It is quite unlikely that the acceptance of Rutherford's nuclear hypothesis would have been so immediate and so uncritical had it not been for the fact that the ground was already prepared for such a hypothesis by the discovery of the electron and of radioactivity, which indicated (1) that particles smaller than atoms exist, and (2) that such particles are ejected by atoms in the radioactive disintegration process. The inference that the atom is a composite structure built up of these sub-atomic entities follows naturally and logically; hence the question which Rutherford and his contemporaries were trying to answer was not the general question of atomic composition, the answer to which they considered self-evident, but the question as to how the electrons and other sub-atomic particles are arranged in the atom.
However, the natural and logical inference on first consideration does not always stand up under more deliberate and thorough analysis, and so it has been in this case. The original argument based on the known characteristics of radioactivity may be summarized as follows:
(b) Electrons are found among the disintegration products.
(c) Therefore, electrons are constituents of atoms.
At first glance this argument may seem sound, and in the formative years of the nuclear hypothesis it was accepted without question. Even today it is still orthodox doctrine. But the true status of the argument can be brought out clearly by stating the analogous argument concerning the photon.
(b) Photons are found among the disintegration products.
(c) Nevertheless, photons are not constituents of atoms.
Here we find that on the basis of exactly the same evidence, the physicist arrives at diametrically opposite conclusions. Because preconceived ideas concerning the electron suggest that it could be an atomic constituent, the evidence from the disintegrations is accepted as proof that it is, whereas similar preconceived ideas concerning the photon suggest that it could not be an atomic constituent, and exactly the same evidence is therefore taken to mean that the photon was created in the process. Actually, of course, the physical evidence does not distinguish between these alternatives, nor does it preclude the possibility that some other explanation may be correct. What the evidence shows is that the electron either
(b) was preexisting within, but not a part of, the atom, or
(c) was derived from the surrounding space, or
(d) was created in the disintegration process, or
(e) originated from some combination of the foregoing, or
(f) had some other origin consistent with the evidence.
At the time the nuclear atom was originally conceived, the existing physical knowledge was not extensive enough to permit visualizing these alternatives that have been listed. The idea that electrons might be created in some physical process, for instance, was probably altogether inconceivable to Thomson or to Rutherford. But today this is commonplace. Such creation is currently being observed in a great variety of processes, ranging all the way from the production of a single electron-positron pair by an energetic photon to the production of a shower of millions of particles by a cosmic ray primary. This new information has made it apparent that the emission of electrons from radioactive material does not necessarily have the significance which was originally attached to it. Current thinking favors the creation hypothesis as the best explanation of this phenomenon, and the textbooks are slowly and reluctantly trying to incorporate this new viewpoint. Kaplan tells us, for example, “... it must be concluded that in beta radioactivity, the electron is created in the act of emission.26
But the same textbook which gives us, on page 154, this conclusion based on up-to-date evidence, still repeats on page 39 the completely contradictory nineteenth-century judgment that the emission of electrons by matter is convincing evidence that electrons exist as such inside atoms,” and it goes on to present the atomic theory based largely on that outmoded idea as if it were fully in accord with present-day factual knowledge. This is not a peculiarity of this particular text. Any other modern text which we might select gives us essentially the same contradictory picture. For instance, another text book tells us, “The disintegration experiments (which indicated emission of protons by atoms) provided definite proof that protons are components of nuclei of all elements.27 Then on the very next page the text goes on to say, “It might be argued that if an electron can be emitted from a nucleus, it must have been there before,” but in spite of the fact that this is exactly the same argument which is characterized as “definite proof” on the preceding page, it is here dismissed with the statement, “This solution... could not, however, be upheld.” Here is a graphic example of what was meant in the introductory chapter when the present-day atomic theory was described as a curious and contradictory mixture of half-century old ideas with up-to-date conclusions. Any theory which is so confused that the textbook authors can “prove” a basic point of far-reaching importance on one page and flatly contradict this proof on the following page, without anyone seeing that there is a conflict, is badly in need of an overhauling.
The information now available makes it quite clear that the electron is not the permanent “building block” type of entity that was envisioned in 1911, but an evanescent particle that can be created or destroyed with relative ease. Recognition of this fact should carry with it the realization that it is not only radioactivity that has ceased to be evidence of the presence of electrons in matter; the appearance of electrons in any physical process can no longer be taken as an indication that these electrons existed prior to the initiation of that process. In fact, the weight of evidence is now strongly in favor of the conclusion that in most cases they are created in the process, and that where electrons do actually have a prior existence, they exist in, and not as a part of, the atoms of matter.
