Origins: Fourteen Billion Years of Cosmic Evolution (10 page)

Until the 1960s, astronomers were content to classify almost all galaxies as spiral, barred spiral, ellipitical, or irregular. They had right on their side, since more than 99 percent of all galaxies fit one of these classes. (With one galactic class called “irregular,” this result might seem to be a slam dunk.) But during that fine decade, an American astronomer named Halton Arp became the champion of galaxies that did not fit the simple classification scheme of the Hubble tuning-fork diagram plus irregulars. In the spirit of “Give me your tired, your poor, your huddled masses,” Arp used the world’s largest telescope, the 200-inch Hale Telescope at the Palomar Observatory near San Diego, California, to photograph 338 extremely disturbed-looking systems. Arp’s
Atlas of Peculiar Galaxies,
published in 1966, became a veritable treasure chest of research opportunities on what can go bad in the universe. Although “peculiar galaxies”—defined as galaxies with such strange shapes that even “irregular” fails to do them justice—form only a tiny minority of all galaxies, they carry important information about what can happen to galaxies gone wrong. It turns out, for example, that many embarrassingly peculiar galaxies in Arp’s atlas are the merged remnants of two once-separate galaxies that have collided. This means that those “peculiar” galaxies are not different kinds of galaxies at all, any more than a wrecked Lexus is a new kind of car.

To track how
such a collision unfolds, you need a lot more than pencil and paper, because every star in both galactic systems has its own gravity, which simultaneously affects all the other stars in the two systems. What you need, in short, is a computer. Galaxy collisions are stately dramas, taking hundreds of millions of years from beginning to end. Using a computer simulation, you can start, and pause as you like, a collision of two galaxies, taking snapshots after 10 million years, 50 million years, 100 million years. At each time things look different. And when you step into Arp’s atlas—batta-bing—here’s an early stage of a collision, and there’s a late stage. Here’s a glancing blow, and there’s a head-on collision.

Although the first computer simulations were done in the early 1960s (and although the Swedish astronomer Erik Holmberg made a clever attempt during the 1940s to recreate a galaxy collision on a tabletop by using light as an analogue to gravity), it wasn’t until 1972 that Alar and Juri Toomre, brothers who both teach at MIT, generated the first compelling portrait of a “deliberately simple-minded” collision between two spiral galaxies. The Toomres’ model revealed that tidal forces—differences in gravity from place to place—actually rip the galaxies apart. As one galaxy nears the other, the gravitational force rapidly grows stronger at the leading edges of the collision, stretching and warping both galaxies as they pass by or through each other. That stretching and warping accounts for most of what’s peculiar in Arp’s atlas of peculiar galaxies.

How else can computer simulations help us to understand galaxies? Hubble’s tuning fork distinguishes “normal” spiral galaxies from spirals that show a dense bar of stars across their centers. Simulations show that this bar could be a transitory feature, not the distinguishing mark of a different galactic species. Contemporary observers of barred spirals might simply be catching such galaxies during a phase that will disappear in 100 million years or so. But since we can’t hang around long enough to watch the bar disappear in real life, we have to watch it come and go on a computer, where a billion years can unfold in a matter of minutes.

Arp’s peculiar galaxies
proved to be the tip of an iceberg, a strange world of not-exactly-galaxies whose outlines astronomers began to discern during the 1960s and came to understand a few decades later. Before we can appreciate this emergent galactic zoo, we must resume the story of cosmic evolution where we left it. We must examine the origin of all galaxies—normal, nearly normal, irregular, peculiar, and knock-your-socks-off exotic—to see how they were born, and how the luck of the draw has left us in our relatively calm location in space, adrift in the suburbs of a giant spiral galaxy, some 30,000 light-years from its center and twenty-thousands of light-years from its diffuse outer edge. Thanks to the general order of things in a spiral galaxy, first imposed on the gas clouds that later gave birth to stars, our Sun moves in a nearly circular orbit around the center of the Milky Way, taking 240 million years (sometimes called a “cosmic year”) for each trip. Today, twenty orbits after its birth, the Sun should be good for another twenty or so before calling it quits. Meanwhile, let’s have a look at where galaxies came from.

CHAPTER 8

The Origin of Structure

W
hen we examine the history of matter in the universe, looking back through 14 billion years of time as best we can, we quickly encounter a single trend that cries out for explanation. Throughout the cosmos, matter has consistently organized itself into structures. From its nearly perfectly smooth distribution soon after the big bang, matter has clumped itself together on all size scales, to produce giant clusters and superclusters of galaxies, as well as the individual galaxies within those clusters, the stars that congregate by the billions in every galaxy, and quite possibly much smaller objects—planets, their satellites, asteroids, and comets—that orbit many if not most of those stars.

