“What is mind?”—no matter

“What is matter?”—never mind

(from Punch magazine, 1885)

That thinking is something which goes on in the brain is a proposition to which we all assent unless we are being deliberately “difficult” about so commonplace a belief. Yet the evidence that it does so is very circumstantial and indirect, and some philosophers have expressed doubts about the matter. Mind, they have argued, is not a “thing” for which a place can be allocated. But from a commonsensical point of view the evidence that makes us think of the brain as the seat of thinking and as the fountain of voluntary action is too persuasive to be dismissed.

The book that prompts these reflections gives us an opportunity to eavesdrop upon an extended dialogue between Karl Popper, whom many regard as the world’s foremost living philosopher, and John Eccles, the Nobel Prize-winning neurophysiologist. Between them they try to clarify a problem that neither thinks is likely to be wholly solved: the problem of the relationship between mind and the various physical performances of the nervous system.

It is a problem upon which two extreme views have been held: at one extreme, that Mind is a thing apart which cannot be said to be in any way embodied—for mind belongs to a quite different “semantic category” from nerve impulses and the like. At the other extreme is the uncompromising materialism that is embodied in Charles Darwin’s question: “Why is thought being a secretion of brain more wonderful than gravity a property of matter?”1 Without going to the other extreme, as Darwin did, I feel confident that the dismissive “category” argument is principally a defense by orthodox philosophers against what they have interpreted as another attempted usurpation of their subject matter by those pesky scientists. It is a poor argument anyway: heredity and high molecular weight polymers also belong to different semantic categories; nevertheless genetic memory is physically embodied in the order of the nucleotides which, strung together, form the giant polymer deoxyribonucleic acid (DNA): I shall try now to explain the notion of a semantic category.

Consider a sentence such as “the cat sat on the mat” and imagine a blank, to be filled in arbitrarily, in place of the word “cat.” Clearly we could substitute the word “dog,” “mouse,” or—meaningfully, though implausibly and perhaps mistakenly—“elephant.” On the other hand we could not substitute “foreign exchange deficit,” which would not be just erroneous or unlikely, but downright meaningless because it belongs to a different semantic category. Some philosophers, led by Gilbert Ryle, take the view that thinking, willing, and other such acts of mind belong to a different semantic category from nerve impulses and other traffic of the brain. To attribute an act of mind to something that goes on in the brain—or to say that a state of mind has no physical effects on the brain—is thus a category blunder as elementary as to say that “the case for proportional representation sat on the mat.”

After the publication of Gilbert Ryle’s Concept of Mind mention of categories and “category mistakes” became painfully common. I don’t think, though, that many who used the term really understood what a “semantic category mistake” was or that they would have been able to give tongue to whatever vague conception of it they may have had. Probably they took Ryle on trust, though to be sure the “category mistakes” to which Ryle refers are in reality simply mistakes—quite straightforward and easily understandable mistakes, too, such as anyone might make.

One example of what Ryle calls a category mistake comes to mind: he envisages a foreigner in Oxford who is shown a number of colleges, libraries, playing fields, museums, and administrative offices, and then asks where the university is—thus making the elementary category blunder of confusing an abstract pedagogic entity, the university, with a piece of ground occupied by bricks and mortar. But this is not a semantic category mistake—it is simply a mistake—one that might easily be made by Americans used to the idea that universities are real material objects situated on a campus—as many American universities at least as good as Oxford are. Popper has criticized Ryle’s argument in detail elsewhere than in this volume2 and I agree that there is nothing in the concept of semantic category mistakes which prohibits our thinking that states or acts of mind can exercise physical effects.

Although Popper’s and Eccles’s opinions differ in several important ways—Eccles believes in God and the supernatural but Popper does not—they share much common ground. Both feel that the materialist “debunking” of man has gone far enough, and neither goes along with “the current intellectual fashions that belittle science and the other great human achievements.”


It is characteristic of Popper’s style of thinking that in repudiating materialism he should acknowledge the great inspiration it has been to science and point out that the leading materialist philosophers “from Democritus and Lucretius to Herbert Feigl and Anthony Quinton” were often great humanists and fighters for freedom and enlightenment. He points out, however, that even at a physical level the “essentialist” theory of matter—that matter is neither capable of further explanation nor in need of it—has been superseded in recent years by explanatory theories of matter. Modern physics, Popper declares, undermines the essentialist theory of matter: “there is no essence which is the persisting carrier or possessor of the properties or qualities of a thing.” Now that the universe has come to be thought of as a theater of interacting events or processes, “the physical theory of matter may be said to be no longer materialist.” La Mettrie’s notion of man as a machine has in recent years undergone a similar transformation: biological materialism transcended itself with the recognition that evolutionary changes of matter have taken such a turn as to lead to self-awareness and purposive behavior.

