L. Bryant Keeling (Ph.D., University of Chicago) is Associate Professor of Philosophy at Western Illinois University, Macomb, Illinois 61455.
The following article appeared in Process Studies, pp. 51-66, Vol. 6, Number 1, Spring, 1976. 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.
Metaphysical principles have been attacked in recent years by analytic philosophers, but Hartshorne maintains that the analytic philosophers have succeeded only in tearing down antiquated metaphysical castles.
Charles Hartshorne shares with most other metaphysically oriented philosophers a strong drive toward the general. Many people seem to have this drive to some degree. They will seek to understand some object or phenomenon by finding out the general class of things to which it belongs. If they are curious about what monotremes are or what causes rainbows, they may look into the matter. When they find out that monotremes are egg-laying mammals and that rainbows are instances of light refraction, they are usually satisfied. Of course, questions may arise about what mammals are or about the principles of light refraction, and answers might be sought in terms of still more general principles of biology or physics. But usually in a short time this process comes to an end. The person accepts a contingent principle just because the data support it or some authority in the field assures him it is true, and not because it is an instance of a still more general principle. Metaphysically oriented philosophers are not usually inclined to stop at such a point. They want to find ultimate principles and categories which are general in an absolute sense. Such principles and categories would apply to and explain the whole of reality, not merely some segment of it. As Hartshorne puts it: "The philosopher is seeking principles so general, so basic, that they are no longer special cases to be explained by the more general principles, but are themselves the most general of ideas, true not only of the actual world but of any conceivable one" (RSP 29). Such principles are called metaphysical principles. They are more general than scientific principles, but they should be understood as being on a continuum with scientific principles.1 Like scientific principles they are intended to give us literal and accurate information about our universe.
This approach has been widely attacked in recent years by analytic philosophers who claim that ultimate principles of the kind set forth by metaphysically oriented philosophers involve misuse of language and are therefore not merely false but nonsensical. This attack, however, has not succeeded in driving the enemy from the field. Charles Hartshorne has maintained that the analytic philosophers have succeeded only in tearing down antiquated metaphysical castles. Their attack, he claims, is effective only against the principles and categories of classical metaphysics on which it is focused and has no force against neoclassical metaphysics.2 It certainly does seem that Hartshorne knows more about analytic philosophy than analytical philosophers know about neoclassical metaphysics! Though my inclination now is toward analytic philosophy, I must admit that Hartshorne’s point has some cogency. His efforts to bring neoclassical metaphysics into dialogue with analytic philosophy have been largely ignored by analytic philosophers, regrettably, I believe, since much could be learned by such a dialogue. I would like to encourage such a dialogue by focusing on one aspect of Hartshorne’s metaphysics from the perspective of analytic philosophy.
My procedure will be to examine the category of feeling, first by sketching briefly Hartshorne’s position that feeling is a metaphysical category. Second, I will attempt to lay bare some of the linguistic assumptions of this position. Third, I will provide a criticism of this position from the perspective of analytic philosophy. Fourth, I will make some suggestions about what might account for the interest in making feeling an ultimate category. Hartshorne’s philosophy is complex, rich, and subtle, and if more than fifteen years of interest in his philosophy have taught me nothing else, they have taught me that effective argument against his position is never easy. So I do not think of what I have to say here as a definitive comment on the subject, but rather as a contribution to a dialogue that Hartshorne himself has sought.
Charles Hartshorne is one of the most vigorous contemporary proponents of panpsychism, the position that every genuine individual has mental attributes. This position perhaps may be developed best by first noting that the entities which make up the universe seem to fit into a hierarchy. As Hartshorne puts it: "Thus it is a reasonable view that all things, so far as they are individuals rather than aggregates, fall upon a single scale . . . running from the least particle of inorganic matter to the great universe itself" (BH 112). We may then attempt to determine just what the variables of this scale are. Some of the variables (e.g., ability to see) will be local variables which apply within a certain range of the scale, but not to the whole scale. The scale itself, however, must be defined by variables whose ranges are coextensive with the whole scale. Obviously such variables, which Hartshorne calls cosmic variables or metaphysical categories, must characterize human beings since human beings are on the scale. It is therefore reasonable to proceed by noting some of the most general characteristics of human beings and examining them to see if they can be generalized to lower and higher individuals on the scale. Some general characteristics of human beings are that they are social animals, they have minds, they feel, and they make decisions. According to Hartshorne, all of the characteristics I have just mentioned are cosmic variables. He maintains that all real entities, from the smallest subatomic particle to the universe, exhibit social characteristics such as feeling and freedom (e.g., LP 191-233; RSP 29-43; BH 111-23). The category of the social is too broad and complex to permit its examination here. I will focus, therefore, on the category of feeling, which is a necessary element of the social.
