitrary that Professor Quine says that he espouses a very fundamental logical pragmatism (LPV 46). He is certainly right that, short of a metaphysical view of the whole, there is no way of making the distinction if one wants something more than a pragmatic result.
Yet to fail to make it is to fail to distinguish between the true and useful. It is not, indeed, that there might not be a pragmatic picture of the world, but rather that such a picture would simply represent the perspective of the particular enquirer in a particular situation. As such, it could have no validity beyond the instant at which it was created. There would be no foundation for any inductive inference. More importantly, there could be no general science, no metaphysics.
Neatly perched on the other horn of the dilemma, however, sits an equally pressing problem. What if one could make a distinction between the necessary and the contingent and could see that there were necessary truths which were other than the simplest tautologies? Suppose one could see that logics of discrete states were or were not necessary?
Then, of course, in the respect in question, the universe could not change. For, if it did, it would have taken some characteristics which nothing in any possible universe could have. Is it possible that this could really be the case? It is worth our while to explore this question for it enables us to see precisely the point at which Whitehead begins his account of the world in Process and Reality.
A difficulty at once occurs. The unchanging characteristics of any world must be universals, forms, or ideas. They cannot be particulars. Particulars always have a reference to places and times. If, as Whitehead insists, a universal is what can characterize many particulars, and a particular is what is characterized by universals, then particulars can have, as such, no features which extend beyond the smallest occupiable place and moment (PR 48/76).
In these terms, there can be no particulars without universals or universals without particulars. And yet any actual world must be one in which particulars have a certain primacy. To see this is a prelude to seeing why it is that Whitehead thinks that there is a certain primacy to the category of events. Ultimately, primacy belongs to events of the special sort which he calls actual occasions. If he is right about this, then, indeed, the dilemma will be very real. I suggest that the dilemma is very real and that Process and Reality is, more than anything else, a device for dissolving it.
The reason for the primacy of particulars is, once again, simply logical. There may be forms of justice and redness. Whitehead insists that there are, though he speaks of them as eternal objects, for a reason which will emerge. But the form of justice is not just, for justice is some sort of relational property which must hold between particulars. And the form of redness is not red, simply because to be red is to occupy a spatial surface in a certain way, and forms do not occupy surfaces. One interesting way to see that this is so is to see that there can be no uni-propertied things. It always takes two or more properties to get anything identifiable. Red things have to exhibit a certain amount of redness, and there is no justice between non-existent persons.3
But it would also seem to be true, and to follow from the same premises, that no particular can ever exhaust the properties which it instantiates. If one has to be red in some particular way in order to be red at all, then there must be other ways of being red, if not in this world, then in some other logically possible one. The old puzzle about how it is that Socrates and Plato can share the same humanity without being the same human being dissolves if one realizes that humanity is such that it can never be wholly exemplified in a single individual.
There is a basic latitudinarianism built into universes in which properties like humanity are in this sense indeterminate. Indeed, a significant part of Whitehead’s point about the clement of freedom in the world we live in can be expressed in just these terms: Things can never be wholly determined by their Platonic or neo-Platonic exemplars.
This lack of determinacy presents an opportunity for freedom. It permits us to think that the universe, in some sense, contains genuine novelty. Yet it also seems to create a gap which is unfillable, not just contingently, but necessarily. If something is red, it is always some particular shade of red. Perhaps it is vermilion or scarlet. If something is just, it is always some specific just act, the righting of a wrong to some oppressed majority or the rendering of the right verdict in a trial. Does one associate each of these particulars with a distinct form? There are indefinitely many such forms, but none is the form of particularity. Those who think Duns Scotus attempted to find such a form, notoriously find it unsatisfying because it seems that there is an implied contradiction in the very idea of a form of “thisness.”4
The primacy of particulars is apparently given by the failure of the forms to specify particularity and it does not appear that this gap can be filled. I think a certain insight dawns on one who reviews this situation against the background of Whitehead’s system: The situation which I have just painted has more radical and deeply puzzling consequences than is usually supposed.
For though the particular must be associated with a place and time, it will follow from its very primacy that it is unique. In that case, if it cannot really be said to be a manifestation of any form or any universal, then it cannot be a thing with a fixed place in space and time. For such a designation implies that it can be identified as a subject and then be said to have precisely whatever characteristics are necessary to fix it in this way, A particular has to be an event, and it cannot itself persist, for to persist it must in some respect remain fixed through time.
