cosmology

Not everybody gets to have something named after them. In everyday life, units of measure such as watts, volts, amps, and celsius degrees are so common that we may miss the reference to the scientists for whom they were originally named: James Watt, Alessandro Volta, André Ampere, and Anders Celsius. An even higher honor is to have your name attached to a physical constant, which is a single measured quantity that happens to reveal itself in many and varied ways throughout the physical world.

The general public may be amused by how much enthusiasm the community of scientists can display over certain physical constants. In a changing world, those things that are the same in time or place bring comfort. Why? The history of science has shown that what is found to be constant often becomes a cornerstone to a fundamental part of our understanding of the universe. Some important constants include Avogadro's number (named for the nineteenth century Italian physicist Amadeo Avogadro), which laid the foundation that enabled us to measure atomic weights, Newton's constant (named for the seventeenth century English physicist Isaac Newton), which enables us to calculate the gravitational force between any two objects in the universe, and Planck's constant (named for the turn-of-the-century German physicist Max Planck), which launched the era of quantum mechanics.

When the next millennium arrives, and a tally of the greatest scientific discoveries of the twentieth century is made, somewhere near the top of the list you will find the expanding universe. Credit the American astronomer Edwin Hubble for the discovery of a constant in 1929, that first described the rate at which galaxies are moving away from each other. Hubble's constant launched the modern era of cosmology.

Some discoveries are amazing even when they are fully anticipated. Finding planets orbiting other stars and finding evidence of past life on Mars are among them. Hubble's discovery of the expanding universe, however, was not inspired by an official prediction, nor was it anticipated in any other way, so it caught everybody by surprise. Nobody thought to think it. Your average astrophysicist had simply assumed (in the complete absence of data) that the universe was static and unchanging—a position embarrassingly reminiscent of the Aristotelian notion that the stars were fixed and unchanging upon the dome of the night sky.

Albert Einstein's equations of gravity, published in 1916 and which superseded those of Isaac Newton, contained a description of the entire universe that allowed for three scenarios: contracting, static, or expanding. The mathematics showed that the static solution was unstable; it described a universe that will spontaneously contract or expand. A similar fate greets a ball balanced on the top of a hill; the ball's position is unstable so it will roll down in one direction or another at the slightest nudge. With a non-static universe staring him in the face, Einstein botched the opportunity to predict it and, out of aesthetic preference, instead favored the static model. This required that he introduce to his equations a balancing constant that countered the effects of gravity to prevent cosmic contraction or expansion. While mathematically legitimate, the constant had no physical basis and Einstein later regretted having introduced it.

In the 1920s, Edwin Hubble had access to the 100-inch telescope on Mount Wilson, California, the most powerful telescope in his day. Fully armed, Hubble was able to deduce that the spiral fuzzy things in the sky were entire spiral galaxies—island universes that were not unlike the Milky Way in shape and appearance. By 1929, he had assembled a list of several dozen galaxies whose velocities were reliably measured by his own efforts and by those of the American astronomer Vesto Slipher of the Lowell Observatory in Arizona. Slipher already noted that most of the spiral systems had very high velocities that were directed away from the Milky Way, but he drew no further conclusions.

Hubble's next task was to calculate the distances of all these objects—measurements that are notoriously difficult to obtain. Unfortunately, we can't just strap on a pedometer and count our paces to the galaxies. We can't (yet) fly to the galaxies and consult our odometer. Nor do we live long enough to wait for a radar signal to be bounced back. Curious about how these galaxies' distances might vary with velocity or position on the sky, Hubble derived the distance to each by relying primarily on Cepheid variable stars as a yardstick. The entire class of Cepheid variable stars, which oscillates in brightness in predictable patterns, is named for its prototype, Delta Cephei, the sixth brightest star in the constellation Cepheus. These stars are not only bright enough to stand out among surrounding stars in distant galaxies, but the time it takes for their brightness to cycle, which is trivial to measure, is correlated with their luminosity. Once you derive the luminosity, the distance to the Cepheid variable (and by association, to its host galaxy) pops out from a simple formula that connects the two.

To give a terrestrial analogy, if light bulbs somehow flickered at a characteristic rate that depended on their wattage, you could deduce the bulb's wattage at any distance by simply noting its flickering rate. You then mathematically ask the question, How far away must a bulb of that wattage be to appear as dim as I see it?

What Hubble found would forever change human conception of the universe. He might have found that all galaxies were haphazardly moving away from the Milky Way. But they weren't. All galaxies might have had recession velocities that decreased with distance. But they didn't. Or, perhaps, all galaxies parked themselves at the same distance from us. But they hadn't. Hubble found that, for nearly all galaxies in his sample, the recession velocities were directly correlated with distance. The farther away a galaxy was, the faster it was moving away from us. In other words, if one galaxy were twice the distance of another, it was moving away from us at twice the speed. Ten times farther, ten times faster, and so forth. The inference from this particular signature, and no other, was that we were part of an expanding universe. The discovery even had a ready-made theoretical context as one of the solutions to Einstein's theory of gravity. Just as Earth was dethroned as the center of the system of planets, a new world order was at hand where now everything was in motion, even the universe itself.

One needn't be a math whiz to understand how to derive a numerical value for the Hubble constant. It's simply the slope of the line through the galaxy data in a plot of velocity versus distance. The Hubble constant's units are therefore velocity divided by distance. Depending on what distance indicators are available, and depending on the reliability of these indicators, the Hubble velocity-distance diagram has yielded very different slopes. Indeed, Hubble's originally derived constant was significantly overestimated due to flawed assumptions about the Cepheid calibrations.

Consider it to have been an omen, because even when finally corrected in the 1950s, there remained just enough uncertainty in the Cepheid distances and in other calibrators to allow two competing camps of astronomers to favor very different values of the Hubble constant by selecting some distance indicators while rejecting others. The low value hovered around 15 kilometers per second per million light years of distance as championed by the American astronomer Alan Sandage and his colleagues, while the high value of 30 kilometers per second per million light years was championed by the French-born, American astronomer Gerard deVaucouleurs, a former mentor of mine. Brief standoffs at the crest of scientific discovery are not uncommon, but measurements of an uncertain quantity are usually scattered over a range of values until better and better experiments serve to narrow the range of uncertainty. If anything other than this convergent process takes place, then nonscientific factors (such as hard-headedness) may be at work.

Its importance and elusiveness has made the Hubble constant somewhat of a Holy Grail of cosmology. A major source of uncertainty is the effect of random galaxy motions relative to one another on the total expansion rate. The farther out you go, the more significant the expansion of the universe compared with the random motions and the more reliable your measurement of the Hubble constant. A handful of the galaxies closest to the Milky Way have random velocities that exceed the general flow from the expanding universe and are thus moving towards us rather than away. The nearby Andromeda Galaxy, a mere 2.2 million light-years distant, illustrates the point well: the expansion of the universe contributes less than fifty kilometers per second to Andromeda's motion away from us, which is inadequate to overcome our general (and typical) 100 kilometer per second motion toward each other.

