Why Psychicalism? Comments on Keeling’s and Shepherd’s Criticisms
by Charles Hartshorne
Charles Hartshorne taught at the University of Texas where he was Ashbel Smith Professor of Philosophy. He had a distinguished career at several other universities, particularly the University of Chicago and Emory University. His most recent book, Creative Synthesis and Philosophic Method, was published by Open Court. The following article appeared in Process Studies, pp. 67-72, Vol. 6, Number 1, Winter, 1974. 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.
Professor Keeling proceeds with a good deal of care and lucidity. He does some justice to my presentation of the case for psychicalism (the term I now prefer to "panpsychism"). With many of his sentences I have no quarrel. However, I have a correction to what he says in the third paragraph of section I. I do not hold that we have "direct access only to human feelings." In human feelings we feel feelings of subhuman bodily constituents, though without distinctness as to the individual constituents concerned. Still, our access to the latter is direct. Human sensation is human awareness of subhuman feelings. The point of bringing in God as able directly to compare feelings in diverse creatures is that we can do the analogous comparison, though not with distinctness, and only with feelings in certain bodily constituents. The aim here was not to make behavioral criteria for ‘feeling’ unnecessary but to supplement the merely behavioral with a more direct criterion in a special case other than the introspective one. In introspection, really retrospection, the direct datum is previous human feeling; in sensation the datum is previous subhuman feeling. This means that direct experience, even apart from behavior, shows that the word feeling applies more broadly than merely to mammals or many-celled animals.
For specific application of ‘feeling’ or ‘sentience’ to nonhuman feelings other than those of our own bodily constituents, the behavioral criteria are all we have. And these criteria must fit also even the special cases mentioned of human and subhuman bodily feelings. However, Keeling seems to have rather forgotten my various discussions of how the behavioral criteria may be generalized beyond common sense. The criteria are generalized, not just the concept. The criteria are: self-initiated activity, including motions (growth and other changes involve motions on the microscale); unity or integration of the motions; influence of the past and the environment; something expressive of anticipation, desire, purpose, satisfaction. These criteria exclude many things, such as typewriters, mountains, trees (but probably not plant cells). They also exclude groups of animals, including termite colonies and the occupants of an airplane.
Contemporary physics no longer expressly denies self-initiated activity of atoms (hence, we may surmise, as one example, the irreducibly statistical half-life law of the transformation of radium atoms into lead). The kinetic theory of heat and the vibratory theory of matter banished merely inert units from science. As Francis Bacon saw, the responsiveness to other things (really to past events) that in our experience appears as memory or perception cannot be denied of the least constituent of matter, since causality would then also be denied. Memory may indeed be absent, since it is the special case in which present and past actualities occur in "personally ordered" linear sequences; but a generalized form of perception is then left as sole representative of causality. To be sure, a door with an electric eye is not a perceiving subject, but then it is "less unitary than its most unified constituents" and hence is to be regarded as a crowd, not an individual.
The full force of the case for psychicalism is too complex to be summed up in the simple account my critic gives. One must, for instance, see that the theory gives a direct "answer" to Hume’s denial of intuitions of causal influence, an answer incomparably clearer than Kant’s and more moderate in its demand for orderliness in nature. Being indeterministic, it does not produce Kant’s antimony with freedom, but does provide for necessary, though not in the classical sense "sufficient," conditions for events, unless by sufficient one means only "sufficing for the possibility of the event." Granted the possibility (which is always less definite than actuality), the event makes itself happen. This is the creativity, the self-initiated activity, of which process philosophers (Peirce, Bergson, James, and others) speak.
