D.S. Clarke, Jr. is Professor of Philosophy at Southern Illinois University, Carbondale, Illinois 62901. He is author of Deductive Logic, Practical Inferences and Principles of Semiotic
The following article appeared in Process Studies, pp. 26-34, Vol. 16, Number 1, Spring, 1987. 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.
The author outlines the two principle reasons why Whitehead is neglected in secular philosophical discussions.
As developed through the writings of Charles Hartshorne, Whitehead’s process conception of God has had an important influence on natural theology as taught in the seminaries. But in the secular philosophical discussion dominated by analytical thought in the English-speaking world he has suffered the worst fate that can befall a philosopher. This is not to be the object of criticism, for all welcome the attention and notoriety it brings and the realization that it usually leads to improved solutions to the problems faced by the community of philosophers. Whitehead’s fate has instead been largely silence and neglect, broken by occasional quotations of his eloquent commentaries on the history and development of civilization. In the sections that follow I want to state briefly what I think are the two principal reasons for this neglect, and then to indicate what I believe to be a promising way to revive the inspiring project that Whitehead undertook in Process and Reality.
I. Analytic Critiques
The first of these reasons is a very obvious one. Whitehead’s metaphysical system can be seen as a successor to the systems constructed by Leibniz and Spinoza and then later by Hegel, Bradley, and Bosanquet in the nineteenth century. At the time Whitehead was writing Process and Reality idealist systems were under attack on methodological grounds, first by C. E. Moore as violating the prescriptions of common sense, then by the school of logical positivism represented chiefly by Schlick, Carnap, and Ayer. As is well known, the basis for the positivists’ attack was the so-called “verifiability criterion of meaningfulness,” the requirement that for a sentence to be meaningful it must be either true by definition of its constituent terms, i.e., analytic, or be a synthetic sentence which is empirically testable. Since the sentences of Whitehead’s metaphysical system violated this criterion, they and the system as a whole were dismissed as meaningless, suitable perhaps for expressing feelings of awe and reverence towards the whole of which we are a part, but not for conveying what is true or false and what can be accepted or rejected on rational, philosophic grounds. Hence, his system could never be the topic of fruitful discussion or development.
Though this positivist critique maybe a substantial cause of the neglect of Whitehead’s philosophy in the English-speaking world, it certainly does not justify this neglect. As has been pointed out many times, to consistently apply the positivists’ verifiability criterion would have the effect of dismissing all of philosophy as meaningless, including epistemology and ethics. Indeed, no sentence in the writings of the positivists seems to satisfy their criterion. Certainly their conclusion ‘Every meaningful sentence is either analytic or empirical’ is not true by definition, for if it were, it would be a trivial definition of the term ‘sentence’. But neither is it empirically testable, for no facts of experience seem to bear on its truth or falsity. Moreover, the analytic-synthetic distinction has been effectively challenged by Quine In recent years.1 Stripped of the possibility of assigning a given sentence in either category as the result of Quine’s critique, there remains little basis for the positivists’ criticisms.
The second reason for the neglect of Whitehead is, I think, much more justifiable. But it is also more complex. To understand it requires first an attempt at stating the central goal of the metaphysical system formulated in Process and Reality. I agree with Hartshorne that this goal was to formulate a philosophy of panpsychism by generalizing fundamental features of human experiencing. The philosophy of organism of Process and Reality is an ambitious attempt to extend descriptions of human experiencing that we give with such terms as ‘sensation’, ‘perception’, ‘sensory image’, and ‘judgment’ to the experiencing of subhuman organisms. Whitehead’s most frequent historical references are to Locke, Hume, and Kant. He agrees with all three that philosophy’s first task is to describe the operations of the human mind based on direct introspection of these operations and the ideas that are their contents. His basic disagreement is only with the phenomenalism that entered philosophy with Berkeley and Hume, with his corrections taking the form of reasserting the realism of Descartes and Locke through his conception of “causal efficacy.” The psychological terminology employed by these and other philosophers of the modern tradition following Descartes can be applied to our own experiencing, Whitehead believes, and then analogically extended to subhuman forms of experiencing of which we can never be directly aware. Whitehead thinks that this extension can only be made by replacing established terminology by an entirely new terminology designed to avoid what he regarded as the old’s misleading associations. Thus, in place of ‘feeling’ be uses ‘prehension’, in place of ‘judgment is concrescence’, and in place of ‘perceiving a sensory image’ is ‘perception in the mode of presentational immediacy’. This new terminology, he thought, could achieve through the definitions he stipulated for it a generality impossible for the terminology inherited from the philosophic tradition.