What we need are good calibrators that can be traced to distances far beyond where the random motions of galaxies matter. For example, Cepheid variable stars have only recently been discovered as far away as the famous collection of galaxies known as the Virgo Cluster—15 million light years away along the line of sight through to the constellation Virgo. Only barely reachable from ground-based telescopes, the discovery of Virgo Cepheids had always been a priority project for the orbiting Hubble Space Telescope. At the Virgo cluster's distance, the expanding universe imparts about a 1,000 kilometers per second recession velocity, further reducing the extent to which random galaxy motions contaminate the measured velocity.

When combined with a half-dozen other distance calibrators, including exotic objects such as supernovae and gravitational lenses, a consensus value for the Hubble constant has now emerged (revealed at a recent Princeton conference titled Unsolved Problems in Cosmology). As you might have guessed, it falls almost exactly between the two previous stubbornly held estimates: 23 kilometers per second per million light-years with an uncertainty of less than 15 percent. Stated in sentence form, the Hubble constant would read: For every million light years distant, an object can be expected to recede from you by an extra 23 kilometers per second with an uncertainty of about 3 kilometers per second either way.

I can live with that.

Incidentally, my professional career recently (and briefly) overlapped the Hubble constant debate when I participated in a multi-author research paper that used a distant supernova's change in luminosity as a distance calibrator. We derived a Hubble constant of 25 kilometers per second per million light years—consider it part of the consensus.

Distances to galaxies beyond the base calibrators can now be computed with confidence directly from the Hubble relation itself:

Distance = Velocity of recession ÷ Hubble constant

At the nearest quasar, 3C273, the recessional velocity provided by the expanding universe is nearly 50,000 kilometers per second. Note that at this expansion velocity, random motions of a few hundred kilometers per second are of little arithmetic consequence. Applying the Hubble relation, we get a distance of about 2,000 million (2 billion) light years, which is nearly a thousand times farther from the Milky Way than the Andromeda Galaxy.

Now that the Hubble Grail has been found, we can turn our attention to the next level of challenging questions that face cosmology, such as, How far away (or equivalently, how far back in time) does the value of the Hubble constant remain valid? If the collective gravity of the universe is serving to slow down the cosmic expansion then the Hubble constant must have been higher in the past. And if the universe one day stops expanding and starts to collapse then the Hubble constant will drop through zero to become negative! For these reasons, the Hubble constant is more correctly identified as the Hubble parameter. These Hubble hijinks are officially described through the deceleration parameter, which contains deeper information about our expanding universe such as answers to the questions: Will we indeed expand forever?, or Will we one day collapse?

And Einstein's original constant—introduced mathematically to stabilize the universe—has recently been exhumed in several circles as a tool to reconcile some peculiar observed features of the universe. Look for that one too. It's called lambda.

As I write this sentence, my newborn daughter, Miranda, turns fourteen days old. She will grow up knowing only a world in which the expansion rate of the universe is well established. Seventy years ago, her grandparents were born when galaxies were thought to be simple fuzzy smudges floating in the Milky Way. Yes, we are living the golden age of cosmic discovery.

An important scientific innovation rarely makes its way by gradually winning over and converting its opponents....What does happen is that its opponents gradually die out and that the growing generation is familiarized with the idea from the beginning.

Max Planck, 1936

What, you might ask, could possibly induce a rational astrophysicist to believe that all the matter, energy, and space of the universe began fifteen billion years ago in a primeval fireball packed into a volume smaller than a marble that has been expanding ever since? The answer is simple: regardless of what you may have read or heard, the big bang is supported by a preponderance of evidence and has become the most successful theory ever put forth for the origin and evolution of the universe.

Scientific evidence in support of a theory sometimes takes you places where your senses have never been. Common sense is that human ability to assess a situation you have never seen before by invoking life experiences derived from your five senses. But twentieth-century science has largely been built upon data that was, and continues to be, collected with all manner of tools that enable us to see the universe in decidedly uncommon ways. As a consequence, while we have always required that a theory make mathematical sense, we no longer require that a theory make common sense. We simply demand that it be consistent with the results of observations and experiments. This posture has enabled profound, yet remarkably counterintuitive branches of physics, such as relativity, quantum mechanics, and big bang cosmology to arise.

Of all the theories about how the physical world works, the general public seems to be most intrigued by the big bang. Who wouldn't? Ideas about the origin of things have always made fascinating science. But I have found some people that vehemently oppose the big bang while being generally uninformed about its fundamental tenets. A well-constructed theory should explain some of what is not understood and, more importantly. predict previously unknown phenomena that can be tested. A successful theory is one where experiments consistently confirm its predictions.

Some like to claim that the big bang is just a theory and should therefore be discounted. Don't be fooled. The beginning of the twentieth century saw the end of labeling successful theories as laws. This change of vocabulary came when new experimental domains revealed the predictions of previous physical laws to be incomplete. The change was the physicist's humble recognition that data from newer and better equipment might provide a deeper realization of the physical world. This is why pre-1900 we had Kepler's laws of planetary motion, Newton's laws of gravity, and the laws of thermodynamics, whereas after 1900 we have Einstein's theory of relativity, quantum theory, big bang theory, and so forth.

Confidence in big-bang cosmology is derived from the strengths of many arguments. Let us start with Edwin Hubble's 1929 observation that we live in an expanding universe, where distant galaxies recede from us faster than the near ones in direct proportion to their distances. Further support came from Albert Einstein's theory of gravity, better known as the general theory of relativity, which predicted an expanding universe as one of its solutions with the precise expansion pattern found by Hubble. Since Einstein's theory preceded Hubble's discovery (by thirteen years), Einstein cannot be accused of putting forth an after-the-fact explanation.

For any theory, one should not hesitate to question every possible assumption, no matter how basic they are. If you happen to have a gripe with the claim that objects with high velocity of recession, are farther away than objects with low velocity of recession then consider the existence of gravitational lenses as a simple test-case. As first predicted by Einstein, the gravity of a high-mass foreground object can distort space in its vicinity so that an object which, by chance, falls along the line of sight in the background, can look as though it is split into two or more images. These optical antics have been observed in dozens of galaxies all around the sky and the lensed object (presumed to be in the background simply because it was the one that got lensed) always has a higher recession velocity than the object whose gravity is serving as the lens itself.

Perhaps it's some kind of an illusion that very distant galaxies have very high recession velocities. We measure the velocities from the increase in wavelength (and associated decrease in frequency) in the spectrum of the light emitted by the galaxy. If indeed the galaxies are receding and have shifted spectral features because of it then they ought to measurably exhibit a stretching of time intervals. Recently, supernovae discovered in distant galaxies have been found to take more time to explode and decline in luminosity than counterpart supernovae in nearby galaxies. That extra time happens to be precisely what you would expect from the wavelength and frequency shifts of the spectral features.

