by James Gleick
Their personal styles spill over into their theoretical work, too. Gell-Mann insists on mathematical rigor in all his work, often at the expense of comprehensibility… . Where Gell-Mann disdains vague, heuristic models that might only point the way toward a true solution, Feynman revels in them. He believes that a certain amount of imprecision and ambiguity is essential to communication.
Yet they were not so different in their approach to physics. Those who knew them best as physicists felt that Gell-Mann was no more likely than Feynman to hide behind formalism or to use mathematics as a stand-in for physical understanding. Those who considered him pretentious about language and cultural trivia felt nonetheless that when it came to physics he was as honest and direct as Feynman. Over a long career Gell-Mann made his vision not only comprehensible but irresistible. Both men were relentless on the trail of a new idea, able to concentrate absolutely, willing to try anything.
Both men, it seemed to a few perceptive colleagues, presented a mask to the world. “Murray’s mask was a man of great culture,” Sidney Coleman said. “Dick’s mask was Mr. Natural—just a little boy from the country that could see through things the city slickers can’t.” Both men filled their masks until reality and artifice became impossible to pry apart.
Gell-Mann, as naturalist, collector, and categorizer, was well primed to interpret the exploding particle universe of the 1960s. New technology in the accelerators—liquid hydrogen bubble chambers and computers for automating the analysis of collision tracks—seemed to have spilled open a bulky canvas bag from which nearly a hundred distinct particles had now tumbled forth. Gell-Mann and, independently, an Israeli theorist, Yuval Ne’eman, found a way in 1961 to organize the various symmetries of spins and strangeness into a single scheme. It was a group, in the mathematicians’ sense of the word, known as SU(3), though Gell-Mann quickly and puckishly dubbed it the Eightfold Way. It was like an intricate translucent object which, when held to the light, would reveal families of eight or ten or possibly twenty-seven particles—and they would be different, though overlapping, families, depending on which way one chose to view it. The Eightfold Way was a new periodic table—the previous century’s triumph in classifying and thus exposing the hidden regularities in a similar number of disparate “elements.” But it was also a more dynamic object. The operations of group theory were like special shuffles of a deck of cards or the twists of a Rubik’s cube.
Much of SU(3)’s power came from the way it embodied a concept increasingly central to the high-energy theorist’s way of working: the concept of inexact symmetry, almost symmetry, near symmetry, or—the term that won out—broken symmetry. The particle world was full of near misses in its symmetries, a dangerous problem, since it seemed to permit an ad hoc escape route whenever an expected relationship failed to match. Broken symmetry implied a process, a change in status. A symmetry in water is broken when it freezes, for now the system does not look the same from every direction. A magnet embodies symmetry breaking, since it has made a kind of choice of orientation. Many of the broken symmetries of particle physics came to seem like choices the universe made when it condensed from a hot chaos into cooler matter, spiked as it is with so many hard-edged, asymmetrical contingencies.
Once again Gell-Mann trusted his scheme enough to predict, as a consequence of broken symmetry, a specific hitherto-unseen particle. This, the omega minus, duly turned up in 1964—a thirty-three-experimenter team had to canvass more than one million feet of photographs—and Gell-Mann’s Nobel Prize followed five years later.
His next, most famous invention came in an effort to add explanatory understanding to the descriptive success of the Eightfold Way. SU(3) should have had, along with its various eight-member and ten-member and other families, a most-basic three-member family. This seemed a strange omission. Yet the rules of the group would have required this threesome to carry fractional electric charges: ? and – ?. Since no particle had ever turned up with anything but unit charge, this seemed implausible even by modern standards. Nevertheless, in 1963 Gell-Mann and, independently, a younger Caltech theorist, George Zweig, proposed it anyway. Zweig called his particles aces. Gell-Mann won the linguistic battle once again: his choice, a croaking nonsense word, was quark. (After the fact, he was able to tack on a literary antecedent when he found the phrase “Three quarks for Muster Mark” in Finnegans Wake, but the physicist’s quark was pronounced from the beginning to rhyme with “cork.”)
It took years for Gell-Mann and other theorists to generate all the contrivances needed to make quarks work. One contrivance was a new property called color—purely artificial, with no connection to everyday color. Another was flavor: Gell-Mann decided that the flavors of quarks would be called up, down, and strange. There had to be antiquarks and anticolors. A new mediating particle called the gluon would have to carry color from one quark to another. All this encouraged skepticism among physicists. Julian Schwinger wrote that he supposed such particles would be detected by “their palpitant piping, chirrup, croak, and quark.” Zweig, far more vulnerable than Gell-Mann, felt that his career was damaged. The quark theorists had to wrestle with the fact that their particles never appeared anywhere, though people did begin a dedicated search in particle accelerators and supposed cosmic-ray deposits in undersea mud.