The most that can be said is what Whitehead actually says: A genuinely basic event, an actual occasion, simply is at its moment, and it perishes with that moment (PR 29/43).5 But even this must seem doubtful, for its uniqueness prevents both determinateness and persistence. If “to be” is to manifest being, it cannot really be said to do that.
We can now state our dilemma more clearly: If metaphysics is to be possible, the world and our statements about it must exhibit some necessary unchanging principles. But there is no way in which, apparently, anything which is genuinely real can manifest any such principle, for reality consists of particulars which are, at most, designatable as unique events. A simpler sub-dilemma, suggests that, surely, no world can exist which has a description which fails to meet the principle of non-contradiction, but no real world can have a description which meets the conditions of any principle.
I shall argue that it is through the doctrine of ideas that this dilemma is to be addressed. Ideas have logical properties, some of which are necessary, but also figure in experience, and so they have properties which must be known empirically.
In the terms which I have sketched, Whitehead’s account of the problem about how the ultimate particulars are related to the forms or universals is the doctrine of prehension. Literally or metaphorically, the actual occasions, as he prefers to call these ultimate particulars, grasp the forms. It has always been difficult to find a felicitous way of expressing Whitehead’s point, and some of the difficulty which interpreters have had in plumbing the conceptual depths of Process and Reality derive from this fact. Most simply, the ultimate particulars, the actual occasions or actual entities as they are alternatively called, are to be regarded as open, as occasions for the ingression (to use, once again. White-head’s own word) of the forms. We can see now why he calls the forms eternal objects; for it is by reference to them, in a particular way, directly and indirectly, that the actual entities acquire their content. They appear, that is, in reference to the object, as the ideals which the actual entities imperfectly realize.
This complex set of relations — ingressions, prehensions, imperfect actualizations — may seem somewhat odd, but they have at least one simple enough explanation in terms of the science of Whitehead’s time and of ours. There is always a degree of approximation in our measurements and a degree of idealization in our theorizing. Whitehead takes this fact to be a manifestation of the truth about the world, not an imperfection in scientific knowledge. Precise measurements, perfect descriptions, and theories which are absolutely true all have their ultimate homes in the domain of eternal objects, not in the domain of things in the world. In the nature of each actual entity “there is always a remainder” which is undetermined (PR 27-28/41).
The actual occasion is open to ingression and has prehensions, positive and negative. It is not, of course, equally open to all ingressions and equally inclined to all prehensions. But how are we to understand this? It is essentially the explanation of the concept of idea that gives us the clue. Whitehead explicitly relates this notion to the concept of idea: “‘Prehensions,”’ he says, “are a generalization from Descartes’s ‘cogitations,’ and from Locke’s ‘ideas”’ (PR 19/29).
This does not make them, for Whitehead, “mentalistic,” for the crucial phrase here is “generalization.” If we can understand, however, in just what sense we have ideas, then we can understand how it is that actual occasions prehend and are open to ingression. The special context in which Descartes finds his cogitations is our minds. But if we insist that cogitations are not necessarily related to minds like ours, we have some ideas of Whitehead’s notion. Locke, Whitehead thought, provided his own grounds for generalization, and it is really Lockean ideas which must have pride of place in our attempts to understand Whitehead.
One may suspect that, in the end, there is no role left for ideas in Whitehead’s conceptual scheme, but this I want to deny. The suspicion develops naturally enough from the fact that the concepts which Whitehead develops out of the concept of idea — especially the concepts of “eternal object” and “prehension” — do most of the work which Whitehead wants done. Furthermore, the word “idea” has complex connotations (especially mentalistic ones) which Whitehead wants to avoid. But I shall argue that it continues to be needed to designate the framework within which these more special concepts function, and this framework does important logical work. But the best way to settle such a matter is to follow Whitehead through some of his reflections on Locke and then through his accounts of “ideas” in Process and Reality.
At the beginning of Process and Reality, Whitehead says, “the writer who most fully anticipated the main positions of the philosophy of organism is John Locke in his Essay, especially in its later books” (PR xi/v). A footnote suggests that by this he means to focus on Book 4, Chapter 6, Section 11 of that work (HU 585-87). Here we find out how a notion of simple and complex ideas introduced by Locke a little earlier bears on the whole question of the unity and intelligibility of the world.
