Quintessence is no other than a quality of which we cannot by our reason find out the cause.

In a way, cosmology is like child’s play.

After all, for decades astronomers have described the universe as an expanding balloon. They knew it was getting bigger. They just weren’t sure whether it would keep growing forever or someday start to deflate.

Now they’re afraid that the balloon might burst.

In a plot twist worthy of Scream 3, the universe shocked cosmologists in the late 1990s with behavior almost as surprising as the original discovery that space is expanding. The new surprise was hyped by the journal Science as the breakthrough of the year in 1998, even though a lot of experts still didn’t believe it. But by the time the

twenty-first century rolled around, the skeptics found it harder and harder to deny the new dogma: the universe-balloon is not only expanding, but it is inflating at a faster and faster rate. The best explanation seems to be that the universe is full of some mysterious cosmic fluid, unlike any known substance or force, utterly invisible, and therefore called “dark energy.”

Nobody really knows what dark energy is or why the universe should be full of it. But if it really is out there, its existence should not have been all that surprising. It had, after all, been prediscovered.

“Yes, we anticipated it,” the cosmologist Lawrence Krauss told me in an interview, “but I never believed it.”¹

Nowadays many people do believe in dark energy, even if it’s too soon to say exactly which prediction it fulfills. Dark energy may represent the realization of an old prediction of Einstein’s—one that he disavowed. Or it may signal a more recent possible prediscovery—a strange, variable energy field that has come to be known as quintessence.

Quintessence is not the name of some cosmic perfume. It is rather the revival of an old idea of Aristotle’s. In his day, mainstream physicists believed earthly substance to consist of four elements (earth, air, fire, and water). But for the heavenly sphere of stars and planets, Aristotle invoked a surreal substance generally known as the ether. Later on, in Latin translation, this “fifth essence” came to be called quinta essentia.

Twenty-three centuries after Aristotle, Einstein’s theory of relativity did away with the ether. But Einstein later found that the universe his theory described wanted to grow or shrink. So he added a term to his cosmic equations, known as the cosmological constant, which now threatens to revive something similar in many ways to the ether that he had done away with. In a sense, what Einstein took away, Einstein gave back.

The cosmological constant story begins with Einstein’s 1917 paper on cosmology, the first to apply the lessons of general relativity to the universe. It’s in this paper that he explored the cosmological ramifications of what I call “Einstein’s other equation.” Almost everybody has heard of Einstein’s first famous equation, E = mc²; but very few outside the physics world could quote his other equation, G_µv = –κT_µv. You can see why.²

So let’s put it in words. In a nutshell the Einstein equation says that geometry of space is determined by all the sources of gravity in that space. (Technically, of course, we should be talking about the geometry of spacetime—space and time combined, as relativity requires. We’ll deal with that distinction later.) The left side of the equation (G_µv) contains the symbol for representing how space is curved. The right side (–κT_µv) contains the total of all the matter and energy (such as radiation) and pressure.

If the notion of curvature of space bothers you, don’t worry, you’re not alone. It is not the sort of thing that is easy to visualize. But you can see the connection between gravity and curvature if you think about something you can visualize, the surface of the Earth.

Earth is, more or less, a big ball, with a curved surface, and that curvature influences the path of objects trying to move in a “straight” line. Picture two ships at sea, steaming along on parallel paths, both heading straight north. As they near the North Pole, the ships would get closer and closer together. A bird flying overhead (let’s call it Isaac) might remark that the ships appeared to be attracting themselves to each other. But then another bird (Albert) comes along and disagrees. No, says Albert, the ships get closer because the water they’re moving on is curved. The curvature of the surface is moving them closer. In fact, Albert and Isaac might get into quite an argument over this, but they might conclude that the curvature of a surface and the attraction of two bodies seem to be just two ways of

describing the same thing. In the end, though, Albert would be able to point out tiny differences in the ships’ motion that differed from what Isaac would have expected.