To understand the origin of the objects that now compose the visible universe, we must focus on the mechanisms that turned the universe’s formerly diffuse matter into highly structured components. A complete description of how structures emerged in the cosmos requires that we meld two aspects of reality whose combination now eludes us. As seen in earlier chapters, we must perceive how quantum mechanics, which describes the behavior of molecules, atoms, and the particles that form them, fits with general relativity theory, which describes how extremely large amounts of matter and space affect one another.

Attempts to create a single theory that would unite our knowledge of the sub-atomically small and the astronomically large began with Albert Einstein. They have continued, with relatively little success, right up to the present time and will endure into an uncertain future, until they achieve “grand unification.” Among all the unknowns that irk them, modern cosmologists feel most acutely the lack of a theory that triumphantly blends quantum mechanics with general relativity. Meanwhile, these seemingly immiscible branches of physics—the science of the small and the science of the large—care not a whit for our ignorance; instead, they co-exist with remarkable success inside the same universe, mocking our attempts to understand them as a coherent whole. A galaxy with 100 billion stars apparently pays no particular attention to the physics of the atoms and molecules that compose its star systems and gas clouds. Neither do the even larger agglomerations of matter we call galaxy clusters and superclusters, themselves containing hundreds, sometimes thousands of galaxies. But these largest structures in the universe nonetheless owe their very existence to immeasurably small quantum fluctuations within the primeval cosmos. To understand how these structures arose, we must do the best we can in our current state of ignorance, passing from the minuscule domains governed by quantum mechanics, which hold the key to the origin of structure, to those so large that quantum mechanics plays no role, and matter obeys the laws laid down by general relativity.

To this end, we must seek to explain the structure-laden universe that we see today as arising from a nearly featureless cosmos soon after the big bang. Any attempt to explain the origin of structure must also account for the cosmos in its present state. Even this modest task has confounded astronomers and cosmologists with a series of false starts and errors, from which we have now (so we may fervently hope) removed ourselves to walk in the bright light of a correct description of the universe.

Throughout most of modern cosmology’s history, astrophysicists have assumed that the distribution of matter in the universe can be described as both homogenous and isotropic. In a homogenous universe, every location looks similar to every other location, like the contents of a glass of homogenized milk. An isotropic universe is one that looks the same in every direction from any given point in space and time. These two descriptions may seem the same, but they are not. For example, the lines of longitude on Earth are not homogeneous, because they are farther apart in some regions and closer together in others; they are isotropic in just two places, the North and South poles, where all lines of longitude converge. If you stand at either the “top” or “bottom” of the world, the longitude grid will look the same to you, no matter how far to the left or the right you turn your head. In a more physical example, imagine yourself atop a perfect, cone-shaped mountain, and imagine that this mountain is the only thing in the world. Then every view of Earth’s surface from that perch would look the same. The same would be true if you happened to live in the center of an archery target, or if you were a spider at the center of its perfectly spun web. In each of these cases, your view will be isotropic, but decidedly not homogenous.

An example of a homogeneous but non-isotropic pattern appears in a wall of identical rectangular bricks, laid in a bricklayer’s traditional, overlapping manner. On the scale of several adjoining bricks and their mortar, the wall will be the same everywhere—bricks—but different lines of sight along the wall will intersect the mortar differently, destroying any claim to isotropy.

Intriguingly (for those who love a certain kind of intrigue), mathematical analysis tells us that space will turn out to be homogeneous only if it is everywhere isotropic. Another formal theorem of mathematics tells us that if space is isotropic in just three places, then space must be isotropic everywhere. Yet some of us shun mathematics as uninteresting and unproductive!

Although cosmologists were aesthetically motivated for assuming the homogeneity and isotropy of the distribution of matter in space, they have come to believe in this assumption enough to establish it as a fundamental cosmological principle. We might also call this the principle of mediocrity: Why should one part of the universe be any more interesting than another? On the smallest scales of size and distance, we easily recognize this assertion to be false. We live on a solid planet with an average density of matter close to 5.5 grams per cubic centimeter (in Americanese, that’s about 340 pounds per cubic foot). Our Sun, a typical star, has an average density of about 1.4 grams per cubic centimeter. The interplanetary spaces between the two, however, have a significantly smaller average density—smaller by a factor of about 1 billion trillion. Intergalactic space, which accounts for most of the volume of the universe, contains less than one atom in every ten cubic meters. Here the average density falls below the density of interplanetary space by another factor of 1 billion—enough to make the mind feel good about the occasional accusation of being dense.

As astrophysicists expanded their horizons, they saw clearly that a galaxy such as our Milky Way consists of stars that float through nearly empty interstellar space. The galaxies likewise group into clusters that violate the assumption of homogeneity and isotropy. The hope remained, however, that as astrophysicists charted visible matter on the largest scales, they would find that galaxy clusters have a homogenous and isotropic distribution. For homogeneity and isotropy to exist within a particular region of space, it must be large enough that no structures (or lack of structures) sit uniquely within it. If you take a melon-ball sample of such a region, the requirements of homogeneity and isotropy imply that the region’s overall properties must be similar in every way to the average properties of any other scoop with the same size. What an embarrassment it would be if the left half of the universe looked different from its right half.

