The Incoherence of Whitehead’s Theory of Perception
by Charles A. Kimball
Charles A. Kimball is professor and chair of religion at Wake Forest University in Winston-Salem. North Carolina. An ordained Baptist minister, he served as Middle East director for the National Council of Churches from 1983 to 1990. The following article appeared in Process Studies, pp. 94-104, Vol. 9, Numbers 3-4, Fall-Winter, 1979. 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.
In his theory of perception, Whitehead attempts to do justice to all our various and apparently conflicting perceptive experiences. By reconciling two seemingly incompatible traditional accounts of perception, he hopes to cut through the Gordian knot of problems which have bound the theory of perception since the seventeenth century. Unfortunately, the boldness of his aims precludes their successful attainment. I shall try to show this not so much by recounting Whitehead’s systematic treatment of perception as by analyzing the reasons he gives and the experiences he cites for the adoption of his theory in the first place. Thus, this paper is not an account of Whitehead’s theory of perception solely in terms of the categories of the philosophy of organism; rather, it is a critique of the coherence of that theory from a point of view outside it.
To put Whitehead’s theory of perception in perspective, let us consider two traditional accounts, each widely held historically and each -- separately -- commanding considerable intuitive appeal. These two are what I shall call (i) the phenomenological (Or sense-datum) and (ii) the causal (or physiological) accounts of perception. (Alternately, rather than as explicit theories, these might be seen as primitive dispositions toward the criteria one chooses to employ in framing or judging a theory of perception.)
(i) The phenomenological or sense-datum account of perception, which Whitehead attributes to Hume, claims to be nothing more than a description of what is given in immediate experience (it being taken for granted that the recognition of "immediate experience" and what is "given" in it present no difficulties). It is not at all concerned to explain how these experiences were caused. Indeed, since Hume and his followers can find no causes in immediate experience, they hold that a search for the causes of experience is in principle mistaken. This is because they believe that what is given in immediate experience is absolutely certain, and any mode of "knowledge" which departs at all from such immediacy (for instance, an inference from experience) is to some degree doubtful. All that can be found in experience, according to (i), are geometrical areas and the sense qualities which inhere in them. Thus a legitimate account of perception can only describe what is perceived, and all that is clearly and distinctly perceived are such areas and sense qualities. Such perceptions arise, therefore, from "unknown causes.
(ii) The causal account of perception, in contrast to (i), purports to be not just a description but an explanation of what is perceived. The exact content of the explanation depends on which sciences are taken as supplying the most relevant information about perception. Usually those are physics and physiology. The standard explanation recounts the existence of a material object, its reflection of light waves of a certain length, the transference of these light waves from the object to the perceptual organs of the observer (in this case, the eyes), the interaction of the light waves with the rods and cones of the retina, the transference of nerve impulses along the relevant nerves to the brain, and, finally, the production of the sensation of sight. Of course this is a crude abstraction from any actual physiological explanation of perception, but from a philosophical point of view, the details of the theory are irrelevant.
What is important is that this type of theory, in whatever form, takes the experience of perception as a result of an extensive series of unperceived antecedent causes, not as something arising from "unknown causes" or perhaps from no cause at all. In this way it goes beyond what is immediately experienced in perception and in fact asserts that what is immediately experienced is relatively unimportant compared to the causal mechanism which contributed to its production. The general strategy, then, is to take one "behind the scenes" to lay bare the mechanism by which the effect was produced. According to (ii) one does not perceive objects directly; such an assumption would be naive. Instead, one perceives through and by means of a complicated medium. In ordinary, unreflective experience this medium goes unnoticed. It is the business of the causal explanation of perception to draw attention to this medium, to show how it transforms objects, and to explain how it produces the effects experienced in perception.
Now it might be thought that (i) and (ii) do not actually conflict but in fact complement one another. Thus, they might be seen not as rival theories about the same subject matter but as complementary accounts of related phenomena. (i) describes the data of perception but does not attempt to explain it. So if clarity about exactly what one perceives is one’s aim, one favors an approach such as (i). On the other hand, (ii) goes beyond what is given in direct perception to find explanations for why it is perceived. In this it does not tamper with (i)’s account of what is perceived but takes it as veritical. If one is interested primarily in an explanation of why one perceives what one does, one looks to a theory such as (ii). Different interests dictate different theories. In either case it seems that there can be no conflict because there is nothing in common to dispute about.
