The 4 Percent Universe: Dark Matter, Dark Energy, and the Race to Discover the Rest of Reality - Richard Panek (2011)

Part IV. Less Than Meets the Eye

Chapter 12. Must Come Down

THEY NEEDED SOMETHING to write on—fast. The discussion had progressed to the point where words wouldn't do. They needed numbers, signs, the propulsive force of mathematical symbols flying across a surface. The table of theorists got up and joined the several other clutches of theorists at work on the only blackboard in the room. Still, there was plenty of space for all. The blackboard was "full wall," as they liked to say at the Perimeter Institute for Theoretical Physics. Blackboards in offices were full wall. Blackboards in the hallways, blackboards in nooks off the hallways, blackboards in outdoor courtyards—all full wall. The blackboard in the café reached floor to ceiling, and stretched the length of the room. The theorists had all turned their backs on the café tables, on the windows, on the view of the sunset over the treeline of a city park. Here, there, along the wall, they hunched forward, peering at the hieroglyphs appearing on the board, gesturing their concerns, voicing their corrections. The new group, however, had no chalk. No matter. They simply bent close to the blackboard and waved their hands, their fingers describing arcs in the air. They didn't need chalk. For them, the equations were there.

From across the café, Brian Schmidt watched. "They're really going at it," he said to nobody in particular. Then he produced a cell phone and took a picture.

As the leader of the original High-z team, Schmidt was one of the dark-energy astronomers whose discovery nine years earlier had sent physicists down the byzantine path leading to this blackboard. Now he had entered the theorists' den. When the weeklong meeting on dark energy began, with a four-day conference at McMaster University, in nearby Hamilton, Ontario, several other astronomers had been in attendance. But today the setting had shifted seventy kilometers northwest to the Perimeter Institute, in Waterloo, Ontario, and the number of participants had thinned considerably. "I'm the last astronomer standing," Schmidt had said to the organizer of the Perimeter event, who answered, "No, you're not. What about Rocky?" Schmidt laughed. Rocky Kolb was as much an astronomer as Schmidt was a theorist, and Schmidt had made a point, in his lecture a day earlier, of identifying himself as a "dyed-in-the-wool astronomer." In the mid-1990s, when Schmidt had delegated the responsibilities for the suite of publications that would present the High-z results, a theorist had told him he would need to include a paper on something called the equation of state; Schmidt had shrugged, said "Okay," and invited his former Harvard officemate Sean Carroll to advise on the topic. Back then Schmidt hadn't even known what the term "equation of state" meant; now it seemed to be all anybody wanted to talk about, not just at the conference this week—in May 2007—but at every other dark-energy conference.

Cosmology had a new number. Just as omega quantified the density of mass, the equation of state quantified the density of energy—specifically, the ratio of pressure to energy density. Cosmologists designated it as w. A cosmological constant would mean that w was exactly equal to -1; Einstein's lambda proposes that a given volume of space should have an inherent amount of energy per unit of volume, and that this energy suffuses the universe and remains constant over time. A w not -1 was quintessence. It would do ... something else.

As the tenth anniversary of the discovery approached, the number of dark-energy meetings was only growing. As a resident of Australia, Schmidt had to travel halfway around the world just to get anywhere; at dinner a couple of nights earlier, he'd joked to his colleagues, "My average velocity for a year is 70 to 80 kilometers per hour." But the setting almost didn't matter. For the participants, the meetings and the message were becoming numbingly similar: the same chorus delivering variations on the same theme—a Mike Turner or a Saul Perlmutter here, a Rocky Kolb or an Adam Riess there, all of them looking for answers and coming up empty: a movable famine.

Still, if you had to attend a conference, the Perimeter Institute at least had a marble-top bar where if you ordered a glass of wine, the bartender produced a wine list. Schmidt—who, in the thirteen years since he'd helped form the High-z collaboration in 1994, had graduated from scrappy postdoc to vineyard owner—approved. The Perimeter Institute began life in 2000 with a $100 million endowment from Mike Lazaridis, the founder of Research in Motion, which created the BlackBerry. The guiding principle was, as at many institutes, to give theorists a place to think free of distractions. The difference with Perimeter was that the freedom came with luxury. The interior design of the building alternated between full-wall windows and exposed concrete. A four-story atrium divided administrators from theorists. For lunch, the theorists could stop by the café, or they could stay in their office and order room service.

