by Gary Zukav
Although these alternatives were unacceptable to Einstein, they are being considered by physicists today. Few physicists believe in telepathy, but some physicists do believe either that at a deep and fundamental level there is no such thing as “independent real situations” of things which have interacted in the past but which are spatially separated from each other, or that changing the measuring device in area A does change “the real factual situation” in area B.
This brings us to Bell’s theorem.
Bell’s theorem is a mathematical proof. What it “proves” is that if the statistical predictions of quantum theory are correct, then some of our commonsense ideas about the world are profoundly mistaken.
Bell’s theorem does not demonstrate clearly in what way our commonsense ideas about the world are inadequate. There are several possibilities. Each possibility has champions among the small number of physicists who are familiar with Bell’s theorem. No matter which of the implications of Bell’s theorem we favor, however, Bell’s theorem itself leads to the inescapable conclusion that if the statistical predictions of quantum theory are correct, then our commonsense ideas about the world are profoundly deficient.
This is quite a conclusion because the statistical predictions of quantum mechanics are always correct. Quantum mechanics is the theory. It has explained everything from subatomic particles to transistors to stellar energy. It never has failed. It has no competition.
Quantum physicists realized in the 1920s that our commonsense ideas were inadequate for describing subatomic phenomena. Bell’s theorem shows that commonsense ideas are inadequate even to describe macroscopic events, events of the everyday world!
As Henry Stapp wrote:
The important thing about Bell’s theorem is that it puts the dilemma posed by quantum phenomena clearly into the realm of macroscopic phenomena…[it] shows that our ordinary ideas about the world are somehow profoundly deficient even on the macroscopic level.7
Bell’s theorem has been reformulated in several ways since Bell published the original version in 1964. No matter how it is formulated, it projects the “irrational” aspects of subatomic phenomena squarely into the macroscopic domain. It says that not only do events in the realm of the very small behave in ways which are utterly different from our commonsense view of the world, but also that events in the world at large, the world of freeways and sports cars, behave in ways which are utterly different from our commonsense view of them. This incredible statement cannot be dismissed as fantasy because it is based upon the awesome and proven accuracy of the quantum theory itself.
Bell’s theorem is based upon correlations between paired particles similar to the pair of hypothetical particles in the Einstein-Podolsky-Rosen thought experiment.* For example, imagine a gas that emits light when it is electrically excited (think of a neon sign). The excited atoms in the gas emit photons in pairs. The photons in each pair fly off in opposite directions. Except for the difference in their direction of travel, the photons in each pair are identical twins. If one of them is polarized vertically, the other one also is polarized vertically. If one of the photons in the pair is polarized horizontally, the other photon also is polarized horizontally. No matter what the angle of polarization, both photons in every pair are polarized in the same plane.
Therefore, if we know the state of polarization of one of the particles, we automatically know the state of polarization of the other particle. This situation is similar to the situation in the Einstein-Podolsky-Rosen thought experiment, except that now we are discussing states of polarization instead of spin states.
We can verify that both photons in each pair of photons are polarized in the same plane by actually sending them through polarizers. On the next page is a picture of this (conceptually) simple procedure.
A light source in the center of the picture emits a pair of photons. On each side of the light source a polarizer is placed in the path of the emitted photon. Behind the polarizers are photomultiplier tubes which emit a click (or an audible electronic equivalent) whenever they detect a photon.
Whenever the photomultiplier tube in area A emits a click, the photomultiplier tube in area B also emits a click. This is because both of the photons in each photon pair always are polarized in the same plane, and both of the polarizers in this arrangement are aligned in the same direction (in this case, vertically). There is no theory involved here, just a matter of counting clicks. We know, and can verify, that when the polarizers both are aligned in the same direction, the photomultiplier tubes behind them will click an equal number of times. The clicks in area A are correlated with the clicks in area B. The correlation, in this case, is one. Whenever one of the photomultiplier tube clicks, the other photomultiplier tube always clicks as well.
Now suppose that we orient one of the polarizers at 90 degrees to the other. On the next page is a picture of this arrangement. One of the polarizers still is aligned vertically, but the other polarizer now is aligned horizontally. Light waves that pass through a vertical polarizer are stopped by a horizontal polarizer and the other way round. Therefore, when the polarizers are oriented at right angles to each other, a click in area A never will be accompanied by a click in area B. The clicks in area A, again, are correlated with the clicks in area B. This time, however, the correlation is zero. Whenever one of the photomultiplier tubes clicks, the other photomultiplier tube never clicks.
There also are correlations between the clicks in area A and the clicks in area B for every other possible combination of polarizer settings between these two extremes. These statistical correlations can be predicted by the quantum theory. For a given setting of the polarizers, a certain number of clicks in one area will be accompanied by a certain number of clicks in the other area.