However, (i) and (ii) are not really as compatible as they seem. Each theory has implications which trespass on the territory of the other. For this reason, it is not possible consistently to hold both (i) and (ii) at the same time.
For instance, suppose that one accepts (i). Then several conclusions follow which are incompatible with (ii). For non-solipsistic versions of a theory such as (i), the subject is inescapably in the world in the sense that it is always tied down to a particular position within it and no other perspective is experienced with an immediacy equal to its own. Thus it follows from the principle which equates immediacy of experience with certainty of knowledge that there can be no legitimate perspective on the world by an individual subject from a point of view outside that subject, a point of view which places the subject within a larger -- "objective" -- context. This rules out causal explanations of perception such as (ii), if it be assumed that there is not immediate awareness of such causal mechanisms within an individual act of perception itself, if one does not have a perception and at the same time experience all the causes which produced it. For from the point of view of an advocate of (i), to hold (ii) would require that the subject wrench itself away from its own immediacy of experience in order to view that same experience within a larger explanatory context. According to (i), even if -- contra Hume -- one could perceive causes, one could concentrate at one time purely on what one clearly and distinctly perceives and at another time on the mechanisms which produced such perceptions, but what one cannot do is to perceive the causal mechanism which is producing the very perceptions one is currently having. There is not enough "room" in consciousness for that. One can occupy only one perspective at a time; the field on which one can focus one’s attention is limited.
Now if it were possible at the same time and in the same act of experience to be aware of both a datum of perception and of the causal mechanisms which produced it, this would still not help the advocates of (ii) in the opinion of the supporters of (i). For in that case there would still be a causal mechanism which was producing the awareness of the original perceptual datum and its cause together. That mechanism would go unnoticed in that act of perception. Perhaps another act could include it, but then a further mechanism would be required, and so on ad infinitum. In any case, the productive mechanisms would always be one step ahead of awareness. Thus there would be no awareness of this cause producing this perceived effect. So the hypothesis of a cause would always be at least somewhat removed from experience. It would be based on an inference, not directly perceived. There could be no direct assurance that this perceptual datum was produced by causes analogous to those experienced in the past, or even that it was caused at all.
Thus someone who consistently held to (i) would have to reject (ii), because either (a) there is never any direct experience of a causal chain in perception, or (b) if there is, it is not of the datum that is now being perceived. In this way, taking (i) seriously precludes accepting (ii) fully.
On the other hand, if one begins by placing primary emphasis on (ii), one is led to reject or at least to discount the claims of (i). Accepting the causal account, one can come to see the perceived datum as only the final link in a long causal chain, only that part of the iceberg which is visible above the surface of the water. With this perspective on the whole act of perception -- the true or "real" perspective, according to the advocates of (ii) -- one is apt to discount the significance of that small part of the causal nexus which is consciously experienced. If one takes (ii) seriously, then, one is led to a belief in the "mereness" of what appears. What "really" happens in perception is that light waves, reflected off physical objects, excite the rods and cones of the retina, etc., etc. We just think the world is as we perceive it; it is actually far different.
Thus (i) and (ii), as here stated, are incompatible. For an advocate of (i), (ii) is trivial to actual experience at best and incoherent at worst. An advocate of (ii) considers (i) naive and illusory, not a correct picture of "reality." One cannot legitimately hold both views together.
Following intuition and tradition Whitehead acknowledges two "modes" of perception: presentational immediacy and causal efficacy. These correspond roughly to (i) and (ii). He differs sharply from these previously discussed views, however, in the way he relates his two modes. This constitutes the uniqueness and, in view of the difficulties, the boldness of Whitehead’s theory of perception.