The first part of the week, at McMaster, had consisted of the usual conference-style presentations: one talk after another in an auditorium seating a hundred or so participants. The Perimeter part of the week, however, would be a workshop: talks open to interruptions, catch-as-catch-can discussions in smaller groups, and the chance to keep on talking in the hallways and in the nooks and on the terrace and, always, over snacks and coffee and meals in the Black Hole Café. At dinner on the first evening at Perimeter, astronomer Schmidt and theorist Christof Wetterich, of Heidelberg University, fell into a discussion about a distinction Kolb had made earlier that afternoon, during the final lecture at McMaster.

Kolb had begun with a meditation on how scientists think about cosmological models. To the astronomers of Copernicus's day (though not necessarily to Copernicus himself), a cosmological model was a representation of a world that made mathematical sense but might bear no relation to reality. Whether the Sun or the Earth was at the center of the cosmos didn't matter; what mattered was which object was more mathematically useful at the center of the cosmological model. Not so for the scientist of today. Over the past four centuries, scientists had learned that the accumulation of evidence could tell them which model was more correct—the one with the Earth at the center or the one with the Sun at the center. Today astronomers regarded the creation of a cosmological model as an attempt to capture "reality itself," Kolb said. "We really think that dark matter is a reality, and that dark energy is a reality." If they somehow turned out not to be, fine. But "that's really what we have to test."

Sitting in the Black Hole Café, Schmidt took a sip of wine, pronounced it palatable, and said that he had to disagree with Rocky. "There is no reality," he said to Wetterich. "There are only predictions."

Wetterich said he had to agree with Rocky. To make his point, he picked up a water glass. "If I drop it," he said, "it will fall to the table."

Schmidt shook his head. Yes, he conceded, every glass throughout history, when released, has fallen. "But once it might not," he said. "You can only predict that it will—to a high degree of confidence," he added.

"I believe it will," Wetterich said.

Schmidt shrugged. "I hope it won't."

The meaning of reality might seem a subject best left to philosophers, but, like philosophy itself, it had always been the domain of physicists, too. The ancients thought they couldn't capture "reality," so they settled for saving the appearances. Once Galileo had provided empirical evidence that Copernicus's Sun-centered system was correct, and once Newton had codified the math, scientists came to understand that equations on paper could do more than approximate reality: If you could find it in the heavens, you could capture it on paper—the point Kolb was trying to make. Then Einstein came along and reversed that process. If you could write it on paper, you could find it in the heavens.* If your equations told you that time passed differently for two observers moving in relation to each other, or that gravity bent light, then that was what nature did. You would, Einstein acknowledged, have to test those predictions: "Experience remains, of course, the sole criterion of the physical utility of a mathematical construction"—the point Schmidt was making to Wetterich. If you found an exception in nature, then you either adjusted or abandoned the theory. But Einstein, speaking from his own experience, then proceeded to argue the point that Wetterich was making to Schmidt: "I hold it that pure thought can grasp reality, as the ancients dreamed."

Schmidt and Wetterich weren't going to settle the debate. It was ancient; it was eternal. For Schmidt, though, it was also personal. In 1998, he'd let go of a glass and it went up.

However transcendent the discovery, the human repurcussions sometimes weighed heavily on Schmidt. This was one of those times. That spring, just prior to the McMaster and Perimeter meetings, Schmidt had heard from the Peter and Patricia Gruber Foundation—a philanthropy made possible by a Wall Street fortune—that he and Saul were the recipients of that year's Gruber Prize in Cosmology, worth $500,000. They were in excellent company. Jim Peebles and Allan Sandage had shared the first cosmology prize, in 2000; other recipients had included Vera Rubin in 2002, and Alan Guth and Andrei Linde in 2004. The apportionment of credit for the 2007 prize was understandable. According to the unwritten rules of science, Brian Schmidt was the big gun on the High-z team. But the team had deliberately tried to rewrite those unwritten rules, and Schmidt was still trying: All week he had been negotiating with the foundation, requesting that Adam Riess be added to the list of recipients because he had been the author of the "discovery of acceleration" paper for the High-z team.

The nuances were more than academic. Once while Perlmutter was making a presentation at a conference, Nick Suntzeff turned to Bob Kirshner and whispered, "Saul thinks there's a Nobel Prize in this."

Kirshner gave Suntzeff a look. "There is!"*

The Grubers were, in a way, the Golden Globes to the Nobel's Academy Awards. The preceding year, 2006, the Gruber Prize in Cosmology had gone to John Mather and the COBE team. Then a few months later the Nobel Prize in Physics went to Mather and his COBE collaborator George Smoot. That the discovery of acceleration was worthy of a Nobel Prize wasn't subject to much debate. What was debated was who made the discovery. "Saul is going to win a Nobel Prize," Alex Filippenko would say, with a shrug. "My only hope is that the Nobel committee will do the right thing and give it to Brian and Adam as well. That would be the fair thing. All right?"—this last said as if there were someone challenging him, which in a sense there was.

