Modern Science and Its Philosophy

Philipp Frank


philosophic misinterpretations of the quantum theory

1. The Origin of Philosophic Interpretations of Physical Theories

As soon as any new physical theory appears, it is used to contribute something toward setting the controversial questions of philosophy, the questions on which philosophers have been working for centuries without coming a single step closer to their solution. Numerous examples of such use suggest themselves. When J. J. Thomson showed that every electrically charged particle possesses inertia just as a mechanical mass does and gave a formula for calculating the mechanical mass of a particle from its charge and size, people deduced from this result arguments to prove that all matter is only a phantom. They found in it an argument for the idealistic world view and against materialism. Similar interpretations were suggested when energetics arose and phenomena were represented as energy transformations rather than as arising from collisions of masses. The theory of relativity then introduced four‑dimensional non-Euclidean space instead of the three‑dimensional Euclidean space in which the directly observable processes of everyday life take place. Later, wave mechanics described physical processes with the aid of the probability concept, which has been often said to be a purely spiritual factor, instead of with the aid of mass particles. Everywhere it appears that the spiritual element is replacing the grossly material.

Such interpretations were attached with especial intensity to Niels Bohr's theory of the complementary nature of certain physical descriptions, from which it was hoped that arguments for vitalistic biology and for free will might be obtained.

If one scans all such interpretations, the empirically establishable fact is found that they all further a movement toward a certain world picture. It is not a case of different world pictures being involved; the same one keeps coming up again and again.

Through the work of Galileo and Newton, anthropomorphic medieval physics was expelled from conscious intellectual life. There remained, however, an unfulfilled longing to bring about the unity of animate and inanimate nature which had been present in medieval physics but was missing in the newer physics. There was left only one problem, for which no satisfactory solution could be envisaged: to understand the processes of life in terms of physics. For that was the necessary condition for a unified conception of nature after the disappearance of the anthropomorphic conception of physics, which had fitted in so well with the vitalistic conception of life.

Every crisis in the history of physical theories is associated with a certain lack of clarity in their formulation, and this unfulfilled longing bursts forth with great strength from the unconscious. Efforts were made to complete the new physical theories by "philosophic interpretations" in order to proclaim the imminence of a return to the anthropomorphic physics of the Middle Ages and a consequent re‑establishment of the lost unity of nature. Spiritualistic physics was to hold the possibility of embracing the living processes also.

The assertion is often heard that there is also a philosophic interpretation of physical theories in the service of a materialistic‑metaphysical picture of the world. However, this symmetrical conception of spiritualism and materialism is a very superficial one. A "materialistic metaphysics," in general, does not exist today as a living intellectual current. At most, it is represented for the domain of physics by those philosophers or scientists who wish to make as wide a gulf as possible between physics and biology, in order to obtain in the field of the living or the social processes free play for a spiritualistic metaphysics.

If, on the other hand, one understands by materialism the belief that all processes of nature can be reduced to the laws of Newtonian mechanics, then this is not a philosophic principle but a physical hypothesis. True, it is a physical hypothesis that has been shown to be wrong, but a physical statement it remains. This false hypothesis is not accepted today by any of the philosophic schools that one is accustomed to designate polemically as "materialistic"—neither by the "dialectical materialism" of Soviet Russia, nor by the "physicalists" that have come out of the Vienna Circle.

The process of the philosophic interpretation of physical theories in the service of the spiritualistic conception of the universe can be analyzed both psychologically and logically. From the psychologic standpoint, the following, roughly, has been established: The physicist, like every other educated person, acquires the remnants of prescientific theories as a "philosophic" world picture, which in our cultural circles consists mostly in a vague idealism or spiritualism as it is usually learned from lectures on general philosophy. The principles of this philosophy are unclear and difficult to understand. The physicist is happy if he finds in his science any propositions that have in their formulation some similarity to propositions of idealistic philosophy. He is often very proud that his field of work helps him throw some light on the general doctrines that are so important for this world picture. Thus even the slightest similarity in the wording is enough to induce the physicist to offer a proposition of his science as support for the idealistic philosophy.

If J. J. Thomson speaks of "real" and "apparent" mass, the philosophically educated physicist is eager to bring this mode of expression into connection with the distinction between a "real" and an "apparent" world. The statement that mechanical mass is only "apparent" mass is then taken as confirmation of philosophic idealism, according to which matter is only an illusion.