To understand feeling as a metaphysical category we must begin at the human level. At this level, feeling can be understood in terms of our own direct experience. We all have known joy, satisfaction, suffering, pain, etc., and therefore we are directly acquainted with feeling at the human level. When we use the term ‘feeling’ as a metaphysical category, however, we must take care not to fall into gross anthropomorphisms. Each level in the scale of entities will exemplify this category in its own way. Hartshorne spells this out very clearly in the following passage:
to say that a creature feels is not to say how it feels; and we have good reason to believe that some creatures feel very differently from human beings. The greater this difference, the harder it will be for us to understand the feeling or to see clearly that it is there. This difficulty is serious with the higher animals, more so with the lower, more so with plant cells, molecules, or atoms -- supposing, for the sake of argument, that these all feel. It is plain enough that atomic feelings must be vastly different from dog feelings, since the measurement of this difference must be the gulf between atom structure and dog structure, between atomic activity and canine activity. (BH 168)
The word ‘feeling’ refers to a certain internal state of affairs whether it is used in reference to human beings or butterflies. But we must remember that even in human beings, feelings vary greatly in terms of intensity and definiteness (BH 117f). These variations are far from exhausted in the known range of human or even animal experience. The feelings of an atom may be of very low intensity and may be very vague compared to the feelings of a rabbit, but there is a continuity that justifies our using the same term in both cases.
Hartshorne argues for this position in numerous places in his writings. Two of the arguments which he uses may throw further light on feeling as a metaphysical category.
(1) All terms which are used to describe reality must initially be drawn from human experience because this is the only sort of experience to which we have direct access (BH 121). If we attempt to describe the passive role of some entity, we will find that we have no categories to use except human categories which in their natural use are closely related to feeling. Scientists speak, for example, of atoms ‘responding’ or ‘being in a satisfied state’. Before we dismiss such expressions as mere metaphors we should consider whether we have an alternative literal" description available which is devoid of implications of feeling. It is Harts-home’s contention that we must describe such entities as atoms in language which implies feeling if we are to describe them at all. If this is the case, what sense does it make to say that they do not actually feel? All of their observable features are compatible with their having (perhaps to a very low degree) feelings. No possible observation could show them to be wholly devoid of feelings because there is no way to "get inside" them and experience the absence of all feeling.3 Since we cannot avoid using categories that imply feelings in our descriptions of the passive role of entities and since we have no conceivable way of knowing entities to be devoid of feelings, there is nothing to be gained by denying that all entities have feelings (RSP 33; of Fl 349). Such a category, according to Hartshorne, is for all purposes an ultimate category.
(2) The second argument is based on the premise that it is impossible to draw a clear line between living and nonliving matter. We know that human beings feel, and we generally believe that other animals feel. A worm or an ameba will respond to a stimulus in a way that strongly suggests the presence of very primitive feeling. And, as Hartshorne puts It: "From man to molecules and atoms we have a series of modes of organization; at no point can one say, below this there could be no experience" (CSPM 6). Contemporary scientists have a difficult time drawing a line between living and nonliving matter. Stephen Toulmin in Foresight and Understanding, for example, points out that "any arguments which justify biochemists in speaking of genes as ‘molecules of extreme complexity’ justify us also speaking of atoms and molecules as ‘organisms of extreme simplicity’" (FU 78). Seen in this light, it may not seem so farfetched to suggest that an atom’s ‘response’ to its environment is really a kind of feeling.
For those who are not familiar with the basic tenets of panpsychism, two objections are likely to arise at this point. (1) It might be objected that it is absurd to speak of books, coffee cups, or planets as having feelings. Hartshorne, of course, agrees that this would be absurd, but denies that his theory implies it. The objects mentioned are not really individuals, they are collections of individuals. A stone, he points out, lacks sufficient unification to be an individual which might be a member of a society, but it may itself be a society (RSP 35f). His position on this point is stated succinctly in the following passage:
Organic monism (panpsychism), in the sense just indicated, includes within itself a limited or relative dualism. The assertion is not that all wholes are purposive or organic; but that, first, all well-unified wholes are organic, and second, that all wholes whatever both involve and are involved in organic wholes. But what wholes are well-unified? My suggestion is that any whole which has less unity than its most unified parts is not an organism in the present sense here in question; though, according to organic monism, its most unified parts, and some unified whole of which it is itself a part, must in all cases be organisms. (LP 192)
(2) It might also be objected that since in human beings and higher animals the ability to feel is dependent upon the functioning of a nervous system, those entities (like atoms) which have no nervous systems would obviously be unable to feel. Hartshorne’s answer is that this analogy does not apply. He writes:
A nervous system is a specialized organ of feeling and volition; as muscles are specialized organs of movement. But as animals without muscles can nevertheless move, so those without nerves may feel, and may move in accordance with those feelings. (RSP 34; cf. BH 201f)
Panpsychism may seem initially to be a position which is sharply at odds with common sense. Upon more careful examination, however, this is by no means as obvious as it might have seemed. The issue is essentially a logical one and purported counterexamples drawn from observations generally have little force against it.