The two teams had long before informally agreed that the discovery was, as Riess would say, "big enough and cool enough to share." The standard construction was that in early 1998 two teams had independently reached the same surprising conclusion—that the expansion of the universe appeared to be accelerating. In June 2006, Perlmutter, Riess, and Schmidt learned that they had won the Shaw Prize in Astronomy, a $1 million award endowed by the Hong Kong media magnate Sir Run Run Shaw in 2002. (Jim Peebles had gotten here first, too, in 2004.) So far, so good. But the following month, July 2006, came the announcement that the Antonio Feltrinelli International Prize in Physical and Mathematical Sciences—awarded once every five years by Italy's Accademia Nazionale dei Lince, or Academy of Lynxes, dating back (albeit with a centuries-long interruption) to Galileo's day, and carrying an award of about $315,000—was going to Perlmutter ... and only Perlmutter.

Members of the High-z team interpreted this news as evidence that the Berkeley Lab "publicity machine" had done its job. They still recalled how George Smoot in 1992 had broken the agreement not to publicize the COBE results before the public announcement; to make matters worse, as Mather wrote in a book about the project, the LBL press release "mentioned NASA only in passing and did not cite a single member of the COBE science working group other than George." Now, members of the High-z team feared, the LBL press office was performing the same beyond-the-call-of-duty (and possibly of ethics) service for Saul, making him "appear to be God and the greatest thing since sliced bread," in the words of Filippenko. It had persuaded someone, somewhere, in a position of influence that in the January 1, 1998, Nature paper, or maybe at the January 8, 1998, AAS press conference, or maybe among the January 9, 1998, AAS posters, the Supernova Cosmology Project had announced the discovery of cosmic acceleration.

Members of the SCP, however, had long thought that the High-z team was trying to discredit them. In 2001 Filippenko's personal account of his experience as the only astronomer who was a member of both teams appeared as an article in Publications of the Astronomical Society of the Pacific; the following year Bob Kirshner published his own personal account as a book, The Extravagant Universe. After Mike Turner reviewed Kirshner's book for Science, LBL's Robert Cahn—still seething from his own experience having to defend the SCP from Kirshner's recommendation to shut it down, followed by Kirshner's appropriation of Perlmutter's Hubble Space Telescope plans—said to Turner, "Well, it tells me something that after their results, Saul set out to design an experiment in space to really understand this, and Kirshner decided to write his memoirs." Or as one member of the SCP collaboration remarked, pretty much summarizing the response of the whole team, "The High-z accounts don't recognize the other side of the Mississippi." For his part Perlmutter asked the LBL publicity department to compile a preemptive history of the discovery, then file it away for the day it might be most useful.

As the tenth anniversary of the discovery grew closer, tensions grew stronger. At one cosmology conference "celebration" of the anniversary, Gerson Goldhaber reviewed the history of the discovery, concentrating on his histogram from the fall of 1997. "I'm mentioning the dates," he said, "because the date of discovery is of some importance." He also talked about the colloquia that Perlmutter had given late that year. "Now, the question is, who remembers all that?" he said. "Well, Saul's talk was videotaped, and I have the videotape."

Riess followed him. He opened his talk by wishing Goldhaber a happy birthday. Goldhaber nodded and accepted the applause of the audience. Then Riess showed some of his team's e-mails from January 1998. And then he showed a clip from his appearance on the McNeil-Lehrer NewsHour on the day that Science published an article on Filippenko's talk at the UCLA meeting.

The following morning it was Perlmutter's turn. He opened by saying he hadn't planned to focus on the past, but he did want to show that on January 9, 1998, his team had their "equivalent of the McNeil-Lehrer" moment with the publication of the front-page article in the San Francisco Chronicle covering his participation in the AAS press conference, and although he didn't have e-mails from that period, he did have their "equivalent"—the minutes of the team's meetings from the fall of 1997, which he also displayed.

"Dear Saul," Bob Kirshner began a letter dated January 12, 2007:

I was dreamily thinking about 2006, as people often do on the final day of the year.

As part of this idle foolishness, I took a look at the Shaw Prize site. I want to say again that I think this award is a great thing and I am very glad the cosmic acceleration is being recognized. You have a lot to be proud of, and I feel the same way about the work that Adam and Brian and the rest of us have done.

But there was one point that jumped out at me and I can't get it out of my mind. That is why I am writing to you.