Whitehead returns to the point in the middle of Process and Reality and insists that “the metaphysical superiority of Locke over Hume is exhibited by his wide use of the term ‘idea.”’ He particularly likes the way in which Locke’s ideas were “determined to particular existents” (PR I 38/209).6 Bear in mind that what we are looking for, in Whitehead’s terms, is an account of just how it is that we can understand the relation between actual occasions and eternal objects. We must understand how actual occasions are related to one another through prehension, and how they are more open to the ingression of some eternal objects than to the ingression of others. In other words, we are looking for an account of the way in which the world holds together and of how it copes with freedom in an orderly way. The passage in Locke’s Essay referred to in Whitehead’s footnote has to do, amongst other things, with the way in which substances hold qualities together. This is another way, I think, of posing the question about just how actual occasions are open to ingressions or tend toward prehensions.
The passage in question, one ought to note, is quite startling in the face of some very common accounts of Locke. It deals with two difficult topics. One, which obviously attracted Whitehead strongly, concerns the unity of the universe. It is here that Locke denies what Whitehead called the “doctrine of simple location.” Locke says, here, that no substance can be understood except in reference to all other substances. Knowledge requires an ultimate unity and, indeed, this is to be the basis of Locke’s argument for the existence of God in Chapter 10 of the Essay (HU 619-21).
The relevance of this issue — or at least of the aspect of it which concerns the unity of the whole — will appear, but the second topic is really the heart of the matter. For it is about the way in which we know substances, and, in the course of telling us about that, it offers clues about how it is that substances do their work. Locke’s account of substance has it that substances are whatever it is that perform the function of holding qualities together. The substance which is the sun holds together properties of being “Bright, Hot, Roundish” (HU 298). Sometimes Locke speaks of “mere Powers” such as the “Color and Taste of Opium as well as its soporific or anodyne Virtues.” These powers depend on opium’s “primary qualities.” Power and primary qualities together “make a great part of our complex Ideas of Substances” (HU 300).
Orthodoxy has it that Locke supposed that substances were hidden from us because they did not appear on the surface of experience and had to be inferred. But that is not really what Locke says. He does say that we know most about secondary qualities and that these qualities depend upon primary qualities. Then he adds “or, if not upon them, upon something more remote from our Comprehension, ‘tis impossible we should know.” And he certainly does insist that we do not know “the Root they spring from” (HU 544-45). In the index he himself made for the Essay, substances are said to be “not very knowable.”
Yet the passage cited by Whitehead (HU 585-87) suggests a different meaning. It explains why we have trouble knowing substances. True we can understand some things in terms of others — secondary qualities through the powers of primary qualities, perhaps in the special sense that we understand color vision through the properties of light and the properties of our optical apparatus. But that only goes a little way. What Locke says in this later passage is that we know very little about substances because substances are to be understood, ultimately, only in terms of the whole universe. That is, we can see how the characteristics of things are held together only when we see the whole system and only when we see it as a system of a certain sort.
What sort of system is it? If we read back as far as Section 5 of Chapter 4 and Sections 7 and 8 of Chapter 6 (HU 564; 582-83), we discover precisely Whitehead’s problem: The system has to be one which expresses itself through individuals whose substance cannot be grasped by us. How, then, are we to understand these individuals? The answer is that we know them through what Locke calls their “simple and complex ideas.”
In Locke’s view, in these passages, simple ideas are literally what is presented to the mind. Complex ideas we make up. The problem is that the complex ideas do not necessarily represent nature — indeed they are not related to nature in the way that the simple ideas are — and yet what we know of substances is necessarily what we can represent by complex ideas. For simple ideas tend to represent simple qualities. If, of course, we could see how all the simple ideas fitted together we would see the truth about substance. We make errors because we put things together wrongly.
A person who thinks there are dragons is right to believe that there are reptiles, flying things, and fires. He has put together some ideas in the right way. But he has then wrongly put together some of his compounds. But everything must have a place somewhere, and so there is some right way of putting things together. If there is, then there is what Locke calls an archetype of things which is really in nature. The fact that he tells us that the archetypes which we create are not in nature does not mean that he thinks that there are no archetypes in nature.
Indeed, the burden of the Book 10 argument about God is surely that knowledge is made possible by the fact that nature exhibits God’s archetypes. And the claim that Locke is a somewhat simple-minded empiricist is, in any case, finally destroyed by a passage from Chapter 19 of Book 4. “Reason is natural Revelation, whereby the eternal Father of Light, and Fountain of all Knowledge communicates to Mankind. . .” (HU 698; italics Locke’s).