Hartshorne’s contention that ‘feeling’ is a metaphysical category is not a mere facade on his metaphysical structure, but an integral part of the structure itself. His metaphysics is unified in his concept of God as "the all-sensitive mind of the world-body" (BH 208). If, as Hartshorne maintains, the universe is made up of sentient, free individuals, there must be some dominant individual in the universe as a whole or there would be total chaos. This dominant individual, God, is not a tyrannical individual ruling the universe from the outside. The universe itself is a single coherent organism with God as its dominant member (MVG 70). All of the metaphysical categories, including feeling, find their primary exemplification in God, God, as Perfect Being, includes all reality within himself. He feels the feelings of every individual in the universe and preserves forever the values inherent in these experiences (RSP 119). It is therefore very important to Hartshorne’s position that even so complex an object as the "world-whole" can be said to have feelings (BH 118f).
Before beginning a critical analysis of Hartshorne’s attempt to generalize feeling and establish it as a metaphysical category, we must get a fuller understanding of his view of metaphysics. Just what are metaphysical categories, and how are they developed? Hartshorne writes that "metaphysics fries to express what all possibilities of existence have in common" (CSPM 162f). This presupposes that there are qualities which all individuals have in common. Metaphysical categories designate those qualities. Metaphysical statements make use of metaphysical categories in giving general explanations of reality. Such statements purport to tell us something about the world, yet they cannot be falsified by any conceivable state of affairs. They state necessary truths and are grounded in the meanings of their terms. Hartshorne writes:
Truths can be necessary only if their denial is absurd, and this can only mean if insight into the meaning of the denial suffices to exhibit it as self-contradictory. In short, the assumptions of philosophy are self-evident upon careful inspection of the terms involved. These terms, like all terms, refer to experience, for there is nothing else for them to refer to. (MVG 70)
Perhaps Hartshorne’s position can be clarified by considering first how he deals with some objections to his point of view and second how a metaphysical category is developed.
There are two possible objections which deserve special consideration because Hartshorne is aware of them and deals with them explicitly. The first is that metaphysical judgments of the kind advanced by Hartshorne do in fact exclude certain possible situations. For example, his principle that all individuals have feeling excludes the possibility of there existing an individual which does not have feeling. It seems to be dogmatic denial of a possible state of affairs. Hartshorne’s reply is: "The purpose of metaphysical theory is not necessarily to cut off what would otherwise be possible alternatives; it is perhaps rather to make us more clearly conscious of the nature of that to which, save through logical confusion or mere verbalism, there can be no alternative" (2:479). In other words, Hartshorne denies that metaphysical statements are in any real sense excluding a conceivable state of affairs. If it could be shown that a metaphysical statement did exclude some conceivable state of affairs, that would be sufficient grounds for holding that the statement it not a metaphysical one. If some category can be shown not to include a conceivable individual, that would be sufficient grounds for holding that it is not a metaphysical category. According to Hartshorne, the metaphysical category of feeling excludes no possible individual. The notion of a nonfeeling being is nonsensical. No positive characteristic which can be observed is incompatible with a being’s having feelings. To assert, therefore, that an individual has feelings is to assert something that cannot be falsified by any possible observation.
But this very characteristic of metaphysical statements raises a second objection. It might be claimed that since Hartshorne’s metaphysical statements exclude nothing positive, they cannot be genuine affirmations of anything. Hartshorne, to the contrary, contends that while it is true that metaphysical statements exclude nothing positive, it is also true that they cannot fail to be positively exemplified (LP 284). Consider the statement that something exists. Hartshorne maintains that such a statement denies no positive state of affairs and therefore is not falsifiable in any conceivable way. Yet it seems clearly to be meaningful and is exemplified by any imaginable situation (CSPM 162f). According to Hartshorne, philosophers who hold that all meaningful statements about the world must negate some conceivable state of affairs are simply begging the question. They have taken a principle which is true for empirical statements and assumed that it is true for all statements. Their principle declares meaningless those statements which might serve as counterexamples to their principle. Statements like "Something exists" violate their principle and yet do not seem to be meaningless on any other grounds than the arbitrary principle they have laid down. If the statement "Something exists" is meaningful even though it excludes no possible positive state of affairs, then perhaps the same is true for "Every individual feels." At least, Hartshorne would claim, the latter is not obvious nonsense.
Let us now see if we can get a clearer picture of what is involved in the development of a metaphysical category. Hartshorne provides the following description of this process:
Since in metaphysics we are seeking ultimate or a priori generality, beyond all contingent special cases, every concept considered as even possibly metaphysical should be freed of limitations which do not seem inherent in its meaning. For example, ‘experience’ is of course not metaphysical if taken as equivalent to ‘human experience.’ The only chance of arriving at metaphysical ultimacy from this concept is if it can be seen that the special characters imposed by ‘human’ are very special indeed. Even ‘animal experience’ could only be an empirical, not an a priori idea. We must ask if there are not dimensions or variables within our experience whose range of possible values in principle infinitely exceed the range of these values found in our, or even in animal, experience. (CSPM 90; of RSP 85)
Feeling is certainly a variable of human experience, so let us consider how feeling might be developed into a metaphysical category following the suggestions in this passage. The starting point seems to present few problems. The meaning of this word is secured, according to Hartshorne, by the experience of joy, sorrow, pain, etc., which are common to all human beings. No one can sincerely deny direct acquaintance with the referent of the word ‘feeling’. In its everyday uses the word ‘feeling’ refers for the most part to the experiences of animals having nervous systems. If this restriction in meaning is retained, then clearly the word ‘feeling’ is not completely general in reference and is not a metaphysical category.