Over the years Brian Schmidt had come to recognize that Kirshner's promotion of Schmidt's thesis work at conferences in the early 1990s—however much the two of them might have disagreed about the apportionment of credit—was part of what a big gun did: get the word out to the community about an acolyte's accomplishments. Now Kirshner had taken on similar duties regarding the High-z team's accomplishments, peppering magazines and newspapers with letters and e-mails objecting to coverage that, in his opinion, favored the SCP.

In this case, he was objecting to the mini-autobiography that Perlmutter had written for the Shaw Prize website, in particular this passage: "We announced these results at the American Astronomical Society January 1998 meeting. Because both our team and Brian's team—including Shaw co-winner Adam Riess—independently announced matching results at conferences in the beginning of the year, by the end of the year most of the scientific community had accepted the startling findings." Over seven single-spaced pages, Kirshner proceeded to quote from (and link to) the LBL's January 8, 1998, press release, contemporaneous press accounts, and books to support his conclusion that Perlmutter "did not announce," "did not announce," " did not announce," and, for good measure, "did not make an announcement" (boldface his) of acceleration in January 1998.

When Kirshner didn't hear back from Perlmutter, he revised the letter, removing the salutation and the references to "you," among other modifications, and on February 27, 2007, he posted it on his Harvard website under the link "Thoughts on the discovery of dark energy."

"Dear Bob," Perlmutter finally replied, in a letter dated June 12, 2007:

Now that the teaching semester is over let me address the 9-page letter that you sent concerning our January 1998 AAS scientific presentation and press conference. As I mentioned in my earlier email I was greatly surprised by your letter, and in fact had previously been thinking I should email a request to a few members of the original High-Z team that they stop referring to our January announcement as "weaker" or "more tentative" than the High-Z team Marina Del Rey announcement, since I think this is incorrect. However, before your email I had never heard the suggestion made that we had not presented *any* substantive results at the January meeting. (Obviously, there is no question about which group's paper got out first, but you are clearly making a broader claim here that I believe misrepresents the history.)

The controversy was coming down to the meaning of the word "announce." A month later, responding to Perlmutter's letter, Kirshner addressed the issue directly: "Did you or did you not 'announce' the accelerating universe at the AAS meeting in January 1998? 'Announcing' is what you claim to have done in your Shaw autobiography. Twice. That's what your letter aims to show. After reading your letter, I am even more convinced that this is not correct." And he went on, for yet another seven single-spaced pages, to cite some of the same press accounts as before, as well as to rebut the references that Perlmutter had included in his June letter, including the front-page article by Charles Petit in the San Francisco Chronicle that ran under the sub-headline "Universe getting bigger and bigger, faster and faster—forever," and reported that the SCP study "seems to indicate" that the "expansion is starting to speed up."*

This time Perlmutter didn't bother to respond.

During the pre-anniversary period, tensions grew stronger within the teams as well. In 2007, even while he was exchanging gritted-teeth pleasantries with Perlmutter, Kirshner managed to alienate some of his collaborators. For a talk entitled "Supernovae and the Accelerating Universe" at the Aspen Center for Physics, Kirshner presented "A Timeline of Important Developments" that, among multiple references to the High-z group, included exactly one mention of Brian Schmidt—a (near) omission that only reinforced the feeling among some team members that Kirshner took "too much credit for himself." Nor was the SCP team immune from this self-cannibalization. Gerson Goldhaber began circulating his own history of the discovery, which underwent several revisions to accommodate the complaints of some SCP collaborators who felt he was claiming the discovery for himself. Privately, though, Goldhaber stuck to his story: "My team found it first, and I found it for my team."

As part of the preparation for the tenth anniversary, STScI—Riess's home turf—sponsored a media day; Perlmutter flew in from the West Coast for the occasion. A week later, Newsweek ran an article on dark energy that led with a re-creation of Riess's 1997 calculation of a universe with negative mass, and went on to quote Kirshner on the cosmological constant. Not only did Perlmutter and the SCP receive no mention, neither did the existence of any discoverer other than Riess (aside from a reference to "and his colleagues").

That did it. Perlmutter contacted the LBL press office: The time had come to release the SCP version of the history of the discovery of dark energy. Shortly thereafter it appeared on the Berkeley Lab website as a three-part series. Part One began: "Saul Perlmutter, leader of the international Supernova Cosmology Project (SCP) based at Berkeley Lab, made the first public announcement of evidence for the accelerating expansion of the universe on January 8, 1998...."

In Cambridge, in an office half a mile up Garden Street from Harvard Square, a quivering hand reached for a keyboard.

This wasn't the legacy Schmidt or any of the other members of the two teams wanted for themselves: bickering eggheads. And now they couldn't even guarantee that they had done their discipline proud.