As for purpose, valuation, satisfaction, it is important to realize that among the differences which the idea of feeling can tolerate are enormous contrasts in time span. An ameba does not look far ahead or vividly recall the distant past, but it may anticipate a fraction of a second and remember over a similar time span. In microphysics it is clear that the time span is vastly different again, as is the space span. I think the criticism that Keeling offers is merely one more attempt to tie us down to belief s which science has already vastly transcended. The physicist uses terms like mass, movement, slow, fast, in ways that would not occur to any ordinary citizen not exposed, at least indirectly, to modem science. Just so psychology, seeking to understand the individuals, the genuinely self-active units, of nature, will have vastly to generalize ordinary concepts of response, stimulus, memory, aversive behavior, and something like preference, hence valuation, satisfaction, dissatisfaction. Of course most creatures do not feel like doing mathematics. But if they are singulars rather than composites they feel like doing something special and prove it by doing that something.
In Keeling’s example about green’s not being a metaphysical generality, he does not stress sufficiently that the reason we know green is not universal is that other positive qualities exclude it; for instance, red does. What excludes feeling? Lack of activity and lack of unity do, but these are not positive in the sense in which "sentient" is so. And all things are either active and unitary, or have constituents or members that are. That this is not so could not conceivably be known.
It is correct that there must be borderline cases where decision is difficult and beyond which feeling is pretty surely inapplicable -- but only in the sense in which it is inapplicable to groups, and there may be doubt as to whether one confronts a singular or a group. There are also extreme singular cases very difficult for our imaginations to deal with. But it is anthropomorphic to derive metaphysical or antimetaphysical conclusions from the fact that, for example, we cannot easily and distinctly imagine how a paramecium, or a fortiori an atom, feels. We still are not talking nonsense if we say that it feels somehow, though only superhuman intelligence could know more than vaguely how. We know a lot about how it does not feel.
Linguistic analysis can rightfully require that "not everything feels." The principle of contrast must be honored. However, crowds are not singular sentient subjects, whether crowds of people, termites, cells, atoms, or molecules. Although a "psychology of atoms" is not at present a recognized branch of science (yet some physicists have talked about it) and though it may never be an especially rewarding one, it does not follow that below the level of microorganisms, say, there just is nothing analogous to what on the higher levels are feeling, memory, anticipation, perception. Absolute negations of this sort explain nothing not better explained otherwise and are incapable of ever being verified.
The attempt to guess my (Peirce’s? Bergson’s? Leibniz’s?) motivation does not impress me. The early experience referred to as convincing me of psychicalism was of the duality "feeling of feeling" as directly given. It was a question of what I took to be evidence, not a question of desirability. I look upon the answer to Hume above referred to as powerful additional evidence. Many other considerations of a similar kind could be mentioned.
The case for psychicalism is so strong, in my view, that I thank anyone for attacking the doctrine. It is one of the two doctrines I am most confident of, the other being the idea of nonclassical, creationist, or indeterminist causality. Absolute exclusion of creativity and absolute exclusion of sentience from entire portions of nature are alike groundless. They are illicit conversions of "not easily knowable" into "known not to be." Nor is it likely that extreme cases of the subhuman (or the extreme case of the superhuman, deity), are open to easy commonsense dogmas, especially negative ones. The opposition to psychicalism has so far taken the posture of casual rejection, as though the truth, at least the negative truth, about what is most remote from ordinary experience were easy to arrive at. I see no ground for this attitude. It is easy to know that atoms do not feel as human beings do. But then neither do apes. Nor do infants feel as you and I do.
One consideration seems to me worth mentioning: if a mind or spiritual being exalted in principle as far above the human mode of awareness as God is defined to be is in any genuine sense conceivable, then a form of awareness as far inferior to the human as an atom is physically might very well also be conceivable. If theology is a difficult subject, then so, for partly the same and partly analogous reasons, is psychicalism. It will take much more than linguistic analysts have so far proposed on this topic to justify negative dogmas as the only reasonable way to view these subjects.
