Computing with Quantum Cats (19 page)

But it is one thing to prove mathematically that the world is either unreal or non-local, quite another to prove it by experiment. Bell realized this, commenting at the end of his paper: “The example considered above has the advantage that it requires little imagination to envisage the measurements involved actually being made.” Little imagination, but a great deal of experimental skill. Astonishingly, it was less than ten years before the first such experiments were carried out—and it might have been sooner had Bell's paper not disappeared into a kind of publishing black hole.

Unlike the first of his two great papers, it was printed fairly quickly, in 1964. But, also unlike the first paper, it did not appear in a widely read or prestigious journal, largely because of Bell's reluctance, as a guest at various American research centers, to impose on his hosts by incurring the “page charges” applied by the more prestigious journals—a fee for
publication based on the number of printed pages occupied by the paper.

Bell did most of the work on the paper while at Brandeis, and completed it in Madison. As he explained to Paul Davies: “Probably I got that equation into my head and out on to paper within about one weekend. But in the previous weeks I had been thinking intensely all around these questions. And in the previous years it had been at the back of my head continually.”
¹⁵
Bell chose to send the fruits of all that thinking to a completely new journal,
Physics
, which had no page charges—in fact, it actually gave contributors a small payment for their papers; but this was no real benefit, since in those pre-Internet days the contributors in turn had to pay the journal for copies of the paper (reprints) to send to friends and colleagues. The two payments more or less canceled each other out. It seems that Bell's paper was accepted for publication (in the very first volume of
Physics
) because the editors mistakenly thought that it refuted Bohm's hidden variables interpretation of quantum mechanics.
¹⁶
Bell's paper did not make a big splash in 1964.
Physics
was not a widely read journal, and was closed down after only four years. Some of the people who did read the paper probably misunderstood it in the same way that the editors had. But the message got through to a tiny number of researchers, who ended up collaborating and competing in the first experiments to test Bell's theorem.

FIRST FRUITS

While Bell was at Brandeis, he gave a talk about his work and distributed a few copies (preprints) of his second paper, which had not yet been published. These had a rather
unprepossessing appearance, produced by a pre-photocopier duplicating process in smudgy purple ink. At first sight, they looked more like the work of a crank than that of a respectable physicist; but one of these smudgy preprints would have a big impact.

Somebody—just who is lost in the mists of time—sent one of the preprints to Abner Shimony, a physicist working at Boston University. But Shimony was not your average physicist. Born in 1928, the same year as Bell, he had graduated from Yale in 1948 with a combined major in philosophy and mathematics, and received his PhD in philosophy from Yale in 1953. Like Bell, he had turned to philosophy to seek the answers to the big questions about life, the Universe and everything; rather later than Bell, but influenced by Born's classic book, he decided that physics was more likely to provide those answers, and after two years of compulsory military service, in 1955 he embarked on a PhD in physics at Princeton. His time in the army, based in a mathematics section where one of the things he did was teach a course on information theory, was very valuable, Shimony recalls, because it gave him time to read up on undergraduate physics.

At Princeton, one of the first things his supervisor did was to tell him to “‘read the paper by Einstein, Podolsky and Rosen on an argument for hidden variables, and find out what's wrong with the argument.' So that was my first reading of the EPR paper, and I didn't think anything was wrong with the argument. It seemed to be a very good argument. I never saw anything wrong with it.”
¹⁷

Alongside his physics research, in 1959 Shimony joined the philosophy faculty of MIT. He lectured there on, among other things, the foundations of quantum physics. As a
part-time physicist he did not complete his second PhD until 1962, after which he took up a joint appointment in physics and philosophy at Boston University. A couple of years later, he received Bell's preprint out of the blue and, resisting the temptation to throw the scruffy document straight into the waste paper bin, read enough of it to realize its importance, then settled down to take a more detailed look. “The more I read it, the more brilliant it seemed.” Already familiar with the EPR argument, he was most impressed by the suggestion that these ideas could be tested by experiment.
¹⁸
What is more, he already knew of an experiment that had been carried out along these lines, and might be adapted to test Bell's theorem.