"Fundamentalist Physics: Why Dark Energy Is Bad for Astronomy." The title alone would have guaranteed that the paper would get attention. That its author was Simon White, one of the directors at the Max Planck Institute for Astrophysics, Germany, guaranteed that it would get serious consideration. That it presented arguments the community had begun finding unavoidable made it a sensation.

The paper appeared online in April 2007, in advance of publication in the journal Reports on Progress in Physics and just prior to the McMaster conference and Perimeter workshop. Like the contentiousness over the Gruber Prize, the paper preoccupied Schmidt that week, not only because everybody else was talking about it, but because he was sympathetic to a lot of what White had to say. The core of White's argument was that astronomy and particle physics constituted two different cultures. Astronomers, White said, were "generalists," exploring the complexities of the universe on a case-by-case basis. Particle physicists were "fundamentalists," wringing the complexities of the universe in the hope of squeezing out an "ultimate foundation"—a "Truth." "Dark Energy," he wrote, "is a unique link between them, reflecting deep aspects of the Fundamental Theory, yet apparently accessible only through astronomical observation."

In the theory-and-observation, call-and-response system of investigating nature that scientists had refined over the previous four hundred years, the dark side of the universe represented an irruption. Copernicus's heliocentric theory anticipated Galileo's observations of Jupiter and Venus, which inspired Newton's theory of universal gravitation, which anticipated more than two centuries' worth of moons, planets, and stars, which inspired Einstein's theory of general relativity, which anticipated the observations of the expanding universe, which inspired the Big Bang theory, which anticipated the observations of the cosmic microwave background, which inspired the revival of Einstein's theoretical cosmological constant, which anticipated the observations of Type Ia supernovae, which inspired ... what? Not a theory, exactly. Just a name for a theory—and not even a theory. A theory-to-be: dark energy.

"We're desperate for your help," Schmidt had called to the theorists in the audience at another cosmology meeting a couple of years earlier. "You tell us what you need, we'll go out and get it for you."

To which the most succinct response was one his old officemate theorist Sean Carroll offered at yet another cosmology meeting: "We have not a clue."

Not a clue, yet no end of ideas. Every day Adam Riess checked an Internet site where scientists posted papers; he was hoping for the paper that would finally present a "deep theory," but he found most of them "pretty kooky." Saul Perlmutter liked to begin public talks with a PowerPoint illustration: papers on dark energy piling up, one on top of the next, until the on-screen stack ascended into the dozens. Schmidt had looked online at how many papers cited the original dark-energy papers and found three thousand—of which twenty-five hundred were theories. In his talk at McMaster, he had included a list of prospective candidates for dark energy that a friend had culled from the recent literature:

Tracker Quintessence, single exp Quintessence, double exp Quintessence, Pseudo-Nambo-Goldstone Boson Quintessence, Holographic dark energy, cosmic strings, cosmic domain walls, axion-photon coupling, phantom dark energy, Cardassian model, brane cosmology (extra-dimension), Van Der Waals Quintessence, Dilaton, Generalized Chaplygin gas, Quintessential inflation, Unified Dark matter and Dark energy, superhorizon perturbations, Undulant Univese, various numerology, Quiessence, general oscillatory models, Milne-Born-Infeld model, k-essence, chameleon, k-chameleon, f(R) gravity, perfect fluid dark energy, adiabatic matter creation, varying G etc, scalar-tensor gravity, double scalar field, scalar+spinor, Quintom model, SO(1,1) scalar field, five-dimensional Ricci flat Bouncing cosmology, scaling dark energy, radion, DGP gravity, Gauss-Bonnet gravity, tachyons, power-law expansion, Phantom k-essence, vector dark energy, Dilatonic ghost condensate dark energy, Quintessential Maldacena-Maoz dark energy, superquintessence, vacuum-driven metamorphosis

"Time to get serious." The PowerPoint slide, teal letters popping off a black background, stared back at a roomful of cosmologists at yet one more conference. Sean Carroll had taken it upon himself to give his fellow theorists their marching orders. The "heyday for talking out all sorts of crazy ideas," as Carroll explained, was over—that heady, post-1998 period when Michael Turner might stand up at a conference and call for "irrational exuberance." Now had come the metaphorical morning after.

The observers had done their job. They had used supernovae, weak lensing, BAO, galaxy clusters, and the cosmic microwave background to find more and more evidence for acceleration until the community agreed: The effect was genuine. Then the observers continued doing their job, trying to figure out if dark energy is quintessence or the cosmological constant. And they continued doing their job. And continued. "Dark Energy is the Pied Piper's pipe," White wrote, "luring astronomers away from their home territory to follow high-energy physicists down the path to professional extinction."

