. . . universes other than our own are believed to be completely unobservable, so one can question whether it makes any scientific sense to talk about them. I would argue that it is valid science. . . .

Freshmen in college, exposed for the first time to academic rigor, often complain about the demands of their classes. I can still remember one of my classmates describing the difficulty of one professor’s tests.

“His questions are all like ‘define the universe, and give three examples,’” my classmate said. And in those days, it was an aptly ironic comment. Back then it was considered correct to say that if you’ve seen one universe, you’ve seen them all. Since the universe is all there is, there can be only one. It’s common sense. A test question asking for three examples was asking the impossible.

Similarly, scientists used to say you shouldn’t ask what happened before the big bang, the cosmic explosion that gave birth to the universe. It was a meaningless question, like asking who won the 1994 World Series. There was no World Series in 1994, and there was no time or space “before” the big bang—the very idea of “before time” makes no more sense than the idea of more than one universe.

Traditionally, cosmologists have scoffed at amateurs who dared to ask such questions. Imagine my fascination, then, to discover that some of those very cosmologists themselves secretly harbored the same wonders about what happened “before the beginning.”

I remember when this realization hit me, while listening to the cosmologist Andrei Linde deliver a talk at a workshop in 1991, in which his discussion led up to asking what happened before the big bang. “It is impossible to ask the question,” he said. “But it is impossible not to be curious about this.”

After his talk, I interviewed him and asked for details. “I would say that what we are seeing now perhaps was not the big bang but was one in a sequence of bangs,” he said. “There are many small bangs. The universe not only produces galaxies, it reproduces itself many times.”¹

So if Linde is right, it does make sense to ask what happened before our big bang. And it does make sense to talk about more than one “universe.” The universe we see may be just one member of an extended ultracosmic family. As Martin Rees, the Astronomer Royal of Great Britain, puts it, “Our entire universe may be just one element— one atom, as it were—in an infinite ensemble: a cosmic archipelago.”² It’s as dramatic a shift in human thinking, Rees writes, as the Copernican revolution and the subsequent realization “that the Earth is orbiting a typical star on the edge of . . . just one galaxy among countless others.” In a similar way, the universe may be just one “bubble” of space in a megafroth of cosmic carbonation extending far beyond the view of any conceivable telescope.

In a way, I shouldn’t have been surprised when I heard Linde

express these ideas—there had been whispers, after all, from time to time about such forbidden questions. In the 1980s, Linde and others had published papers describing a multiplicity of universes. And it was a logical enough deduction—I could have thought of it myself. I knew about inflation, the popular idea that in a split second after the universe was born, it had been disrupted by an enormous burst of ultrarapid expansion (as you might recall from Chapter 4). During inflation, one tiny patch of space almost instantly enlarged itself to a vastly larger scale, like a little wrinkled balloon transforming into the Hindenburg. (Or to be a little more precise in size, a pinch of space much smaller than an atom growing into the size of a basketball.) If some small patch of space inflated to make our universe, what happened to the rest of the space that didn’t inflate? Was it just “out there,” out of view? Well, maybe. But if one part of it ballooned to make our universe, why couldn’t other parts of it blow up, too? For that matter, why couldn’t any old patch of space, say, somewhere in New Jersey, decide to puff up into a new universe just for the fun of it? It would be an interesting twist for a future Sopranos episode.

Imagining all this would not have been possible, however, without imagining that the universe is expanding to begin with. That possibility seemed somehow beyond the scope of human imagination until Einstein, in 1917, realized that such a thing was thinkable. But Einstein didn’t like the idea. The equations for his brand-new theory of gravity, general relativity, told him that the universe should be either getting bigger or getting smaller, when everybody knew that it was just sitting still, as it always had been. So Einstein “fixed” his equations, adding a term that made sure the universe stayed the same size for all time.

It seems that the first scientist to really appreciate the possibility, and to show in a concrete way that the universe may in fact be expanding, was the Russian mathematician Alexander Friedmann. And I think I know why Friedmann was the one. It had something to do with his taste in literature.

In brief accounts of the history of cosmology, you will sometimes see a reference to Friedmann as a Russian meteorologist. I think that’s how I first heard him described, and found it odd that a weatherman would have been concerned with general relativity. Later I came across an allusion to Friedmann the mathematician, and wondered whether that earlier reference to meteorologist had been a sloppy error. It turns out that Friedmann was both—a mathematician at heart, with weather forecasting as a day job.

Friedmann’s story is compelling and tragic; he’s an intriguing character who somehow seemed destined not to succeed but did anyway, sort of. His early death deprived the world of future prediscoveries, I suspect, and also limited the recognition he received for the main insight into nature that he did succeed in offering, the idea that the universe can expand. “He introduced both motion and development firmly in to the science of the Universe . . .,” wrote his biographers, “and overcame and destroyed the centuries-old paradigm of the static nature of the Universe.”³

He was born in 1888, son of a ballet dancer and composer (his father) and a pianist and music teacher (his mother). As a student in St. Petersburg, in the days before Lenin ruled the land and St. Petersburg became Leningrad, Friedmann was something of a student activist, involved in staging student strikes to protest government policies. He became a first-rate mathematician anyway, started in graduate school in 1910, and passed his master’s exam in 1914.