This conclusion applies not only to electrons, but to electric charges in general, irrespective of whether or not they can be definitely identified with the presence or absence of electrons. A century ago it was considered that the appearance of positively and negatively charged ions when a material goes into solution constituted definite proof that the charges also exist in the undissolved substance, and even today we find the chemistry textbooks making such statements as this, “We now know that ionic compounds exist as ions even in the crystalline state.28 But again the advance of knowledge has invalidated the prevailing conclusion. It has been found that many substances which form ions in solution are definitely not of the “ionic” nature in the solid, and the same textbook from which the foregoing statement was taken tells us a few pages later, “If ions are not present in an electrolyte before it is dissolved, they must be formed from the molecules of the compound as it dissolves.
This is precisely the same kind of a situation which we encountered in connection with the question of the origin of the electrons which make their appearance in radioactive disintegration. Many substances break up into ions, at least partially, on going into solution, and if the substance is of a type which, according to currently accepted theory, could be composed of ions in the solid state, the formation of ions in solution is commonly interpreted as proof that the substance is thus composed. But where there are reasons why the existence of ions in the solid state is incompatible with present-day theory, exactly the same evidence is taken to mean that the charges are created in the ionization process. Here again, when we examine the situation carefully, it is clear that the physical evidence does not distinguish between these alternatives, but as long as it is necessary to assume that some ions are created in the process, it is obviously quite possible, and even probable, that all ions are thus created; that is, this is the way in which ions are formed. Thus the hypothesis that the ions exist in the solid prior to solution is not only without the proof that is claimed; it is not even the most probable of the readily available explanations of the observed facts. The creation explanation has the distinct advantage that it applies the same ionization mechanism to all substances, whereas the alternative and generally accepted explanation requires two different mechanisms.
Summarizing the foregoing, it is now apparent that electrons, and electric charges in general, are easily created in physical processes of various kinds, and hence the emission of electrons from matter during such processes can no longer be considered as proof, or even as good evidence, that the electrons, as such, existed in the matter before the process took place.
At this juncture someone will probably point out that even though the emission of electrons from matter can no longer be considered as proof that the electron is a constituent of matter, the emission is still consistent with such a hypothesis, and definite proof from this source is no longer necessary in view of the large amount of supporting evidence now available elsewhere. The bald truth is that this other evidence is chimerical; the whole history of the development of the concept of the atomic electron is a story of piling one unsupported assumption on top of another, and without the definite and positive proof which the emission of electrons from matter was presumed to furnish, the whole structure collapses. Bohr's original postulates, for example, are simply ridiculous if he first has to assume that the electron is a constituent of matter, and then goes on to postulate behavior characteristics for these hypothetical atomic constituents totally unlike anything ever observed. If his action in abandoning the solid ground of established physical facts and striking out on an uncharted course of pure hypothesis can be justified at all, which is questionable, it can only be justified on the ground that he thought that definite and positive proof that the electron is a constituent of matter already existed, and hence if the behavior of these atomic electrons could not be explained in a normal manner, it was reasonable to presume that they must follow some different laws.
As long as there is any question at all as to whether or not the electron is actually a constituent of matter, the fact that the atomic electron cannot be reconciled with known physical laws is a strong argument against the existence of any such entity, not a justification for formulating new physical laws. Had Bohr been in possession in 1913 of the experimental knowledge of the present day, including the now well-established fact that electrons are transient entities that can be readily produced or destroyed, it undoubtedly would have been obvious to a scientist of his competence that the reason for his inability to fit the atomic electron into the existing framework of physical laws was not that this constituent of the atom is governed by a different set of laws, but that no such atomic electron exists.
One of the characteristics of a sound physical theory is that it leads in an easy and natural way “with the appearance of a certain inevitableness,” as Bridgman puts it, to explanations of physical phenomena other than that for which it was originally developed. Planck's original Quantum Theory, for example, was developed to explain the behavior of radiation from an energy distribution standpoint, but one of its first important consequences was a simple and logical explanation of the photo-electric effect: a related but totally different phenomenon. Similarly, we could expect that if the concept of the electron as a constituent of matter were valid, we would find it leading easily and naturally to solutions of other related problems. But the whole history of this concept has been just the opposite. Nothing has developed easily and naturally; every step that has been taken has been forced and artificial, and each advance into new territory has been made only by sacrificing some part of existing physical knowledge, so far as its application to the atom is concerned.