Whatever the valid results of linguistic analysis, putting topics "under the rug" because they are remote from ordinary and prescientific ways of thinking and therefore especially difficult can hardly be one of them. Ordinary good sense does not exhaust possible wisdom. Nor should a philosopher take the agenda of current science as definitive for all the future. The question, what are the lower limits of comparative psychology, or even whether there are any such limits, is important for the future both of science and of philosophy. One of the uses of philosophy is to help future scientists to transcend mere common sense, as Greek atomism and Platonic cosmology, (but hardly Aristotle, with his commonsense physics) helped early modern physicists. The inhibiting role akin to that of Aristotle is assumed by many philosophers today. It is well that some of us should take a different role.
Dr. John J. Shepherd’s criticism of panpsychism (PS 4:3-10) is ingenious and too complex for adequate consideration in brief compass. The principle of parsimony is not the final court of appeal. In metaphysics the decisive question is not whether an account is the simplest possible but whether it is possible, provided criteria for the coherence and positive meaning of the concepts are accepted. Psychicalists, from Leibniz down, have challenged the ability of materialism or dualism to meet these criteria. Concrete realities in the form of experiences are directly given; so this concept of concreteness has positive meaning. But concrete reality in the form of mere matter differs from the psychicalist concept only by its negative "mere." No positive alternative is provided.
Shepherd asserts that mental images and brains have "different and incompatible spatio-temporal patterns" and concludes that one must choose between the correctness of neurophysics and the correctness of the psychicalist theory. However, (1) images, like thoughts, are abstractions, not concrete actualities. They are neural states so far as these appear in human experiences. The neural states consist of subhuman experiences and are actual; actual also are the human experiences. The images are not additional actualities, but are the former as appearing to the latter. In so far psychicalist and central-state materialist can partly agree: images are appearances. (2) To count against psychicalism, any incompatibility of spatio-temporal patterns must, I should think, be between the patterns assigned to the neural actualities and the human experiences of which images are abstract aspects. Of course a sequence of human experiences has a different spatio-temporal pattern from the (nonlinear) sequences of neural events. But where is the incompatibility, if this means contradiction, in asserting both patterns? They do not qualify the same actuality. (3) Perhaps the argument is that physics purports to give a complete account of the spatio-temporal structures in nature. However, since physics does not even consider the question, what is the nature of experiences (which certainly have temporal and, I hold, spatial structures), the justification for this claim is not apparent. (4) Is the argument that there is not room in the region of space occupied by the neurons for both the types of structures mentioned above? This might, I suppose, be held, yet I cannot help wondering with what cogency in a world that accommodates particles, photons, and other kinds of radiant energy, atoms, molecules, cells, and animals.
It may be that my speaking of "the spatial patterns in some of our sensations" as evidence that our experiences are not point-like has been taken as implying that, contrary to (1) above, sensations or images are extended actualities. I realized even in appealing to this evidence that there was danger of ambiguity. "The sensations are extended" is elliptical. The immanent data of the sensations are in diverse parts of space, and the mass of data is extended in the sense in which a group of actualities, each with a different locus in the spatial system, is extended. The experience of which a sensation is an aspect is extended in the sense in which a single actuality is so. Leibniz almost succeeded in making this distinction and failed only because he denied intrinsic dependence of successive states of his monads upon previous states of other monads. To be spatial as a single actuality is to depend upon a plurality of immediate prior data that are contemporary with (independent of) each other. The spatiality "in" the sensations expresses this form of spatiality. To be spatial as a single actuality is also to condition immediately a plurality of subsequent data that are contemporary with one another. An event occurs when and where it is most directly conditioned and conditioning, and the directly conditioning and conditioned events are always plural and in different loci. This, I take it, is what is meant by saying that the locus of an event is a volume, not a point.
The reader will now see why I do not find a conflict between the physics of the nervous system and my psychicalism. It is indeed true that the faith of some, by no means all, atomic physicists in the literal and complete truth of quantum physics when applied to animal organisms contradicts the forms of conditioning mentioned above. But Heisenberg and Bohr suggested and Wigner argues that quantum physics in its present form cannot be the whole and literal truth of organic behavior. The issue here is not between physics and a certain philosophy; it is an issue within physics and an unsettled one.