There was a reality problem, distinctly more intense than the problem posed by more familiar entities such as electrons. Zweig had a concrete, dynamical view of quarks—too mechanistic for a community that had learned as far back as Heisenberg to pay attention only to observables. Gell-Mann’s comment to Zweig was, “The concrete quark model—that’s for blockheads.” Gell-Mann was wary of the philosophical as well as the sociological problem created by any assertion one way or the other about quarks being real. For him quarks were at first a way of making a simple toy field theory: he would investigate the theory’s properties, abstract the appropriate general principles, and then throw away the theory. “It is fun to speculate about the way quarks would behave if they were physical particles of finite mass (instead of purely mathematical entities as they would be in the limit of infinite mass),” he wrote. As if they were physical particles; then again, as if they were conveniences of mathematics. He encouraged “a search for stable quarks”—but added with one more twist that it “would help reassure us of the nonexistence of real quarks.” His initial caveats were quoted by commentators again and again in the years that followed. One physicist’s typically uncharitable interpretation: “I always considered that to be a coded message. It seemed to say, ‘If quarks are not found, remember I never said they would be; if they are found, remember I thought of them first.’” For Gell-Mann this became a permanent source of bitterness.
Feynman, meanwhile, had disregarded so much of the decade’s high-energy physics that he had to make a long-term project of catching up. He tried to pay more attention to experimental data than to the methods and language of theorists. He tried, as always, to read papers only until he understood the issue and then to work out the problem for himself. “I’ve always taken an attitude that I have only to explain the regularities of nature—I don’t have to explain the methods of my friends,” he told a historian during these years. He did manage to avoid some passing fashions. Still, he was turning back to a community after having drifted outside, and he had to learn its shared methods after all. It was no longer possible to approach these increasingly formidable, specialized problems as an outsider. He had stopped teaching high-energy physics; in the late sixties he began again. At first his syllabus contained no quarks.
By the late sixties and early seventies a new accelerator embedded in the rolling hills near Stanford University in northern California had taken the dominant role in the strong-interaction experiments that were so central to the search for quarks. The Stanford Linear Accelerator Center (SLAC) made a straight two-mile cut in the grassy landscape. Aboveground, cows grazed and young physicists in jeans and shirts—nearly a hundred of them—sat at picnic tables or walked
in and out of the center’s many buildings. Below, inside a knife-straight evacuated copper tube, a beam of electrons streamed toward targets of protons. The electrons achieved energies far greater than theorists had ever had to manage. They struck their targets inside an end station like a giant airplane hangar and then, with luck, entered a detector inside a concrete blockhouse, lined with lead bricks, riding on railroad tracks and angled upward toward the ceiling. Sometimes high-speed motion-picture cameras recorded the results, and elsewhere in the laboratory teams of human scanners guided an automatic digitizer that could read the particle tracks from—for a given monthlong experiment—hundreds of millions of filmed images. A single bubble chamber at the end of the particle beam, in its five-and-a-half-year useful lifetime, saw the discovery of seventeen new particles.
It was a tool for exploring the strong force—so called because, at the very short distances in the domain of the nucleus, it must dominate the force of electromagnetic repulsion to bind protons and neutrons (hadron was now the general term for particles that felt the strong force). Feynman had been thinking about how to understand the working of the strong force in collisions of hadrons with other hadrons. These were complex: at the high energies now available for studying short distances, hadron-hadron collisions produced gloriously messy sprays of detritus. The hadrons themselves were neither simple nor pointlike. They had size, and they seemed to have internal constituents—a whole swarming zoo of them. As Feynman said, the hadron-hadron work was like trying to figure out a pocket watch by smashing two of them together and watching the pieces fly out. He began visiting SLAC regularly in the summer of 1968, however, and saw how much simpler was the interaction offered by electron-proton collisions, the electron tearing through the proton like a bullet.
He stayed with his sister; she had moved to the Stanford area to work for a research laboratory, and her house was just across Sand Hill Road from the accelerator center. The physicists who would gather on the outdoor patio to listen to his stories that summer would see him slamming his open hands together in a boisterous illustration of a new idea he had. He was talking about “pancakes”—flat particle pancakes with hard objects embedded in them.
The Caltech connection was important to experimenters at SLAC, and by the late sixties the connection meant Gell-Mann far more than Feynman. Gell-Mann had created the scientific subculture of current algebra, the mathematical framework surrounding his quarks, and SLAC theorists thought of themselves as trying to generalize these tools to smaller distances, higher energies. At accelerators like SLAC, most of the thinking focused on the simplest reactions—two particles in, two particles out—although most of the actual collisions produced enormous flashes of many more particles. Experimenters wanted the most precise possible data, and precision was impossible in these bursts of detritus. Feynman chose a different point of view. He introduced a formalism in which one could look at the distributions of twenty or fifty or more particles. One did not have to be able to measure the momentum of each particle; in effect one could sum over all the possibilities. A Stanford theorist, James D. Bjorken, had been thinking along similar lines. An electron hits a proton; an electron comes out, along with a burst of immeasurable fragments. The emerging electron was a common factor. Bjorken decided to set aside the miscellaneous spray and simply plot the distribution of the energies and angles of the emerging electrons, averaged over many collisions.