But by then, he’d become intrigued with the weather, especially with applying math to the dynamics of the atmosphere. When World War I interrupted, Friedmann volunteered to help Russian pilots drop bombs on targets at the Austrian front. Applying his knowledge of math and the atmosphere, he developed some equations for predicting the proper release points to achieve the desired trajectories. “I have recently had a chance to verify my ideas during a flight over

Przemysl,” Friedmann wrote to his friend Steklov in 1915. “The bombs turned out to be falling almost the way the theory predicts.” In fact, goes the legend, the Germans somehow found out about Friedmann. When the Russian bombs were hitting their targets, German soldiers would say “Friedmann is in the air today.”⁴

Friedmann wrote of several harrowing flights and dangerous landings, but he survived the war and came away with a deepened knowledge of the way math described how the atmosphere worked. In this way, I suspect, he learned to appreciate the physical meaning of his squiggles on paper. By figuring out how to relate mathematics to the physics of the air, he prepared himself to understand the connection between Einstein’s equations and the behavior of the whole universe, of space and time.

In fact, as I read about Friedmann, I had concluded that it was this physical intuition for the meaning of math that made him the right person to realize that the universe can expand. And probably that did have something to do with it. Much later, though, I discovered a clue that hinted at another reason why Friedmann had such insight.

He was the first scientist, perhaps, but not really the first person to suggest that the universe expands. That idea dates back at least to the middle of the nineteenth century, and it came from the troubled but creative mind of the American poet Edgar Allan Poe.

Although familiar with “The Raven” and “Annabel Lee,” I’d never known that Poe had dabbled in cosmology. My first inkling of that came at a party in Santa Monica, given for me by my friends K. C. and Rosie⁵ to celebrate the publication of my first book, The Bit and the Pendulum. That title, of course, alluded to Poe’s short story “The Pit and the Pendulum.”

Musical entertainment at the party was provided by a singer named Lynda Williams, known in the world of science as the Physics Chanteuse. Her songs evoke themes of primordial nucleosynthesis and the laws of motion. In discussing my book’s title, she commented on the allusion to Poe and asked if he had been the one to solve Olbers’ paradox—the mystery of why the sky is dark at night. I had no idea. An astrophysicist at the party, Kenneth Brecher of Boston University, said he didn’t know either, but if Poe had discussed anything like that, it would be in an essay he wrote on cosmological issues called Eureka.

It wasn’t hard to find references to Eureka, and I came across the full text on the World Wide Web. Sure enough, Poe had offered an explanation for Olbers’ paradox, and much more. He envisioned objects that sound suspiciously like black holes, and described the universe as exploding outward from a point, expanding in size, and then contracting again.

Poe spoke of “the primordial particle,” the embodiment of unity, that unity conferring upon it “infinite divisibility.” “From the one Particle, as a center, let us suppose to be irradiated spherically—in all directions—to immeasurable but still to definite distances in the previously vacant space—a certain inexpressibly great yet limited number of unimaginably yet not infinitely minute atoms,” Poe wrote.

Here was the big bang theory of the universe, in 1848, more than seven decades before Friedmann worked out the math. Talk about prediscovery. I was amazed. It was a great example of an imagination capable of discerning deep truths about existence, far in advance of their real discovery.

But of course, Poe was not a scientist. In retrospect, much that’s right about the big bang can be seen in his writings, but some of it doesn’t really fit so well. (The matter from his big bang blasted out into “previously vacant” space; in the modern view the space did not exist in advance of the big-bang explosion.) And he did not provide

the critical ingredient needed to get credit for his foresight, namely, the mathematics that made it all quantitative and precise. It was just literary imagery mixed with physical philosophy, interpretable from a modern perspective as anticipating the work of Friedmann, perhaps, but not really related to it.

Except for one thing. Poe, it turns out, was one of Friedmann’s favorite authors.

Not many people have made this connection. Among cosmological commentaries I have found several mentions of Poe’s cosmological speculations, but never any mention of a connection to Friedmann. (I did find one obscure paper by a Poe scholar noting Friedmann’s fondness for Poe.) And in truth, I don’t know for certain that Friedmann read Eureka specifically or that if he did, it influenced his cosmological research. But I do know that Friedmann read Poe, from a passing mention in a Friedmann biography published in English in 1993.

One of the authors of Friedmann’s biography, Viktor Frenkel, related a discussion in the small Russian town of Tim with Valentina Doinikova, a woman who had known the physicist Paul Ehrenfest. Ehrenfest, born in Austria in 1880, had married a Russian and spent time in St. Petersburg, where he knew Friedmann. So when Frenkel was interviewing Valentina about Ehrenfest, Friedmann’s name came up. And it turned out that Valentina had at one time been engaged to Friedmann. So Frenkel asked about Friedmann, eliciting such information as that he usually wore a bowler hat, always carried an umbrella, and oh yes, his favorite authors included Dostoyevsky and Poe.⁶

It seems to me quite possible, then, that Friedmann was conditioned by Poe’s imagination to see the true meaning of Einstein’s equations, whereas others, Einstein included, did not.⁷

So it came to be that in 1922, Friedmann published “On the Curvature of Space,” his first paper on relativity. “The purpose of this

note is . . . to demonstrate the possibility of a world in which the curvature of space is independent of the three spatial coordinates but does depend on time,” Friedmann wrote.⁸ That is, in modern language, the size of the universe can change as time goes by.