Of greater scientific interest is the logical structure of these philosophic misinterpretations. The process of thought leading to them consists of two steps. First, physical propositions that are really statements about observable processes are regarded as statements about a real, metaphysical world. Such statements are meaningless from the standpoint of science, since they can be neither confirmed nor contradicted by any observation. The first step is therefore the transition to a meaningless metaphysical proposition. In the second step this proposition, by means of a rather small change in wording, goes over into a proposition which again has a meaning, but is no longer in the realm of physics; it now expresses a wish that people should behave in a certain way. This proposition is then no longer metaphysical, but has become a principle of morality, of ethics, or of some other system of conduct.

One can adduce numerous examples of such processes involving two steps. As the simplest, we choose the well‑known example of the electromagnetic mass. J. J. Thomson formulated the purely physical proposition that every electrically charged body possesses mechanical inertia, which can be calculated from the charge. To this has been added the hypothesis, likewise physical, that the entire mass of the body can be calculated in this way. Philosophers then expressed this as a metaphysical principle by saying: "In the real world there is no mechanical mass at all." This principle obviously has no scientific content. From it there follow no observable facts. As the second step, it was asserted that the material world, as a mere illusion, is unimportant in comparison to the world of the spirit, and that therefore man in his actions can or should neglect any changes in the material world and should devote himself to his spiritual perfection.

When influential groups express such wishes, the fact has a great importance for human life, of course, and possesses a meaning, but there evidently exists no logical connection with the electromagnetic theory of matter, and the whole thing arises only through this misinterpretation with its two steps.

The essential part of the misinterpretation is the passage through the "real" metaphysical world. The misinterpretation can therefore be avoided only if one tries to set up a direct short circuit between the physical principle and the moral principle. This can be done, for example, through the consistent use of the "physicalistic language," which Neurath and Carnap have suggested as the universal language of science.

As understood in Carnap's "logical syntax," the source of these misinterpretations is always the use of the "material mode of speech." The contrast between "apparent" and "real" mass is made to appear as a statement about a fact of the observable world, whereas it is really a syntactic rule about the use of the word "real." Only the formula for the connection between electric charge and inertia is a statement about the observable world.

Quite the same logical structure is possessed by the misinterpretations of the relativity and quantum theories. The first has been employed to provide a basis for the belief in predestination, the second to give scientific arguments for "spontaneity of action" and "freedom of the will."

2. The Complementary Conceptions of Quantum Mechanics and Their Interpretations

The philosophic misinterpretations of quantum mechanics can be best understood if we remember that the same tendencies are at work here as in the interpretation of previous theories, and that the process takes place along exactly the same lines, both psychologically and logically.

First we must make clear the meaning of the complementarity conception in physics.

One often reads the following formulation: "It is impossible to measure the position and the velocity of a moving particle simultaneously." The world, therefore, just as it is according to classical mechanics, is filled with particles having definite positions and velocities; unfortunately, we can never attain a knowledge of them. This presentation, in which the states of the particles play the role of the "thing in itself" in idealistic philosophy, leads to innumerable pseudo problems. It introduces physical objects, namely, particles with definite positions and velocities, about which the physical laws of quantum mechanics say nothing at all. These objects play a role similar to that of the reference system that is absolutely at rest, which some wish to add to the theory of relativity but which never occurs in any physical proposition. In both cases the reason for this addition is that such expressions were found useful in the earlier state of physics, and the school philosophy had made of them constituents of the "real world"; therefore they must be kept forever.

Another way of representing the situation consists in saying that particles "in general do not possess definite positions and velocities simultaneously." This mode of expression appears to me to have the difficulty that the combination of words "particle with an indefinite position or velocity" transgresses the syntactic rules according to which the words "particle," "position," and "indefinite" are ordinarily used in physics and everyday life. Of course, there would be no objection if a new syntax were introduced for these words for the purposes of quantum mechanics. In that case, expressions like "particle with an indefinite position" could be employed inside of physics without any danger. And there exist many correct works on the quantum theory in which this is the case. However, gross misunderstandings arise as soon as this way of speaking is used in matters where it is no longer a question of the quantum theory. We can bring about this transition to other fields only by regarding the particle with an indefinite position as a constituent of the "real world"—and then we are right in the midst of the philosophic misinterpretations that were described in Section I.

I believe that, as a starting point for a correct formulation of the complementarity idea, one must retain as exactly as possible the formulation set forth by Bohr in 1936.

Quantum mechanics speaks neither of particles the positions and velocities of which exist but cannot be accurately observed, nor of particles with indefinite positions and velocities. Rather, it speaks of experimental arrangements in the description of which the expressions "position of a particle" and "velocity of a particle" can never be employed simultaneously. If in the description of an experimental arrangement the expression "position of a particle" can be used, then in the description of the same arrangement the expression "velocity of a particle" can not be used, and vice versa. Experimental arrangements, one of which can be described with the help of the expression "position of a particle" and the other with the help of the expression "velocity" or, more exactly, "momentum," are called complementary arrangements, and the descriptions are referred to as complementary descriptions.