The notions of emergence and of creativity play an important part in Popper’s thinking. A clue to understanding Popper’s use of these ideas is that he regards man as part of the universe. Man is creative, ergo the universe is creative. Popper reasons:

With the emergence of man, the creativity of the universe has, I think, become obvious. For man has created a new objective world, the world of the products of the human mind, a world of myths, of fairy tales and scientific theories, of poetry and art and music.

Popper, who argued the case for the objective existence of this world in his Objective Knowledge (1972), calls this “World 3,” in contradistinction to the physical World I and the subjective or psychological World 2 (see below).

The existence of the great and unquestionably creative works of art and of science shows the creativity of man, and with it of the universe that has created man.

Popper’s principal argument—in my opinion completely convincing—turns upon his acceptance of the notion of “emergence.” Let us follow Popper in schematizing the hierarchy of what the natural world is made of in the form of a table starting at the top with, say, political and ecosystems and ending at the bottom with subatomic particles, or whatever the lowest analytical level may be.

(12) Level of ecosystems

(11) Level of populations of many-celled animals

(10) Level of metazoa and multicellular plants

(9) Level of tissues and organs (and of sponges?)

(8) Level of populations of unicellular organisms

(7) Level of cells and of unicellular organisms

(6) Level of organelles (and perhaps of viruses)

(5) Liquids and solids (crystals)

(4) Molecules

(3) Atoms

(2) Elementary particles

(1) Sub-elementary particles

(0) Unknown: sub-sub-elementary particles?

We can see by inspection that there is a progressive enrichment of empirical content at each level as we go from bottom to top; we can see also that new properties and characteristics “emerge” as we ascend: sexuality and fear for example emerge at a biological level. Moreover every statement that is true and meaningful at one level is also true at every level above it: it is a truth of politics no less than of chemistry that NaOH + HCl = NaCl + H2O.

“Reductionism” is the ambition, valid as a research program, to interpret higher levels in terms of lower levels—to interpret sociology in the language and with the concepts that apply to the behavior of individuals, to interpret biology according to the laws of chemistry, and chemistry according to the laws of physics. Reductionism has been a highly successful research stratagem: it is the way of interpreting the world that makes it easiest to see how, if need be, the world might be changed. On the other hand, the ambition it embodies may be impossible to fulfill: thus it is merely silly to say that political concepts such as proportional representation or the “foreign exchange deficit” can be “interpreted in terms of” physics and chemistry. But it should be possible to say this if the axiom of reducibility were unconditionally true, because the phrase “interpretable in terms of…” indicates a transitive relationship—i.e., if A can be interpreted in terms of B and B in terms of C then A can be interpreted in terms of C.

In the light of these concepts Popper thinks the most reasonable view of consciousness is that it is an emergent property which has arisen under selective pressures and is not therefore the result of an intervention by any psychic force or other supernatural agency.

In his discussion of objections to this theory of emergence it seems to me that Popper (though he mentions it) might have elaborated upon F.A. von Hayek’s point that our knowledge of the working of the brain must always be incomplete because brain function is both the subject of the investigation and the means by which it is investigated. In much the same way and for somewhat similar reasons a painter can never complete a painting which includes the painter himself painting the picture and the canvas upon which he is working. Popper uses this example to illustrate the cognate limitation that is embodied in Goedel’s Theorem, which applies for example to the attempt to demonstrate by deduction that a vast deductive system such as that of Principia Mathematica (which Joergen Joergensen described as a “deductive theory of deduction”) is free from self-contradictions. To describe a system fully we must be able to stand outside it: it is hopeless if we are part of it ourselves.


Popper is well known to believe in the “real” existence of a world—which he calls World 2—of states of mind such as awareness, consciousness, anxiety, embarrassment, etc. Popper is an interactionist moreover: he believes that World 2 interacts with the ordinary world of physical objects and events, called World 1. My own favorite illustration of the truth that World 2 and World 1 interact is blushing, a state of affairs in which a mental state—embarrassment—brings about the closure of arteriovenous anastomoses of the skin of the face, thus flooding the capillaries with blood: the existence of this interaction is not a solution of the brain/mind problem so much as a challenge to seek and appraise relevant evidence, discuss alternative views, and look for causal connections.