Bell discovered that no matter what the settings of the polarizers, the clicks in area A are correlated too strongly to the number of clicks in area B to be explained by chance. They have to be connected somehow. However, if they are connected, then the principle of local causes (which says that what happens in one area does not depend upon variables subject to the control of an experimenter in a distant space-like area) is an illusion! In short, Bell’s theorem shows that the principle of local causes, however reasonable it sounds, is mathematically incompatible with the assumption that the statistical predictions of quantum theory are valid (at least valid in this experiment and in the Einstein-Podolsky-Rosen experiment).*
The correlations which Bell used were calculated, but untested predictions of the quantum theory. In 1964, this experiment was still a hypothetical construct. In 1972, John Clauser and Stuart Freedman at the Lawrence Berkeley Laboratory actually performed this experiment to confirm or disprove these predictions.8 They found that the statistical predictions upon which Bell based his theorem are correct.
Bell’s theorem not only suggests that the world is quite different than it seems, it demands it. There is no question about it. Something very exciting is happening. Physicists have “proved,” rationally, that our rational ideas about the world in which we live are profoundly deficient.
In 1975, Henry Stapp, in a work supported by the U.S. Energy Research and Development Administration, wrote:
Bell’s theorem is the most profound discovery of science.9
The deduction of superluminal communication from the results of the Clauser-Freedman experiment rests upon an important assumption: namely, that the states of the measuring devices prior to the arrival of the photons in area A and area B do not matter. This is, after all, a reasonable assumption. Normally we say that the orientation of a measuring device prior to a measurement is not relevant to the result that we get at the time of a measurement. The result of an experiment depends upon the state of the measuring device at the time that the particle is detected by it, and not on the state of the measuring device before the particle gets there. However, superluminal communication cannot be deduced from the results of the Clauser-Freedman experiment without this assumption. Even though the pho
tons in the photon pair cannot exchange information via light signals while they are in flight (each is traveling away from the other at the speed of light), the measuring device in area A and the measuring device in area B, which are set prior to the beginning of the experiment, may have exchanged information in the conventional manner (via light signals propagating within space-time). In other words, in the Clauser-Freedman experiment the information about the setting of the measuring device in either region has sufficient time, traveling at the speed of light or less, to reach the other region before the particle arrives.
In 1982, Alain Aspect, a physicist at the Institute of Optics, University of Paris, in Orsay, France, conducted an experiment which was similar to the Clauser-Freedman experiment, but with one important difference: the settings on the measuring devices in Aspect’s experiment could be changed at the last minute (or, more precisely, at the last microsecond).10 Changing the settings on the measuring devices at the last minute insures that information about the setting of the measuring device in either area does not have sufficient time, traveling at the speed of light or less, to reach the other region before the particle arrives.* In other words, Aspect, in effect, performed Bohm’s thought experiment.
Like the Clauser-Freedman experiment (and several Clauser-Freedman-type experiments which had been performed in the meanwhile),11 Aspect’s experiment verified the statistical predictions of quantum mechanics. Because Aspect’s experiment, however, satisfied the conditions upon which the logical analysis leading to the phenomenon of superluminal transfer of information is based (that area A and area B are space-like separated) physicists were able to deduce this phenomenon solely on the basis of Aspect’s experimental results. This lent considerable credence to the conclusion which Stapp had reached five years previously. Wrote Stapp:
Quantum phenomena provide prima facie evidence that information gets around in ways that do not conform to classical ideas. Thus, the idea that information is transferred superluminally is, a priori, not unreasonable.
Everything we know about Nature is in accord with the idea that the fundamental process of Nature lies outside space-time…but generates events that can be located in space-time. The theorem of this paper supports this view of Nature by showing that superluminal transfer of information is necessary, barring certain alternatives…that seem less reasonable. Indeed, the reasonable philosophical position of Bohr seems to lead to the rejection of the other possibilities, and hence, by inference, to the conclusion that superluminal transfer of information is necessary.12
Thus, eighty-two years after Planck presented his quantum hypothesis, physicists have been forced to consider the possibility, among others, that the superluminal transfer of information between space-like separated events may be an integral aspect of our physical reality.*, †
Bell’s theorem showed that either the statistical predictions of quantum theory or the principle of local causes is false. It did not say which one is false, but only that both of them cannot be true. When Clauser and Freedman confirmed that the statistical predictions of quantum theory are correct, the startling conclusion was inescapable: The principle of local causes must be false! However, if the principle of local causes fails and, hence, the world is not the way it appears to be, then what is the true nature of our world?
There are several mutually exclusive possibilities. The first possibility, which we have just discussed, is that, appearances to the contrary, there really may be no such thing as “separate parts” in our world (in the dialect of physics, “locality fails”). In that case, the idea that events are autonomous happenings is an illusion. This would be the case for any “separate parts” that have interacted with each other at any time in the past. When “separate parts” interact with each other, they (their wave functions) become correlated (through the exchange of conventional signals) (forces). Unless this correlation is disrupted by other external forces, the wave functions representing these “separate parts” remain correlated forever.* For such correlated “separate parts,” what an experimenter does in this area has an intrinsic effect upon the results of an experiment in a distant, space-like separated area. This possibility entails a faster-than-light communication of a type different than conventional physics can explain.