If one adheres strictly to this terminology one will never run the risk of falling into a metaphysical conception of physical complementarity. For it is clear that nothing is said here about a "real world," nor about its constitution, nor about its cognizability, nor even about its indefiniteness.

A great seduction to metaphysical interpretations lies in the frequently occurring formulation of complementarity according to which the "space‑time" and the "causal" descriptions are said to be complementary. In this way the fact is often hidden that this again only means the complementarity of position and momentum, or of time and energy. By "causal description" we understand here only the description by means of the principles of conservation of energy and of momentum, which does not quite agree with what is usually understood by causality. In popular presentations, among which are those of some physicists, this is not always set forth clearly. This lack of clarity arises from the use of the expressions "space," "time," and "causality," which as a kind of trinity play a somewhat mysterious role in idealistic philosophy. If by "space‑time description" is meant simply the assignment of coordinates and time, by "causal description" the application of the conservation principles, then this beloved terminology can be retained, of course. But it then loses the charm of the mysterious and can no longer be used to pave the way for a transition from physics to idealistic philosophy, thereby favoring those misunderstandings described in Section I.

If we are once in the midst of metaphysical formulations, we can easily come to rather crass misinterpretations. As an example, I shall give one by a very prominent physicist. A. Sommerfeld says (Scientia, 1936):

If we treat the human body physiologically, we must speak of a corpuscular localized event. To the psychic principle we can assign no localization, but must treat it—and this is also the opinion of psycho‑physiologists—as if it were present more or less throughout the body, just as the wave is connected with the corpuscle in an unspecifiable way.

Here we may see with great clarity how every metaphysical formulation of a statement of physics can be used with great ease to support a statement of idealistic philosophy that only sounds somewhat similar.

To express the idea of complementarity for physics in closest association to Bohr's formulation, so that it will not lead to any metaphysical misinterpretations but yet can be carried over to fields outside of physics, one will have to proceed somewhat as follows:

The language in which occur statements like "The particle is at this place and has this velocity" is suited to experiences involving gross mechanical processes and cannot be employed satisfactorily for the description of atomic processes. However, one can give a group of experimental arrangements for the atomic domain in the description of which the expression "position of a particle" can be used. In the description of these experiments—and in this consists the idea of Bohr—the expression "velocity of a particle" can not be used. In the atomic domain, therefore, certain parts of the language of gross mechanics can be used. The experimental arrangements, however, in the description of which these parts can be used, exclude each other.

Meaningless metaphysical propositions immediately arise if one says that "reality" itself is "dual" or displays "different aspects."

3. Complementarity as an Argument for Vitalism and Free Will

Many physicists and philosophers have tried to make use of Bohr's doctrine of the complementarity of physical concepts in order to obtain arguments for the impossibility of an understanding of biology and psychology in terms of physics. Here we can distinguish something like a psychologic and a biologic argument. The first runs approximately as follows: If one seeks to describe a psychic state in terms of introspective psychology, the state is so strongly altered by self‑observation that it is no longer the original state. It is not possible to be angry and at the same time to observe and describe one's anger. The existence of a psychic state is incompatible with its observation.

The second runs something like this: If one wishes to describe the state of a living organism by means of physical quantities, the measurement of these quantities requires such a severe disturbance of the organism that it must be killed. The description of a living being through physical variables is incompatible with its life.

The psychologic argument is basically a good one. It is a long-recognized doctrine of every positivistic conception of science, including that of A. Comte, that one cannot found any logically connected psychology on principles obtained through self‑observation. One must go over to an objective observation of human actions and movements of expression, as required by American behaviorism, and in accordance with the logical analysis given by Carnap, and Neurath of the statements concerning psychic processes.

If psychology is formulated in terms of behaviorism or physicalism, the psychologic argument coincides with the biologic one.

If one applies the Bohr idea of complementarity, one can formulate the role of self‑observation in psychology somewhat as follows: There are certain experimental arrangements in the field of psychology that can be described with the aid of propositions and expressions obtained from self‑observation. There are other situations in our life that cannot be described with these expressions. In this there is no contradiction. As in physics, so in psychic life there are complementary situations, and complementary languages for their description.