According to Hartshorne, however, this restriction is arbitrary. No observations we have made or could possibly make could prove the absence in a cell or a molecule or even an atom of the same sort of thing that the word ‘feeling’ refers to when it is applied to human beings. The restriction on the use of the word ‘feeling’ in everyday language seems to be an accident and can be removed without generating any obvious absurdities. Its meaning has been altered in the sense that it may now refer to a broader range of reality. What could possibly be gained by restricting its range of application in such a way that some actual or possible being would fall outside its scope? If nothing could be gained by this, then the word should be completely generalized and recognized as a metaphysical category. The word ‘feeling’, thus generalized, cannot fail to have a referent regardless of the form reality may take.
One might initially think that this process admits of a reductio ad absurdum. Why cannot any word be made into a metaphysical category by this process? Consider what would happen if one attempted to use this process, for example, on the color green. The word "green" refers to a certain range of the visible spectrum. Its most primitive meaning involves the exclusion of the other colors. If it were generalized so that its range of application were the entire visible spectrum (while still retaining its basic meaning), a contradiction would result. Something could then be green and not green at the same time and in the same respect. The restrictions on the application of the word "green" in ordinary language turn out not to be arbitrary ones, but are the necessary consequence of the meaning or reference of the word.
One might also argue that in the expansion of the range of application of a word to make it into a metaphysical category the original meaning would be lost. Hartshorne denies that this is so on the basis of his general theory of meaning. I cannot give this theory the treatment it deserves at this point, but crudely stated the theory is that the meaning of a predicate word is its referent. If the meaning of a word is such that it ordinarily refers to x as an element of A and B and we expand its range of application so that it refers to xy as an element of A and B and C and D, it still retains its original meaning because it still refers to x. The only check on the process of generalization which is needed, according to Hartshorne, is the sort of contradiction which we observed developing when we attempted the metaphysical generalization of the word "green." The continued meaningfulness of the term generalized is guaranteed by the principle that the meaning of a predicate word is its referent.
I approach this section with some uneasiness. The purpose of the section is to provide a criticism of Hartshorne’s use of feeling as a metaphysical category from the perspective of analytic philosophy. But what is that perspective? Analytic philosophy is a diverse and complex movement rather than a position. It would be foolhardy indeed for me to claim to speak for analytic philosophers. On the other band, it would be dishonest to claim the basic method I use here as my own. I have been deeply influenced by the later Wittgenstein, Austin, Strawson, Ryle, Searle, and other contemporary philosophers who are sometimes referred to as "ordinary language philosophers." The basic idea to which I shall appeal has been stated (perhaps somewhat vaguely) by William P. Alston, one of the outstanding contemporary philosophers of language, in the following passage:
There is a certain conviction about linguistic meaning which is widely shared today. This conviction might be expressed as follows. Somehow the concept of the meaning of a linguistic expression is to be elucidated in terms of the use of that expression, in terms of the way it is employed by the users of the language. (1:141)
In support of this claim, Alston cites passages from the writings of Ryle, Nowell-Smith, Evans, Strawson, and Warnock. I want to try to show the implications of this view of meaning for Hartshorne’s concept of feeling as a metaphysical category. My strategy will be to argue that Hartshorne’s use of the word ‘feeling’ deviates substantially from its ordinary usage while at the same time presupposing that usage and that this makes its use as a metaphysical category impossible.
To begin with let us note that Hartshorne was correct when he stated that in ordinary language the word ‘feeling’ is applied primarily to human beings. When I try to think of natural non-philosophical uses of the word ‘feeling’, I come up with examples like: "She feels happy," "He feels sick," "I feel like playing chess," "In which tooth do you feel the pain?" "I feel sorry for her," "He feels remorse over what he did," "She feels hopeful about the outcome," etc. It seems that in most ordinary cases where the word ‘feel’ is used in connection with a human subject there is a specification of what is felt. What is felt may be a relatively simple sensation or emotion, or it may be something logically tied to social conventions, like the conventions which make it possible to call the movement of pieces on a board "playing chess."
When we extend the use of the word ‘feel’ to animals, we tend to observe this distinction. In general, people are more likely to be comfortable saying that a dog feels happy than saying that the dog feels like playing chess. Some people, to be sure, have a strong tendency toward anthropomorphisms in the case of certain favored animals like dogs, cats, and horses. They might not hesitate to say that such animals feel remorse over what they have done. But would not most people be a little uneasy with the notion that an earthworm feels remorse over what it has done? Even in the case of simple emotions and sensations we tend to make distinctions among the kinds of animals to which we would apply them. We would probably speak of a bird feeling happy or feeling pain, and we might be inclined to speak of an ant being in pain, but I suspect that most people would find it a little odd to speak of an ant being happy. Although we notice that an ameba responds to acid, we might very well hesitate before saying that the ameba feels pain.