In a similar way, Albert Einstein’s theory of gravity makes subtle predictions that differ from Isaac Newton’s. Objects move the way they do through space not because they are tugging on one another by way of an invisible gravitational force, but because space is curved—moving objects follow the curvature. Einstein’s brilliant insight was that massive objects themselves cause the curvature of the space. A star or planet warps the space around it the way an obese gymnast distorts a trampoline. In the words of John Wheeler, “mass grips spacetime, telling it how to curve; spacetime grips mass, telling it how to move.”³ Einstein’s other equation captures the mathematics of that insight.

When I first encountered that equation, I was baffled by the part about pressure. But I think it was because nobody ever bothered to tell me in school that pressure is part of the force of gravity. And no doubt nobody bothered because pressure rarely matters. But in principle, space’s curvature at any location is affected not only by the density of matter sitting in it but also by the pressure exerted by any matter or radiation in the vicinity. Einstein’s other equation takes all that into account.

Ordinarily, pressure is negligible. At the surface of the Earth, for example, the air pressure is trivial compared to the density of the mass in the Earth itself. And what really matters in curving space is the density of the energy, and to get the energy equivalent of a mass you multiply it by the speed of light squared. Consequently matter’s mass-energy density contributes vastly more than its pressure to the curvature of space. Similarly, radiation pressure is usually pretty small—you can’t lift an object against the force of gravity by shining a flashlight on it from below.

Under certain circumstances, though, pressure can be high

enough to be important—as in the center of a neutron star, where not only mass density but pressure as well is extremely high. Similarly, when the universe was very young and still a glowing fireball, the density of radiation was high and contributed a significant amount of pressure. In fact, in those days, the radiation-energy density exceeded the matter-energy density. Therefore cosmologists say that the universe was “radiation dominated” back then. But as the universe expanded, it cooled, and the radiation density diminished. Eventually the radiation’s declining density dipped below the density of matter, and the universe became “matter dominated.”⁴

One of Einstein’s great realizations in the 1917 paper was that some third sort of component could compete for dominance with matter and radiation in affecting the curvature of space. In fact, Einstein believed he needed some such component to make sense of the universe. Without a new ingredient, the math told him, the universe would not stay still but would either expand or contract (depending on what initial conditions you applied the equations to).⁵ Einstein, trapped in the paradigm of a static universe, and unaware of any evidence to the contrary, decided that the equations describing spacetime’s geometry must be modified to prevent the universe from expanding or collapsing. He did not want the balloon to deflate or burst. So he added a fudge factor to his equation, thereby preserving a universe characterized by lack of character—an utterly changeless, finite universe in which the motions of stars were slow. This fudge factor, which came to be known as the cosmological constant, represented some sort of repulsion in space that prevented matter from collapsing because of gravitational attraction.

Einstein was vague about what this term actually represented physically. But later it was identified with some sort of energy residing in the vacuum of space. In other words, space without matter and radiation in it wasn’t exactly empty. Some sort of energy inherent in the balloon’s elastic kept it from changing in size.

A strange feature of this energy was the way it affected the pressure part of Einstein’s other equation. The math said that this mystery component of the cosmos must exert negative pressure.

This idea can be very confusing, because it seemed like Einstein wanted a repulsive force to keep the universe from collapsing. You might think a positive pressure would exert a repulsive force. A negative pressure sounds more like a suction, pulling stuff in rather than pushing it out. But that’s not the way Einstein’s other equation works. Pressure (positive pressure) contributes to the attractive gravitational force. Negative pressure does the opposite, so it has the effect of repulsion. (Sometimes you see this effect referred to as “antigravity,” but that doesn’t seem quite proper, since pressure is a legitimate part of the gravitational equation. Negative pressure is just a part of gravity that affects the overall result in an unusual way.)

In any event, because positive pressure enhances gravitational attraction, negative pressure exerts a repulsive effect. So the natural tendency of attractive gravity to pull things together—and make the universe collapse—would be countered by the repulsive effects of a negative pressure, and the universe could maintain its static state.