In a paper published in 1957, David Bohm and his student Yakir Aharonov had discussed entanglement and drawn attention to an experiment carried out back in 1950, for a completely different reason, which seemed to show entanglement at work. That experiment had been carried out by Chien-Shiung Wu and her student Irving Shaknov, and involved monitoring gamma rays (very high-energy photons, even more energetic than X-rays) produced when an electron meets a positron and annihilates. The relevant property of photons that is measured in such experiments is their polarization, which is analogous to the spin of an electron. The relevant point is that a photon can be polarized in any direction across its line of flight, like the baton twirled by a majorette. The Wu-Shaknov data suggested a correlation between the polarizations of separated photons (implying entanglement), but were not conclusive—which was hardly surprising, since the experiment had not been set up to measure such things.

Shimony did not follow up the implications immediately, but a little later he was visited by a graduate student, Michael Horne, looking for a problem to work on for his PhD. Shimony showed Horne Bell's paper, along with those of Bohm and Aharonov, and Wu and Shaknov, and suggested that he might try to design an experiment to test Bell's theorem. As a preliminary, Horne quickly found that he could construct a simple hidden variables model that would account for the Wu-Shaknov results, but that some much more sophisticated experiment would be needed to provide a proper measurement of Bell's inequality, using polarization measurements at a variety of different angles on pairs of photons produced by a single source. The good news, though, was that you didn't need gamma rays to do the job—ordinary photons of visible light would suffice. Indeed, it is easier to measure the polarization of such “ordinary” photons. All of this formed part of Horne's PhD thesis, accepted in 1970; but by then the experimental side was moving on.

In 1968, Shimony and Horne learned of experiments carried out at the University of California, Berkeley, by Carl Kocher and Gene Commins. They had measured the polarization of photons produced from calcium atoms in a process known as a cascade. They had taken measurements for just two polarizations at right angles to each other, in a simple experiment originally intended as a demonstration for an undergraduate physics course, and were unaware of Bell's work, so their results were inconclusive as far as testing hidden variables theory went; but clearly such an experimental setup could be adapted for testing Bell's theorem. Kocher and Commins weren't interested in pursuing this possibility, but Shimony and Horne now had a clear idea of
what kind of experiment they wanted done. All they had to do was find a laboratory with the right kind of apparatus and an experimenter willing to do the job. They found the apparatus (actually using a mercury cascade rather than a calcium cascade) in the laboratory of Frank Pipkin, a professor at Harvard; and they found their experimenter in the form of Richard Holt, a graduate student at Harvard. With Pipkin's approval, Holt (who already knew of Bell's papers) set out on what turned out to be a mammoth project to measure Bell's inequality. But he had scarcely started when the team was hit by a bombshell, in the form of the program for the spring 1969 meeting of the American Physical Society. Included in that bulletin was the abstract of a paper to be presented at the meeting by John Clauser, of Columbia University, who, the abstract revealed, was already working on the design of a similar experiment.

Clauser had graduated in physics from Caltech in 1964, then moved on to Columbia to work for an MA (1966) and PhD (1969). He was an experimenter, and his PhD work was in astrophysics—specifically, radio astronomy—but he was also interested in quantum physics, and had been astonished by Bell's papers, which he came across in 1967. His reaction was that the implications could not possibly be true, and he tried to find a counter-argument, but failed.
¹⁹
Realizing that the puzzle could only be definitively resolved by carrying out an experiment, he began scouring the scientific literature for papers describing such experiments. He also wrote to Bell, Bohm and de Broglie asking if they knew of any such experiments; the answers, of course, were all “no.” Bell later said that this was the first communication he had ever received about the paper he had published in 1964.
Having failed to find any relevant experiments (although he did find the paper by Kocher and Commins), Clauser began to work out how he could do such an experiment himself, to the frustration of his thesis supervisor, who told him he was wasting his time and should concentrate on his radio astronomy work. From the outset, his objective was to find evidence that Bell's inequality was
not
violated, and that the world really did operate in accordance with local reality. So he prepared a paper describing the kind of experiment he planned to do, and arranged to present it at the spring 1969 meeting of the American Physical Society.