For more than twenty years, particle physics had been pursuing one prey: the Higgs boson, a hypothetical particle that would explain the presence of mass in the universe. The Tevatron at Fermilab had been trying to create it; soon the Large Hadron Collider in Geneva would try to manufacture it. By the time the discovery of cosmic acceleration was ten years old, astronomers were beginning to wonder if they, too, were practicing a science in search of one result: w.

Schmidt recognized that he had done his part to usher astronomy into the arena of Big Science. Together the two supernova search teams had relied on the efforts of more than fifty collaborators. But the changes that astronomy was experiencing would have been happening anyway. Not only were the areas of study becoming more specialized—supernovae, CMB, gravitational lensing, and on and on—but so were the means of studying them, the narrow bands of the electromagnetic spectrum. The South Pole Telescope, for instance: the Sunyaev-Zel'dovich effect required astronomers to use a specific sub-millimeter wavelength in order to detect the "holes" left by photons that had moved out of that frequency. In one generation astronomy had gone from the lone observer on a mountaintop taking photographs in visible light to dozens of collaborators around the globe pursuing a variety of specializations by looking at increasingly narrow bands along the electromagnetic spectrum. Even if the supply of funding globally stayed the same, the demand for it wouldn't. Increasing specialization of areas of research and increasing diversification of research methods were creating an effective shortage of resources. And nothing was demanding those resources like dark-energy research. ADEPT—the space telescope that Chuck Bennett and Adam Riess had conceived in response to Perlmutter's SNAP—had eventually adopted baryon acoustic oscillations as its primary observing strategy, though it would also incorporate Type Ia supernovae. SNAP had expanded its mission from Type Ia supernovae alone to weak lensing. The cost for either satellite would be at least $1 billion—except NASA budgeted only $600 million. "It's not worth it!" one astronomer pleaded with a NASA representative at a dark-energy conference. "Either you do it right, or you don't do it. And if you're not going to do it right, then give us the money back and we'll do other things."

"Hear, hear," piped up Mike Turner from a seat near the front of the auditorium.

Schmidt was married to an economist, and he took an unapologetically bottom-line approach to the idea of a space telescope dedicated to dark energy: Was the mission worth the expense? What were the trade-offs? How much good science wasn't going to be done because of the community's concentration on dark energy? The answers would have been easier if the evidence so far were indicating that dark energy was not only quintessence but unmistakably quintessence—"standing out like dog's balls," as Schmidt would say, adopting the Australian vernacular. Studying something that creates a lot of changes in the cosmos, even subtly, would be more challenging, more satisfying, and probably more revealing of the universe's secrets than studying something that stays the same. If the magic number for omega had been 1, then the un-magic number for the equation of state was -1 because, as Jim Peebles said, "then it's a number, and we have nothing to do."

But as the first decade of dark energy drew to a close, the evidence seemed to be pointing to the cosmological constant. The more that observers continued to do their job, the closer they got to -1. The question then became, as the title of a session at one of the dark-energy conferences put it, "How Well Do We Need to Do?"—as in, "How close do we need to get to w equals -1 before we agree that w does equal -1?" Lawrence Krauss, a theorist and panelist for the session, laid out his argument on a PowerPoint slide:

The most reasonable theoretical prediction is w = -1.
Observations suggest w = -1.
Measuring w approximately = -1 therefore tells us nothing.

So observers shouldn't pursue w at all, right? Wrong. "How well do we need to do?" Krauss said, repeating the title of the session. "Better than we'll be able to do! We need to do better than anything you're ever going to be able to do in the lifetime of the people in this room, I expect, experimentally. And that's just life. In spite of the fact that you're likely to spend the rest of your lives measuring stuff that won't tell us what we want to know, you should keep doing it. But you should be prepared to have a standard model for twenty or thirty years that you don't understand." Speaking on behalf of the theorists, he was basically telling observers, Keep on doing your job; we'll catch up.

Until they did (assuming they did!), science was stuck with—to invoke a term that Carroll helped popularize through articles, lectures, papers, and a blog—a "preposterous universe." It was a universe that had the benefit of a seemingly perfect match between observation and theory—the workings of the heavens and the equations on paper. Take the observations of supernovae and the cosmic microwave background, apply the theory of general relativity, and you had a universe that did indeed add up to the magic omega number of 1. But it was also a universe that didn't add up. Take the observations of supernovae and the cosmic microwave background, apply the other cornerstone of twentieth-century physics, quantum theory, and you got gibberish—an answer that was 120 orders of magnitude off.

Which didn't mean that dark matter and dark energy were the modern equivalents of epicycles or the ether. But it did mean that theorists had to confront the same problem that had consumed Einstein for the final three decades of his life: how to reconcile the physics of the very big—general relativity—with the physics of the very small—quantum mechanics.