Obviously there was no physical evidence whatsoever at the time for such an energy field, or substance, or fluid, or whatever it was. Not only that, it was not something that emerged naturally from Einstein’s theory. He had to add it into his equation, simply because he thought the universe differed from what his original equation suggested. Einstein acknowledged as much when he noted that his added term was not required by anything known about gravity. “That term is necessary only for the purpose of making possible a quasi-static distribution of matter, as required by the fact of the small velocities of the stars,” Einstein wrote.⁶ In other words, since stars weren’t flying toward or away from each other, spacetime must be static. (At the time, it wasn’t yet clear that stars are clumped into galaxies.)

Einstein’s use of a mistaken conception of the cosmos to deduce

the existence of vacuum energy strikes me as a most unusual route to prediscovery. Dirac prediscovered antimatter because the equations told him to. In this case, Einstein seems to be telling the equations what to do. The theory itself did not require the addition of this fudge factor, but Einstein appended it to his equation anyway. Perhaps this is a good example to keep in mind in the effort to understand how prediscovery is possible.

In any event, Einstein’s term was challenged a few years later by Friedmann, who (as perhaps you’ll remember from Chapter 5) managed to escape the static universe dogma and establish the mathematical basis for contraction and expansion. Einstein was evidently not impressed, as he clearly retained belief in a static universe until the end of the 1920s. But then Hubble analyzed the motion of distant galaxies and found that they did move away from one another very rapidly—the farther away they got, the more rapidly they receded. The universe, it seemed, was not static after all.

Thereupon Einstein, according to George Gamow, called the cosmological constant the “biggest blunder he ever made in his life,”⁷ a story retold perhaps more often than any other anecdote in the history of science. As far as I have been able to tell, there is no evidence for this assertion other than Gamow’s remark in his autobiography, and Gamow’s reputation for anecdotal accuracy is somewhere near the same level as Ronald Reagan’s. But whether Einstein actually used the phrase “biggest blunder” or not, he did abandon the cosmological constant after Hubble’s discovery.⁸ In a later edition of his book on relativity, Einstein noted that he would never have invented the cosmological constant if Hubble’s work had been done sooner. “If Hubble’s expansion had been discovered at the time of the creation of the general theory of relativity, the cosmologic member would never have been introduced.”⁹

While Einstein attempted to bury his mistake, it was never quite forgotten. His original intent, to keep the universe safely static, no longer made any sense. But other cosmologists saw that the cosmological constant might still matter, even in an expanding universe. Students of quantum physics knew that even a vacuum could be full of energy, simply because quantum physics does not allow the energy of anything to be exactly zero. And while Einstein wanted vacuum energy to exert a repulsive force to keep the universe from contracting, such energy could just as easily alter the rate at which the universe is expanding. Vacuum energy might therefore influence measurements of the universe’s age and size. In fact, an apparent mismatch between the universe’s age and expansion rate was just the problem that many astronomers hoped that vacuum energy could solve. So the cosmological constant, the fudge factor in Einstein’s equations representing vacuum energy, came back from the grave.

The age-expansion mismatch was a tough problem. For the universe, just as for a person, only certain combinations of size, age, weight, and growth rate make any sense. A 25-year-old man should weigh more than 40 pounds and not be rapidly growing, for example. But through much of the 1990s, the mass, expansion rate, and age of the universe seemed out of sync. Some measures of the expansion rate even indicated that the universe might be younger than the oldest stars, a rather blatant paradox. Adding Einstein’s cosmological constant into the mix altered the calculations of the universe’s age, resolving the discrepancy.

Vacuum energy could also solve the problem of the universe being “underweight.” Most theorists believed that the geometry of space described by Einstein’s other equation should conform to the standard Euclidean geometry taught in school. In other words, space would be flat, and two parallel light rays zipping through empty

space would remain a constant distance apart. But Einstein’s equation allows for space to be curved. Space might be curved like a ball (a “closed” universe with a lot of mass) or curved inward like a saddle (a lightweight, “open” universe). In a closed universe, parallel lines would converge, like longitude lines on the surface of the Earth. In an open universe, parallel lines would diverge.