As I have said, there is another real world too, Popper believes: the world of the products of the human mind—the world of theories, hypotheses, stories, myths, arguments, and so on. The interactionist position is clearly supported by the self-evidence of the interaction between Worlds 2 and 1; but in spite of these interactions World 1 is a closed world: physical processes can be and must be explained and understood by physical theories, whether or not these self-sufficient physical processes are in some way linked to World 2. A simple solution of the body/mind problem is that which Popper describes as “radical physicalism,” according to which mental processes and states of mind do not exist: only physical states of the brain can be said to do so. The other extreme—panpsychism—contends that nothing else exists: “the stuff of the world is mind-stuff” (it was once said), of which matter is some kind of exterior manifestation. Popper finds grave objections to this view.

Eccles’s section of the joint work begins by recounting the greatest triumph of the cell theory: the notion that—unlikely though it seemed at first—the nervous system has a cellular structure; and by means of clear descriptions and diagrams Eccles describes the localizations of sensory and motor functions and faculties in the brain.

Although Eccles’s account of the matter has the authoritative tone and expertness to be expected in a scientist of his distinction, it is not philosophically sophisticated (there is no consideration, for example, of the Kantian coloration of much modern sensory physiology). At the same time, his opinions have a bluff straightforwardness too often shunned by the philosophers of mind:

When thought leads to action, I am constrained, as a neuroscientist, to conjecture that in some way my thinking changes the operative patterns of neuronal activities in my brain. Thinking thus eventually comes to control the discharges of impulses from the pryamidal cells of my motor cortex…and so eventually the contractions of my muscles, and the behavioral patterns stemming therefrom.

Eccles takes the view that a certain part of the brain—e.g., the cerebellum—is responsible for the normal execution of a physiological performance which can be shown to go wrong when that part is damaged or otherwise interfered with. Geneticists take much the same view about genes: if a mutant gene is responsible for the failure to synthesize a particular enzyme such as phenylalanine hydroxylase then the normal (i.e., non-mutant) counterpart of that gene is automatically taken to be the one that is reponsible for the normal synthesis of the enzyme. These habits of thought are so deeply ingrained that any attempt to criticize their logic will probably be ignored; besides, they may be right.

The most original and illuminating part of this book is without doubt the long section occupied by the dialogue between the two authors. It is a very special pleasure to read this grave and measured discussion, each man learning from the other and both above all else anxious to get at the truth of the difficult matters they discuss. There is nothing quite like it in any other philosophic work of comparable stature.

My only criticism of the dialogue as dialogue is that the natural friendliness and good manners of the participants may have inclined both of them to declare that they are more closely in agreement with each other than they really are—particularly over the role of sensory information in our knowledge: Popper attaches more importance than Eccles does to the role of expectation, predisposition, and the interpretative element generally in the way in which we turn sensation into sense.

Both authors believe in the reality of the existence of the state of consciousness and both believe it to be an emergent property; as to whether lower animals enjoy conscious states, both agree that the existence of degrees of consciousness even in human beings is very relevant. At the same time, I was surprised by Eccles’s skepticism of the tool-making capabilities of chimpanzees, the evidence for which I think convincing. Popper for his part regards tool-making as an advanced manifestation of a faculty to be found in all living organisms: “that living organisms in a sense select and fashion their own environment.”

In the outcome the authors agree on the interactionist position: that acts and states of mind can exercise physical effects and that the physiological activities of the brain can affect the mind. Both believe that physiological research will progressively deepen our knowledge of this interaction, even if the problem is not likely to be completely solved.

Among the most attractive features of this book are the authors’ lack of dogmatism and their determination from the beginning not to dismiss the brain/mind problem as a nonproblem or as a pseudo problem—by declaring for example that “brain” and “mind” are different categories and that never the twain shall meet. “Semantic categories” have been something of a nuisance in philosophy, because their existence—and they do exist—has more often been used to evade problems than to solve them.

The very substantial merits I have called attention to will probably be judged to outweigh the occurrence of a misprint on p. 562.

In evaluating the interactionist position we may legitimately retreat into a pragmatic stronghold: the concept works and leads to fruitful ideas and actions. Here is the part of the brain that has to do with speech; there, with sound. Stimulate the hypothalamus here there or elsewhere and the subject will feel enraged, elated, or ravenously hungry, as the case may be. There is nothing more offensive in the idea that these faculties or states of mind have a material basis than in the idea that the optic nerve has to do with sight and the auditory with sound. Such notions as these make sense of the behavioral consequences of damage to the central nervous system, and put us in the way of finding out what we can do about them.

Even if we never know exactly how brain and mind interact the interactionist position is methodologically a most fruitful one: as time goes on natural and contrived experiments will progressively enrich our understanding of the physical basis of mind—very likely in ways that will be medically useful, so that in spite of the disillusioned dialogue with which this article began it will one day certainly become true to say that what mind is does matter.

This Issue

November 8, 1979