Why are we inclined to discriminate the way we do in extending the use of the word ‘feel’ to apply to animals? My suggestion is that this is because the criteria for the ordinary use of this word involve observable circumstances. We would not say that a dog feels like playing chess no matter what his behavior because the circumstances of that behavior would not include the appropriate social conventions. We might say that a dog feels happy at times because his behavior is rather like that of human beings when they feel happy. The behavior of an injured ant might remind us somewhat of the behavior of an injured person, but I have never seen an ant do anything that reminded me of the kinds of things people do when they are happy. The avoidance reaction of an ameba when dilute acid is introduced into its environment may remind me a little of the way a person jerks his hand away from a hot stove, but it also reminds me of the way people blink when a bright light is flashed in their eyes. I think I can see how someone might want to stretch the criteria a little and speak of an ameba’s feeling pain, but it certainly strikes me as a borderline case of the ordinary use of the word.
I have been trying to build at least a prima facie case that the criteria for the ordinary use of the word ‘feel’ include observable behavior on the part of the animal which is said to feel something which is fairly similar to the behavior of human beings when they are said to feel the same thing. This case can perhaps be made stronger still by calling attention to some features of the way we use the word ‘feel’ in connection with human beings. It is obvious that a person can feel happy and not show it in his behavior, so to say that a person feels happy is not to make a statement about the person’s behavior. On the other hand, I can imagine circumstances in which a person’s behavior is such that if anyone observed it and denied that the person felt happy I would be inclined to think the observer simply did not understand the meaning of the expression ‘feel happy’. It is only because such behavior is possible that we can say that a person might feel happy but not show it. Interestingly enough, when we speak of animals as feeling happy or feeling pain, we are inclined to emphasize the behavioral criteria of the use of these expressions even more. There is something a little odd about saying that a dog feels happy but is not showing it. I suggest that there are two reasons for this: (1) the paradigm cases for the ordinary use of the word ‘feel’ are those in which human beings exhibit ‘feeling behavior’, and situations where someone could be said to feel but not show it are highly complex situations which are parasitic on the paradigm cases; and (2) there is a general recognition that animal behavior differs from human behavior, and we are suspicious of applications that diverge too much from the paradigm cases.
My contention is that there is no possible behavior on the part of an electron or the universe as a whole which is sufficiently analogous to human behavior to make it possible to apply the word ‘feel’ in its ordinary sense to either. What behavior of an electron or the universe as a whole is even remotely like the behavior of a human being who feels happy or feels pain? At one point Hartshorne writes: "It is to be observed that physiology can as yet furnish no reason for denying feeling even to so complex an object as the world-whole, for we understand too little how feeling is possible to an animal organism to be able to infer that it is impossible that a different type of whole, even one so vast as the world, should feel" (BH 118). The problem, however, is not that we know on the basis of physiological evidence that the universe (Or an electron) does not feel; the problem is rather that the appropriate behavioral criteria for the use of the word in this setting are absent, and so we do not know what it means to use the word ‘feel’ here.
It will not, I think, solve the problem to point out that there is no clear place to draw the line between the sorts of beings that are ordinarily said to have feelings and the sorts of things that are not ordinarily said to have feelings. There are paradigm cases of things that can be said to have feelings (human beings) and paradigm cases of things that cannot be said to have feelings (electrons), and there are borderline cases where we are uncertain about whether to use the word ‘feel’ or not (amebas). But this is true of any word, and we are not ordinarily bothered by this. A fictional narrative three hundred pages long could be a novel, and one that is ten pages long could not. What about one that is seventy pages long? Or forty? There are borderline cases where we are unsure whether to call a fictional narrative a short novel or a long short story, but that is not taken as a justification for extending the use of the word "novel" to cover all fictional narratives.
My argument so far has focused on third person uses of the word ‘feel’. After all, when we speak of an atom or the universe as having feeling, we are using third person language. But perhaps it might be objected that Hartshorne avoids the difficulties I raise by his emphasis on first person language as the basis for the meaning of the word ‘feel’. If the meaning of the word ‘feel’ is established by introspectively connecting the word with the internal reality it designates, then I have a meaning for the word which I can imaginatively generalize in applying it to atoms or the universe. This objection is difficult to answer briefly, but I have two reasons for thinking that Hartshorne cannot successfully meet this criticism. First, if the meaning of the word ‘feel’ is established solely by introspection, then the word would be a word in what analytical philosophers call a "private language." A private language is one in which the words used refer to that which can only be experienced by the speaker of the language and therefore in principle cannot be taught to another person since the other person could never know the referents of the terms. The concept of a private language involves notorious difficulties.4 It is widely held that the meaning of a word cannot be established privately.