In the simplest cases, the overall geometry of space would be determined by how much matter it contained. More than a certain “critical” amount, and the universe would be closed, eventually to collapse. Less than the critical amount, and the universe would be open and expand forever. A “borderline” universe, with precisely the right amount of mass, would expand forever, but at a slower and slower rate, and the geometry of space would be essentially flat. (Euclid’s geometry would still be no good in the presence of a massive body, but overall, on average, space would seem to be Euclidean.)

If Alan Guth’s inflation idea really did describe the birth of the universe correctly, we’d live in the borderline universe, and space’s geometry should be very, very close to flat. If so, the universe must contain precisely the right amount of matter to bring the expansion almost to a halt. But various methods of measurement strongly suggested that the universe did not contain that much matter. Adding vacuum energy to the mix, though, could make up the difference. (Remember, all the energy adds up to affect space’s curvature.) It seemed that if you wanted the universe to be flat, you needed to revive Einstein’s cosmological constant, or something like it.

On the other hand, there was no real evidence that the universe was flat—that was more or less a theoretical prejudice. And the theoretical basis for vacuum energy didn’t seem so sound, either. Calculations based on theory indicated that empty space should be teeming with an immense amount of vacuum energy—much more than observations allow. The equations predict the amount of vacuum energy to

be higher than it actually is by a factor equal to 1 followed by 120 zeros. Nobody knows why the prediction is so far off.

“I think it’s the biggest unsolved problem in cosmology today, and it’s our biggest embarrassment,” says Josh Frieman, a cosmologist at Fermilab.¹⁰

For a long time, most cosmologists favored the idea that there must be some reason why all the vacuum energy gets canceled out somehow, leaving zero. It was easy to imagine that some simple (but unknown) law of physics was at work, eliminating all the unwanted vacuum energy. It was not as easy to imagine that some mystery process would eliminate most of it, but not quite all. In short, the problem seemed too hard to solve. Therefore everybody went merrily on their way, assuming that the cosmological constant must be zero and therefore they could ignore it.

Along the way came some observations that provided comforting support for that point of view. I remember an astronomy meeting in 1992 where astronomers reported a study of the light from distant quasars using the Hubble Space Telescope. Quasars are the bright beacons on the universe’s edges (from our point of view) shining beams of light across space like powerful cosmic flashlights. We can learn a lot about what’s between the quasars and us by studying the light that they emit, because on its way to us the light is affected by what it passes through. Or around. Large masses, for example, exert gravitational force that bends the light as it goes by, the phenomenon known as gravitational lensing. A large mass could “lens” a distant quasar in such a way as to generate a separate image, so astronomers would see two images of the same quasar.

Now there’s no point in going into the details, but vacuum energy would influence how many such double images you would see when surveying the sky for quasars. At the 1992 meeting, reports of such a study suggested that the amount of lensing was just what it should be, with no discernible effect from any vacuum energy. John

Bahcall, a prominent astrophysicist from the Institute for Advanced Study in Princeton, was emphatic. There might be a very small amount of vacuum energy, below the threshold of the telescope’s capabilities to detect, but not enough to be cosmologically interesting. “It’s too small to help you with any known astronomical puzzles,” he said. “It’s too small to be any good for anything.”¹¹

Other experts, however, said that this result was uncertain enough to allow different interpretations. So when I wrote about it, I suggested that it was unwise to jump to conclusions. “Don’t be surprised,” I wrote, “if the cosmological constant comes back from the dead again.” And it did.

Throughout the 1990s, cosmologists sought ways of solving their problems without adding dark energy into the cosmological mix. Some measurements of the expansion rate became more compatible with the ages of the oldest stars. And maybe, some experts suggested, the universe just didn’t have enough matter to make space flat because space wasn’t really flat. But as the twentieth century came to an end, new observations challenged all those attempts to avoid the need for dark energy.

Eventually, decisive evidence in favor of flatness came from the cold microwave background radiation, which contains an imprint of how the tiny seeds of matter were arranged in the very early universe. Taking the sky’s temperature reveals the starting place for the evolution of the huge galaxies and galactic clusters that grew from those tiny seeds. Satellites, telescopes, and instruments on balloon missions can measure how much the radiation temperature differs between points separated by different angles on the sky. At some angles, the temperature difference is greater than at others.