It is because the archetypes from the mind of God form a coherent system and form the basis of the substance of things that knowledge after all is possible. Notice, however, that what I just wrote is meant quite literally: The archetype in question is the basis of the substance of things, i.e., it is the basis of any possible account of how the qualities of the world fit together.
The things themselves have their own independence which, if we consider them as particulars, seems to put them outside the scope of knowledge. But they do have a structure which makes some Platonic qualities more relevant to them than others. Locke insists in Book 4, Chapter 11, Sections 13 and 14 that we can only know particulars through their ideas (HU 637-38), i.e., through seeing that the best interpretation of the idea is that the thing exists. And this is because the idea, in the whole context of what Whitehead calls Locke’s philosophy of organism, gives it a position in a system which makes certain properties relevant to it and others irrelevant.
In the Essay, Locke is reluctant to tell us what ideas are. John Norris chided him for this omission and, in his replies to Norris and to Norris’s acknowledged master, Malebranche, Locke finally admits a measure of reality to ideas.7 In those replies, the suggestion about the existence of ideas is of a very special kind. Ideas are only real, Locke says, in the way that motion is real. They are not another kind of thing. They are what we interpret when we get knowledge. Though Norris, in Locke’s view, wrongly attributed to them a reality which was independent, he rightly estimated their importance. They are the source of both right and wrong interpretations of the world. What we should mean by the world is whatever counts as the correct interpretation. What it is for something to be material is for it to be properly interpretable as a material substance. If so, the problem cannot be about “inference” to something unknown.
It is this notion which Whitehead wants to generalize. I do not know that he ever read Locke’s reply to Norris but he says “Locke’s ‘particular ideas’ are merely the antecedent actual entities exercising their function of infusing with their own particularity the ‘passing on,’ which is the primary phase of the ‘real internal constitution’ of the actual entity in question” (PR 213/324). Whitehead’s generalization thus includes a restatement of the position I was ascribing to Locke, a restatement from the perspective of the object. But that this perspective should exist is entailed by the Lockean theory, too.
Thus generalized — and so freed from the limitations imposed by the association of ideas with the conscious states of reflective beings — ideas provide a way of escaping the dilemma. They can be eternal and unchanging, but can only exist in a world in which they are expressed through particulars which instantiate their properties. Those particulars in turn are necessarily perfectly particular. They are momentary and perish when the moment they have defined has finished. This, in turn is compatible with explanations for things only because what remains is the structure which determines a measure of relevance.
To understand fully what Whitehead intended — and how he developed these Lockean notions — one must follow him through the nine passages of Process and Reality which contain important discussions of ideas — and beyond into Adventures of Ideas and Modes of Thought. On first reading, one might even take Whitehead’s initial account to have disposed of the concept of idea in favor of the concept of prehension. Descartes and Locke are mentioned, but all that is said about them and about ideas is that the Whiteheadian doctrine of “prehension” is a generalization of their theses (PR 19/29). The same suggestion is repeated at the next mention, but at once a complication sets in. For we are told that ideas are associated with propositions, and also that, in addition to what we might ordinarily take to be propositions — whatever it is, for instance, that is expressed in statements or sentences — there are real propositions, and that these are “lure[s] for feeling.”
What is being expressed in this section of Process and Reality is, evidently, the theory which I have already discussed, the theory that there is a structure to the world such that its particulars have more natural affiliation to some eternal objects than others. (In Whiteheadian terms, one might, insisting that particulars have a measure of real freedom, say that they are more inclined to some eternal objects than to others.) The same theory has it that particulars are also primary and this, too, is again emphasized in the discussion of fact and form in which Whitehead insists that Locke is right to hold that “there are not first the qualities and then the conjectural things, but conversely” (PR 53/83).
At the next mention of ideas, however, this thesis is expanded and becomes a major key to understanding Whitehead’s doctrine. Section IV of the chapter on “the extensive continuum” introduces the notion of the idea as the key to the dual existence of things in the world and in knowledge (PR 76/118). The example has to do with the sun as an Object of knowledge and as an Object in nature.