Theorists could use the two theories simultaneously—for instance, Hawking radiation, Stephen Hawking's idea that while quantum mechanics dictated the existence of pairs of virtual particles at a black hole's horizon, general relativity dictated that sometimes one of those pairs would slip over the edge into the black hole and the other would rebound back into "our" universe. But theorists hadn't yet figured out a way to make the two theories work together—to make an observation of 0.7 consistent with a prediction of 10120. What made the two theories incompatible—where the physics broke down—was the foundation of the past four centuries' worth of physics: gravity itself.

In physics, gravity is the ur-inference. Even Newton admitted that he was making it up as he went along. In one of his letters to Richard Bentley about the stability of a universe operating under the influence of gravity, Newton wrote that the notion of a force of attraction existing between two distant objects is "so great an Absurdity that I believe no Man who has in philosophical Matters a competent Faculty of thinking can ever fall into it." Nearly two centuries later, the German philosopher-scientist Ernst Mach wrote, "The Newtonian theory of gravitation, on its appearance, disturbed almost all investigators of nature because it was founded on an uncommon unintelligibility." Now, he went on, "it has become common unintelligibility." Einstein endowed gravity with intelligibility by defining it not as some mysterious force between two objects but as a property of space itself, and he refined Newton's equations so that the presence of matter and the geometry of space are interdependent. Most of the interpretations of dark matter and dark energy arose from the right-hand side of the equation for general relativity, the side where Einstein put matter and energy. But there are two sides to every equation, and in this case what was on the other side was gravity.

Most astronomers had dismissed Modified Newtonian Gravity, or MOND, as an explanation for dark matter back in the early 1980s; after the 2006 release of the Bullet Cluster data—the photograph that "showed" dark matter separating out from regular matter in the collision of two galaxy clusters—even its defenders began to distance themselves. The Bullet data could accommodate MOND, but you would still need some kind of dark matter to explain the rest. And at that point MOND began to lose the attraction of simplicity.

But just because MOND might not be valid didn't mean that general relativity was. Back in the 1950s Bob Dicke had organized the Gravity Group at Princeton in part to put Einstein to the test, and his efforts helped inaugurate a generation of experiments. The discovery of evidence for such general-relativistic phenomena as black holes, pulsars, and gravitational lensing had only accelerated those efforts. Dark matter and dark energy, however, endowed those efforts with a sense of urgency.

Physicists were testing gravity on the scale of the very large. In the Sacramento Mountains of New Mexico, the Apache Point Observatory Lunar Laser-ranging Operation (APOLLO) aimed a pulse at the Moon twenty times every second. If a cloud happened to pass, then a green dot appeared, Bat Signal-like, in the purpling twilight over Alamogordo. Otherwise the beam traced a clear path to a target 238,000 miles away: one of three suitcase-size mirrors that Apollo astronauts had planted on the lunar surface four decades earlier specifically to facilitate this kind of experiment. With every laser burst, a few of the photons from the beam would bounce off the reflecting surface and complete their return journey to New Mexico. Total round-trip travel time: 2.5 seconds, more or less.

That "more or less" made all the difference. By timing the speed-of-light journey, researchers were measuring the Earth-Moon distance moment to moment and mapping the Moon's orbit with exquisite precision. As in the apocryphal story of Galileo dropping balls from the Leaning Tower in Pisa to test the universality of free fall, APOLLO treated the Earth and the Moon as two balls dropping in the gravitational field of the Sun. If the orbit of the Moon exhibited even the slightest deviation from Einstein's predictions, scientists might have to rethink those equations.

Physicists were also testing gravity on the scale of the very small. Until the turn of the twenty-first century researchers didn't have the technology to measure gravity at ranges of less than a millimeter, in part because testing gravity isn't simply a matter of putting two objects close to each other and quantifying the attraction between them. All sorts of other things may be exerting a gravitational influence. In a series of jewel-box experiments at the University of Washington, researchers had to take into account the metal of nearby instruments, the soil on the other side of the concrete wall that encircled the laboratory, the changing water level in the soil after a rainfall (and since the experiments were taking place in Seattle, that was a lot of changing), a nearby lake, the rotation of the Earth, the position of the Sun, the dark matter at the heart of our galaxy. Nonetheless, they narrowed their measurements of gravitational attraction down to a distance of 56 microns, or 1/500th of an inch.