The key issue is finding what angle has the greatest temperature

difference. That peak difference depends on the influence of the cosmic equivalent of sound waves. In the early moments of the universe, gravity would compress radiation particles (or photons), but the pressure of the photons would eventually fight back and rebound. The compression-rebound cycle generates oscillations, or waves, that boost the temperature differences. The angle of maximum temperature difference can be calculated based on the assumption that space is flat. Any deviation from the expected angle would suggest that the universe is not flat after all. But by 2001, several experiments had measured the angle precisely enough to confirm that the universe was, in fact, very close to flat. So some sort of funny dark energy would seem necessary to explain everything else.

Even before the compelling evidence for flatness came in, another key advance had generated support for dark energy. That advance involved finding a way to measure the expansion rate of the universe in the distant past. If the universe in fact contained a lot of dark energy, the negative pressure would play havoc with the standard ideas of how the universe had expanded. For one thing, the universe would be expanding faster today than it was a billion years ago. But before the mid-1990s, astrophysicists had no good tool for measuring the universe’s expansion history. Then they found a (standard) candle in the darkness.

At an astronomy meeting in San Antonio, in January of 1996, I first heard of a new effort to measure cosmic distances by exploiting the brightness of exploding stars, or supernovas. A particular version of supernova, known as Type 1a, seemed especially good for this purpose. In theory, they should all explode with the same brightness. Thus their distance could be inferred by how bright or dim they appeared.

Of course, it isn’t that simple. In real life Type 1a supernovas do vary some in intrinsic brightness. And what’s worse, distance alone doesn’t make them dim. Dust gets in the way, too. A bright nearby

supernova shrouded in dust might appear to be far, far away. And if you don’t know how far away those objects are, using them to calculate the expansion rate is impossible.

But at the San Antonio meeting, scientists from the Harvard-Smithsonian Center for Astrophysics described a way to correct the calculations when dust interferes. The trick was watching how rapidly a supernova dims over time. An intrinsically bright supernova grows dim rather slowly; dimmer ones fade away more rapidly. And the dimmer ones tend to be redder than the bright ones, which are bluer. Dust’s effect can be inferred by its effect on color. Dust blocks blue light but lets red light through, as in those reddish West Texas sunsets. By comparing the observed color with the expected color (based on how rapidly the star dims), the presence of dust can be detected and compensated for in the calculations.

At the same time, new computer-controlled telescope search strategies made it possible to track numerous supernovas in distant galaxies. So the race was on. Two separate teams embarked on such computer-aided supernova searches to find as many Type 1a’s as possible. Two years later, the first results were in, and the shocking conclusion was revealed: it looked as though the universe really was expanding faster today than yesterday.

Many experts remained skeptical. But all of a sudden the idea of dark energy had to be taken seriously. In May 1998, dozens of leading figures in the field gathered at Fermilab for a workshop to discuss the implications of the accelerating universe, and dark energy became the topic of the day.

It is hard to relate the sense of excitement that flowed through that meeting, as though the cosmologists themselves were animated by a new form of energy. At a lunch with reporters, several of the scientists gushed over the new findings, like kids challenged to figure out how a wonderful new toy worked.

“We have no real clue for what this stuff is,” said Josh Frieman.

“It’s a monumental issue both for fundamental physics and for cosmology,” said Paul Steinhardt.

“We were thrown a curve ball,” said Michael Turner. “If it holds up, it’s a surprise that the universe is accelerating rather than slowing down.”¹²

In his talk at the workshop, Turner coined the name “funny energy” for whatever was in the vacuum and suggested that it was nothing other than the cosmological constant itself. “It was good enough for Einstein,” Turner said. “It ought to be good enough for us.”¹³

But if funny energy is the vacuum energy described by Einstein’s cosmological constant, it suggests a curious coincidence. Vacuum energy arises from space in a roughly constant amount. As space stretches out, creating more space, more vacuum energy is created, too. So the density of vacuum energy in the universe remains constant at all times. For matter and other forms of energy, though, the situation is much different. There’s only so much matter around, so as space expands, the density of matter diminishes.