In fact, Friedmann found within Einstein’s equations the possibility that the universe could grow and then shrink. Perhaps, Friedmann suggested, the universe would grow and shrink just once. Or perhaps it could then grow again—an image of a periodic or oscillating universe that would have a beginning, an end, and then a new beginning. As Friedmann pointed out, the equations described possibilities but did not determine which of them the real universe actually chose. “Our knowledge is insufficient for a numerical comparison to decide which world is ours,” he said.⁹ But he noted that for a reasonable guess of the mass of the universe, its lifetime would be on the order of 10 billion years, not much off from current estimates of how long the universe has been around. What’s significant, of course, is not the precision of that estimate, but that Friedmann thought in terms of a universe with possibly finite lifetime, rather than the everlasting cosmos that Einstein and almost everybody else believed in.

Einstein read Friedmann’s paper but was unimpressed; he believed Friedmann had committed a mathematical error. In fact, Einstein had committed the error, and after two years of effort on Friedmann’s part (with the help of a friend’s visit to Einstein), Einstein relented and published an apology. (He still wasn’t ready to admit that the universe expanded, however.)

Friedmann returned to the issue in a second major paper, published in 1924. In his first paper, he had treated the possibility that the universe was finite, meaning that the curvature of spacetime would be positive (like the curvature of the surface of a sphere). In his 1924 paper he explored the possibility that space was negatively curved (like the surface of a saddle). In that case the universe would expand forever and could be infinite in extent. (It would not necessarily be infinite, though, as we’ll see in Chapter 9.)

For most of the rest of the twentieth century, cosmologists wondered which of the universes that Friedmann identified was the type we live in. But it was Friedmann who first clearly realized that those multiple possibilities existed.

I should mention that the science historian Helge Kragh has argued that Friedmann didn’t really care about the possible physical manifestations of his universes but was only interested in the math.¹⁰ Maybe that’s right. But it sure seems to me that Friedmann must have believed his squiggles on paper could have physical meaning. He may not have known exactly what that meaning was, since the equations did allow different possibilities. But he did warn that interpreting the physical meaning of those equations required assumptions about the way that points in space are connected (the mathematical discipline known as topology). That also suggests to me that he did give some thought to his math’s physical consequences.

In any event, it’s impossible to know what Friedmann would have thought about the subsequent development of the big bang theory, or whether he would have contributed more himself, for he died in 1925. He had embarked on a record-setting balloon flight to record conditions in the upper atmosphere and overshot the landing zone, requiring an arduous trip back home. Shortly after his return, Friedmann became ill; the doctors diagnosed typhoid fever, and he died on September 16. (In the Friedmann biography, a relative alludes to his drinking some unboiled water on the return from the balloon trip.)

Though his career was cut short, Friedmann did live long enough to see—and to reveal to others—that Einstein’s squiggles described a universe that need not be the static, unchanging, dull foreverlasting space of traditional science. Friedmann did, in fact, prediscover the expansion of the universe.

Credit for the actual discovery, of course, goes to Edwin Hubble. In his famous paper of 1929, Hubble deduced that galaxies fly apart from each other, and the farther apart they are to begin with, the

faster they fly apart. The data he used to make that deduction were rather sketchy, but his interpretation survived over the decades as more and more observations were made.

The apparent explanation for the galaxies’ higher velocities at greater distances was that the space between the galaxies was getting bigger. Hubble was very cautious in his original paper, though, couching the idea of expansion in technical cosmobabble. “The velocitydistance relation may represent the de Sitter effect . . . [including] a general tendency of material particles to scatter,” he wrote.¹¹

Translation of Hubble’s discovery into the big bang theory of the universe’s birth came from Georges Lemaître, a Belgian astrophysicist and clergyman, with later embellishments from George Gamow. (Gamow, in fact, had been one of Freidmann’s students. I don’t know if Gamow ever read Edgar Allan Poe.)

The big bang theory got its name in 1950, when Fred Hoyle, a British astronomer who didn’t like the idea at all, used that phrase as a slur on a radio show.¹² It turned out to be a sound bite that came back to bite him; it gave the theory a vivid imagination-snatching label that made it easier to become popular. It made it possible for anybody, scientist or not, to think they had some understanding of how the universe began, even if they really didn’t.

Most people, no doubt, think that some big explosion is all there is to the idea. There is little appreciation that the big bang theory is a complex mathematical framework, rooted in Einstein’s theory of relativity and the equations that Friedmann and others developed from it. Still less do most people understand what cosmologists really mean when they say the universe began with a bang.