The basic framework for the philosophy of organism is provided, then, by the introspective psychological tradition of modern philosophy, with an artificial terminology introduced as the vehicle for generalizing to subhuman forms of experiencing. But starting with Peirce and Frege in the nineteenth century and continuing with Wittgenstein, Ryle, Austin, and a host of others in the twentieth, the fundamental assumptions of this framework came under consistent and, I think, effective attack. We can describe with some accuracy our sensations and feelings, but our thought processes — our judgments, decisions, suppositions, emotions, etc. — seem incapable of direct description.
We can, however, describe the forms of the sentences we use in expressing these judgments, decisions, suppositions, and emotions, and attempt to describe how these sentences are used in communication situations and how their constituent subjects and predicates function. Indeed, human thinking just is the interpretation and use of language, and no adequate account of our mental operations can ignore their linguistic base. The lesson taught by the linguistic philosophers is that the more we become self-conscious of the language we use the more we realize how linguistic distinctions have all along been imported into what we believed were direct descriptions of psychological processes. We are like beings forced to see the world through spectacles, and linguistic categories are the lenses that determine how this world appears to us.
II. Images as Objects
Further, many of the foundations of post-Cartesian philosophy begin to crumble when we examine the linguistic bases on which they were constructed. One of these foundations is the view that when we see a table, hear a trumpet, taste sugar, etc., what we really see, hear, or taste are not ordinary objects, but instead our own ideas or images — the colored shape directly apparent to us, the loud blare, the sugary taste. These so-called appearances or “immediate ideas” are said to be objects of a kind of direct, error-free perception. The so-called “external objects” — the table, the trumpet, the sugar — are not themselves perceived. They are instead the “things in themselves” or “noumena” behind the appearances. We can be certain of how these things appear to us, the fact that we are directly aware of a brownish rectangular shape, a high pitched blare, a sugary taste.
But we can be mistaken as to whether the objects are as they appear to us. For post-Cartesian philosophy it is even uncertain whether there are existing physical objects to which the appearances correspond, with idealists divided from realists over the issue. The dominant view in modern philosophy was the version of realism espoused by Descartes and Locke, a version which took the form of a “representative theory perception.” According to this theory, what we perceive are particular ideas or images. Causing these “subjective” or “internal” images are “external” objects. What we call the perception of these latter objects is in fact an inference we make to them from images as their representations.
Criticisms of this view of perception have been given by a number of philosophers in the past 30 years and have taken a variety of forms.2 The chief criticism is that sensory images can be seen to be fictitious objects once we look carefully at the language we use to describe them. Such descriptions do seem liable to error, as when we describe a certain color by the word mauve’ and later correct ourselves as having misdescribed it. Further, the language used in these descriptions seems to be dependent on language used to directly describe ordinary things. To describe a taste as salty, for example, is to say that the taste is like that experienced when we taste salt. The conclusion reached is that the modern philosophical tradition was mistaken in postulating sensory images as objects of perception. Instead, what we should say is that we perceive ordinary things like tables and chairs by having these images. The images are not “objects” at, all, but simply means by which we perceive objects.
The extent to which Whitehead is himself vulnerable to these same criticisms is open to debate. His choice of a novel terminology makes it often difficult to determine the extent to which he shares views of past philosophers. He does use the terms ‘sensum’, ‘sense presentation’, and ‘image’ to stand for the immediate objects of awareness, the ideas or appearances of the tradition.3 These sensa are the direct object of what he terms “perception in the mode of presentational immediacy.” His discussion of “delusive perceptions seems to indicate that he regarded the direct objects of perception in this mode to be sensa or images, with error introduced when we make an incorrect inference to the nature of the material things causally related to it. When we look at a chair, he says, delusive perception results from an incorrect inference from a certain geometrical shape directly apparent to us. A non-delusive case arises, in contrast, “when we see a chair-image and there is a chair” (PR 100). As we shall see in the next section, Whitehead’s principal version of symbolic reference becomes his means of relating images or sensa as directly perceived to the objects which cause them.
Whitehead’s use of assumptions dating back to Descartes and Locke in his account of perception leaves him vulnerable to the criticisms introduced by the revolution in philosophic method taking place at the time he was writing his major works, one in which the analysis of the functioning of language was replacing psychological introspection as the principal method for understanding human thought. The fact that many of the assumptions fundamental to Whitehead’s starting point in human experience were thrown into question by those undertaking this revolution is the main reason, I think, for the subsequent neglect of his philosophy in the English-speaking world.