Back in the universe’s youth, the vacuum energy could not have been very significant. If the vacuum energy density exceeded the matter density from the outset, space would merely have blown apart and no chunks of matter would have been able to coalesce into stars or planets or anything else. Obviously, therefore, matter was denser than vacuum energy in the past. But now, the supernova observations suggest, the vacuum energy density is more important, driving an accelerating expansion. In cosmological terms, the switchover wasn’t so long ago. Humans seem to have come upon the scene soon after the time that vacuum energy density and matter density were about the same. (In this case, “soon” means a few billion years, but for cosmologists that’s like the day before yesterday.)

And so we have what Turner called the “Nancy Kerrigan problem.” Why me? Why now?¹⁴ In more technical language, it is called the cosmological coincidence problem.

At first glance, it does seem to be an amazing coincidence. Why should human astronomers be so lucky to be around at just the epoch when a changing number (the matter density) is just about equal to a constant number (the vacuum energy density)? At the Fermilab meeting, Steinhardt and colleagues suggested that the coincidence could be explained. Perhaps, they argued, the solution lies in returning not to Einstein, but to Aristotle. Maybe the dark energy is not Einstein’s cosmological constant after all, but another mysterious energy form that they called quintessence.

Quintessence, supposedly, would be a “field,” some sort of mysterious invisible fluid permeating all of space, like Aristotle’s fifth essence. In fact, it made sense to call this new field a fifth essence as well. In Aristotle’s day, remember, the standard model of ancient physics proposed that everything was made of four elements: earth, air, fire, and water. After two millennia or so of advances, physicists can still divide the matter and energy in the universe into four new categories: ordinary matter (made mostly of protons and neutrons); radiation; neutrinos (fast, light, ghostly particles); and cold dark matter, exact identity unknown. If there’s something else—namely, dark energy—it seems logical to call it a fifth essence—or quintessence.

Steinhardt and colleagues coined the term quintessence in a 1998 paper, but the basic idea had been around for a while. Josh Frieman and Chris Hill, another Fermilab physicist, and collaborators had proposed something very much like it in 1995.¹⁵ But nobody paid much attention until dark energy screamed out for an explanation.

Like the cosmological constant, quintessence would exert “negative pressure,” kind of like the way a rubber sheet pulls in on itself. Or you could think of it as a stretched-out spring that wants to pull itself back in. Unlike the cosmological constant, though, it could interact

with other stuff, and that property provides a possible explanation for the Nancy Kerrigan problem.

If quintessence exists, it fills all of space with some bizarre form of matter-energy that differs from Einstein’s cosmological constant in an important respect—it isn’t constant. Quintessence could solve the cosmological coincidence problem by being changeable; its strength could differ from place to place and time to time, being “thicker” in some places than others. “It’s not like any other kind of matter that we’re aware of,” said Robert Caldwell, one of Steinhardt’s collaborators.¹⁶

Of course, it’s one thing to say that space is filled with energy that changes in strength over time in just the right way to solve all your problems. It would be much more satisfying scientifically to know why it changed in such a way as to solve your problems. The answer wasn’t obvious at first, but by a year or so later Steinhardt had worked out an attempted explanation. The Nancy Kerrigan problem might be solved, he said, by something called “tracker fields.”

At the risk of getting a little technical, and with the cost of a brief digression, it helps to know that quintessence would be what physicists call a scalar field. And so it seems to me that the time has come to try to explain what a field is. Physicists throw the term field around as though they were talking about farms or baseball. But it isn’t obvious to outsiders what the term field really means. Basically you can think of a field as something sitting in space—inseparable from space— that affects other things in space. In modern physics, almost everything comes from a field. Particles of matter are not tiny little balls, but rather are twists or knots in a matter field of some sort. But those fields are quantum fields, more complicated than the fields we need to talk about now. We can stick to classical fields.