The most powerful supporting argument for the big bang derives from the cosmic microwave background. Shortly after the Second World War, and shortly after the notion of a hot, explosive origin for the universe was proposed by the physicist George Gamow, the physicists Ralph Alpher and Robert Herman invoked simple principles of thermodynamics and particle physics to infer that the density of matter and energy of the universe must have been higher in the past, concluding that there should be a leftover signal from an earlier time, when the ambient temperature of the universe was thousands of degrees. That leftover signal, by virtue of the expanding universe, should have cooled appreciably and would appear today as an omni-directional bath of microwave energy with a characteristic temperature of a few degrees on the Kelvin absolute temperature scale. In 1965 a part of this background signal serendipitously revealed itself in data obtained by the microwave antennae of two Bell Labs physicists, Arno Penzias and Robert Wilson, for which they were jointly awarded the 1978 Nobel prize in physics.

If you have a gripe with the claim that some accidentally discovered microwaves are the cooled remnant of a youthful, hot universe, then consider that the big bang predicts a specific mixture of energy for this bath of microwaves that characterizes a single temperature. By similar reasoning, the specific mixture of energy emitted by the Sun (including the relative amounts of infrared, visible and ultraviolet light), characterizes a single temperature (6000 kelvins) at its surface. In 1990, the COBE satellite (COsmic Background Explorer) measured this background and indicated a single temperature (2.726 kelvins) to an accuracy of two-tenths of one percent.

You might be skeptical about whether this single-temperature assortment of microwaves actually came from the early universe. You might prefer to think they were created by your neighbor's microwave oven or a police radar gun or by some microwave-emitting wall of interstellar material nearby in space. But we know that the gravity of galaxy clusters slightly reduces the energy of light that passes through them. And when we look for what the microwave background does in the line of sight to these distant clusters we see a slight drop in energy, implying that the microwave background indeed hails from beyond these clusters and not in front of them.

You may not be convinced that the universe was hotter in the past than it is today, as it must have been in the big bang picture. Consider distant galaxies, which, because of the light travel time between the galaxy and us, we see not as they are but as they once were. If big bang cosmology is correct, these distant galaxies should be bathed in a hotter cosmic background than what is measured in the present. Sensitive measurements of molecules that react differently to different background temperatures have allowed us to infer a temperature for the cosmic background from distant galaxies that is in precise accord with the predicted temperature of the universe at the time the light that we measured left these galaxies.

Just for fun, let's turn back the big bang clock, and use current laws of physics to extrapolate the behavior of the universe to a time when it was much smaller, denser, and hotter—when the background was upwards of a trillion degrees. (Our current theories of physics actually allow us to describe the behavior of the universe starting from the first 0.0000000000000000000000000000000000000000001 seconds of its existence all the way up to 15 billion years and beyond. Times earlier than this 10^-43 seconds have no meaning in quantum mechanics.) At these early times and high temperatures, all atoms were broken apart into their component nuclear particles. Combining all that we know of quantum mechanics, particle physics, and all we have learned from busting atoms to smithereens in particle accelerators, we conclude that as the cosmic soup expanded and cooled, nuclear particles recombined to make a specific and predictable assortment of atoms: the universe was born with 75 percent of its mass as hydrogen and about 25 percent as helium. These are bold extrapolations, but surveys of the most helium-deficient galaxies (those that have undergone very little star formation and hence suffered very little contamination) routinely find between 22 and 27 percent helium, in good agreement with big bang predictions.

A few other light elements are predicted to have formed in trace amounts during the first several moments of the universe. Among these are heavy hydrogen (which is simply a proton and a neutron), light helium (which is simply helium that is missing a neutron from its nucleus), and lithium (the third lightest element on the periodic table of elements). The measured quantities of these light elements in the universe are also consistent with the predictions from the big bang.

We didn't just make this stuff up. It represents an unprecedented marriage of astrophysics and particle physics where a coherent cosmic picture has emerged from a minimum of assumptions that tells us the galaxy velocities are real, the galaxy distances are real, the expanding universe is real, relativity is real, quantum mechanics is real, and the big bang is real. Whenever different sub-branches of a science support the same theory then the confidence you bestow upon the theory is greatly enhanced.

But alas, all is not perfect in paradise. There remains a few holes in big bang theory.

Most importantly, the density of mass in the universe today implies an initial value that is remarkably close to the critical density, which is the density that packs just enough mass for the universe to live at the boundary between one that will ultimately recollapse and one that will expand forever. The fine-tuning that this requires among the values for many of the cosmological parameters in the early universe could not have happened randomly.

And going deeper than the simple extrapolations of the big bang we find that the microwave background is far too uniform from one patch of the sky to the next to have emerged from the conditions thought to have been present in the early universe.

Unfortunately, the early, rapid expansion of the universe does not leave enough time for the galaxies to form as we think they should form; and the big bang cannot tell us what happened before 10^-43 seconds, or for that matter, what happened before zero seconds—or why the laws of physics are what they are.

Do we throw away the big bang along with the bath water because of these complications? Or do we retain the big bang's successful predictions and see if there is room to modify the theory's details in an attempt to solve these problems? These sorts of questions have arisen before. In the mid-sixteenth century, the Polish astronomer Nicolaus Copernicus proposed a model of the known universe with the Sun as the center of all motion rather than Earth. This heliocentric model was much, much simpler than the competing geocentric model because it removed the need for complex epicycles to account for the motions of the planets in the sky, especially during their occasional retrograde motion. But there was a problem. The predicted paths of the planets in the heliocentric model continually deviated from the actual paths of the planets in the sky. Should Copernicus have therefore discarded the entire idea of a Sun-centered universe, or should he have modified some of the model's details? Copernicus' heliocentric view was, of course, basically correct. The problems arose because he naively assumed that the planets orbited the Sun in perfect circles rather than in ellipses the concept of gravity was not yet invented. It would be two hundred years before Isaac Newton's universal law of gravitation supplied a bigger picture that modified and completely subsumed Copernicus' view of the world.

Progress has already been made to resolve some of the problems with the big bang model. The most significant modification is known as inflationary cosmology, where the energetics of the very early universe passes through a phase that spontaneously triggers an period of extremely rapid expansion. Inflation naturally accounts for what was thought to be an embarrassingly fine-tuned critical density. It also allows the cosmic microwave background to be as uniform as it is measured to be. Introduced in the early 1980s by the American physicist Alan Guth, inflation is a natural consequence of the principles of quantum mechanics when applied to the fabric of space and time in the early universe and thus has no household analog. Inflation's main prediction is that the universe was born with its mass density equal to the critical value and continues today have the critical mass density. Current observations have recovered anywhere from 20 to 40 percent of the mass necessary to reach the critical density. Inflation enthusiasts are fervently looking for the rest.