Presentational immediacy, as the name implies, indicates that component of consciousness which is most vividly present now. It is immediate awareness. It is "our perception of the contemporary world by means of the senses" (PR 311/ 474). Whitehead is careful to point out, however, that perception in the mode of presentational immediacy is not identical with the entire content of our present consciousness. In this he differs from (i). It is only one aspect or "part" of consciousness, though it is that part which is most clear and distinct and, hence, most noticeable. Perception in the mode of presentational immediacy is confined to an awareness of spatial areas and the sense qualities inhering in them (PR 121/ 185). Presentational immediacy discloses nothing about the past or the future: it shows us only the world experienced now. The only relations it exhibits are geometrical; no causal connections between the various spatial regions are evident. Nor is there any awareness of the causes of that immediate experience. Fortunately, Whitehead thinks, our knowledge is not limited to what is provided by presentational immediacy. This saves it from being a "barren aesthetic display" (PB 324/ 494).
Causal efficacy, Whitehead’s other pure mode of perception, makes up for these deficiencies of presentational immediacy, but what it gains in relatedness and explanatory power it loses in vividness and sharpness of definition. As its name indicates, perception in the mode of causal efficacy is supposed to give some kind of awareness of causal relationships. Exactly what kind of awareness it is supposed to give, however, is far from clear. Sometimes, especially when he is speaking pre-systematically and citing examples from ordinary experience, Whitehead seems to claim that causal efficacy merely discloses relations among the data of presentational immediacy and to indicate that "something is going on in nature and some things are affecting other things." This is Christian’s interpretation: "What is felt [in causal efficacy] is an activity going on in that situation, an activity which relates the entities . . . in the situation in a dynamic way" (IWM 147). If this were all Whitehead meant by perception in the mode of causal efficacy, he would merely be describing a commonplace experience of which we are all aware, and his doctrine would be philosophically innocuous; in fact it would not be able to sustain the weight of the philosophical edifice he builds upon it.
At other times, however, especially when he is speaking systematically and not illustrating or supporting his argument with concrete examples, Whitehead seems to make a stronger claim for causal efficacy. He seems to indicate not only that it presents a vague sense of activities and dynamic relations in nature, but that it also discloses in considerable detail the causal mechanism by which the perceptions in the mode of presentational immediacy were produced. In this sense, causal efficacy provides the basis for the physiological account of perception. In human beings, this type of causal efficacy is exhibited in the phenomenon of the "withness" of the body: according to Whitehead, we know that we touch with our hands, taste with our palates, see with our eyes, etc. In this way, he thinks, we have direct awareness of our bodies as the cause of our perception in the mode of presentational immediacy.
In either case, a perception in the mode of causal efficacy occurs in the same act of awareness as a perception in the mode of presentational immediacy, though it is distinguishable from it. We do not alternate between the two modes but experience both at the same time. In the former case (Christian’s interpretation), causal efficacy merely relates dynamically the percepta of presentational immediacy without being an additional percept of the same type. In the latter case, a vague awareness of the causes which produced a perception in the mode of presentational immediacy is present with it, though not in the same way. In this sense, causal efficacy represents an awareness of the culmination of the past routes of causation in the present act of awareness while that present act continues with undiminished vividness. In just what sense and with what legitimacy the past (qua cause) can be said to be in the present, it is my purpose to determine.
Presentational immediacy and causal efficacy complement one another; each makes up for the deficiencies of the other. Presentational immediacy presents vivid, clearly defined data, but exhibits no significant connections among them; causal efficacy provides causal relationships, but its data are vague. It follows "that what we want to know about, from the point of view either of curiosity or of technology, chiefly resides in those aspects of the world disclosed in causal efficacy: but that what we can distinctly register is chiefly to be found among the perception the mode of presentational immediacy" (PR 169/ 257). This produces something of a dilemma: for unless presentational immediacy and causal efficacy overlap in some way, unless there is a common ground between them, there is no assurance that they are giving information about the same entities and the beneficial effects of their complementary relation would be lost. Whitehead argues that in fact there is such a common ground, that the two pure modes are related by symbolic reference."