Classical fields come in three flavors, scalar, vector, and tensor. Gravity (or, a gravitational field) is a tensor field. Multiple factors determine its magnitude, or strength, at any one point. A tensor field is actually just a more complicated version of a vector field, which has a strength and a direction at any point in space. Electromagnetic fields are vector fields—they have a strength and a direction at every point. If you put a little magnet at a point in a magnetic field it will orient itself in a specific direction. This is how a compass works, why the needle points north, because it is influenced to align itself in that direction by Earth’s magnetic field.

Scalar fields, on the other hand, do not have a direction, just a magnitude. That is, a scalar field can merely be “thicker” at one place than another, or at one time or another. That’s why the amount of energy residing in a quintessence scalar field can change over time.

Picturing the energy in a scalar field is not very easy, since scalar fields and energy are both invisible. But you can picture a graph of the amount of energy in a scalar field, and think about the field as representing an object moving along the graph. So let’s picture the scalar field as a bowling ball rolling over an “energy terrain” and see if that helps.

Now all you need to know is the difference between potential energy and kinetic energy. The bowling ball’s position on the terrain tells us its potential energy. On a high peak, potential energy is high, because as the ball rolls down the side of the hill it will pick up speed, and its potential energy will be transformed into kinetic energy, the energy of motion. If this bowling ball represents a scalar field, then the scalar field is thick on peaks and thin in valleys.

Now the natural tendency of a scalar field, just like a bowling ball, is to seek a state of zero energy—to give its energy up. Possessing a lot of energy tends to make you unstable, and nature abhors instability. Just as water runs downhill to the lowest point it can reach, energy fields seek their minimum possible value. The potential en-

ergy of the field (bowling ball) is just a measure of how far it is above the zero point it seeks.

The kinetic energy, on the other hand, is related to how fast the bowling ball moves toward the zero point. If it rolls downhill slowly—which means the strength of the field is changing slowly over time—its kinetic energy is low. If it drops rapidly, the kinetic energy is high.

The key point here is that kinetic energy makes positive pressure. A field that changes slowly has very little kinetic energy, too little to matter compared to its negative pressure. But a field that changes rapidly will have a lot of kinetic energy—more than its potential energy. If the kinetic energy exceeds the potential energy, the field will have positive pressure. That won’t work with quintessence, which is supposed to have negative pressure. Therefore quintessence must be a scalar field that does not change very rapidly over time, so that it will have negative pressure. Because it must be changing over time very slowly to keep the pressure negative, quintessence is called a “slowly rolling” scalar field. It’s like a bowling ball rolling down the energy slope very slowly to keep the kinetic energy small.

If quintessence is a slowly rolling scalar field, it would mimic the effect of Einstein’s cosmological constant. But that still doesn’t explain why the magnitude of quintessence is just right to make the universe accelerate now, instead of a long time ago. At the Fermilab meeting in 1998, nobody had a good explanation.

But before long, Steinhardt and colleagues had cooked up a version of quintessence that had the potential to solve the problem. He described it at a meeting for science writers in Hershey, Pennsylvania, in 1999. Suppose, he said, that quintessence is a “tracker” field. In

other words, quintessence interacts with other stuff in space in such a way that its strength can be suddenly changed when other things change.

For instance, it’s plausible that when the universe was very young, quintessence interacted with the radiation in a way that made the quintessence strength decrease as the radiation density decreased. So as the universe expanded, and the radiation density dropped, so did the strength of the quintessence field. (In other words, quintessence “tracked” radiation.)

But then, at some point, the radiation density fell below the matter density. The universe became matter dominated. A tracker field might then respond in a different way—instead of continuing to fall in density, it might just stop wherever it was and then stay constant. But the matter density would continue to drop, eventually falling below the now-constant quintessence density. And the universe would begin to accelerate.

That could explain the Nancy Kerrigan coincidence. If matter domination triggers the quintessence field to stabilize, it makes sense that people would soon be around to talk about it, because matter domination also triggers the formation of galaxies, stars, and planets. It is no longer such a mysterious coincidence, but an obvious coincidence—the same event that triggers the processes needed to make people also makes dark energy take over the cosmos.