One class of inflationary theories describes a mega-universe with multiple areas of expansion where each region looks like a big bang universe from within, and where different regions of expansion can sustain laws of physics that differ from the ones we know. If this model can be tested and supported then inflation will have subsumed the entire big bang into a larger cosmological picture.

If you choose to discard the big bang entirely then step lightly, you will be forfeiting an impressive array of successful predictions—far more than most theories-in-progress enjoy. Nearly everyone in the community of astrophysicists has chosen to work with it, recognizing that our efforts may lead to an even deeper understanding of the universe where the big bang becomes the core idea of something even bigger.

The world has persisted many a long year, having once been set going in the appropriate motions. From these everything else follows.

Lucretius

In the beginning, sometime between 12 and 16 billion year ago, all the space and all the matter and all the energy of the known universe was contained in a volume less than one-trillionth the size of the point of a pin. Conditions were so hot, the basic forces of nature that collectively describe the universe were unified. For reasons unknown, this sub-pin-point-size cosmos began to expand. When the universe was a piping-hot 10³⁰ degrees and a youthful 10^-43 seconds old—before which all of our theories of matter and space break down and have no meaning—black holes spontaneously formed, disappeared, and formed again out of the energy contained within the unified field. Under these extreme conditions, in what is admittedly speculative physics, the structure of space and time became severely curved as it gurgled into a spongy, foamlike structure. During this epoch, phenomena described by Einstein's general theory of relativity (the modern theory of gravity) and quantum mechanics (the description of matter on its smallest scales) were indistinguishable. As the universe continued to expand and cool, gravity split from the other forces. Quickly thereafter, the strong nuclear force and the electro-weak force split from each other, which was accompanied by an enormous release of stored energy that induced a rapid, thirty-power-of-ten increase in the size of the universe. This release of stored energy is loosely analogous to the release of a substance's latent heat upon cooling to its own freezing point. For example, the thermal energy stored in one gram of water at zero degrees exceeds that stored in one gram of ice at the same temperature. The energy difference represents the latent heat of water. The rapid expansion of the universe, known as the epoch of inflation, stretched and smoothed out the cosmic distribution matter and energy so that any regional variation in density became less than one part in 100,000. Continuing onward with what is now laboratory-confirmed physics, the universe was hot enough for photons to spontaneously convert their energy into matter-antimatter particle pairs, which immediately thereafter annihilated each other, returning their energy back to photons. For reasons unknown, this symmetry between matter and antimatter had been broken at the previous force splitting, which led to a slight excess of matter over antimatter. This asymmetry was small but really, really, important for the future evolution of the universe: for every billion antimatter particles, a billion +1 matter particles were born. As the universe continued to cool, the electro-weak force split into the electromagnetic force and the weak nuclear force, completing the four distinct and familiar forces of nature. While the energy of the photon bath continued to drop, pairs of matter-antimatter particles could no longer be created spontaneously from the available photons. All remaining pairs of matter-antimatter particles swiftly annihilated, leaving behind a universe with one particle of ordinary matter for every billion photons—and no antimatter. Had this matter-over-antimatter asymmetry not emerged, the expanding universe would forever be composed of light and nothing else, not even astrophysicists. Over a roughly three-minute period, protons and neutrons assembled from the annihilations to become the simplest atomic nuclei. Meanwhile, free-roving electrons thoroughly scattered the photons to and fro, creating an opaque soup of matter and energy. When the universe cooled below a few thousand degrees kelvin—about the temperature of fireplace embers—the loose electrons moved slowly enough to get snatched from the soup by the roving nuclei to make complete atoms of hydrogen, helium, and lithium the three lightest elements. The universe is now (for the first time) transparent to visible light, and these free-flying photons are visible today as the cosmic microwave background. Over the first billion years, the universe continued to expand and cool as matter gravitated into the massive concentrations we call galaxies. Between fifty and a hundred billion of them formed, each containing hundreds of billions of stars that undergo thermonuclear fusion in their cores. Those stars with more than about ten times the mass of the Sun achieve sufficient pressure and temperature in their cores to manufacture dozens of elements heavier than hydrogen, including the elements that compose planets and the life upon them. These elements would be embarrassingly useless were they to remain locked inside the star. But high-mass stars fortuitously explode, scattering their chemically enriched guts throughout the galaxy. After seven or eight billion years of such enrichment, an undistinguished star (the Sun) was born in an undistinguished region (the Orion Arm) of an undistinguished galaxy (the Milky Way) in an undistinguished part of the universe (the outskirts of the Virgo super cluster). The gas cloud from which the Sun formed contained a sufficient supply of heavy elements to spawn a system of nine planets, thousands of asteroids, and billions of comets. During the formation of this star system, matter condensed and accreted out of the parent cloud of gas while circling the Sun. For several hundred million years, the persistent impacts of high-velocity comets and other leftover debris rendered molten the surfaces of the rocky planets, preventing the formation of complex molecules. As less and less accretable matter remained in the solar system, the planet surfaces began to cool. The one we call Earth formed in a zone around the Sun where oceans remain largely in liquid form. Had Earth been much closer to the Sun, the oceans would have vaporized. Had Earth been much farther, the oceans would have frozen. In either case, life as we know it would not have evolved. Within the chemically rich liquid oceans, by a mechanism unknown, there emerged simple anaerobic bacteria that unwittingly transformed Earth's carbon dioxide-rich atmosphere into one with sufficient oxygen to allow aerobic organisms to emerge and dominate the oceans and land. These same oxygen atoms, normally found in pairs (O₂), also combined in threes to form ozone (O₃) in the upper atmosphere, which served (and continues to serve) as a shield that protects Earth's surface from most of the Sun's molecule-hostile ultraviolet photons. The remarkable diversity of life on Earth, and we presume elsewhere in the universe, is owed to the cosmic abundance of carbon and the countless number of molecules (simple and complex) made from it. How can you argue when there are more varieties of carbon-based molecules than all other molecules combined. But life is fragile. Earth's encounters with large, leftover meteors, a formerly common event, wreak intermittent havoc upon the ecosystem. A mere sixty-five million years ago (less than two percent of Earth's past), a ten-trillion-ton asteroid hit what is now the Yucatan Peninsula and obliterated over 90 percent of Earth's flora and fauna—including dinosaurs, the dominant land animals. This ecological tragedy pried open an opportunity for small, surviving mammals to fill freshly vacant niches. One big-brained branch of these mammals, that which we call primates, evolved a genus and species (Homo sapiens) to a level of intelligence that enabled them to invent methods and tools of science; to invent astrophysics; and to deduce the origin and evolution of the universe.