As it is most commonly articulated by cosmologists today, the standard big bang theory merely contends that long ago, probably

13 billion to 14 billion years before the present, the visible universe was very small, very hot, and very dense. That is to say, all of space and all the matter and energy it contains were wrapped up so compactly that ordinary ideas for describing space and time were rather meaningless. It’s not the case that a small, hot egg burst into the space surrounding it; instead, space itself was confined inside the “egg.” And then for some reason the tiny, hot, and dense space rapidly began to get bigger—that is, it exploded—everywhere at once. As this space grew bigger it cooled, eventually allowing familiar objects to form—first atoms, and then stars and galaxies, and ultimately planets and people.

So the picture of a gigantic explosion is a little misleading. In an ordinary explosion, some matter “blows up” and scatters itself into space. In the big-bang explosion that launched the universe, the matter and space were all tangled up together as they exploded into existence. In the standard view, there was no space, no time, and no matter “before” the big bang. The very idea of “before” the bang has no meaning, since there was no time until the big bang occurred.

Anyway, that’s the standard view. But keep in mind, the idea of the big bang as an explosion is just a metaphor. Still, the evidence all points to a very hot, dense phase in the young universe, very much like the fireball of a big explosion. As the universe aged, it got bigger, just as material spreads out from the center of an explosion. So “big bang” really is a pretty good shorthand label.

For many years a competing view, championed by Hoyle and friends, was given equal space in many science textbooks. It was known as the steady-state universe. Its advocates accepted the expansion of the universe but could not believe it had begun suddenly one day with a big explosion. Instead, they surmised, somewhere in space matter was being continually created, out of nothing, in order to keep the overall appearance of the universe the same even as it grew bigger. To some, that scenario seemed as plausible as the big

bang. But the steady-state view was slain in 1964, when Bell Labs scientists Arno Penzias and Robert Wilson detected the cool afterglow of the big-bang explosion, the cosmic microwave background radiation introduced in Chapter 4.

It is by now a well-known story: Penzias and Wilson wanted to study radio signals from space, but they could not eliminate all the sources of static from the antenna dish they were using as a radio telescope. Eliminating all possible sources of interference, including pigeon droppings on the satellite dish, left them with a faint noise from microwaves at a temperature barely above absolute zero. The microwaves came from all directions—no signal here from aliens or an enemy spy satellite. Somehow the background of space itself had a slight temperature.

Soon Penzias and Wilson learned from other physicists at nearby Princeton that the microwave radiation might be the afterglow of the big bang. (The Princeton scientists were making plans to search for the radiation themselves, but Penzias and Wilson beat them to it.) This cosmic microwave background radiation turned out to be the smoke left over from the big-bang gun. Objects throughout the universe today sit in a bath of radiation left over from the cosmic explosion that got the universe going to begin with.

Years earlier, Gamow had foreseen that the big bang should have generated such a relic radiation. He referred in a 1948 paper to the “high intensity radiation which remained from the original stage of expanding universe” before stars formed. Calculations of the radiation’s temperature today were made by Gamow’s collaborators Ralph Alpher and Robert Herman, who found that this microwave background should measure about 5º Kelvin, that is, 5 Celsius-sized degrees above absolute zero. And that turned out to be pretty close to the temperature that Penzias and Wilson found, 3.5º Kelvin, give or take a degree. (Today’s best measurements give a temperature of 2.7º.) Considering the degree of difficulty in making this prediction, Alpher and Herman’s result was phenomenal. Astrophysicist J. Rich

ard Gott has compared it to predicting that a 50-foot flying saucer would land “on the White House lawn” and then watching as a 27-foot one actually does.¹³

Penzias and Wilson were not aware of the predictions, though. They heard about the explanation from Robert Dicke and colleagues at Princeton, who published a companion paper in the Astrophysical Journal providing the cosmic microwave explanation for Penzias and Wilson’s data.

Even after those publications appeared, in 1965, many steady-state theorists refused to surrender. Still today sometimes they try to revive the steady-state corpse. But a consensus rapidly grew among most cosmologists that the big bang was the best idea to pursue for understanding the universe’s origins. And from the mid-1960s on, several new developments strengthened the big bang’s case. It would not take a Perry Mason to win a courtroom verdict in the theory’s favor. The evidence is compelling:

Point 1: As Hubble showed, and subsequent work confirms, the universe is expanding. This is known because galaxies fly away from each other at a rate that depends on how far apart they are. (The farther apart, the faster they get even farther apart.) These speeds can be measured by shifts in the color of light emitted from the galaxy’s stars. It gets redder the farther away a galaxy is, the way a train whistle or ambulance siren gets lower in pitch as it moves away from the listener.

The natural explanation for these observations of galaxies receding from one another in this way is that space itself is expanding— the universe is getting bigger every day. If the universe will be bigger tomorrow than it is today, it was smaller yesterday, and smaller still the day before that. Run the film of universal history backwards, and it looks as though the universe must have been a mere speck of its present self at some point in the past, roughly 14 billion years ago.