This can be achieved in two ways. (1) The two modes may share a common spatial region. This is defined sharply in presentational immediacy and vaguely in causal efficacy, but in the latter case the definition is apparently precise enough to be recognized as the same area as is presented in the first mode. (2) Or they may exhibit a common sensum (eternal object). The same sense quality which makes its appearance in presentational immediacy is recognized in the route of causal inheritance presented in causal efficacy. In this way "symbolic reference is the acceptance of the evidence of percepta, in the mode of immediacy, as evidence for the localization and discrimination of vague percepta in the mode of efficacy" (PR 179/ 272).
In ordinary experience the pure modes of perception rarely occur in isolation; they are normally "unified by a blind symbolic reference" (PR 180/ 273). "In order to find obvious examples of the pure mode of causal efficacy we must have recourse to the viscera and to memory; and to find examples of the pure mode of presentational immediacy we must have recourse to so-called ‘delusive’ perceptions" (PR 121f/ 186).
The epistemological consequences of Whitehead’s theory of symbolic reference are much more far reaching and revolutionary than his protestations of its commonplaceness would suggest. For symbolic reference clearly requires that presentational immediacy and causal efficacy be present together in a single, unitary act of awareness. Consider what this means. It means that an awareness of the cause of a perception -- in the mode of causal efficacy -- must be copresent with the experience of the effect -- in the mode of presentational immediacy. We must be aware of the causes of our sensations while we are experiencing them. As I pointed Out in section I, some traditional views regard this as impossible. It is clear, then, that in his doctrine of symbolic reference Whitehead parts company with the Cartesian epistemological tradition and attempts to mediate the dualities which have plagued it.
Thus Whitehead’s doctrine acknowledges that we perceive through a medium while denying that this amounts to a representative theory of perception. According to Whitehead’s view, the original datum of perception is transformed by the time we perceive it: one actual occasion is in another not formaliter -- as the original occasion is in itself -- but objectively. Yet the first occasion is really in the second, not its surrogate or representative. This rather implausible result is made possible, Whitehead thinks, because we perceive in the modes both of causal efficacy and of presentational immediacy. Presentational immediacy shows the datum objectively, and causal efficacy shows us how it was transformed to that state. Thus, awareness both of the effect and the causes that produced it enable Whitehead to maintain both that perception is mediated and that we perceive actualities. In this way he combines the direct realism of common sense with the causal theory of physiology. He makes room both for the vividness of immediacy and for the explanatory power of science. He acknowledges both what we see -- sense qualities -- and what we "really see" -- molecules, light waves, etc. According to Whitehead’s theory we are in effect able to experience the world from the perspective of the ideal scientific observer -- whose standpoint is outside the world -- and from our own limited point of view within the world. And we are able to do both at the same time.
I now turn to a consideration of the evidence Whitehead offers for this extraordinary and daring theory. It seems obvious that he wants -- or should want -- to place most stock in the arguments which attempt to show how an awareness of causal efficacy can be elicited from ordinary experience. For on his own principles if there were not some kind of awareness of causal efficacy, there would be no warrant for making statements about it: "Nothing is to be received into the philosophical scheme which is not discoverable as an element in subjective experience" (PR 166/253). "We can only discuss experiences which have entered into conscious analysis" (PR 179/273). "The elucidation of immediate experience is the sole justification for any thought" (PR 4/6). Unfortunately, those things of which we are most acutely conscious are not the most significant. Our perspective on the world is inverted. "Those elements of our experience which stand out clearly and distinctly in our consciousness are not its basic facts" (PR 162/ 245). "It must be remembered that clearness in consciousness is no evidence for primitiveness in the genetic process: the opposite doctrine is more nearly true" (PR 173/ 263f).
There need be no conflict between these two positions, however, if there is some original awareness, however faint, of "the basic facts of experience" from what later turns out to be the inverted perspective of consciousness. In that case, the task of the philosopher is to elicit into prominence those elements of experience whose importance was obscured by the distorted perspective of unreformed consciousness. He is to effect a re-formation of consciousness, rearranging what had always been there so as to highlight what was previously only dimly perceived. This is in fact the procedure that Whitehead seems to be following in his account of the man who knew the light caused him to blink (PR 175/265) and in his discussions of the "withness" of the body.