Yes, the universe had a beginning. Yes, the universe continues to evolve. And yes, every one of our body's atoms is traceable to the big bang and to the thermonuclear furnace within high-mass stars. We are not simply in the universe, we are part of it. We are born from it. One might even say we have been empowered by the universe to figure itself out—and we have only just begun.

Physics describes the behavior of matter, energy, space, and time, and the interplay among them in the universe. From what scientists have been able to determine, all biological and chemical phenomena are ruled by what those four characters in our cosmic drama do to one another. And so everything fundamental and familiar to us earthlings begins with the laws of physics.

In almost any area of scientific inquiry, but especially in physics, the frontier of discovery lives at the extremes of measurement. At the extremes of matter, such as the neighborhood of a black hole, you find gravity badly warping the surrounding space-time continuum. At the extremes of energy, you sustain thermonuclear fusion in the ten-million degree cores of stars. And at every extreme imaginable, you get the outrageously hot, outrageously dense conditions that prevailed during the first few moments of the universe.

Daily life, we're happy to report, is wholly devoid of extreme physics. On a normal morning, you get out of bed, wander around the house, eat something, dash out the front door. And, by day's end your loved ones fully expect you to look no different than you did when you left, and to return home in one piece. But imagine arriving at the office, walking into an overheated conference room for an important 10:00 a.m. meeting, and suddenly losing all your electrons—or worse yet, having every atom of your body fly apart. Or suppose you're sitting in your office trying to get some work done by the light of your desk lamp, and somebody flicks on the overhead light, causing your body to bounce randomly from wall to wall until you're jack-in-the-boxed out the window. Or what if you went to a sumo wrestling match after work and saw the two spherical gentlemen collide, disappear, then spontaneously become two beams of light?

If those scenes played out daily, then modern physics wouldn't look so bizarre, knowledge of its foundations would flow naturally from our life experience, and our loved ones probably would never let us go to work. Back in the early minutes of the universe, though, that stuff happened all the time. To envision it, and understand it, one has no choice but to establish a new form of common sense, an altered intuition about how physical laws apply to extremes of temperature, density, and pressure.

Enter the world of E = mc².

Albert Einstein first published a version of this famous equation in 1905 in a seminal research paper titled On the Electrodynamics of Moving Bodies. Better known as the special theory of relativity, the concepts advanced in that paper forever changed our notions of space and time. Einstein, then just twenty-six years old, offered further details about his tidy equation in a separate, remarkably short paper published later the same year: Does the Inertia of a Body Depend on Its Energy Content? To save you the effort of digging up the original article, designing an experiment, and testing the theory, the answer is Yes. As Einstein wrote,

If a body gives off the energy E in the form of radiation, its mass diminishes by E/c2 . . . . The mass of a body is a measure of its energy-content; if the energy changes by E, the mass changes in the same sense

Uncertain as to the truth of his statement, he then suggested,

It is not impossible that with bodies whose energy-content is variable to a high degree (e.g. with radium salts) the theory may be successfully put to the test.

There it is. The algebraic recipe for all occasions when you want to convert matter into energy or energy into matter. In those simple sentences, Einstein unwittingly gave astrophysicists a computational tool, E = mc², that extends their reach from the universe as it now is, all the way back to infinitesimal fractions of a second after its birth.

The most familiar form of energy is the photon, a massless, irreducible particle of light. You are forever bathed in photons: from the Sun, the Moon, and the stars to your stove, your chandelier, and your night light. So why don't you experience E = mc² every day? The energy of visible light photons falls far below that of the least massive subatomic particles. There is nothing else those photons can become, and so they live happy, relatively uneventful lives.

Want to see some action? Start hanging around gamma-ray photons that have some real energy—at least 200,000 times more than that of visible photons. You'll quickly get sick and die of cancer, but before that happens you'll see pairs of electrons—one matter, the other antimatter; one of many dynamic duos in the particle universe—pop into existence where photons once roamed. As you watch, you will also see matter-antimatter pairs of electrons collide, annihilating each other and creating gamma-ray photons once again. Increase the light's energy by a factor of another 2,000, and you now have gamma rays with enough energy to turn susceptible people into the Hulk. But pairs of these photons now have enough energy to spontaneously create the more massive neutrons, protons, and their antimatter partners. High-energy photons don't hang out just anywhere. But the place needn't be imaginary. For gamma rays, almost any environment hotter than a few billion degrees will do just fine.

The cosmological significance of particles and energy packets transmuting into each other is staggering. Currently the temperature of our expanding universe, calculated from measurements of the microwave bath of light that pervades all of space, is a mere 2.73 degrees Kelvin. (On the Kelvin scale, zero is the temperature at which molecules have the lowest possible energy, room temperature is about 295 degrees, and water boils at 373 degrees.) Like the photons of visible light, microwave photons are too cool to have any realistic ambitions to become a particle via E = mc²; in fact, there are no known particles they can spontaneously become. Yesterday, however, the universe was a little bit smaller and a little bit hotter. The day before, it was smaller and hotter still. Roll the clocks backward some more—say, 13.7 billion years—and you land squarely in the primordial soup of the big bang, a time when the temperature of the cosmos was high enough to be astrophysically interesting.

The way space, time, matter, and energy behaved as the universe expanded and cooled from the beginning is one of the greatest stories ever told. But to explain what went on in that cosmic crucible, you must find a way to merge the four forces of nature into one, and find a way to reconcile two incompatible branches of physics: quantum mechanics (the science of the small) and general relativity (the science of the large). Spurred by the successful marriage of quantum mechanics and electromagnetism in the mid twentieth century, physicists set off on a race to blend quantum mechanics and general relativity (into a theory of quantum gravity). Although we haven't yet reached the finish line, we know exactly where the high hurdles are: during the Planck era. That's the phase up to 10^-43 seconds (one ten-million-trillion-trillion-trillionths of a second) after the beginning, and before the universe grew to 10^-35 meters (one hundred billion trillion-trillionths of a meter) across. The German physicist Max Planck, after whom these unimaginably small quantities are named, introduced the idea of quantized energy in 1900 and is generally credited with being the father of quantum mechanics.

Not to worry, though. The clash between gravity and quantum mechanics poses no practical problem for the contemporary universe. Astrophysicists apply the tenets and tools of general relativity and quantum mechanics to very different classes of problems. But in the beginning, during the Planck era, the large was small, and there must have been a kind of shotgun wedding between the two. Alas, the vows exchanged during that ceremony continue to elude us, and so no (known) laws of physics describe with any confidence the behavior of the universe during the brief interregnum.