Taking this complementarity into consideration, one will easily see what can be gained for the understanding of free will from the analogy to the quantum theory. Even before Bohr's discovery of complementarity, M. Planck had advanced the following argument for the compatibility of free will with physical causality: If a man could calculate his future actions from the present pattern of the physical world, this knowledge would react on his present state—for example, on the molecules of his brain—and thus change his state. Hence there is no predictability of the future. Hence free will cannot be in contradiction to the physical causality of occurrences in the human body.

From this it only follows that a man cannot calculate his future actions from the results of self‑observation. It might still be possible, however, for one to calculate beforehand the actions of other men, and to do so even from purely physical observations.

If one applies here Bohr's idea of complementarity, one can give to the whole matter a firmer logical structure. One can then say: Certain situations of human behavior are described with the help of the expression "free will"; under other experimental conditions this expression cannot be used. We are, therefore, dealing here with complementary situations and with complementary descriptions, but not with any contradictions. Bohr himself pointed out that his considerations of complementarity cannot be used to provide an argument for "free will"; they can only yield a useful representation of the epistemologic status of the problem.

It seems to me, however, that there is also a certain objection to the use of the words "free will" for the description of certain situations, corresponding to the experimental arrangements in physics. Expressions like "position of a particle" are expressions taken from the physics of everyday life which, because of complementarity, remain suitable for atomic physics only in certain special situations. Likewise, "free will" would have to be an expression from the psychology of daily life which in scientific psychology could be employed only under certain experimental conditions. This, however, seems to me not to be the case. "Free will" is not an expression from the psychology of daily life; it is rather a metaphysical or theological expression. In everyday life "freedom" is never anything other than "freedom from external coercion," or at most "freedom from intoxication and hypnosis." This has nothing to do with the philosophic conception of freedom of will. If it is correct to say, following Bohr, that the expression "free will" can be used advantageously for the description of certain situations, this expression can refer only to the quite unphilosophic concept drawn from the psychology of everyday life. Hence from this use no conclusions can be drawn about the philosophic freedom of will. It is only necessary to put to oneself the question whether, for the general situation in which the concept of free will is used in practice, any change has been created by quantum mechanics and the complementarity concept. By this I mean, of course, the application of the freedom concept to the question of the responsibility of a criminal, and to the related question of the harshness or lightness of the punishment. One need only formulate precisely the whole idea of complementarity and follow through carefully the whole chain of ideas up to the punishment of the criminal to see at once that no consequences follow here for the problem under consideration. It is therefore very questionable whether it is appropriate to use the expression "free will" in the applications of the complementarity idea to psychology.

If, however, in accordance with the new conceptions of behaviorism and physicalism, psychology is based on principles containing, not statements about self‑observation, but statements about the behavior of experimental subjects, then the complementarity considerations in psychology as just described drop out, and psychology becomes a part of biology. In that case the psychologic argument of Bohr reduces to the biologic one. It is, therefore, a question of whether the behavior of living organisms can be represented by laws in which only physical variables occur.

If one wishes to describe a living being physically, one must specify the state of each of its atoms: this is Bohr's starting point. The observations required for this description, however, involve physical disturbances of the organism that are so great as to be fatal. The states of the atoms of an inanimate body can be specified within the limits imposed by the Heisenberg uncertainty relations, whereas the large protein molecules with which life is associated are destroyed by disturbances that would allow atoms to continue to exist.

Experiments by which the living organism may be described in terms of the functions that characterize it as living are therefore carried out under experimental conditions quite different from those of experiments on the organism as a physical system. According to Bohr, it is a question here of "complementary" experimental arrangements, which are described in "complementary languages." Therefore, to describe the phenomena of life in a language which is not that of physics or chemistry is logically free from objections and does not constitute a lapse into a spiritualistic vitalism.

This way of putting the matter, as given by Bohr, is very different from that of most of his "philosophic interpreters," and is certainly tenable. In so far as its usefulness is concerned, some remarks can be made. The whole argument derives its force from the fact that it is an analogue of the argument that led from classical to quantum physics and justified the statement that atomic processes cannot be described in the language of classical physics. In order to establish limits for the appropriateness of this analogy we shall therefore compare two lines of thought.

First, in the transition to quantum physics one reasons as follows: According to classical physics, one must be able, in principle, to devise experiments permitting the measurement of the positions and the velocities of individual particles with arbitrary accuracy. But our knowledge of atomic processes—for example, the Compton effect—shows on closer analysis that the possibility of such measurements is contradicted by experience. Hence atomic phenomena cannot be described in the language of classical physics.