Whitehead counters Hume’s contention that we have no direct experiences of causes by drawing attention to this common phenomenon: "In the dark, the electric light is suddenly turned on and the man’s eyes blink" (PR 174/ 265). "The man will explain his experience by saying, ‘The flash made me blink’; and if his statement be doubted, he will reply, ‘I know it, because I felt it’" (PR 175/266). In such a case, Whitehead thinks, we have direct experience that causes are operative in nature and, moreover, that they are operating on us. Because of this, our experience is not limited to a mere succession of data which have no causal connections with one another. Christian has argued that this experience of causality is not just another percept in the same way that the flash and the blink are percepts; instead it is an awareness of a dynamic relationship between the flash and the blink (see section II above). Also, Christian thinks Whitehead is merely pointing out that experience is richer than Hume said it was, and he does not intend this example to prove that there actually is anything like a causal nexus operative in nature and that we have direct awareness of its functioning. I believe this particular example really does not show anything more than Christian claims it does. Whitehead probably thought it proved more. This is a moot point, however, since Whitehead provides other, more convincing examples which he definitely does believe give a direct awareness of the operations of causal mechanisms.
These experiences are of what he calls the "withness" of the body. We see with our eyes, hear with our ears, touch with our hands, etc. We know, says Whitehead, that these organs are the means by which our various sensations are produced. We know that we experience the world through the medium of our bodies. We know this, not because we have all been exposed to physiology, but because we have direct experience of it. "For instance, we see the contemporary chair, but we see it with our eyes. . . . Thus colors objectify the chair in one way, and objectify the eyes in another way, as elements in the experience of the subject" (PR 62/ 97).
The datum transmitted from the stone becomes the touch-feeling in the hand, but it preserves the vector character of its origin from the stone. The touch-feeling in the hand with this vector-origin from the stone is transmitted to the percipient in the brain. Thus the final perception is the perception of the stone through the touch in the hand. In this perception the stone is vague and faintly relevant in comparison with the hand. But, however dim, it is there. (PR 120/183)
Thus, according to Whitehead’s theory, each perception contains within itself the vector marks of its origin: we are directly aware of the routes by which we have inherited our feelings. In this way we are conscious at the same time of our feelings and of their causes. Causal efficacy and presentational immediacy coincide in the same act of awareness: symbolic reference is achieved.
With respect to this evidence, I shall now proceed to argue (1) that the experiences Whitehead cites do not prove what they need to prove in order to substantiate his theory, (2) that we do not actually have such experiences anyway, at least not in the form Whitehead describes them, and (3) that such experiences of causes and effects together, when they involve a reference to the subject experiencing them, seem in principle to be impossible.
(1) In order to give direct evidence for the full physiological account of perception, which Whitehead clearly intends it to do, the experience of the "withness" of the body must exhibit all the links in the causal chain which are acknowledged by physiological theory. In none of the passages in which he is drawing attention to the withness of the body does Whitehead describe this phenomenon in anything approaching the detail of a physiological theory; he hits only the high points. Indeed, he admits that some points along the causal route of inheritance are vividly felt (e.g., the hand when holding the stone), while others fade to the fringes of consciousness (e.g., the nerves between the hand and the brain). Now if these intermediate links are so dimly "felt" as not to be evident to consciousness at all, then the "evidence" of this experience clearly does not prove enough to establish the full physiological account of perception. It would provide evidence only for a direct stone-hand-consciousness sequence. The intermediate links are certainly not felt, though they may be inferred.
(2) Now what about the "parts" of the causal chain of which we are vividly aware? Do we know for certain that we feel with our hands, see with our eyes, hear with our ears, etc.? We certainly think we do, but what evidence do we have for believing it? If our awareness of the efficacy of our bodily organs in producing certain sensations is not direct and indubitable, then Whitehead cannot count the experience of their efficacy in favor of his theory. That is to say, if this is the case, we must fall back on Christian’s weaker thesis that Whitehead is not proving that any real causal nexus exists and that we have direct awareness of it.