In this picture, what happens here is intimately and immediately connected to what happens elsewhere in the universe, which, in turn, is intimately and immediately connected to what happens elsewhere in the universe, and so on, simply because the “separate parts” of the universe are not separate parts.
“Parts,” wrote David Bohm:
are seen to be in immediate connection, in which their dynamical relationships depend, in an irreducible way, on the state of the whole system (and, indeed, on that of broader systems in which they are contained, extending ultimately and in principle to the entire universe). Thus, one is led to a new notion of unbroken wholeness which denies the classical idea of analyzability of the world into separately and independently existent parts…13
According to quantum mechanics, individual events are determined by pure chance. We can calculate, for example, that a certain percentage of spontaneous positive kaon decays will produce an antimuon and a neutrino (63%), a certain percentage will produce a positive pion and a neutral pion (21%), a certain percentage will produce two positive pions and a negative pion (5.5%), a certain percentage will produce a positron, a neutrino, and a neutral pion (4.8%), a certain percentage will produce an antimuon, a neutrino, and a neutral pion (3.4%), and so on. However, quantum theory cannot predict which decay will produce which result. Individual events, according to quantum mechanics, are completely random.
Said another way, the wave function which describes spontaneous kaon decays contains all of these possible results. When the decay actually happens, one of these potentialities is converted into an actuality. Even though the probability of each potentiality can be calculated, which potentiality actually happens at the moment of decay is a matter of chance.
Bell’s theorem implies that which decay reaction occurs at a certain time is not a matter of chance. Like everything else, it is dependent upon something which is happening elsewhere.*
In the words of Stapp:
…the conversion of potentialities into actualities cannot proceed on the basis of locally available information. If one accepts the usual ideas about how information propagates through space and time, then Bell’s theorem shows that the macroscopic responses cannot be independent of far-away causes. This problem is neither resolved nor alleviated by saying that the response is determined by “pure chance.” Bell’s theorem proves precisely that the determination of the macroscopic response must be “nonchance,” at least to the extent of allowing some sort of dependence of this response upon the far-away cause.14
Superluminal quantum connectedness seems to be, on the surface at least, a possible explanation for some types of psychic phenomena. Telepathy, for example, often appears to happen instantaneously, if not faster. Psychic phenomena have been held in disdain by physicists since the days of Newton. In fact, most physicists do not even believe that they exist.*
In this sense, Bell’s theorem could be the Trojan horse in the physicists’ camp; first, because it proves that quantum theory requires connections that appear to resemble telepathic communication, and second, because it provides the mathematical framework through which serious physicists (all physicists are serious) could find themselves discussing types of phenomena which, ironically, they do not believe exist.
The failure of the principle of local causes does not necessarily mean that superluminal connections actually exist. There are other ways to explain the failure of the principle of local causes. For example, the principle of local causes—that what happens in one area does not depend upon variables subject to the control of an experimenter in a distant space-like separated area—is based upon two tacit assumptions which are so obvious that they are easy to overlook.
First, the principle of local c
auses assumes that we have a choice about how we perform our experiments. Imagine that we are doing Clauser and Freedman’s photon experiment. We have before us a switch which determines how the polarizers will be positioned. If we throw the switch up, the polarizers are aligned with each other. If we throw the switch down, the polarizers are oriented at right angles with each other. Suppose that we decide to throw the switch up and align the polarizers. Normally, we assume that we could have thrown the switch down and oriented the polarizers at right angles if we had wanted to. In other words, we assume that we are free to decide whether the switch before us will be up or down when the experiment begins.
The principle of local causes assumes (“…variables subject to the control of an experimenter…”) that we possess and can exercise a free will in the determination of how to perform our experiment. Second, and this is even easier to overlook, the principle of local causes assumes that if we had performed our experiment in a different way than we actually did perform it, we would have obtained some definite results. These two assumptions—that we can choose how to perform our experiment and that each of our choices, including those that we did not select, produces or would have produced definite results—is what Stapp calls “contrafactual definiteness.”
The fact, in this case, is that we decided to perform our experiment with the switch in the “up” position. We assume that, contrary to this fact (contrafactually), we could have performed it with the switch in the “down” position. By performing the experiment with the switch in the “up” position, we obtained some definite results (a certain number of clicks in each area). Therefore, we assume that if we had chosen to perform the experiment with the switch in the “down” position, we likewise would have obtained some definite results. (It is not necessary that we be able to calculate what these other results are.) Odd as it may seem, some physical theories, as we shall see, do not assume that “what would have happened if…” produces definite results.