Hartshorne himself has acknowledged this, recognizing a private language of feeling as a paradox "so powerfully attacked by Wittgenstein" (LP 154). His method of avoiding this paradox will not, I think, work here. He writes:
Accordingly, the question of privacy is answered: to God all emotions and impulses are fully open to enjoyment, inspection, and comparison. We thus eliminate the paradox: "How can your feelings (not just your behavior) be like or unlike mine, since comparison can never take place"? It can take place -- in God. (LP 151)
But such a move can be made without circularity in reasoning only after the meaningfulness of attributing feelings to God has been established, and that is what my criticism calls into question.
My second reason for thinking that Hartshorne cannot meet the criticism I have made by an introspective derivation of the meaning of the word ‘feel’ is that if its meaning could be established in this way it could not, with the meaning so established, have any third person uses. Unless there are public criteria for the use of a word like ‘feel’ which involve observable behavior and the circumstances of that behavior, there can be no such thing as a correct use of that word as applied to others. Norman Malcolm makes this point in the following way:
A proponent of the privacy of sensations rejects circumstances and behavior as a criterion of the sensations of others, this being essential to his viewpoint. He does not need (and could not have) a criterion for the existence of pain that he feels. But surely he will need a criterion for the existence of pain that he does not feel. Yet he cannot have one and still hold to the privacy of sensation. If he sticks to the latter, he ought to admit that he has not the faintest idea of what would count for or against the occurrence of sensations he does not feel. His conclusion should be, not that it is a contradiction but that it is unintelligible to speak of the sensations of others. (KG 105)
Unless we have public criteria for the use of the word ‘feel’, what sense does it make to say that an atom or the universe feels anything? Why, for example, is it "plain enough" that the feelings of an atom must be vastly different from the feeling of a dog because their structures are different (BH 168)? If the meaning of the word ‘feel’ is established solely by introspection, then the connection of feeling to structure is Just as contingent as the connection of feeling to behavior. On this basis there would be no reason to say that it could not be the case that an electron has all the feelings a human being has. I am not even sure that one could not just as meaningfully say that chairs and typewriters have feelings. Perhaps they are wholes which are less well unified than their most unified parts (LP 192), but why does this prevent me from saying they have feelings if feeling is only contingently connected to structure and behavior? When we abandon circumstances and behavior as criteria for feeling, we are in an "Alice in Wonderland" kind of world.
It may still be maintained, however, that there is a serious weakness in my criticism of Hartshorne. I have tried to show that to make feeling a metaphysical category requires that the word ‘feel’ be used in ways which are very unlike the ways it is used in ordinary language. The immediate response is likely to be: "and why is ordinary language sacred?" Almost every academic discipline develops its own terminology, and this usually involves taking words from ordinary language and making them technical terms by modifying their meanings. Problems need not arise from this practice so long as we keep in mind when the word is being used in its ordinary sense and when it is being used as a technical term. Hartshorne contends that this practice is necessary for philosophy. He writes:
Ordinary modes of speaking, for ordinary purposes, are to be accepted as making sense, and as an important source of philosophical insight -- provided it be borne in mind that philosophical purposes differ from ordinary ones. They differ especially in the degree of generality sought, and this implies an unusual concern with extreme cases, such as things radically smaller or larger than, or in some other way radically different from, man and the things man commonly deals with. (CSPM 71)
It is quite clear from this passage and others that Hartshorne feels considerable respect for ordinary language. It does the job for which it is intended quite well, but the job for which it is intended is not the job the philosophers must do. The philosopher may use ordinary language as a source of insight, but he must also recognize its limitations. One of the clearest expressions of this notion occurs in the following passage:
However, it is not apparent that ‘ordinary language’ needs to recognize, in any obvious way, the pervasive abstractness of our pragmatic concepts of physical instrumentalities. Normally we simply use ‘inanimate’ physical things, and perceive them chiefly with an eye to their use; we do not ordinarily even ask what they are apart from our uses, or apart from the human species, or ‘in themselves’ or for God. These questions are somewhat abnormal, hence their answers will not lie on the surface of ordinary language. (CSPM 142)
To say that Hartshorne’s metaphysics involves a modification of ordinary language is therefore merely to point out what he himself not only admits, but even insists upon.
Let us examine briefly the implications of this view of the relation between philosophical and ordinary language for feeling as a metaphysical category. The metaphysical use of the word ‘feeling’ is supposed to be an extension of its ordinary use, but not a violation of it. This view is clearly stated in a passage in which Hartshorne defends Whitehead’s use of the word in the following passage:
‘Feeling’ has been held to be misused in Whitehead. I hold that his defense against this criticism is valid. He is by no means saying that everything feels, e.g., trees or planets. Many ‘things’ do not feel, and yet feeling is pervasive. For if a thing does not feel, its actual entities can be held to do so. Thus tree cells, in their singular actualities, are credited with some radically subhuman form of sentience. If this doctrine is wrong it cannot, so far as I can see, be on any merely linguistic ground. Nor is the question empirical in the proper sense. As I have often argued, there is no nonquestion-begging criterion of the absence of feeling in concrete singular realities. In other words, the absence of feeling cannot in principle be distinguished from the absence of concrete singularity. This is a logical problem, or a linguistic one if you like; but it is not Whitehead, in my opinion, who is misusing words, but his critics. Ordinary language does not limit feeling to human beings, and the question of the physical in nature radically transcends the concerns of common sense or linguistic good sense, in any philosophically relevant sense of these phrases. My arguments on this point, like those of Pierce, Whitehead, and others, have not, so far as I can see, been refuted by the anti-Whiteheadians. (WP 179f)
The metaphysical use of the word ‘feeling’, according to Hartshorne, extends its ordinary use, but does not contradict that ordinary use. Ordinary language does make a distinction between things which do feel (such as trees and planets), but that distinction is respected in using feeling as a metaphysical category. Ordinary language, however, is neutral on the question of whether electrons or the universe as a whole can have feelings. This is not a common sense issue but a metaphysical one, so there is no reason to think that the issue can be settled by an examination of ordinary language.