At the end of the Planck era, however, gravity wriggled loose from the other, still-unified forces of nature, achieving an independent identity nicely described by our current theories. As the universe aged through 10^-35 seconds it continued to expand and cool, and what remained of the unified forces split into the electroweak and the strong nuclear forces. Later still, the electroweak force split into the electromagnetic and the weak nuclear forces, laying bare the four distinct forces we have come to know and love—with the weak force controlling radioactive decay, the strong force binding the nucleus, the electromagnetic force binding molecules, and gravity binding bulk matter.

By now, the universe was a mere trillionth of a second old. Yet its transmogrified forces and other critical episodes had already imbued our universe with fundamental properties each worthy of its own book.

While the universe dragged on for its first trillionth of a second, the interplay of matter and energy was incessant. Shortly before, during, and after the strong and electroweak forces parted company, the universe was a seething ocean of quarks, leptons, and their antimatter siblings, along with bosons, the particles that enable their interactions. None of these particle families is thought to be divisible into anything smaller or more basic. Fundamental though they are, each come in several species. The ordinary visible-light photon is a member of the boson family. The leptons most familiar to the nonphysicist are the electron and perhaps the neutrino; and the most familiar quarks are. . . . well, there are no familiar quarks. Each species has been assigned an abstract name that serves no real philological, philosophical, or pedagogical purpose except to distinguish it from the others: up and down, strange and charmed, and top and bottom.

Bosons, by the way, are simply named after the Indian scientist Satyendranath Bose. The word lepton derives from the Greek leptos, meaning light or small. Quark, however, has a literary and far more imaginative origin. The physicist Murray Gell-Mann, who in 1964 proposed the existence of quarks, and who at the time thought the quark family had only three members, drew the name from a characteristically elusive line in James Joyce's Finnegans Wake: Three quarks for Muster Mark! One thing quarks do have going for them: all their names are simple—something chemists, biologists, and geologists seem incapable of achieving when naming their own stuff.

Quarks are quirky beasts. Unlike protons, each with an electric charge of +1, and electrons, with a charge of –1, quarks have fractional charges that come in thirds. And you'll never catch a quark all by itself; it will always be clutching on to other quarks nearby. In fact, the force that keeps two (or more) of them together actually grows stronger the more you separate them—as if they were attached by some sort of subnuclear rubber band. Separate the quarks enough, the rubber band snaps and the stored energy summons E = mc² to create a new quark at each end, leaving you back where you started.

But during the quark-lepton era the universe was dense enough for the average separation between unattached quarks to rival the separation between attached quarks. Under those conditions, allegiance between adjacent quarks could not be unambiguously established, and they moved freely among themselves, in spite of being collectively bound to each other. The discovery of this state of matter, a kind of quark soup, was reported for the first time in 2002 by a team of physicists at the Brookhaven National Laboratories.

Strong theoretical evidence suggests that an episode in the very early universe, perhaps during one of the force splits, endowed the universe with a remarkable asymmetry, in which particles of matter barely outnumbered particles of antimatter by a billion-and-one to a billion. That small difference in population hardly got noticed amid the continuous creation, annihilation, and re-creation of quarks and antiquarks, electrons and antielectrons (better known as positrons), and neutrinos and antineutrinos. The odd man out had plenty of opportunities to find someone to annihilate with, and so did everybody else. But not for much longer. As the cosmos continued to expand and cool, it became the size of the solar system, with a temperature dropping rapidly past a trillion degrees Kelvin. A millionth of a second had passed since the beginning.

This tepid universe was no longer hot enough or dense enough to cook quarks, and so they all grabbed dance partners, creating a permanent new family of heavy particles called hadrons (from the Greek hadros, meaning thick). That quark-to-hadron transition soon resulted in the emergence of protons and neutrons as well as other, less familiar heavy particles, all composed of various combinations of quark species. The slight matter-antimatter asymmetry afflicting the quark-lepton soup now passed to the hadrons, but with extraordinary consequences.

As the universe cooled, the amount of energy available for the spontaneous creation of basic particles dropped. During the hadron era, ambient photons could no longer invoke E = mc² to manufacture quark-antiquark pairs. Not only that, the photons that emerged from all the remaining annihilations lost energy to the ever-expanding universe and dropped below the threshold required to create hadron-antihadron pairs. For every billion annihilations—leaving a billion photons in their wake—a single hadron survived. Those loners would ultimately get to have all the fun: serving as the source of galaxies, stars, planets, and people.

Without the billion-and-one to a billion imbalance between matter and antimatter, all mass in the universe would have annihilated, leaving a cosmos made of photons and nothing else—the ultimate let-there-be-light scenario.

By now, one second of time has passed.

The universe has grown to a few light-years across, about the distance from the Sun to its closest neighboring stars. At a billion degrees, it's still plenty hot—and still able to cook electrons, which, along with their positron counterparts, continue to pop in and out of existence. But in the ever-expanding, ever-cooling universe, their days (seconds, really) are numbered. What was true for hadrons is true for electrons: eventually only one electron in a billion survives. The rest get annihilated, together with their antimatter sidekicks the positrons, in a sea of photons.

Right about now, one electron for every proton has been frozen into existence. As the cosmos continues to cool—dropping below a hundred million degrees—protons fuse with protons as well as with neutrons, forming atomic nuclei and hatching a universe in which 90 percent of these nuclei are hydrogen and 10 percent are helium, along with trace amounts of deuterium, tritium, and lithium.

Two minutes have now passed since the beginning.

Not for another 380,000 years does much happen to our particle soup. Throughout these millennia the temperature remains hot enough for electrons to roam free among the photons, batting them to and fro. But all this freedom comes to an abrupt end when the temperature of the universe falls below 3,000 degrees Kelvin (about half the temperature of the Sun's surface), and all the electrons combine with free nuclei. The marriage leaves behind a ubiquitous bath of visible-light photons, completing the formation of particles and atoms in the primordial universe.

As the universe continues to expand, its photons continue to lose energy, dropping from visible light to infrared to microwaves.

Today, everywhere astrophysicists look we find an indelible fingerprint of 2.73 degree microwave photons, whose pattern on the sky retains a memory of the distribution of matter just before atoms formed. From this we can deduce many things, including the age and shape of the universe. And although atoms are now part of daily life, Einstein's equilibrious equation still has plenty of work to do—in particle accelerators, where matter-antimatter particle pairs are created routinely from energy fields; in the core of the Sun, where 4.4 million tons of matter are converted into energy every second; and in the cores of every other star.

It also manages to occupy itself near black holes, just outside their event horizons, where particle-antiparticle pairs can pop into existence at the expense of the black hole's formidable gravitational energy. Stephen Hawking first described that process in 1975, showing that the mass of a black hole can slowly evaporate by this mechanism. In other words, black holes are not entirely black. Today the phenomenon is known as Hawking radiation, and is a reminder of the continued fertility of E = mc².

But what happened before all this? What happened before the beginning?