But the effect has been to leave our philosophy impoverished, for Whitehead’s project was an important correction to the anthropocentric bias characteristic of post-Cartesian philosophy. The ancient Greeks saw all of nature as alive and themselves as but the most advanced stage in what Lovejoy was later to term the “great chain of being.” The rationalization of this conception came with Aristotle’s Metaphysics and De Anima. The effect of the new method of philosophy instituted by Descartes and Locke was to make human mental operations and their contents the sole object of study, with the rest of nature the “unknown something” behind the veil of appearances. Whitehead’s philosophy of organism is an attempt to restate the Creek conception by extending features of human experiencing to subhuman forms. But while it may be possible to extend certain psychological capacities to lower organisms, it does not seem possible to extend to them our capacity to use language. We humans seem uniquely endowed with this capacity, with only very primitive anticipations among some of the higher primates. By making language use its central object of study contemporary philosophy seems to have committed itself to an even more extreme form of that same anthropocentric orientation that Whitehead saw himself as combating.
III. Two Versions of Symbolic Reference
There is, however, a possible way of overcoming this difficulty if certain very basic features of language can be found also in what is interpreted at lower stages of organic life. I will use the generic term ‘sign’ from the semiotic tradition revived by Charles Peirce at the end of the nineteenth century to stand for any object of interpretation. For this tradition a linguistic sentence is regarded as a sign of a relatively high level of complexity sharing certain common features with more primitive signs interpreted by lower forms of life. The problem of restating Whitehead’s panpsychism becomes then one of isolating such common features as meaning and reference found in the use of simple sentences that can be analogically extended to these primitive signs.
Whitehead’s account of the functions of what he terms “symbols in Symbolism and Process and Reality offers a suggestive attempt at such an extension.4 In the first of these works two basic forms of symbols are distinguished: linguistic expressions in the form of perceived words and sentences and our sense presentations correlated to natural objects. Whitehead seems to give two versions of the functioning of sense presentations as primitive types of symbols. In one version sense presentations such as a colored shape are symbols of other elements of our experience, as the visual appearance of a flash of lightning may be a symbol of the sound of thunder with which it has been correlated in our past experience. We thus see the lightning flash and expect the thunder, and such primitive types of interpretation of natural events seem to be present in any organism capable of learning from experience. By being able to anticipate these associated events organisms can behave in ways designed to secure a benefit or avoid a harm. Thus, for the deer in the forest an odor may be a symbol of the sight of a predator; to interpret this symbol is both to anticipate the sight and to flee to avoid harm. For Whitehead says, “coloured shapes seem to be symbols for some other elements in our experience, and when we see the coloured shapes we adjust our actions towards those other elements” (5 4). These “other elements,” Whitehead says in Symbolism, constitute the “meaning” of the symbol, and the transition from symbol to meaning is called here “symbolic reference.”
The human mind is functioning symbolically when some components of its experience elicit consciousness, beliefs, emotions, and usages, respecting other components of its experience. The former set of components are the ‘symbols,’ and the latter set constitute the ‘meaning’ of the symbols. The organic functioning whereby there is a transition from the symbol to the meaning will be called ‘symbolic reference.’ (5 8)
This version of symbolic functioning allows a comparison between the interpretation of a sense presentation such as the sight of lightning and linguistic expression such as the word ‘tree’. Just as the flash that appears has as its meaning the thunder with which it is associated, so the word ‘tree’ has as its meaning the visual, olfactory, and tactual experiences of trees.