Of course, some serious details remain to be worked out, and quintessence has become a source of countless new research papers in the past couple of years. One especially interesting question is what quintessence implies for the future of the universe. If the dark energy is Einstein’s cosmological constant, the universe’s fate is sealed. For if it’s constant, and will never diminish, once it dominates it will always dominate, and the universe will expand forever. But if the dark energy is quintessence, there is hope. If it changed in strength once, it could change again.

Turner and Krauss explored this issue in a 1999 paper. The problem of the fate of the universe, they said, had been badly oversimplified in textbooks. For years astronomers had promised that the answer to the universe’s fate was just around the corner. All they had to do was measure how much matter the universe contained. With a little help from some equations provided by Einstein, knowing the mass of the cosmos should have made it simple to say what its future held in store. Too much matter, and the universe would someday stop expanding and start shrinking, ultimately crushing itself—and everything in it—into nothingness. (The movie version of such a future would be called The Big Crunch.) Too little matter, on the other hand, and the universe would grow, cooling as it expanded, getting bigger and colder forever—more like The Big Chill.

More careful expositions pointed out that, technically, it’s not the amount of matter that matters, but its density: A dense universe dies young, like James Dean; a not-so-dense universe just keeps getting older, like Strom Thurmond.

But dark energy changed the rules of the game. Density, Krauss and Turner proclaimed, does not determine destiny. Now it seems that the fate of the universe not only remains unknown, it may be forever unknowable. In other words, it’s impossible to say how long forever will be.

If the dark energy is Einstein’s cosmological constant, and is therefore always the same strength, it will inevitably exert more repulsive force than the attractive gravity of all the universe’s matter. As the universe expands, the matter density naturally diminishes. Even if astronomers measured a matter density that seemed high enough to crunch the universe, they could never be sure that a tiny amount of cosmological constant might not be lurking in space below the threshold of detection. Vacuum energy paltry enough to

escape detection today would still dominate the universe in the distant future, ensuring endless expansion.

But if the dark energy is quintessence, it might not be constant at all, but could change over time. In that case its future is unpredictable. “How do you know it’s not going to disappear?” Turner said.¹⁷

Consequently today’s measurements of matter density cannot determine the universe’s future. “Basically you may just as well throw your hands up in the air,” Turner said.

Even if it seemed the universe had far too little matter to collapse (and that seems to be the case today), there are no guarantees. Just as a tiny cosmological constant would someday take over and make the universe expand forever, a tiny but negative vacuum energy would someday cause expansion to stop and collapse to begin. The negative energy would suck space in on itself.

Negative energy sounds bizarre, but it shows up frequently in theory and sometimes in the lab. And a constant amount of negative dark energy filling the universe—a “negative cosmological constant”—might very well be produced by various physical processes. Physicists Je-An Gu and W-Y. P. Hwang of the National Taiwan University in Taipei made that point in a 2001 paper. If the universe keeps expanding, any cosmological constant, positive or negative, eventually becomes the most important factor in determining the universe’s fate, they pointed out.

“A negative cosmological constant, even if nearly zero and undetectable at present, can make the universe collapse eventually,” the physicists wrote.¹⁸

Krauss made the same point in his book Atom. “Once we acknowledge the possibility that empty space can have energy, our ability to unambiguously predict the future of the universe goes out the window,” he wrote. “A negative energy in empty space could still result in an extra-attractive force. . . . This would eventually stop the current expansion.”¹⁹

“We really have no direct way of knowing the future,” Krauss told me in an interview. No observation can be accurate enough to say for sure that a small amount of dark energy isn’t hiding from view.

So scientists cannot say what the future of the universe will be. The universe may end in a hot crunch or a cold, endless expansion, and scientists will not be able to answer Robert Frost’s question about fire or ice ahead of time.

Unless, of course, they discover the ultimate “theory of everything,” which specifies precisely how much energy space must contain. “Having the ultimate theory is the only way we’ll know,” Krauss said. “I’m just not sure we’ll have the ultimate theory.”²⁰

But many physicists believe that such an ultimate theory is almost at hand—if the universe turns out to be made of string.