Point 2: The age of the universe, based on the expansion rate, is

about the same as the age of the oldest objects within the universe as measured by other methods. If the universe had been around forever, it should have objects within it that are much older than 14 billion years, but it doesn’t. Therefore the universe must have sprung into existence at some point in the past, as the big bang theory suggests.

Incidentally, the fact that the universe began at a specific time in the past and is now expanding explains Olbers’ paradox about why the nighttime sky is dark. If the universe extended out infinitely, there would be enough stars to light up every point on the sky. So anybody claiming the big bang theory is wrong ought to keep quiet unless they can explain why it’s dark at night.

Point 3: If there really was an explosion engulfing all of space, the radiation from that explosion should still be around, kind of like the glowing embers of a dead campfire. And sure enough, the faint glow of microwave radiation discovered by Penzias and Wilson is just such radiation. A hot primordial explosion explains this glow, as the initial heat would have prevented atoms from forming until the universe was perhaps 400,000 years old or so. Before then, photons of light would have mingled with the electrically charged particles of matter in an endless game of bumper cars, going nowhere. But as space expanded and cooled, the temperature eventually dropped enough to allow electrons to join with atomic nuclei, and the photons that had been bouncing around in the fireball would be free to stream through space. In the time since then, that hot radiation should have cooled to about 3º above absolute zero, just as the latest measurements have found.

Furthermore, if this picture is right, the radiation should arrive from all directions, with almost exactly the same temperature in all directions, and it does. And it should have a range of intensities at different wavelengths corresponding to what the big bang theory predicts. On all counts, the cosmic background radiation comes in right on the money. (By the way, this radiation causes some of the static you’d see on a TV tuned to a channel with no signal. If you

eliminated every other possible source of that static, some would remain—further evidence for the big bang.)

Point 4: If the universe was as hot and dense as the equations of the big bang theory indicate, then the original soup of matter should have been cooked up into a specific mix of different chemical elements. The big bang theory can be used to calculate how much hydrogen, helium, and other light chemical elements should have been cooked up in that soup shortly after the universe was born. Those calculations suggest that the universe should be made of about three-fourths hydrogen, a little less than one-fourth helium, and traces of other elements. And in fact, observations of gases in space and the most pristine stars (uncontaminated by elements cooked up inside stars that later exploded) indicate that the amounts of hydrogen, helium, and other light elements are just what the big bang theory says they should be.

Point 5: Related calculations suggest that the neutrino, the subatomic particle predicted by Wolfgang Pauli, should be found in three distinct types, or flavors, but no more than three. Experiments at atom smashers independently find just that same limit on the number of neutrino flavors.

Point 6: If the big bang theory is right, it should explain why the universe is full of big clusters of galaxies. For those galaxies to be there, some little lumps of matter must have been around shortly after the beginning so they could pull more matter toward them by the force of gravity and in so doing grow bigger and bigger. If those lumps existed, they would have distorted the radiation of the big-bang explosion, producing little blips in the temperature of the microwaves found throughout space today. These very tiny temperature differences were detected by the COBE satellite in 1992.

While a few heretic astronomers have tried, nobody has succeeded in constructing any other reasonable theory that explains all these things. Explaining any one of the above points in any other

way is hard enough. Explaining the whole package is something that only the big bang theory can do. Nothing else remotely comes close.

In other words, the big bang theory is not simply a plausible idea. It is a precise theory that explains, in a quantitative way, many features of the universe we see. It is hard to imagine that the ultimate theory of the universe will not incorporate the basic notions of the big bang in some form or another. And yet, the basic big bang cannot be the whole story. It explains a lot, but it doesn’t explain everything.

Again, in the standard view, there is no “before” the big bang. One way of looking at it is to think of the universe as beginning as a point—no size at all—where the matter-energy content of everything that would become the universe was infinitely dense. The universe began—time began—when that point started to grow.

It’s hard to visualize three dimensions of space growing out of a point, but it’s a little easier to imagine two dimensions growing in this way, as on the surface of the Earth. So let’s just say the Earth represents the universe, keeping in mind that there’s really an extra dimension we’re not visualizing.

Start with a point, with all the matter and energy of the universe compressed to an infinitely high density. Then let the Earth’s surface grow out from the point, curving away just as Earth really does curve away from the North Pole. As time goes by, the Earth gets bigger, growing ever southward. A trip around the Earth at its widest point keeps getting longer as time goes by.

Of course, the real Earth gets to a point of maximum width—the equator—and then starts to shrink again. If the universe is really like the Earth, it will expand only so much, reach a maximum size, and then get to be smaller again, like the Earth south of the equator. That is one possibility.

On the other hand, the universe could keep on getting bigger.

Instead of ending up like a sphere, the ever-expanding universe would look like an endless badminton bird, growing ever wider the farther down you went from the top. Or, the universe could keep expanding but at a slower and slower rate, eventually expanding so slowly that it would appear for all practical purposes to have stopped expanding. In that case the badminton bird would taper down to a nearly constant width.

Remember, this growth represents the passage of time. So what happens if you get down by the equator and decide to turn around and head north—the equivalent of going backward in time. Eventually you reach the North Pole and can go no farther. There is nothing north of the North Pole. And there is no “before” preceding the big bang.