Whitehead’s second version of symbolic functioning is very different from the first. In this version the sense presentation as symbol is said to stand for the physical object with which it is causally correlated, not to another experience. “Symbolism from sense-perception to physical bodies,” Whitehead says, “is the most natural and widespread of all symbolic modes” (5 4). In Process and Reality this differing conception of the symbol is developed in considerable detail in the form of a revised statement of “symbolic reference.” Such reference is regarded here as being the process by which sense presentations perceived in the “mode of presentational immediacy” are projected on to regions of space external to the percipient. These regions are those from which the causal chains terminating in the sense presentations or sensa were initiated. In this way there is a combination of presentational immediacy with what Whitehead terms “perception in the mode of causal efficacy,” a direct perception of the causal relation between the sense presentation and the object which it stands for. Rasvihary Daz gives the example of seeing a friend approaching. What is immediately apparent is some shape and color. For Whitehead, Daz correctly points out,
the coloured shape is used as the symbol of the actual person, and the perception of it blends itself with that of the actual being given in causal efficacy. . . . This is how symbolic reference works. We project the sensum on to the physical nexus causally felt, and take it (the sensum) as the representative, in clear consciousness, of what is vaguely, but deeply, felt in causal.5
In this version the “meaning” of the symbol as sensum would seem to be the object causally related to it.6
In the first version of symbolic functioning a kind of “error” seems to arise when expectations are disappointed, for example, when the sound of thunder fails to follow the flash of lightning. But for this second version of symbolic reference error occurs when the object being symbolized either fails to exist, as for a mirage, or exists in a much different way than is being represented, as for the straight stick immersed in water which looks bent, the red book which looks yellow to the person suffering from jaundice, etc. Errors are thus due to what Whitehead terms “delusive appearances” of the kind discussed in the previous section where a sensum fails to match what it symbolizes or represents.7
The contrast between these versions of the functioning of primitive signs can be found in the British empiricist tradition. The first phenomenalist version was clearly stated by Berkeley. Just as linguistic expressions lead us to anticipate certain experiences, so for Berkeley certain experiences or immediate ideas are “natural signs” of others, as the sound of the coach is a sign of the sight of the approaching coach.8 The second realist version can be found in the writings of Thomas Reid, for whom sensations are signs of external objects. For Reid it is by a “natural kind of magic” that we take them to stand for these objects; there are no grounds in experience for making this association.9 As we have seen, Whitehead disagrees with Reid in that he holds that we have a direct intuition or “feeling” of external objects as causes of sensations. Nevertheless, his second version of signs shares some central features with Reid’s, and can be regarded as its successor.
IV Analogies Between Natural Signs and Language
To which of these alternative versions of natural signs can features of language be most adequately extended? A complete answer would be very complicated,10 but in general terms we can see that Whitehead’s second version is inadequate for the task. A sense presentation or sensum is for Whitehead a causal effect of a sequence of events correlated with the physical object that it stands for. But what meaningful analogy can hold between such a causal effect and a linguistic expression such as the word ‘tree’ or the sentence ‘The stove is hot’? Utterances of these expressions are clearly not causal effects of the objects or the states of affairs they represent. It is not the tree which causes an utterance of ‘tree’, nor the hot stove which causes a person to say ‘The stove is hot’. Instead, utterances of these expressions lead us to anticipate future events and possibly as a consequence orient our actions. If someone says ‘Tree’ and points, I anticipate at the location to which he points that I will see a tree. If someone says ‘The stove is hot’, I believe that the stove he is using the subject ‘the stove’ to refer to will be hot if I touch it. To avoid the painful touch, I may remove my hand. These obvious features of the use of language are quite consistent with the first version of natural signs (or symbols) that Whitehead presents, that for which a sense presentation has as its meaning other experiences and leads to our altering our actions.
There is also at least a general analogy between judgments of falsity directed towards utterances and disappointment of expectations for natural signs. To judge ‘This stove is hot’ false is first to identify the referent of the subject and then recognize that the predicate fails to be instantiated. Similarly, when the thunder fails to follow the lightning, there is recognition of a nonoccurrence of what was expected. But so far as I can see, these features are inconsistent with the second version requiring the sense presentation to be the causal effect of what it signifies. The cause of a sense presentation as sign lies in the past, while in all the examples just cited the anticipation is of a future event, and it is in the future that realization of error takes place.
Nevertheless, Whitehead is understandably reluctant to endorse the phenomenalist implications of his first version, since it seems to create a schism between the philosophic account of sign interpretation given in terms of correlations between experiences and the world as described by physics and the other sciences. His introduction of causal efficacy as a mode of perception is designed to overcome this schism, what he refers to as the “bifurcation of nature” introduced into philosophy by the dualism of Descartes and Locke. It is not necessary, however, to employ the sign-signified distinction in order to explain the basic facts of perception. What I see when I look up in the sky is the flash of lightning at a certain location. As the linguistic philosophers have pointed out (cf. Section II), I do not see a yellowish sensum and then interpret this as standing for some physical event causing the sensum, for having the yellowish sensum is but an aspect of seeing the lightning. The perception is direct and immediate and itself involves nothing analogous to error. To interpret what I see as a sign is to expect to hear thunder at a time and place directly indicated by the lightning.