It is my contention, however, that the metaphysical use of the word ‘feeling’ does not merely extend its range of application, but violates the criteria which give the word meaning in ordinary language. I have tried to show that one of the criteria for the use of the word ‘feel’ in ordinary language is that the thing which is said to feel must at least be capable of observable behavior which is very similar to human behavior. To assert that something feels is not merely to assert that such behavior is possible, but the possibility of such behavior is at least a necessary condition for such an assertion. When Hartshorne extends the range of application for feeling the way he does, it is no longer clear what the word means since it cannot mean the same thing it does when it is used in ordinary language.
Consider the ability to do arithmetic. The attribution of this ability clearly involves behavioral criteria. We can test a person’s ability to do arithmetic by presenting him with certain problems and observing the way he goes about solving them. If a person regularly solves arithmetical problems which he has never seen before, then we say he has this ability. If a person regularly fails to solve such problems, then we say he lacks this ability. There may, of course, be borderline cases where we are unsure whether to say a person has the ability to do arithmetic or not. Also, it is perfectly possible for a person to have the ability to do arithmetic and never demonstrate his ability to solve problems. But it would be very odd indeed to attribute this ability where the entity to which it is attributed is in principle incapable of exhibiting problem-solving behavior.
I can, for example, imagine that someone might claim that his cat had the ability to do arithmetic because I can at least imagine the cat’s exhibiting the requisite behavior. We might therefore allow that it is meaningful to say that a cat has the ability to do arithmetic even though it is almost certainly false. But what would our reaction be if someone said that perhaps a hydrogen atom has the ability to do arithmetic? Here we are at a loss to know what is being claimed. The person might point out that this is not a problem for common sense and that we cannot solve the problem by an examination of ordinary language. He might say that he was using the expression "ability to do arithmetic" in roughly the same way it is used in ordinary language and is merely extending its range of application. In reply, we might point out that in ordinary language our use of this expression is governed by the requirement that the entity to which this ability is attributed at least be capable of exhibiting behavior like that of human beings when they solve problems in arithmetic. His use of the expression clearly violates this requirement, and it is therefore no longer clear what he means. It is not that we have some kind of evidence that hydrogen atoms lack this ability; it is simply that we do not know what it means to attribute this ability to them.
I believe that Hartshorne is wrong when he claims that ordinary language leaves open such questions as whether or not atoms can feel. Given the criteria that govern the word ‘feel’ in ordinary language, the question cannot even arise. Of course, it is open to Hartshorne to provide some different set of criteria for the use of the word ‘feel’ in accordance with which it would make sense to ask whether or not atoms can feel, but he has not done so. If he did, my guess would be that the question would become philosophically uninteresting. The importance of the issue lies in the claim that human beings are at one with the rest of the universe in terms of their inner experiences as well as in terms of their sheer physical characteristics. A definition of the word ‘feeling’ which is different from what the word means as it is now used in relation to human beings would probably have little charm for Hartshorne and other panpsychists.
My exposition and critique of Hartshorne’s effort to make feeling a metaphysical category is now complete, but I am haunted by a question which Hartshorne raises in the following passage:
Let us suppose, for the purpose of this discussion, that there are necessary metaphysical truths, denials of which are in some fashion absurd. Is language, with its basis in everyday communication equipped to express such truths? Presumably any answer must be qualified. If there were no difficulty in expressing metaphysical necessities, would not more agreement have long ago been reached? If there were no possibility of expressing them, would the attempt have been persisted in so long by so many superior intellects? (GSPM 139)
I have argued that what Hartshorne asks his readers to suppose in the passage is wrong, at least with respect to ‘feeling’ as a metaphysical category. I have tried to show that his treatment of feeling does violence to the ordinary use of the word which it presupposes. But why does he make this mistake? Why is it important for him to argue that the universe feels and is composed entirely of entities that feel? Why are other intelligent and sensitive philosophers and theologians attracted to this position? I do not think of the arguments I have presented as being especially subtle or profound, and Hartshorne has certainly read with care many of the works of the philosophers from which I derived them. Perhaps my arguments are unsound, but if they are not, why have not Hartshorne and other panpsychists arrived at my conclusion before? These questions are of deep concern to me, and I must confess that I have no answer which I feel is fully satisfactory and defensible. I do, however, have a tentative suggestion which may at least be a step in the direction toward an answer.