Astrophysicists have no idea. Or, rather, our most creative ideas have little or no grounding in experimental science. Yet certain type of religious person tends to assert, with a tinge of smugness, that something must have started it all: a force greater than all others, a source from which everything issues. A prime-mover. In the mind of such a person, that something is, of course, God.

But what if the universe was always there, in a state or condition we have yet to identify—a multiverse, for instance? Or what if the universe, like its particles, just popped into existence from nothing?

Such replies usually satisfy nobody. Nonetheless, they remind us that ignorance is the natural state of mind for a research scientist on the ever-shifting frontier. People who believe they are ignorant of nothing have neither looked for, nor stumbled upon, the boundary between what is known and unknown in the cosmos. And therein lies a fascinating dichotomy. The universe always was goes unrecognized as a legitimate answer to What was around before the beginning? But for many religious people, the answer God always was is the obvious and pleasing answer to What was around before God?

No matter who you are, engaging in the quest to discover where and how things began tends to induce emotional fervor—as if knowing the beginning bestows upon you some form of fellowship with, or perhaps governance over, all that comes later. So what is true for life itself is no less true for the universe: knowing where you came from is no less important than knowing where you are going.

Back at the beginning of everything, when the universe was a fraction of a second old, a ferocious trillion degrees hot, and glowing with an unimaginable brilliance, its main agenda was expansion. With every passing moment the universe got a little bit bigger. But it also got a little bit cooler and a little bit dimmer. Meanwhile, matter and energy cohabited in a kind of opaque soup, in which free electrons continually scattered light to and fro. For 380,000 years things went on that way, until the temperature dropped below 3,000 degrees. Right about then, electrons slowed down enough to be captured by passing protons, thus bringing atoms into the world, and a cosmic background of visible light was set free.

The cosmic background is the incarnation of the leftover light from a dazzling, sizzling early universe. It's a ubiquitous bath of photons—massless vehicles of light energy that are as much waves as they are particles. As the cosmos continued to cool, the photons that had been born in the visible part of the spectrum lost energy to the expanding universe and eventually slid down the spectrum, morphing into infrared photons. Although the visible light photons had became weaker and weaker, they never stopped being photons.

What's next on the spectrum? Today, more than 14-billion years after the beginning, the photons of the cosmic background have cooled to microwaves, giving us the modern moniker cosmic microwave background, or CMB for short. Continue to expand and cool the universe, and astrophysicists some 50 billion years from now will be writing about the cosmic radio wave background.

In the 3,000 degree youthful universe, photons didn't travel far before scattering every which way off an electron. Back then, if your mission had been to see across the universe, you couldn't have. Any photons you saw would, just nano- and pico-seconds earlier, have bounced off an electron right in front of your face, and so you would have seen only a glowing fog everywhere you looked. The entire luminous, translucent, reddish-white sky would have been nearly as bright as the surface of the Sun.

Every photon travels at the speed of light until it slams into an electron. Then it bounces off and travels in another direction at the speed of light, slams into another electron, bounces off, and so on. Only when photons are no longer slamming into electrons can they freely traverse the universe. And electrons didn't cease their obstructionism until they made some more or less permanent commitments to atomic nuclei. That's when the the universe became transparent, and that's when the cosmic background was formed.

The temperature of the universe is directly related to the size of the universe. It's a physical thing. If a photon lost half its original energy, it's because the universe grew to twice its original size. As the universe grows, the photon's wavelength gets longer, stretching with the expanding fabric of space and time.

A photon's wavelength is the simple separation between one crest and the next, a distance you can measure with a ruler. A photon's energy is measured by its frequency, or the number of wiggles it makes during a given interval of time. Photons of lower frequencies carry less energy than photons of higher frequencies.

When an object glows from being heated, it emits radiation (light) in all parts of the spectrum, but will always peak somewhere. For household lamps, the light bulbs all peak in the infrared, a part of the spectrum we detect only in the form of warmth on our skin. But bulbs also, of course, emit plenty of visible light. That's why when you put your hand near your bedside lamp, you feel the light as well as see it. Currently the peak wavelength of the cosmic background is about 1 millimeter, which is smack dab in the microwave part of the spectrum. If you open a channel on your walkie-talkie where there isn't a strong signal, you hear static. That static is ambient background microwaves, a few percent of which are the CMB; the rest comes from the Sun, cell phones, police radar guns, leaky microwave ovens, and so on.

Being the remnant of something that was once brilliantly aglow, the CMB has the wavelength profile we expect of a radiant but cooling object: it peaks in one part of the spectrum but radiates in other parts of the spectrum as well. In this case, besides peaking in microwaves, the CMB also gives off some radio waves and a vanishingly small number of photons of higher energy.

The existence of the CMB was predicted by the Russian-born U.S. physicist George Gamow and colleagues during the 1940s, culminating in a 1948 paper that laid out the physics of the early universe. The foundation of these ideas came from the 1927 work of the Belgian physicist and Jesuit priest Georges Lemaître, who is the generally recognized father of big bang cosmology. But it was U.S. physicists Ralph Alpher and Robert Herman, in 1948, who estimated what the temperature of the cosmic background ought to be. They based their calculations on a of Einstein's 1916 general theory of relativity, Edwin Hubble's 1929 observations that the universe is expanding, and the particle and nuclear physics developed before, and during the Manhattan Project of WWII.

With hindsight, the argument is relatively simple. The fabric of space-time would have been smaller yesterday than it is today, and if it was smaller basic physics requires that it must have been hotter. So they turned the clock backward, and imagined a time when the universe was so hot that all the atoms would have been completely ionized, laying bare with all their electrons roaming free. And if that was the case, light would not always have arced, uninterrupted, across the universe. The photons' free ride couldn't have begun before a certain stage in the cooling of the cosmos.Not before the electrons would have combined with atoms and stopped interfering with the light. This transition occurs at a few thousand degrees.

While it was Gamow who suggested that the universe was once hotter, and that you could know the physics of the early universe, it was Herman and Alpher, who proposed a temperature of 5 degrees. Yes, they got the wrong temperature—the CMB is 2.7 degrees Kelvin—but together, those three physicists made an extrapolation unlike any other in the history of science. To take some basic atomic physics from a slab in the lab, and then deduce the largest-scale phenomenon ever measured, was nothing short of extraordinary. Writing about this feat, the U.S. physicist J. Richard Gott says, Predicting that the radiation existed and then getting its temperature correct to within a factor of 2 was a remarkable accomplishment—rather like predicting that a flying saucer 50 feet in width would land on the White House lawn and then watching one 27 feet in width actually show up.