Actually, in the act of sight we do not see our eyes; they are the one thing we do not see. As Wittgenstein says, "Nothing in the visual field allows you to infer that it is seen by an eye" (TLP 5.633). The visual field stops somewhere around where our eyes are presumed to be. But the region where the visual field stops is vague: we might with equal warrant say that we see with the whole top half of our faces as that we see just with our eyes. Moreover, physiology tells us that it is the rods and the cones that are efficacious in sight, but if we were ignorant of physiology, we would have no way of knowing this. Thus, there is no direct experience of the efficacy of just that part of the eye which physiology tells us is actively productive of sight. In other words, the "knowledge" that "we see with our eyes" is learned or inferred and is not a direct experience.
In addition, the sensation of sight (or any other sensation) could be produced in a nonstandard way without our being aware of it. This is a common phenomenon with amputated limbs and toothaches. In such cases the sensory datum complete with the feeling of bodily efficacy is present. Therefore, the feeling of bodily efficacy, since it can be delusive, provides no certain knowledge about the operation of a causal nexus in nature.
(3) Moreover, direct awareness of the physiological causes of our contemporary experience seems to be impossible in principle.
Whitehead asserts that "colors objectify the chair in one way, and objectify the eyes in another way" (PR 62/ 97, emphasis mine). We do not see the eyes in the act of sight in the same way that we see the chair. We see the chair and we see it by means of the eyes. The eye is the medium through which we see the chair. In traditional language, the chair is the object of sight.
Despite his admission that we do not perceive the medium of perception in the same way that we perceive the object of perception Whitehead’s theory still depends on a merging of these two modes. We must somehow be directly aware of the causal mechanism of sight, i.e., this causal mechanism must be objectified for us, to escape from the "barren aesthetic display" of presentational immediacy. But while one is concentrating on the chair, the feeling of the efficacy of the eye fades from consciousness; and while one is concentrating on the feeling of the eye, the chair fades away. We can be conscious of objects and conscious of ourselves, but not with equal vividness at the same time.
But even if one could so crowd one’s consciousness as to find room both for the object and the objectified medium of perception at the same time, a further difficulty would remain. Perception for Whitehead is mediated: we do not perceive objects directly but only through a medium. This means that we do not perceive things as they are in themselves but only modified versions of them: "objectification relegates into irrelevance, or into a subordinate relevance, the full constitution of the objectified entity" (PR 62/ 97).
Yet Whitehead still opts for realism, because he thinks we have direct awareness of how these objectified entities have been changed by the time we perceive them. This is an idle assumption, however, unless we really do have some direct (unmediated) perception of the original datum of objectification, so that we can tell what it is that has been changed and that it has been changed. Otherwise, we will take modified objects for things as they are in themselves. But this violates Whitehead’s insistence on mediated perception: one actual occasion never experiences the full immediacy of another actual occasion; it feels it only objectively. B feels only a selection of A. But with no direct awareness of A as it is fully in itself, B has no legitimate way to tell that it is sampling only a selection of A and not A in itself. Once again, we see that Whitehead’s attempt to maintain two perspectives at once does not succeed, because he needs B to feel A as A is in itself (subjectively) and as A is for B (objectively). But if B can feel A as A is in itself, there is no need for B to feel only a selection of A; and if B can feel only a selection of A, then it is not possible for B to feel A as A is fully in itself. Direct realism renders mediated perception unnecessary, and mediated perception makes realism impossible.
Thus Whitehead’s theory of perception is unable to reconcile its opposing tendencies of realism and mediatism. Whitehead does not provide sufficient evidence for the unification of his two pure modes of perception, and his theory fails to overcome the traditional difficulties which prevent the consistent unification of theories (i) and (ii). The causal explanation of perception and the description of its immediate data are both legitimate and useful activities, each providing an essential perspective which is overlooked by the other. The mistake is to think these two perspectives can be enjoyed at the same time.
IWM -- Christian, William A. An Interpretation of Whitehead’s Metaphysics. New Haven: Yale University Press, 1959.
TLP -- Wittgenstein, Ludwig. Tractatus Logico-Philosophicus. Translated by D. F. Pears and B. F. McGuinnes. London: Routledge and Kegan Paul, 1961.