This way of thinking has made most astrophysicists happy enough. Stephen Hawking (and his collaborator Jim Hartle) developed this idea fully into what they call the “no boundary” proposal— just as the Earth is round and has no boundary, the universe just “is,” in a four-dimensional sense, with three dimensions of space and one of time. Time is merely one of the dimensions that is zero at the big bang the way the North Pole is 90º latitude. (We could just as well have called the North Pole 0º latitude, with the equator at 90º and the South Pole 180º.)

One way of viewing all this is to say that time is just an illusion based on the way our senses and brains perceive things. Everything, all of space and all of time, exists all at once, everywhere. We just move through different points on the time dimension the way a ship moves through the latitudes and longitudes on the ocean. When a ship is near the North Pole, the circumference of the Earth is very small; for anybody observing the universe at a time coordinate shortly after the big bang, the universe would be very small.

Personally, I don’t find this view very satisfactory, but it does seem to conform pretty well to the picture of the universe that Einstein’s theory of relativity provides, and it’s a pretty good theory.

However, it is probably not a perfect theory. The assumption we made way back, about starting the universe out as a sizeless point with matter-energy of infinite density, is bogus. Remember, this theory is based on Einstein’s equations of general relativity, and those equations break down when you get to a sizeless point. So we need a new, or at least modified, theory.

Sadly, nobody really knows exactly what that new theory is. There are many, many ideas about it, most beyond the realm of anything that can now be tested. But there has been one significant modification of the big bang theory that has achieved great observational successes and has offered as a bonus an astounding potential prediscovery: universes other than ours. The universe as we know it may not be the only such region of space that exists.

It may be true that studying cosmology is a lot more expensive than it used to be, but that’s not the origin of the theory of inflation. Inflation was born from the need to solve a few problems with the standard big bang theory. For example, on the largest scales, the matter in the universe seems to be spread out in a perfectly smooth way. It doesn’t look like that to us, because we are very small compared to the universe.

If we look at a bucket of sand, it seems pretty much as though the sand grains are all smoothly distributed. But if we were very tiny little bugs crawling around in there, so tiny that one sand grain seemed huge, we’d notice that some of the grains are packed a little more tightly or spaced out a little more than others—in much the way that the galaxies out in space seem clumped a little more some places than others, with different-sized gaps between the clusters. From the point of view of someone who could get the big picture, however, the galaxies are spread out pretty uniformly. Besides, the microwave back

ground is almost exactly the same temperature in all directions, further indicating that the universe at the beginning must have been a smoothly mixed soup.

But the original big bang theory cannot explain how the universe could get that smooth. It’s as if something had thoroughly stirred it up at the beginning. (You could assume that everything just began all stirred up, but an assumption is not an explanation.) Cosmologists called this the “horizon problem.” Think of the ordinary idea of a horizon—the point that represents as far as you can see. That means the point from as far away as light can reach your eyes. Translated to cosmology, the horizon indicates the distance that light can reach in the time available, or in other words, the distance that any physical influence can reach, since no physical influence can travel faster than light.

Even when the visible cosmos was very small, it was too big for light to travel all the way across it in the fraction of a second available before everything got blasted far apart. Yet the universe looks smooth. So what did the mixing? No physical influence could have mixed up the matter that was beyond the horizon in the early universe. A stirring rod would have had to be moving faster than light. That’s the horizon problem.

Another conundrum unresolved by the big bang theory was known as the flatness problem. It referred directly to issues that Friedmann would have understood perfectly—namely, whether the universe would expand forever or not. Friedmann had worked out the math for various scenarios. If the amount of matter in the universe was sufficiently small, there wouldn’t be much gravity to pull things together and the universe would expand forever. Too much matter, and gravity would eventually win over the expansion initiated by the big-bang explosion. The universe would someday stop expanding and begin to contract, eventually crushing everything back into a point—a scenario that eventually earned the nickname

of “big crunch.” Maybe the universe would then “bounce,” and start expanding again, but that would be of no consolation to everyone who had been crushed to death. (Although that’s not as bad as it sounds. As the astronomer Virginia Trimble once pointed out to me, the growing temperature as the universe collapsed would wipe everybody out long before the collapse was over. “You’d be fried before you’d get crunched,” she said.)

There was, of course, a third possibility: that the universe would continue expanding, just at an ever slower rate, so that eventually it would be growing larger so slowly that for all practical purposes it would not be growing at all. And that scenario corresponded to an amount of matter that would make the geometry of space, on average, just like Euclidean geometry—the geometry of flatness.

In 1980 it was not possible to say for sure which scenario described the universe. But it was clear that the universe was pretty close to flat. It could not have very much less or very much more than the amount of matter that would place it on the borderline between eternal expansion and eventual contraction. And that realization posed a problem. For any slight variation from this critical density of matter would have been magnified immensely during the 14 billion years that the universe had been expanding. So at the beginning, the amount of matter must have been precisely fine-tuned to just the right amount.