Whitehead returns to Descartes. In a famous but difficult text (part of the Replies to the First Objections), Descartes said that “the idea of the sun will be the sun itself existing in the mind, not indeed formally, as it exists in the sky but objectively, i.e., in the way in which objects are wont to exist in the mind.”8
In Science and the Modern World, Whitehead had said that he had difficulty in squaring this sentence with the rest of Descartes’s doctrine. (SMW 73-74) and in Process and Reality he chides both Descartes and Locke for failing to hold firmly and consistently to the doctrine expressed in it. Whitehead understands this duality as a structure which enables us to identify a thing correctly. The sun can only be an object in nature insofar as it is a center of focus for an entire system which parcels out the properties of the universe so as to provide a unique sequence of actual occasions which we are able to grasp through our scientific theories. The idea in the mind must exhibit the same logical structure. Though the same structure organizes different actual occasions, the two systems mirror one another in such a way that we can be said to have knowledge. Descartes, of course, is not so inconsistent as Whitehead supposes since, for him, objective reality is the property which ideas have by virtue of which they are able to refer to things, while formal reality refers to the form which is shared between an idea and an actual thing.
It would appear that, for Descartes, an idea would have to have ideational reality as well — would have to exist in at least the minimal way which was admitted by the later Locke (see Alquié’s footnote, OP 520). But this it would seem must be true for Whitehead as well. For it now turns out that prehensions account for only one dimension of ideas. There is also a kind of ideational form which permits that kind of mutual reflection through which knowledge becomes possible. Indeed, much later on in Process and Reality, Whitehead castigates Locke for having too thin a notion of idea. Ideas cannot be, he says, as Locke claimed they were, “mere qualifications of the substrate mind” (PR 147/223). That is indeed so — and the later Locke who wrote in response to Norris and Malebranche would, as I have already suggested, have agreed. But if ideas were merely the immanent tendency of the system of particulars to favor some qualities over others, they would be mere qualifications of a substrate object. Whitehead’s doctrine in his chapter on the extensive continuum is, after all, that ideas must have this immanent tendency and be the structures which permit intelligent reflection, the structures which permit what exists at one place and time to be intelligible anywhere in the system. The presence of adequate structures for intelligibility is a general condition for all knowledge whatsoever and a particular condition for the possibility of science. Physics has to be the same everywhere, even if no two actual occasions can be identical.
The uniformity of physics is possible only if there are structures which permit one thing to be reflected in another. If we ask how any such reflection is possible, we shall discover that ideas have still another dimension. Ideas are also eternal objects. Whitehead explains that he does not call them forms or essences because he does not want to be mistaken for a Platonist or for one of the “critical realists” of his own time who tended to use the word “essence” in ways which he felt were undesirable. He does not himself use the word “idea” here, he suggests, because he is afraid that it will be understood in a subjective way (PR 44/70).
But quite a bit of confusion is, I think, caused by the fact that one may not notice that what he says in this passage — in a chapter on “fact and form” — is that eternal objects are, in any case, a dimension of whatever it is that has figured in his account of prehension and of the common substrate of things and knowledge. The word “idea” tends to get rather rough treatment, yet, if one looks closely. Whitehead has, in fact, spelled out what needs to be spelled out. He makes it very clear that the eternal object is not a distinct entity. “The actual subject which is merely conceiving the eternal object is not thereby in direct relationship to some other actual entity. . . . An eternal object is always a potentiality for actual entities (PR 44/70).
A few pages later, Whitehead describes the logical difficulties which would confront one who thought that eternal objects were really distinct entities which could exist in abstraction from their counterpart prehensions. There is not and cannot be, for instance, “One entity which is merely the class of all eternal objects.” If there were, there would be yet other objects, he says “which presuppose that class but do not belong to it” (PR 46/73). He is referring here to an analogue of Russell’s paradox. For instance, if there were such a class there would be another class which would include it and all the other eternal objects. The totality would be an object, and so distinct from the rest. But then it could not be the totality.
What there is, he says, is a complex — but a complex can only be understood as something expressed through a plurality. Idea is surely the proper name for this complex.
Most revealing of all the references to the word “ideas,” however, are the final paragraphs of the chapter on “the subjectivist principle.” No doubt Whitehead was always uneasy about the word “idea.” But in these paragraphs he insists that there is something which has a unity of its own and which shows itself as exhibiting the dimensions which he refers to as “eternal object,” “prehension,” and “the structural conditions for knowledge.” There, Whitehead, explaining the principle of concrescence, whereby eternal objects ingress into actual occasions — or whereby actual occasions prehend eternal objects — says “this development is nothing other than the Hegelian development of an idea” (PR 167/254).