He isolated a remarkable regularity in the data, a phenomenon he called “scaling”—the data looked the same at different energy scales. He did not know just how to interpret this. He had a variety of guesses, most framed in the language of current algebra. When Feynman arrived, Bjorken happened to be away; Feynman saw the graphed data without hearing a clear explanation of its origin. He suddenly recognized it, however, and he calculated long into the evening. It could be viewed as a graph of his pancake theory, the theory he had been toying with all summer on his own.
He had decided to cut through the incalculable swarming muddle of proton pieces by positing a mysterious new constituent that he called a parton, a name based inelegantly on the word part. (Finally he had an entry of his own in the Oxford English Dictionary.) Feynman made almost no assumptions about his partons except two: they were pointlike, and they did not interact meaningfully with one another but floated freely about inside the proton. They were an abstraction—just the kind of unobservable entity that physicists hoped not to have to fall back on—yet they were tantalizingly visual in spirit. They were pegs on which to hang a field theory of the old, manageable sort, with wave functions and calculable probability amplitudes. By analogy, quantum electrodynamics had its partons, too: the bare electrons and photons.
Feynman showed that collisions with these hard nuggets inside the proton would produce the scaling relations in a natural manner, unlike collisions with the puffy whole proton. He chose not to decide what quantum numbers they did or did not carry, and he most emphatically decided not to worry one way or the other about whether his partons were the fractionally charged quarks of Gell-Mann and Zweig.
By the time Bjorken returned, he found the theory group awash in partons. Feynman buttonholed him. He had idolized Feynman ever since taking an old-fashioned, historically organized quantum electrodynamics course at Stanford. “When Feynman diagrams arrived,” he said, “it was the sun breaking through the clouds, complete with rainbow and pot of gold. Brilliant! Physical and profound!” Now here was Feynman in the flesh, explaining Bjorken’s own theory to him with a new language and a new visual image. As he could instantly see, Feynman’s essential insight was to place himself once again in the electron, to see what the electron would see at light speed. He would see the protons flashing toward him—and they were therefore flattened relativistically into pancakes. Relativity also slowed their internal clocks, in effect, and, from the electron’s point of view, froze the partons into immobility. His scheme reduced the messy interaction of an electron with a fog of different particles to a much simpler interaction of an electron with a single pointlike parton emerging from the fog. Bjorken’s scaling pattern flowed directly from the physics of this picture. The experimenters grasped it instantly.
The parton model was oversimplified. It explained nothing that Bjorken could not explain, although Bjorken’s explanation seemed less fundamental. Partons required considerable hand-waving. Yet physicists clutched at them like a lifeboat. Three years passed before Feynman published a formal paper and many more before his partons finally and definitively blended with quarks in the understanding of physicists.
Zweig’s aces, Gell-Mann’s quarks, and Feynman’s partons became three paths to the same destination. These constituents of matter served as the quanta of a new field, finally making possible a field theory of the strong force. Quarks had not been seen or detected in the direct fashion of more venerable particles. They became real nonetheless. Feynman took on a project in 1970 with two students, assembling a vast catalog of particle data in an effort to make a judgment about whether a simple quark model could underlie it all. He chose an unconventional model once again, using data that let him think in terms of the electromagnetic field theory of the last generation, instead of the hadron-collision data that interested most theorists. For whatever reason, he was persuaded—converted into a quarkerian, as he said—although he continued to stress the tentativeness of any one model. “A quark picture may ultimately pervade the entire field of hadron physics,” this paper concluded. “About the paradoxes of the quark model we have nothing to add, except perhaps to make these paradoxes more poignant by exhibiting the mysteriously good fit of a peculiar model.” Younger theorists learned how to explain confinement—the quark’s inability to appear as free particles—in terms of a force that grew rapidly with distance, in strange contrast to forces such as gravity and electromagnetism. Quarks became real not only because ingenious experiments gave an indirect look at them, but because it became harder and harder for theorists to construct a coherent model in which
they did not figure. They became so real that Gell-Mann, their inventor, had to endure the after-the-fact criticism that he had not fully believed in them. He never understood why Feynman had created his own alternative quark and maintained a distinction that faded in the end. He missed no opportunity to call Feynman’s particles “put-ons.” Like Schwinger years before, he disliked the fanfare over a picture that he thought was oversimplified—anyone could use it.
Quarks were real, at least to physicists of the last years of this century. Partons were not, in the end. What is real? Feynman tried to keep this question from disappearing into the background. In a book assembled from his lectures, Photon-Hadron Interactions, he concluded:
We have built a very tall house of cards making so many weakly based conjectures one upon the other… . Even if our house of cards survives and proves to be right we have not thereby proved the existence of partons… . On the other hand, the partons would have been a useful psychological guide … and if they continued to serve this way to produce other valid expectations they would of course begin to become “real,” possibly as real as any other theoretical structure invented to describe nature.
Once again Feynman had placed himself at the center of modern theoretical physics. His language, his framework, dominated high-energy physicists’ discourse for several years. He wanted to move on again, or so he told himself. “I’m a little bit frustrated,” he said to a historian soon after he published his first parton paper.