But what physical process could have been responsible for that fine-tuning? Nobody had a clue. Except for Alan Guth.

In 1979, Guth was a young physicist at Cornell University, trying desperately to solve an obscure problem about a hypothetical particle that behaved like half a magnet. Now every child knows (or should know—based on some informal surveys, this isn’t as widely known as it used to be), you can’t have half a magnet. If you take a bar magnet and break it in half, the result is two magnets. Any magnet has two poles, designated north and south, and if you break it the new ends become new poles.

In the early universe, however, strange particles corresponding to only one magnetic pole might have been created. In fact, Guth calculated, huge numbers of these “magnetic monopoles” should have been created. But nobody had ever seen the slightest evidence of even one. So Guth and fellow Cornell postdoc Henry Tye tried to figure out why.

Developing an idea of the physicist John Preskill, Guth and Tye realized that the problem would go away if the process producing monopoles could be delayed somehow. Let’s skip the details, except for one—Tye pointed out to Guth that the solution assumed that the expansion rate of the universe was not affected by the delay.

In December 1979, Guth began to explore that issue and shocked himself with the realization that the expansion rate of the universe would change because of the delay—dramatically. In a hundred billionth of a trillionth of a trillionth of a second, the universe would double in size 100 times. And a hundred doublings meant it ended up a million trillion trillion times bigger than it had started. Guth began referring to this tremendous burst of expansion as inflation, and he saw that it would solve both the monopole problem and the finetuning or flatness problem. Soon he discovered that his inflation also solved the horizon problem (although he had been unaware of that problem when he started). Such a rapid blast of expansion would have smoothed out any lumps just the way blowing up a balloon eliminates any wrinkles it possessed while uninflated.

Guth’s original version of inflation was flawed in one critical respect—there was no apparent way for inflation to end. But soon an improved version of the theory came from Andrei Linde in the Soviet Union and from two young physicists in Pennsylvania, Paul Steinhardt and Andreas Albrecht.¹⁴ Inflation rapidly became the most popular game in cosmology, with new versions popping up in the literature like new universes in the void. Within a decade, you couldn’t tell the inflation theories apart without a scorecard—there was old inflation, new inflation, chaotic inflation, double inflation, hybrid

inflation, and slow-rolling inflation. The proliferation of inflation versions suggested to some that the idea was wacky. But others believed that the basic idea was simply very robust—it worked so well that in whatever precise version of the big bang you liked, inflation of some sort would have to play an essential role.

Anyway, nowadays the evidence for inflation is very strong. Many precise observations of the microwave background, from satellites, balloons, and ground-based telescopes, have confirmed inflation’s predictions. The inflated lady may not yet have sung, but most cosmologists believe inflation will win the game of explaining the way the visible universe looks today.

Some cosmologists, though, are not content just to explain the universe we see. They’d like to explore some of inflation’s other ramifications—notably the possibility of other bubble universes that might have inflated somewhere out of sight. If such additional universes do in fact exist, they would surely represent one of the most amazing of prediscoveries. You could ask for no clearer example of discovering something before there was any physical evidence of its existence, because one of the properties of other universes is that there is no way to physically detect them.

In this picture, Earth’s universe lives in utter isolation from the others, one of countless separate bubbles of space somehow embedded in a vastly larger “metauniverse” or “multiverse.” These other universes would be out of reach of any possible communication—no airline service, no e-mail, no message in a bottle could ever go there. And so, it would seem, all those far-off cousin universes are good for nothing. So why bother? Well, some cosmologists assert, maybe these bubbles are good for something. They may also help explain why the good old standard universe is the way it is.

Even with inflation, there are questions about the universe that

astronomers still have trouble answering. Such as, why is it so big? And why do certain numbers from physics take the precise, constant value that they do? Why, for example, is the speed of light 300 million meters per second instead of 55 miles per hour?

One answer, popular with a few scientists but generally held in contempt, is that only a universe with certain features could contain people to ask such questions. Interest in this notion, known as the anthropic principle, began to grow in the 1970s, when British physicist Brandon Carter pointed out that small changes in many of the basic numerical constants of physics would render life impossible. He proposed a “weak” anthropic principle holding that the universe could not be much different and still support life. Other scientists proposed a “strong” anthropic principle suggesting that the universe has properties that require life to exist.

Most scientists think the weak form is pretty self-obvious, and the strong form is pretty self-stupid. Nevertheless, “anthropic” reasoning is alive and well in discussion of the possibility of multiple universes. After all, our “bubble” does possess properties hospitable to life. The question is, why? Perhaps the answer has something to do with the fact that it might be only one of many bubbles, all with different properties. Ours would naturally be the best of all possible bubbles—the one bubble with just the right mix of features to contain anybody to worry about it.

In this picture, the multiverse offers a bunch of different physics laboratories, each with different laws of nature. In some of those labs, under some of those laws, it is possible to create life. Assuming that the raw materials for making life must be cooked up in stars and spit out by supernova explosions, a very narrow range of values are possible for things like the masses of elementary particles and the strengths of elementary forces and other physical constants. Our universe has the physical constants it has, then, because it is the lab where life like us is possible.