Professor Keeling, to return to his essay, makes use of the argument, ostensibly derived from Wittgenstein, that psychicalism implies the possibility of a private language or implies that mind is more than behavior, although only one’s own mind, at most, is knowable except as behavior. Part of my reply here is that psychicalism is in a genuine respect as far possible from the concept of mind as merely private. That our own feelings are not the only feelings is known to us (mostly subconsciously to he sure) not solely by analogy and from the behavior of other people or animals, but by direct givenness in at least one privileged case, the feelings of our own bodily microconstituents. (There is another such case, which I shall not here discuss, the givenness of deity.) The bodily constituents are given mostly subconsciously, but some of us think we can consciously intuit our physical pleasures and pains as direct participations in feelings enjoyed or suffered by our bodily constituents. "Other minds" are at all times directly experienced, even though indistinctly so far as the individual other subjects are concerned. (I hold with Leibniz, Peirce, Bergson, and many others that human direct intuitions are pervasively indistinct as to the data, taken one by one or as singulars. But "indistinctly given" is not equivalent to "indirectly given." Intuitions both direct and wholly distinct are the prerogative of deity. It is a persistent illusion that we human beings can have them.)
Psychicalism implies that there is nothing concrete besides one’s own experiences except experiences not one’s own. And so, of course, all experience furnishes examples of "other minds." In memory the other mind is ones own past mind; in perception it is mind not one’s own at all. The argument from behavior and by analogy comes in to get us beyond other mind in the limited forms, one’s own past mental states, mental states of one’s bodily constituents (also divine mind). Here I agree with Plantinga; the argument by analogy, whatever its defects, is all we have (GOM 269). But it does not have to bear the burden of teaching us that "other mind" in some sense is a reality. In any case, until technology produces much better robots, there is only trivial doubt about the basic validity of ordinary inferences of this kind, granting a measure of vagueness and inaccuracy. And of course the primary learning to interpret the behavior of other animals as indicative of minds more or less like our own takes place in infancy, and we have no need of appealing to it every time we deal with another person or a dog or horse. One can, even a psychicalist can, include the social nature of language as a special and very significant form of the behavioral analogy.
Another part of my reply (and here I follow a suggestion of Lewis Ford) to the private language argument is that metaphysics is concerned with generic features of feeling, not its specific or individual nuances. Any privacy of the latter does not prevent metaphysical generalizations. Still another part concerns the givenness of deity, which is more relevant, I think, than Keeling allows.
One final point: the "introspection" by which our own experiences are given is not a mysterious function additional to perception and memory. It simply is memory, especially in its very short-run and most immediate form. (Here Ryle and Whitehead are in notable agreement.) In introspecting I keep remembering how I have just felt and thought, also how I have just perceived and remembered. To know one’s own knowing is to remember it. To introspect is to use memory in a special way, above all to use the memory of just previous not yet forgotten experience as contrasted to memory of experience for a time "forgotten." Those who fail to find experience as itself a datum are those whose conscious attention is to the data of previous perception rather than to the previous perceiving -- and remembering -- itself. They are extreme extroverts or objectifiers, or they philosophically affect to be. They fail to notice or admit their remembering of remembering or remembering of perceiving. However, the givenness of mind is not limited to remembering, for perceiving is also the intuiting of mind, but of subhuman (or superhuman) mind, rather than of the animal mind which is either one’s own or a constituent of one’s own body.
Thus psychicalism has a fairly complete theory of how its own truth is known.
May I say again that I appreciate criticisms of psychicalism, especially those of the high quality of Dr. Shepherd’s. I hope on some other occasion to say something about his objections to the Buddhist-Whiteheadian event-theory of individuals.
GOM -- Alvin Plantinga. God and Other Minds. Ithaca: Cornell University Press, 1967.