So far, Einstein was holding, both across the universe and across the tabletop. And with every narrowing of the range where Einstein might have been wrong, another hypothesis died, and Brian Schmidt's PowerPoint slide lost one more esoteric name. Even if Einstein didn't hold, researchers would first have to eliminate other possibilities, such as an error in the measure of the mass of the Moon or the Sun, or in the level of groundwater after a spring shower, before conceding that general relativity required a corrective.

Even so, astronomers knew that they took gravity for granted at their own peril. Why was it so weak? Why—in the example that scientists commonly cited—could the gravitational pull of the entire Earth on a paper clip be counteracted with a dime-store magnet?

Because, some theorists proposed, gravity was a relic from a parallel universe. Theorists commonly called these universes "branes," as in membranes. If two branes were close enough to each other, or even occupying the same space, then they might interact through gravity. Gravity would be on the scale of the other three forces if we had access to those other universes. What was a powerful force in a parallel universe might be the source of the effects of dark matter or dark energy in ours. The problem with these theories, at least from an astronomer's point of view, was how to test them. A theory needs testable predictions or it's not a truly scientific theory; its validity must come down to observations. But how can you observe a universe beyond your own?

Scientists are always wincingly aware that they are prisoners of their perceptions. It was true, for example, that if dark energy was the cosmological constant, then a hundred billion years from now all that cosmologists would see would be a handful of galaxies. But it was also true that we didn't need to wait a hundred billion years to confront a similar perceptual obstacle. Inflation already ensured that certain traces of the universe's initial conditions would be forever out of reach.

Those conditions would put even branes to shame, in terms of defying perceptions. If inflation can pop one quantum universe into existence, why not many? In fact, according to quantum theory, it should. It would, if inflation actually happened. In that case, our inflationary bubble would be one of an ensemble of 10500 inflationary bubbles, each its own universe. That's 100,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000 universes. Our universe would just happen to be the one with a value of lambda suitable for the existence of creatures that can contemplate their hyper-Copernican existence.

This scenario was, in a way, a logical extension of the argument that Vera Rubin had tried to make with her master's thesis. The Earth is rotating, the solar system is rotating, the galaxy is rotating. Why not the universe? Similarly, the Earth turned out to be just one more planet orbiting the Sun, the Sun one more star in an island universe, the island universe one more galaxy in the universe. And the universe?

By 1973, scientists had named this idea the anthropic principle, and by and large they hated it. Called it "the 'A' word." Wouldn't even discuss it. If any scientific speculation, no matter how wild, ultimately must come down to a prediction, then what prediction could the anthropic principle make? How could you falsify the claim that an unfathomable number of universes exist outside our own? If you can't, critics said, then you had to file the idea under metaphysics—overlooking, perhaps, that it was the metaphysics of midcentury cosmology that had gotten them to a lambda-CDM-plus-inflation universe.

The ongoing resistance to the anthropic principle following the discovery of lambda was similar to an earlier era's discomfort with a homogeneous and isotropic universe following the discovery of the CMB. But then homogeneity and isotropy got an explanation: inflation. A universe that underwent a brief period of extraordinary growth would indeed appear the same wherever you were and wherever you looked.

The anthropic principle was similarly ad hoc—and so what? "Have you heard a better idea?" So wrote no less an authority on initially resisting a homogeneous and isotropic universe than Jim Peebles, in 2003. "I hear complaints that this anthropic principle has been introduced ad hoc, to save the phenomenon. But the same is true of A. The cosmological constant is now seen to save quite a few phenomena." As for another explanation for a low value of lambda: "Something may turn up." But it didn't have to; inflation itself had elevated the homogeneity and isotropy of the universe out of the ad hoc and into the inevitable. Maybe inflation did the same for the existence of 10500 other universes.

Which didn't mean what critics of the anthropic principle often accused it of meaning: the end of physics. While inflation might predict a menagerie of universes, it didn't explain the mechanism that would allow lambda to vary, universe to universe. Theorists would still have to try to work out the physics for that understanding of existence.

And that would be their legacy—the legacy of Brian Schmidt, Saul Perlmutter, Adam Riess, and the dozens of other discoverers of evidence for the acceleration of the universe. It wouldn't be personal acrimony, and it wouldn't be changes to their profession's sociology. It would be the revolution in thought that dark energy mandated. Almost certainly this revolution would require the long-awaited union of general relativity and quantum theory. It might involve modifying Einstein's equations. It could feature parallel, intersecting, or a virtually infinite ensemble of universes.

But whatever this revolution wound up being or doing, it would need what speaker after speaker at conference after conference acknowledged, adopting the same shrugging grace and gratitude as the Dicke birds when they learned they'd been scooped, as Vera Rubin when she realized that astronomy had been overlooking most of what was out there: a "new physics."

What greater legacy could a scientist leave a universe?