One must be careful, as I shall argue, about the reference to Hegel, but what Whitehead is talking about is basically the notion that one cannot understand actual occasions except in terms of the idea they instantiate, or understand the ideas except in terms of their expressions. This is not a vicious circle for two reasons. One is that ideas develop through their instantiations just as their instantiations are shaped by ideas, and so the unfolding of the idea is a process. Ideas are thus not simply eternal objects nor are they simply prehensions. Eternal objects and prehensions are aspects of ideas. The other escape from the vicious circle depends on the fact that ideas are not intelligible in isolation. We may think we have an idea of “redness.” But nothing is red unless it occupies some space and has a place in some system. We cannot imagine anything’s being ‘just red” any more than we can imagine anything’s being ‘just good.” Even abstract mathematical properties have their meanings within the context of some system which provides a distinct place for each of them — which explains, perhaps, why Kant supposed that “7 + 5 = 12″ is not really an analytic proposition.
Whitehead picks up the notion of idea I have been talking about again in Adventures of Ideas and in Modes of Thought. One of his basic themes is that, as he says, “human life is driven forward by its dim apprehension of notions too general for its existing language. Such ideas cannot be grasped singly, one by one in isolation” (AI 16). As usual, if one substitutes “prehension” for “apprehension” one gets a universal principle, and, indeed, later in Adventures of Ideas Whitehead suggests this generalization in his account of prehension (AI 176-77). and then he goes on to say:
The universe is dual because in the first sense it is both transient and eternal [and]…because each final actuality is both physical and mental…. Throughout the universe there reigns the union of opposites (AI 19).
In Modes of Thought, as I noticed, Whitehead relates these concerns to the nature of logic itself. The basic notion of the proposition with variables expresses an idea with roots in reality (MT 106-7).
Generalized to its extreme, this relation is the relation between any eternal object and any actual occasion and it is built into Whitehead’s world. But he notes that “logic presupposes metaphysics” (MT 107) — not metaphysics logic.
Given all this, I think we should agree with Whitehead’s assessment that his notion of idea is the notion of something which is closely related to “the Hegelian development of an idea.” Yet one must make some distinctions and not jump instantly to the notion that it is “the Hegelian Idea” that Whitehead is talking about.
The Whiteheadian idea is, I think, in one of its aspects (but only one) a particular species of the Hegelian concrete universal. If one asks for the idea of the Prussian state, the Poland China pig, or Barnard’s Star, one is asking, in Whiteheadian terms, for a certain kind of account of development. Each of these examples involves a point to which we ought to attend.
The Prussian state had its origins in 1226 when a Polish Duke, Conrad of Mazovia, invited the Teutonic Order to come and bring what he took to be civilized government to the local heathens. The idea developed as the organization of the state developed. The knights originally admitted none of the local nobility into their ranks and, as the state took shape, so did the sense of identity amongst the locals. The state did not really take on its full identity until after 1609 when it fell into the hands of the electors of Brandenburg. At each state, the idea influenced events and the events influenced the idea. As people began to think of themselves as Prussians, events began to take on a Prussian character.
One could try to bring order out of this oscillating development in Whiteheadian terms by noticing that the actual occasions, the events, had to take on one shape or another. Whatever happened there could not have been a shapeless or formless event. But this is just to say that there is a conceptual necessity for something to happen — this is the foundation of the idea. It is equally a necessary truth, as we have seen, that the conceptual necessity cannot determine everything.
The Poland China pig developed neither in Poland nor in China but out of pigs introduced by the Shakers into the Miami Valley of Ohio where there was a need to find a quick-fattening animal to eat up the surplus corn. The modern neat black pig with white feet, white face, and white tail tip developed out of a spotted animal and only gradually did it become clearly identifiable — the embodiment of an idea and an ideal which could be identified across the generations. No one had a clear idea of the Poland China pig when the process started in 1816 or indeed until after 1870, and no one can really say that either the idea or the events had precedence.
Ultimately, a state or a pig is, in Whiteheadian terms, a vast society of actual occasions. Ultimately, too, at one extreme are eternal ideas which resist further analysis, and actual occasions which are the simplest units any universe could exhibit. But the whole universe — states, pigs and stars — is an interlocking system in the sense that the events can be open to some and not other basic ingredients, and their openness is determined by the whole system. Pigs depend on stars being what they are, and states may rise or fall on grains of sand as anyone who has followed the history of silted ports knows full well.
When we come to stars, though, we may suppose that the story must be different. The minds of knights and nobles went into the shaping of the Prussian state, and the farmers who came after the Shakers shaped the Poland China pig. But no one shaped Barnard’s star.