As one observer expresses it, “Bohr solved the problem of the stability of a system of moving electric charges simply by postulating that the cause of the instability... did not exist.29 To the layman his might seem to involve a rather drastic redefinition of the word “solve,” but be that as it may, the ensuing history of the Bohr atom and its lineal descendants is one long series of problems for which there seems to be no solution other than to postulate that they do not exist. The orbits which Bohr postulated for the electrons could not be located specifically, hence it was postulated that no definite orbits exist; the theoretical momentum and position of an individual electron could not be reconciled, and a “Principle of Uncertainty” was therefore formulated, asserting that the electron could not have a definite momentum and a definite position at the same time; even with the benefit of this extraordinary principle, identification of positions was found to be impossible, so it was postulated that the impossibility was inherent and that the best that could be done was to calculate a probability that the electron might be found at a certain location; some of the theoretical consequences were inconsistent with the usual cause and effect relationships, and it was therefore postulated that causal relations are not operative at the subatomic level. Now in relatively recent years, the long list of assumptions and postulates has been climaxed by the assumption, sponsored by the Copenhagen school of theorists (who represent the “official” viewpoint of present-day theoretical physics), and expressed by Heisenberg in the previously quoted passage, that this atomic electron does not even “exist objectively.
All of these “solutions” of the problems that have been encountered in the development of the concept of the electron as an atomic constituent have, of course, modified the characteristics of the atomic electron very drastically. As the nuclear atom was originally conceived, the negatively-charged constituent was presumed to be the same electron that is observed experimentally. This experimental electron is a definite and well-defined thing, notwithstanding its impermanence. We can produce it at will by specific processes. We can measure its mass, its charge, and its velocity. We can control its movement and we have methods by which we can record the path that it takes in response to these controls. Indeed, we have such precise control over the electron movement that we can utilize it as a powerful means of producing magnified images of objects which are too small for optical magnification. In short, the experimental electron is a well-behaved and perfectly normal physical entity. But such an electron cannot even begin to meet the requirements which have been established step by step for the atomic electron, as the concept of this particle has been gradually modified to “solve” one problem after another. The atomic electron, as it is now portrayed, is not a definite and tangible entity such as the experimental electron. It does not conform to the usual physical laws in the manner of its experimental counterpart, but has some unique and unprecedented behavior characteristics of its own, including a strange and totally unexplained ability to jump from one orbit to another (or to do something entirely incomprehensible which has the same effect) with no apparent reason and, so it seems, complete immunity from all physical limitations. We can deal with it only on a statistical basis, and even then, as Herbert Dingle points out, we can make our statistical methods for dealing with such particles effective “only by ascribing to the particles properties not possessed by any imaginable objects at all.30 Furthermore, as already mentioned, the leading theorists of the present day tell us that the atomic electron cannot be accommodated within the three-dimensional framework of physical space; it must be regarded merely as a symbol rather than as an objectively real particle.
In view of this fact that the atomic electron no longer has even a remote resemblance to the experimental electron, it is manifestly absurd to continue basing physical theory on the fiction that the two are identical. The previous conclusion that there is no proof that the electron is a constituent of the atom must therefore be extended to assert specifically that the electron as known experimentally is definitely not a constituent of the atom. The hypothetical negatively-charged atomic constituent currently sharing the name electron” with the experimental particle is something of a totally different character, a purely theoretical creation, unrelated to anything that has ever been observed and itself not capable of being observed: an “abstract thing, no longer intuitable in terms of the familiar aspects of everyday experience,31 as Margenau describes it.
It should be emphasized that the conclusion just stated-the conclusion that the negatively charged constituent of the atom (if such a constituent exists) is a purely hypothetical entity unrelated to the experimental electron-is not something that has been developed in this work, or that depends in any way on any of the arguments presented herein. It is simply a necessary consequence of an obvious fact that modern physicists have chosen to ignore: the fact that two physical entities are not identical if they have little or no resemblance to each other. If the theorists wish to contend that the hypothetical negative constituents of the atom are identical with the experimentally observed particles that we call electrons, then they must accept, with no more than reasonable modifications justified by the environment, the properties of the experimental electrons; if they find it necessary to invest their hypothetical atomic constituents with a totally different set of properties, then they cannot identify them with the experimental electrons. Even the physicists, who in these days are permitted to “get away with murder,” as James R. Newman expresses it, must be required to conform to some of the elementary rules of logic.
Long years of effort have convinced the theoretical physicists that the first alternative, the construction of an atom in which the experimental electron is the negatively charged constituent, is impossible. As brought out earlier in this discussion, the concept of a nucleus composed wholly or in part of a group of positively charged particles is likewise untenable in the light of present-day knowledge. Assumption (2) in the list previously given is therefore invalid; that is, if the atom is constructed of “parts,” these parts are not known subatomic particles; they are purely hypothetical concepts of which no independent experimental evidence exists. This is a difficult pill to swallow: a conclusion that the scientific world will find very hard to accept, not only because it invalidates many of the cherished ideas and concepts of modern physical science, but also because it is in direct conflict with the seemingly natural and logical inference which is immediately drawn from the existence of radioactivity.