Clauser's abstract prompted Shimony to telephone him with the news that his team was thinking along the same lines, and a suggestion that they might join forces. Clauser was not keen on the idea, until he learned that Holt already had the basis of the experimental apparatus needed to do the job. The four scientists (Clauser, Horne, Shimony and Holt) met up at the American Physical Society gathering, and got on well enough together to produce a joint paper (usually referred to in the trade by their initials, CHSH). This paper presented a generalization of Bell's theorem and gave practical details of the kind of experiment needed to test it, using polarized photons. But before the collaboration could proceed further, Clauser was offered a job at Berkeley to work in radio astronomy under the laser pioneer and Nobel Prize winner Charles Townes. This was a plum position in itself, but Clauser was just as interested in the fact that the Kocher and Commins apparatus was still at Berkeley, and might be adapted to carry out a proper test of Bell's theorem. Commins was not keen on the idea, because he regarded entanglement as a known feature of quantum mechanics and saw no need to
test it; but Townes, whose opinion carried more weight, was supportive, in spite of the fact that Clauser was supposed to be working on radio astronomy. The upshot was that Commins allowed a new graduate student, Stuart Freedman, to work on the project with Clauser, and Clauser never actually did any radio astronomy work worth mentioning. “Without Townes,” says Clauser, “I could never have done that experiment.” Clauser and Freedman were now competing with Holt, in a race to carry out the first proper test of Bell's theorem. But both teams had seriously underestimated the amount of time and effort this would take.

Fortunately, we do not have to go over all the trials and tribulations that the experimenters encountered. The bottom line is that in April 1972 Freedman and Clauser were able to publish a paper in the journal
Physical Review Letters
reporting that Bell's inequality was violated. “We consider,” they said, “these results to be strong evidence against local hidden-variable theories.” Remember that this was the
opposite
of what Clauser had set out to prove; it is, somehow, more compelling when experimenters find their expectations wrong than when they find what they hoped to find. Freedman was awarded his PhD for his part in this work in May 1972.

Meanwhile, Holt had found the opposite result! His experiment implied, but not very strongly, that Bell's inequality was not violated. A wealth of other experiments have since confirmed that he was wrong and Clauser and Freedman were right, but nobody has ever found out exactly what went wrong with Holt's experiment; the most likely explanation is that a glass tube in the apparatus had an un detected curve in it which affected the polarization of the photons passing through it. Nevertheless, Holt received his
PhD in 1972. Clauser, in a heroic effort to resolve the confusion, carried out his own version of Holt's experiment and found results in disagreement with local realism. These results were published in 1976. The same year, another researcher, Ed Fry at Texas A&M University, had carried out a third test of Bell's theorem, using a laser-based system, and also found a violation of Bell's inequality.

In 1978, Clauser and Shimony published a review summarizing the situation, and concluding: “It can now be asserted with reasonable confidence that either the thesis of realism or that of locality must be abandoned…The conclusions are philosophically startling: either one must totally abandon the realistic philosophy of most working scientists or dramatically revise our concepts of space-time.”

As these words make clear, by then it was almost impossible to find a loophole which would allow for the possibility of local reality. Almost, but not quite. One notable loophole remained; but it was about to be closed by an experiment carried out in Paris by a team headed by Alain Aspect.

CLOSING THE LOOPHOLE

The essence of the experiments to test Bell's theorem is that photons from a single source fly off in opposite directions, and their polarizations at various angles across the line of sight are measured at detectors as far away as possible from the source. The angle of polarization being measured can be chosen by setting one detector—a polarizing filter—at a particular angle (let's call it filter A), and another filter (filter B) at another carefully chosen angle on the other wing of the experiment. The number of photons passing through filter A
can be compared with the number of photons passing through filter B. The results of the first-generation experiments, including those of John Clauser, showed that the setting of filter A affected the number of photons passing through filter B. Somehow, the photons arriving at B “knew” the setting of A, and adjusted their behavior accordingly. This is startling enough, but it does not yet prove that the communication between A and B is happening faster than light (non-locally), because the whole experimental setup is determined before the photons leave the source. Conceivably, some signal could be traveling between A and B at less than the speed of light, so that they are in some sense coordinated, before the photons reach them. This would still be pretty spooky, but it would not be non-local.