Whitehead’s point, however, is that one cannot draw an ultimate line between stars and pigs. One cannot do so, for they are intelligible only as a system. If the pig depends upon the star which must stay in its place for the firmament to stay firm and must not emit too much radiation for reliable pig breeding, they both simply represent different points or moments of focus in the same system. Stars, pigs, and Prussian knights all prehend, though stars do not do so consciously. It is a mistake, Whitehead believed, to think that what characteristically attracts our attention in consciousness only exists in consciousness.
Stars take on the characteristics which make them graspable in ideas only as a result of a long process. But every such process requires some determinate form. Whitehead says “The spatio-temporal relationship, in terms of which the actual course of events is to be expressed, is nothing other than a selective limitation within the general systematic relationships among eternal objects” (SMW 161). Naturally, this relationship is reciprocal. “An eternal object, considered as an abstract entity, cannot be divorced from its reference to other eternal objects, and from its reference to actuality generally” (SMW 159).
The Hegelian Idea — of which there is, in principle, only one — is something rather stronger. Though the world and the Hegelian Idea are supposed, ultimately, to become identical, at least if one reads Hegel in the most traditional way, Hegel’s Idea has a greater reality than any yet actualized world, and it serves, finally, as the explanation of things. Nature and history alike are the unfolding of the Idea. The reciprocity so characteristic of Whitehead’s developing ideas is at least less clear in Hegel. If one has the perspective of the Absolute, would one be able to see the whole of reality as determined by the Idea? Not necessarily, of course, for, in Hegel’s view, too, there is a process of events which are sublated (or ablated) into its later phases, and the particulars of those developments. since they lack perfect rationality, are also not fully determined. But the Idea does have a kind of precedence for Hegel which it does not have for Whitehead. Whitehead would never have written the Logic of the Encyclopedia as a distinct enterprise to be completed before the Nature Philosophy and the Philosophy of Spirit.
Perhaps this point needs a little further explanation. Could Whitehead have written a book about the nature of the eternal objects conceived as logically quite independent of anything else? Whitehead’s eternal objects do not and cannot change. But, as he insisted in Science and the Modern World, “the metaphysical status of an eternal object is that of a possibility for an actuality” (SMW 159). They are incomprehensible without an actuality. They are only intelligible within a larger conceptual frame. Such a larger conceptual frame which exhibits the relations of eternal objects to actual occasions and vice versa is what I am calling an idea. An “idea,” initially, as Locke thought, is what we take as the object of our interpretation when we confront our experiences. If we follow Whitehead, we can specify its conceptual ingredients — eternal objects, prehensions and so forth. But an idea is not another thing in the world. The things in the world are what our interpretations of ideas lead us to believe in. They are ordinary objects — stars, mountains, houses, trees, animals, and people.
Eternal objects really do not have a final conceptual priority. And this makes a difference. For the Whiteheadian picture is one in which rational structures of things and whatever is ontologically primary necessarily reflect one another. It is precisely this necessity which is the necessity of ideas to be expressed through actualities which can be understood as ordinary objects. The necessity for actual occasions to have structures permits us to resolve the problem about how one can have a philosophy which uses conceptually necessary systems of ideas to represent an open universe.
But let us now, finally, come to the point. Does this account of things make metaphysics possible? It is useful to turn the problem around so as to see it more clearly in the context of the central strands of metaphysics. The Whiteheadian tension between eternal objects and actual occasions is a version of the traditional problem created by the duality of concept and thing. Any such duality, if it is not resolved, creates a rupture between thought and being. And indeed, what I have been calling Whitehead’s account of ideas is a generalization of the notion of thought, freeing it from its parochial associations with our particular fields of consciousness. But, freed or not, whatever view one takes of thought, one must be able to give an account of the relation of thought to being. This relation is, on the face of it, disastrously paradoxical, and one way to answer the basic question is to ask whether or not the account which I have been ascribing to Whitehead does enable us to deal with these paradoxes.
Logically, assuming that we have and ought to have both concepts and things or both thought and being, the initial possibilities are only five: (1) Thought and being overlap, but thought is more extensive than being; (2) Thought and being overlap, but being is more extensive than thought; (3) Thought outruns beings and being outruns thought (each in some distinct aspect); (4) Thought and being are wholly separate; and (5) Thought and being exactly coincide; Possibilities (4) and (5) cannot both be true. But, though (1) (2) and (3) each contradict (4) and (5), they do not contradict each other. Indeed, (3) reminds us it can both be true that some elements of thought Outrun being and that some elements of being Outrun thought. If these five options include all the possibilities, however, skepticism seems to be inevitable.