Further, what is correlated with the sign is not simply another sense presentation, but a locatable sound. It is at a given spatial-temporal location that I recognize whether or not what I have expected in fact occurs. The description of the perception of both the lightning and the following thunder does not therefore have to be given in phenomenalist terms, nor does abandoning the phenomenalist description require Whitehead’s second version of symbolic reference as the projection of a sensum on a spatial region to which it is causally related. The reference of a natural sign would seem instead to be to that spatial and temporal region at which the signified event is expected, for example, the region at which the thunder is expected after seeing the lightning.
We can find, therefore, very general similarities between the use and interpretation of sentences at the linguistic level and primitive natural signs as characterized in Whitehead’s first version of symbolic reference. These similarities are sufficient to allow the analogical extension of features of language to signs interpreted by subhuman forms of life, certainly those “higher grade of organisms, as Whitehead terms them, capable of learning. The most primitive of these would seem to be the amoeba learning to respond differentially to events in its environment. But between these primitive forms of sign interpretation and the interpretation and use of language there are many intermediate levels. There are, for example, signals used in animal communication, nonconventional signals which include some gestures, and single-word sentences such as the word ‘Tree’ used by children in the early stages of language acquisition. Only as a relatively recent evolutionary development do we have developed natural languages with their sentences and combinations of sentences as forms of discourse. Whitehead has little to say of these intermediate levels, though he was certainly well aware of them and notes the use of signals both to convey information and to express emotions. “Speech in its embryonic stages as exemplified in animal and human behavior,” he says in Modes of Thought, “varies between emotional expression and signaling” (MT 52). Clearly, any complete extension of features of language must include these intermediate signs along with the natural signs that have been our focus.
I noted above the existence of semiotic as the discipline which attempts to isolate general features of signs. I believe the most promising way of restating Whitehead’s project of generalizing beyond human experiencing is in terms of the principles outlined in this discipline. In this way the sound reasons for contemporary philosophy’s attention to language use can be acknowledged. At the same time recent philosophy’s anthropocentric bias can at least partially be overcome. The generality and scope of Whitehead’s philosophy continues to be an inspiration in the contemporary world. Despite its neglect his vision of a metaphysical system as comprehensive as Aristotle’s, one that embraces all forms of life, has enduring appeal. But at the same time it seems to me that we cannot turn our backs on advances undertaken by philosophy in the past 50 years, and that an attempt must be made to apply the methodology dominating contemporary philosophy to Whitehead’s central aim. I have tried to outline in this paper what I think is the proper direction this attempt should take and those aspects of Whitehead’s account of symbolic reference which can serve as a promising starting point.
1For the original statement of this criticism see W. V. O. Quine, “‘Two Dogmas of Empiricism” in From a Logical Point of View, 2nd edition (Cambridge: Harvard University Press, 1961).
2The most influential criticisms are those by J. L. Austin in Sense and Sensibility, (London: Oxford University Press, 1962). See also Gilbert Ryle, The Concept of Mind (London: Hutchinson. 1949), Chapter VI.
3See PR 98, where Whitehead uses the term ‘chair-image’ to stand for the effect of a sequence of physical causes. In contrast, a ‘sense datum’ is used, following the early Russell, in the sense of a universal or “eternal object” instantiated by a particular image.
4I follow Whitehead in using his term ‘symbol’ in a generic sense applying to both linguistic expressions and sense presentations as natural signs. In the semiotic tradition ‘symbol’ has since Aristotle been restricted to conventional signs and ‘sign’ used as the generic term.
5Rasvihary Das, The Philosophy of Whitehead (New York: Russell and Russell, 1964). p. 138. See also Donald Sherburne’s summary account of symbolic reference in A Key to Whitehead’s PROCESS AND REALITY (New York: Macmillan, 1957), p. 246f.
6Whitehead does allow in PR 274f. for the possibility of his first version of symbolic functioning: “The species from which the symbolic reference starts is called the ‘species of symbols,’ and the species with which it ends the ‘species of meanings.’ In this there can be symbolic reference between two species in the same perceptive mode.” Thus, the symbol can signify another sensum, with both perceived in the mode of presentational immediacy. But, he adds, “the chief example of symbolism . . . is that between the two perceptive modes,” where the symbol is a sensum and the meaning the causally related object.
7See S 54f. This interpretation of Whitehead’s conception of error can be found in Nathaniel Lawrence, Whitehead’s Philosophical Development (Berkeley: University of California Press, 1956), p. 343.
8See Three Dialogues between Hylas and Philonous, I.
9Thomas, Reid, An Inquiry into the Human Mind, Pt. II, Ch. 5. Sec. III.
10For a detailed analysis of signs at different levels see my Principles of Semiotic (Boston: Routledge and Kegan Paul, forthcoming, 1987).