Of course, many physicists reject the validity or the significance

of the anthropic argument. Nevertheless, they still see in the equations of cosmology the possibility of multiple bubbles. And as these ideas have developed over the years, they’ve begun to exert influence outside of cosmology. They seem especially intriguing to anyone interested in the interface between science and religion.

Multiple universes were a prime topic of discussion at a science-religion conclave I attended in 1999 in Washington at the Smithsonian’s National Museum of Natural History.

Prominent scientists, theologians, philosophers, and historians gathered there to discuss the grandest of questions: Did the universe have a beginning? Was the universe designed? Are we alone? Among the participants was Guth, the originator of the inflation idea, now working on the details of what he called “eternal” inflation—the idea of a never-ending series of bangs producing new bubble universes.

Guth outlined the basics of inflation, telling how it essentially blew up a small patch of the whole universe into a bubble that became a universe unto itself. He reminded everyone that inflation solves some serious problems for cosmology, such as where matter comes from (it is created from the energy of empty space that drove the inflation) and why the universe scientists see today looks roughly the same in all directions (because it grew from one small homogeneous patch of space).

And if inflation could create one entire universe from a tiny patch of space, it could do it again, Guth said. And again. In fact, he asserted, countless other universes may continue to burst into existence—eternal inflation. If so, the question becomes whether the universe, or metaverse, or this series of bubbles, ever had a beginning, pushing the issue of creation of the cosmos into a new context. If the visible universe is just a bubble in a preexisting space, did the

preexisting space have a beginning? Guth seeks the answer in his equations.

“It looks to me that probably the universe had to have a beginning,” he said, “but I wouldn’t bet everything on it.”¹⁵

In either case, if eternal inflation is correct, the universe of today’s textbooks is just one of countless many. To some of the theologians at the Washington conference, that sounded like bad news. It invites the conclusion that life exists simply because out of so many bubbles, one must be the Goldilocks universe—just the right temperature (plus other properties) for life to thrive. In other words, no specially designed habitat for humanity, just a lucky cosmic break. Instead of arising by design, human existence would be a happy coincidence.

In fact, some theologians try to argue that it’s not “scientific” to propose all these other unobservable universes, and therefore creation of merely one universe by a god is a simpler—and therefore a more scientific—explanation. I find this reasoning to be rather transparently faulty. Scientists generally prefer the simpler of two scientific explanations, but a simple nonscientific explanation does not become scientific just because it is simple.

A more scientific (at first glance) objection, offered by some scientists, is that other universes are inherently unobservable. And that runs counter to the common scientific concept that what can’t be observed is scientifically meaningless. In other words, whether other universes exist is more a philosophical question than a scientific one.

But many scientists argue that such a vision of science is much too limited. Scientists try to explain what they see, and sometimes the only explanation that works requires the existence of things that can’t be seen. In other words, these extra universes might be a necessary part of whatever theory ultimately explains why our universe is the way it is. If a theory that explains everything that can be observed requires these universes to exist, then there would be a sound scientific basis for accepting their existence.

Martin Rees has made this point on many occasions. “The ques-

tion ‘Do other “universes” exist?’ is one for scientists—it isn’t just metaphysics,” he writes. If a theory including numerous universes explains many hard-to-explain observable things, then that theory should be taken seriously. “If it predicts multiple universes we should take them seriously too, just as we give credence to what our current theories predict about quarks inside atoms, or the regions shrouded inside black holes.”¹⁶

Science takes plenty of unobservable things seriously, he argues. For example, some galaxies may lie beyond the range of current telescopes, but nobody would argue, therefore, that those galaxies did not “really” exist. A new telescope might bring some of them into view tomorrow.

On the other hand, what about galaxies that are not visible because they are beyond the horizon of space and time? Some galaxies are so far away that light traveling from them hasn’t yet had time to reach Earth. No matter how powerful your telescope, you can’t see them. But you might be able to see them someday—if the universe’s expansion is slowing down, the day would eventually come when their light reached Earth. It would seem silly to say that they aren’t “real” now but only will become real when their light arrives.

On the other hand, maybe the universe’s expansion rate is not slowing down, but accelerating (thereby giving away the big surprise of Chapter 6). In that case, light from those too-distant galaxies will never reach Earth. They will never be seen. But does that make them less real than they would have been if the expansion rate were slower?

In the same vein, Rees contends that galaxies in some other universe deserve to be just as real as those in our own bubble that we cannot see. Sure, these extra universes complicate the cosmic picture, but they may very well be the necessary accoutrements of a cosmos that produces one universe that looks like the one we live in—with properties that allow us to live in it. (And, perhaps, with properties that allow us to prediscover things like unseen additional universes.)

“The multiverse concept might seem arcane, even by cosmological standards, but it affects how we weigh the observational evidence in some current debates,” Rees notes. “Our universe doesn’t seem to be quite as simple as it might have been. . . . Some theorists have a strong prior preference for the simplest universe and are upset by these developments. It now looks as though a craving for such simplicity will be disappointed.”¹⁷