If we wish to extend this reasoning from the inanimate to the animate, we must accept it as an experimental fact that an observation by physical means, sufficiently accurate to enable one to describe exactly the physical state of the individual atoms of a living body, represents so great a disturbance that it kills the organism. It follows then that classical physics, aided by quantum physics (of inanimate atoms), is inadequate for the description of the phenomena of life, since it is incompatible with the application of physics to the living organism that the latter should be killed by every act of exact measurement.

The strength of the quantum theory lies in the fact that no hypothesis about the atom based on classical physics could be found that was in agreement with the experimentally testable behavior of observable bodies. If the testing of a hypothesis about atoms through direct measurements of their mechanical state (position and velocity) had not been in contradiction with empirical facts, the hypothesis would have remained within the framework of classical physics. Since, however, quantum mechanics does involve contradiction, it goes beyond classical physics.

If we wish to retain the same chain of ideas for the transition from inanimate to living bodies, then empirical evidence must be presented to show that the exact physical observation of the atoms of a living body is incompatible with the known empirical laws for the behavior of living bodies and with the physical hypothesis about their atomistic structure. As long as this evidence has not been submitted, it follows only from Bohr's train of thought that in biology, in the present state of our knowledge, the complementarity mode of expression is possible and perhaps even desirable. In contrast, for the transition from classical physics to quantum mechanics one can conclude that in atomic physics the complementarity mode of expression is necessary.

4. Summarizing Remarks

From all that has been said, it is clear that Bohr's complementarity theory does not provide any argument for free will or vitalism. Likewise, one cannot derive from it any new conception about the relation between the physical object and the observing subject, if we understand the words "object" and "subject" in the sense in which they are used in empirical psychology. In presentations of quantum mechanics in which reference is made to this new role of the observing subject, the word "subject" is understood in quite another sense. By "subject" is always meant the measuring arrangement, which can be described in terms of classical physics. What was shifted by the quantum theory was the relation between the object of atomic theory—the atom or electron—which cannot be described by means of classical physics, and the measuring instrument, which can be described classically. The observing subject, in the sense of empirical psychology, has no other task than to read off the measuring instrument. The interaction between measuring instrument and observing subject can be described classically, as far as we can say from the present state of physics. The boundary line between the classical and the quantum‑mechanical descriptions lies between the electron and the measuring instrument. Since within the region of classical description it can be displaced arbitrarily, the boundary line can also be drawn between the measuring instrument and the observer. But thereby nothing new is expressed, since within the classical region the position of the section is arbitrary.

The great importance of Bohr's complementarity theory for an branches of science, especially for the logic of science, seems to me that it starts out with a language that is generally understood and accepted, the language used to describe the gross mechanical processes of motion. Its significance lies in the fact that in its use all men are in harmony. In physics this language is used in such expressions as "position of a particle," in the sense of gross mechanics. Atomic processes, however, cannot be described in this language, as the new physics has shown. Bohr has demonstrated in a careful analysis of modern physics that certain parts of the language of everyday life can nevertheless be retained for certain experimental arrangements in the field of atomic phenomena, although different parts are required for different experimental arrangements. The language of daily life thus possesses complementary constituents which can be employed in the description of complementary experimental arrangements.

There is no doubt that this idea is also a fruitful one for logical syntax in general and deserves to be applied to other branches of science. One would have to start out in psychology with the language of everyday life and see whether, in the transition to more subtle problems, this language could be retained. One might perhaps start with the "physicalistic" "protocol language" of Carnap and Neurath and see whether any parts of it are particularly suitable for describing certain situations. Perhaps the symbol language of psychoanalysis is a suggestion of such a partial language. The phenomenal language of which Carnap often spoke in his earlier works must be dropped as a general language, but perhaps, as a constituent of a general language in the sense of the Bohr conception, it can provide a satisfactory description for certain experimental situations.

SOURCE: Frank, Philipp. Modern Science and Its Philosophy. Cambridge, MA: Harvard University Press, 1949. Reprint: New York: George Braziller, 1955. Chapter 8, Philosophic Misinterpretations of the Quantum Theory, pp. 158-171.

Modern Science and Its Philosophy: Contents

Vienna Circle, Karl Popper, Frankfurt School, Marxism, McCarthyism & American Philosophy: Selected Bibliography

Positivism vs Life Philosophy (Lebensphilosophie) Study Guide

Salvaging Soviet Philosophy (1)

Home Page | Site Map | What's New | Coming Attractions | Book News
Bibliography | Mini-Bibliographies | Study Guides
My Writings
| Other Authors' Texts | Philosophical Quotations
Images & Sounds | External Links

CONTACT Ralph Dumain

Sign Registry View Registry

Uploaded 8 January 2006

Site ©2006 Ralph Dumain