At the time Gamow, Herman, and Alpher were coming up with their prediction, physicists were still undecided about the beginning of the universe. The year 1948 also saw the publication of a rival steady state theory of the universe by the English physicist Hermann Bondi and the American Astronomer Thomas Gold. The CMB proved to be the turning point. It indicated clearly that the universe had once been different—smaller, and hotter—whereas the rival, steady state model implied that the universe, while expanding had always looked the same. The universe could not be hotter yesterday And they provided for matter to pop into our universe from someplace else, at just the right rate, to leave the expanding universe with the same average density. And. The first direct observation of the CMB ware the nails in the coffin of the steady-state theory, and that observation was made inadvertently in 1964 by Arno Penzias and Robert Wilson of Bell Laboratories. In 1978, they won the Nobel Prize for it.

In the 1960s everyone knew about microwaves, but almost no one had the technology to detect them. Back then most wireless communication was with radio waves, which are longer, and so the existing receivers and detectors and transmitters weren't useful. You needed a shorter-wavelength detector and a suitable antenna to capture them. Bell Labs, pioneer in the communications industry, had a beefy horn-shaped antenna that could focus the microwaves down to the detectors.

If you're going to send or receive a signal, you don't want too many things contaminating it. Penzias and Wilson wanted to get a measure of the background interference—from the Sun, from the center of the galaxy, from terrestrial sources, from whatever. So they made an innocent measurement. They weren't cosmologists; they were physicist-technologists looking for microwaves, unaware of the Gamow, Herman, and Alpher predictions. What they were decidedly not looking for was the cosmic microwave background; they were just trying to open up a new channel of communication for Bell Laboratories, and before doing that, they needed to characterize all the things that would contaminate a signal.

So Penzias and Wilson run their experiment, and correct their data for all the sources of interference they know about, but there's one part of the signal that doesn't go away, and they just can't figure out how to eliminate it. Finally they look inside the dish. They see pigeons nesting inside. And so they're worried that a white dielectric substance (pigeon poop) might be responsible for the signal, because the signal comes from every direction, and it doesn't change, and the only thing that's all over their fancy horn-shaped antenna and doesn't change is the pigeon poop. So they clean it up, the signal drops a little bit, but there's still something left. The paper they publish in 1965 in the Astrophysical Journal talks about the unaccountable excess antenna temperature.

But at the same time Penzias and Wilson were scrubbing bird droppings off their fancy antenna, a team of physicists at Princeton, led by Robert Dicke, were building a detector specifically to find the CMB. They, however, didn't have the resources of Bell Labs, so their work went a little slower. as the moment Dicke and his colleagues heard about Penzias and Wilson's work, they knew they'd been scooped. The Princeton team knew exactly what the observed excess antenna temperature was. Everything fit: the temperature, the fact that the signal came from every direction, and that it wasn't linked in time with Earth's rotation or position around the Sun.

Because light takes long stretches of time to us from distant places in the universe, if we look out in space we are actually look back in time. So if the intelligent inhabitants of a galaxy far, far away were measuring the temperature of the cosmic background radiation at the moment captured by our gaze, they would get a temperature higher than 2.7 degrees, because they are living in a younger, smaller, hotter universe than we are.

Can such an assertion be tested? It turns out that the molecule cyanogen (CN)—best known to convicted murderers as the active component of the gas administered by their executioners—gets excited by exposure to microwaves; if the microwaves are warmer than the ones in our CMB, they excite molecule a little more. The cyanogen in distant, and thus younger, galaxies is exposed to a warmer cosmic background than the cyanogen in our galaxy. Indeed, their cyanogens lives at a higher exited state than ours. Although CN is a powerful cosmic marker, its behavior was all learned in the lab

You can't make this stuff up.

But why should any of this be interesting? The universe was opaque until 380,000 years after the big bang, so you could not have witnessed matter taking shape even if you'd been sitting front row center. You couldn't have seen where the galaxy clusters and voids were starting to form. Before anybody could have seen anything worth seeing, photons had to travel, unimpeded, across the universe.

The spot where each photon began its cross-cosmos journey is where it smacked into the last electron that would ever stand in its way—the point of last scatter. As more and more photons escape unsmacked, they create an three-dimensional, expanding surface of last scatter, some 120,000 years deep. That surface is where all the atoms in the universe were born: an electron joins an atomic nucleus, and a little pulse of energy in the form of a photon soars away into the wild red yonder.

By then, some regions of the universe had already begun to coalesce by the gravitational attraction of its parts. Photons that last scattered off electrons in these regions developed a different, slightly cooler profile than those scattering off the less sociable electrons sitting in the middle of nowhere. Where matter accumulated, the strength of gravity grew, enabling more and more matter to gather. These regions seeded the formation of galaxy superclusters while other regions were left relatively empty.

When you map the CMB in detail [see Sharper Focus, by Charles Liu, Natural History, May 2003], you find that it's not completely smooth. It does have spots that are slightly hotter and slightly cooler spots than average. And by studying these temperature variations in the CMB—that is to say, by studying the structure of the surface of last scatter—we can infer what the structure and content of the matter was in the early universe. We know what the structure of matter is today, because we see galaxies and clusters and superclusters. To figure out how those systems arose, we use out best probe, the cosmic microwave background. The CMB is a remarkable time capsule a way to reconstruct history in reverse. Studying its patterns is like performing some sort of cosmic phrenology. We're looking at the skull bumps of the grown-up universe and inferring its behavior as an infant.

When constrained by other observations of the contemporary and distant universe, the CMB enables you to find out all sorts of fundamental cosmic properties. Compare the distribution of sizes and temperatures of the warm and cool areas and you can infer how strong the force of gravity was at the time and how quickly matter accumulated, allowing you to then deduce how much ordinary matter (4%), dark matter (23%), and dark energy (73%) there is in the universe. From here, it's then easy to tell whether or not the universe will expand forever.

Ordinary matter is what we are all made of. It has gravity and interacts with light. Dark matter is a mysterious substance that has gravity but does not interact with light in any known way. Dark energy is a mysterious pressure that acts in the opposite direction of gravity, forcing the universe to expand faster than it otherwise would. What our phrenological exam says is that we understand how the universe behaved, but that most of the universe is made of stuff for which we have no clue what it is.

Our profound areas of ignorance notwithstanding, today, as never before, cosmology has an anchor, because the CMB reveals the portal through which we all walked: the surface of last scatter. It's a point where interesting physics happened, and where we learned about the universe before and after it's light was set free.

The simple discovery of the cosmic microwave background turned cosmology into something more than mythology. But it was the accurate and detailed map of the cosmic microwave background that turned cosmology into an experimental science. Cosmologists have plenty of ego—how can a person not be ego-driven when it's your job to deduce what brought the universe into existence? But without data, their explanations were just tall tales. In this modern era of cosmology, each new observation, each morsel of data wields a two-edged sword: it enables cosmology to thrive on the kind of foundation that so much of the rest of science enjoys, but it also constrains theories that people thought up when there wasn't enough data to say whether they were wrong or not. No science achieves maturity without it.

Let there be cosmology.