If being outruns thought there are things we cannot know, and we do not know how these things influence what we seem to be able to know. Indeed, on Whitehead’s account or on the account which Whitehead draws from Locke, everything must be organically related. If thought and being were wholly separate, one would get the same result. In neither of these cases can the line between thought and being be represented in thought. If thought outruns being then, again, unless there is something in thought which represents the distinction, we are lost so far as all knowledge goes, and, again, the line between the two presumably cannot be represented, for there is no determinate thought which counts as the thought of non-being. If, however, thought and being exactly coincide we know that something is wrong, for then it would seem that error would be impossible. Everything we could think of would be an actuality. And, if we know anything at all, it is that we make mistakes.
The way out, it would seem, is to hold that thought and being permeate one another. If one already supposes that this must be the case — on the ground again that, if it is not, we cannot finally talk sense in and about the world, then the solution which Whitehead suggests becomes at least plausible. On such a view, thought must occur in the guise of ideas in and through the actual occasions, and the actual occasions must grasp or prehend the ideas. Thought and being are not perfectly identical, for actual occasions are particulars and ideas are not. But it is not that there is something in the actual occasions that thought cannot grasp. It is rather that the particularity of the actual occasions is represented as a place in a system of ideas. And it is not that the ideas are not expressible in actual occasions. It is only that nothing counts as their completed and full expression.
LPV — Willard Van Orman Quine. From a Logical Point of View. New York: Harper. 1963.
Q — Willard Van Orman Quine. Quiddities. Cambridge, MA: Harvard University Press, 1987.
HU — John Locke. Essay Concerning Human Understanding. Ed. Peter H. Nidditch. Oxford: The Clarendon Press, 1975.
OP — Rene Descartes. Oeuvres Philosophiques. Ed. Ferdinand Alquié. Paris: Gamier, 1967.
1Three other editions of Process and Reality exist. One published by Cambridge appeared shortly after the original. Another, published by the Free Press in 1969, did not correct the many errors in the original. The fourth edition, which was edited by David Ray Griffin and Donald W. Sherburne in 1978, did correct all the obvious errors. This edition was reproduced in paperback in 1979 and has become the commonly used text. But neither Whitehead’s manuscript nor the typescript from which Macmillan’s worked seems to have survived. Hence the corrections have, inevitably, an element of speculation about them. Fifty years of secondary literature refers to the original edition and the corrected edition includes its page numbers in the text. For convenience I have included page numbers to both the original and the corrected text in the 1979 paperback version.
2WiIliam Kneale and Martha Kneale, The Development of Logic, Oxford: The Clarendon Press. 1962, p.741. This particular section was written by William.
3Possible persons may have rights, and they can be treated unjustly when, for instance, we act so as to deprive the next generation of clean water. or of contacts with elephants or whales. But the injustice is caused by real, existent persons.
4The reading often given to Dons Scotus is unfair because what he means to designate by haecceitas is the individuating difference. In addition to the forms as we usually conceive them, there must be a principle of individuation. It may be a kind of principle of emanation through which thc forms express themselves as distinct things. What is involved is a form which, when it informs things, results in their individuality.
5Whitehead ascribes this notion to Locke. He repeats the idea in a number of contexts, reminding his readers of it again on the last page, PR 351/533.
6The passages to which Whitehead calls attention in his references to Locke’s “wide use” of the term “idea” seem to me to refute quite clearly the notion occasionally put forward that Locke thought ideas were mental images. Jonathan Bennet ascribes this view to himself and to Michael Ayers (Times Literary Supplement, March 20, 1992, p. 9). In fact Ayers (Locke, London: Routledge, 1991. Vol. I, p. 68. says this is only one side of the story and that “much of Locke’s language and thinking was found to reflect a duality in the notion of an idea recognized by Descartes” and that Lockean ideas also have a kind of “objective being.” Ayers confines himself to the Essay, but this is what emerges most strongly in Locke’s Remarks upon some of Mr. Norris’s Books. (See below).
7John Locke, An Examination of P. Malebranche’s Opinion of seeing all Things in God, in Posthumous Works of Mr. John Locke, London: W. B. for A. and J. Churchill, 1706; and Remarks upon some of Mr. Norris’s Books, wherein he asserts P. Malebranche’s Opinion of seeing all Things in God, in A Collection of several Pieces of Mr. John Locke, London: J. Bettesworth for R. Franklin, 1720.
8I cite Alquié’s edition because it has useful notes on this passage (See OP 520).