The original concept of the atom was that it is the indivisible ultimate particle of matter; the word atom actually means indivisible. But the discovery of radioactivity showed that the atom is not indivisible, as this is a process of disintegration, in which particles are ejected and the original atom is transformed into one of a different kind. The natural conclusion to be drawn from this new knowledge-the conclusion that was drawn when the knowledge was new, and which is still one of the principal supports for the present-day theory of the atom-is that the atom is a complex structure composed of sub-atomic particles. The validity of this conclusion in its general aspects will be discussed later. For the moment we are dealing only with the question of the nature of these particles.
Just as it is natural to conclude that the existence of radioactive disintegration proves that the atom is composed of individual parts, so it is natural to conclude that the particles ejected from the atom in the process of disintegration are the parts of which the atom is composed. In fact, this conclusion seems to be implicit in the first. But this second of the natural and seemingly obvious conclusions turns out to be entirely erroneous. Three types of particles emanate from the disintegrating atom, and existing knowledge indicates that not one of these three existed as such in the atom prior to the disintegration. The alpha particles are positively charged helium atoms, and it was quickly realized that they could not be primary atomic “building blocks; present-day opinion, as previously noted, is that the beta particles, which are electrons, are created in the disintegration process; and the gamma particles (if we stretch the definition of “particle” far enough to include them) are photons, units of radiation, and have always been considered to be products of the disintegration, not as pre-existing entities.
This throws an entirely different light on the picture. If we were able to show that the particles ejected by the atom were of such a character that it could be logically concluded that they were the “building blocks” of which the atom is constructed, then we could take the stand that radioactivity furnishes a satisfactory explanation of the general nature of atomic structure. But the physicists cannot and do not contend that this is true; when we boil their statements down to the essence, we find that they are, in effect, advancing the curious contention that the emission of certain particles from the atom during radioactive disintegration is a proof that the atom is constructed of certain other particles. This is a far cry from the conclusions which seemed so natural and logical on first consideration of the phenomenon of radioactivity. The physicists' present stand is neither natural nor logical, and it destroys the whole force of the original argument. Not only does it leave the question of the identity of the atomic constituents entirely up in the air, but the fact that it has been necessary to conclude that all of the particles ejected by the radioactive atom are created in the disintegration process also raises some serious questions as to the validity of the basic assumption that the atom is composed of “parts.
In any event it is now clear that the electron or any other particle that is proposed as an atomic constituent will have to stand on its own feet without any support from radioactivity. The conclusions of the early 1900s to the contrary will simply have to be rewritten, in the light of modern knowledge, no matter how reluctant the theorists are to take this step. From the viewpoint which the advances in experimental knowledge have given us, the textbook statement that the emission of electrons by atoms is convincing evidence that electrons exist as such inside atoms” must be rewritten to read that “the creation of electrons in physical processes such as radioactivity is convincing evidence that electrons exist as such inside atoms,” which, of course, reduces it to an absurdity. If the electron is to be advocated as an atomic constituent, then some consistent picture of an atom constructed wholly or in part of electrons will have to be devised and, as has been brought out earlier in the discussion, it is now admitted that this cannot be done if the atomic electron has the properties of the electron which is observed experimentally. Hence we come back to the fact that if there is a negatively charged constituent of the atom, it is not the experimentally observed electron; it is a purely hypothetical particle of a much different nature.
At this point, then, we can say that the nuclear atom, as currently conceived, is impossible. It has been shown that the two items which are relied upon to furnish proof of the validity of the basic assumptions on which the nuclear theory rests not only do not supply any such proof, when they are carefully analyzed, but actually furnish strong evidence to the contrary. It has also been shown that without the pro(JNS1)of which these two items, radioactivity and Rutherford's scattering experiments, are supposed to furnish, the entire structure of the nuclear theory collapses. Every one of the eight major assumptions of this theory, as previously listed, topples in this general collapse, except assumption (1), which we have not yet considered.
A most heretical conclusion? Perhaps so. But consider the following statement by Erwin Schroedinger, one of the principal architects of modern physical theory, who can hardly be classified as a scientific heretic, and ask yourself whether he is not saying exactly the same thing in more cautious words:
Once we have become aware of this state of affairs, the epistemological question: “Do the electrons really exist in these orbits within the atom?” is to be answered with a decisive No, unless we prefer to say that the putting of the question itself has absolutely no meaning. Indeed there does not seem to be much sense in inquiring about the real existence of something, if one is convinced that the effect through which the thing would manifest itself, in case it existed, is certainly not observed. Despite the immeasurable progress which we owe to Bohr's theory, I consider it very regrettable that the long and successful handling of its models has blunted our theoretical delicacy of feeling with reference to such questions. We must not hesitate to sharpen it again, lest we may be in too great haste to content ourselves with the new theories which are now supplanting Bohr's theory, and believe that we have reached the goal which indeed is still far away.32