My suggestion is that a metaphysical description of the universe may be a sort of "picture preference" which serves to reinforce a person’s general orientation toward life.5 If this is true, then a person might have a strong inclination to believe that the picture preference which supports his general outlook on life is actually descriptive of the universe. Poetry and mythology may also serve this purpose to some extent, but a sophisticated person might accept this function without assuming that they provide a literal description of reality. But if this drive for a supportive picture of the universe were present in a person who also had a strong rationalistic bent, there might be a tendency to project the picture in question on the screen of the universe and assume that the projected picture is the reality.
Hartshorne himself indicates that his basic view of the scope of feeling was arrived at early in his life: "My own conviction that reality is indeed an ‘ocean of feelings’ (Whitehead’s phrase) was reached at an early age, without conscious reference to any philosophical writer or teacher, and was based, as I know from conversation Whitehead’s was, upon an attempted analysis of immediate experience, where alone reality can be encountered" (WP 148). Does this suggest that his view of feeling fulfills other needs of his in addition to his need to know the truth about reality? At any rate, there does seem to be a kind of general outlook on life which Hartshorne’s view of reality as an ‘ocean of feelings’ would support.
I do not assume that Hartshorne or any other process philosopher necessarily has just this particular outlook nor that their having it would be their reason for accepting this metaphysical position. I am only suggesting that if a person held Hartshorne’s view of feeling, it would tend to reinforce this sort of outlook on life. This outlook on life would involve sympathy and tenderness toward other human beings and animals; a kind of reverence for nature in general; a tendency to think of oneself as belonging in this world rather than as alienated from it; and a recognition that warmth and love are fundamental values for life. A person with such an outlook, it seems to me, might very well find the view that reality is an "ocean of feelings" very supportive. By projecting the most fundamental values of his own life onto the universe he would reinforce his feeling that he is really a part of this world rather than an intruder in it. His belief would provide a very important sense of security and confidence.
I do not mean to imply that if Hartshorne’s view of feeling does tend to reinforce a certain orientation toward life, this would in itself constitute the slightest reason for doubting that his view correctly describes reality. Rather, I am suggesting that if my criticism of Hartshorne’s view is correct, then this might be at least a partial explanation of why his position on the issue of feeling is attractive to him and to many philosophers and theologians.
BH -- Charles Hartshorne. Beyond Humanism. Chicago; Willett, Clark and Co., 1937.
CSPM -- Charles Hartshorne. Creative Synthesis and Philosophic Method. LaSalle, Ill.: The Open Court Publishing Co., 1970.
FU -- Stephen Toulmin. Foresight and Understanding. New York; Harper Torchbooks, 1961.
KC -- Norman Malcolm. Knowledge and Certainty. Englewood Cliffs, N.J.: Prentice-Hall. Inc., 1963.
LP -- Charles Hartshorne. The Logic of Perfection. LaSalle, Ill.: The Open Court Publishing Co., 1962.
PI -- Sidney and Beatrice Rome, eds. Philosophical Interrogations. LaSalle, Ill.: The Open Court Publishing Co., 1970.
RSP -- Charles Hartshorne. Reality as Social Process. Glencoe, Ill.: The Free Press, 1953.
WP -- Charles Hartshorne. Whitehead’s Philosophy. Lincoln: The University of Nebraska Press, 1972.
1. William P. Alston. "Meaning and Use." The Theory of Meaning. Edited by G. H. R. Parkinson. London; The Oxford Press, 1968.
2. Charles Hartshorne, "Causal Necessities: An Alternative to Hume," Philosophical Review, 63 (1954), 479-99.
1Hartshorne praises Whitehead for his understanding that "mathematics, physics, and biology have progressed because they have never been content simply to accept a handful of separate principles, but have looked for higher principles to unite the others" (WP 136).
2See, e.g., Charles Hartshorne, "The Structure of Metaphysics: A Criticism of Lazarowitz’s Theory," Philosophy and Phenomenological Research, vol. 19. no. 2 (December, 1958). Here Hartshorne indicates that he is in substantial agreement with much that Lazarowitz says in his attack on the principles of classical metaphysics. But he points out that Lazarowitz has not taken any examples from those representatives of modern metaphysics (presumably process philosophers) which Hartshorne would regard as the best.
3How would one experience the "absence of all feeling" anyway? Hartshorne has generally insisted that there are no purely negative facts. To experience the absence of all feeling one would have to experience something else that is incompatible with feeling. And what could that possibly be?
4For a brief but clear exposition of the notion of a private language and the difficulties it involves, see Norman Malcolm, "Wittgenstein’s Philosophical Investigations" in Wittgenstein: The Philosophical Investigations. edited by George Pitcher (New York: Anchor Books. 1966), pp. 65-103.
5An approach somewhat similar to this has been suggested by John Wisdom and Morris Lazarowitz. They have, however, tended to develop this approach in Freudian terms. I am inclined to think that an Adlerian perspective is likely to prove more fruitful because of its teleological orientation.