How large a region must we examine to find a homogeneous and isotropic universe? Our planet Earth has a diameter of 0.04 light-seconds. Neptune’s orbit spans 8 light-hours. The stars of the Milky Way galaxy delineate a broad, flat disk about 100,000 light-years across. And the Virgo supercluster of galaxies, to which the Milky Way belongs, extends some 60 million light-years. So the coveted volume that can give us homogeneity and isotropy must be larger than the Virgo supercluster. When astrophysicists made surveys of the galaxies’ distribution in space, they discovered that even on these scales of size, as large as 100 million light-years, the cosmos reveals enormous, comparatively empty gaps, bounded by galaxies that have arranged themselves into intersecting sheets and filaments. Far from resembling a teeming, homogenous anthill, the distribution of galaxies on this scale resembles a loofah sponge.

Finally, however, astrophysicists made still larger maps, and found their treasured homogeneity and isotropy. Turns out, the contents of a 300-million-light-year scoop of the universe does indeed resemble other scoops of the same size, fulfilling the long-sought aesthetic criterion for the cosmos. But, of course, on smaller scales, everything has clumped itself into distinctly non-homogeneous and non-isotropic distributions of matter.

Three centuries ago, Isaac Newton considered the question of how matter acquired structure. His creative mind easily embraced the concept of an isotropic and homogeneous universe, but promptly raised an issue that would not occur to most of us: How can you make any structure at all in the universe without having all the matter of the universe joining it to create one gigantic mass? Newton argued that since we observe no such mass the universe must be infinite. In 1692, writing to Richard Bentley, the master of Trinity College at Cambridge University, Newton proposed that

if all the matter in the universe were evenly scattered throughout all the heavens, and every particle had an innate gravity toward all the rest, and the whole space throughout which this matter was scattered was but finite, the matter on the outside of the space would, by its gravity, tend toward all the matter on the inside, and by consequence, fall down into the middle of the whole space and there compose one great spherical mass. But if the matter was evenly disposed throughout an infinite space, it could never convene into one mass; but some of it would convene into one mass and some into another, so as to make an infinite number of great masses, scattered at great distances from one to another throughout all that infinite space.

Newton presumed that his infinite universe must be static, neither expanding nor contracting. Within this universe, objects were “convened” by gravitational forces—the attraction that every object with mass exerts on all other objects. His conclusion about gravity’s central role in creating structure remains valid today, even though cosmologists face a task more daunting than Newton’s. Far from enjoying the benefits of a static universe, we must allow for the fact that the universe has been expanding ever since the big bang, naturally opposing any tendency for matter to clump together by gravity. The problem of overcoming the cosmic expansion’s anti-convening tendency becomes more serious when we consider that the cosmos expanded most rapidly soon after the big bang, the era when structures first began to form. At first glance, we could no more rely on gravity to form massive objects out of diffuse gas than we could use a shovel to move fleas across a barnyard. Yet somehow gravity has done the trick.

During the early days of the universe, the cosmos expanded so rapidly that if the universe had been strictly homogenous and isotropic on all size scales, gravity would have had no chance of victory. Today these would be no galaxies, stars, planets, or people, only a scattered distribution of atoms everywhere in space—a dull and boring cosmos, devoid of admirers and objects of admiration. But ours is a fun and exciting universe only because
in
homogeneities and
an
isotropies appeared during those earliest cosmic moments, which served as a kind of cosmic soup-starter for all concentrations of matter and energy that would later emerge. Without this head start, the rapidly expanding universe would have prevented gravity from ever gathering matter to build the familiar structures we take for granted in the universe today.

What made these deviations, the inhomogeneities and aniso-tropies that provide the seeds for all the structure in the cosmos? The answer arrives from the realm of quantum mechanics, undreamt of by Isaac Newton but unavoidable if we hope to understand where we came from. Quantum mechanics tells us that on the smallest scales of size, no distribution of matter can remain homogeneous and isotropic. Instead, random fluctuations in the distribution of matter will appear, disappear, and reappear in different amount, as matter becomes a quivering mass of vanishing and reborn particles. At any particular time, some regions of space will have slightly more particles, and therefore a slightly greater density, than other regions. From this counterintuitive, airy-fairy fantasy, we derive everything that exists. The slightly denser regions had the chance to attract slightly more particles by gravity, and with time the cosmos grew these denser regions into structures.

In tracing the growth of structure from times soon after the big bang, we can gain some insight from two key epochs we have already met, the “era of inflation,” when the universe expanded at an astounding rate, and the “time of decoupling,” about 380,000 years after the big bang, when the cosmic background radiation ceased to interact with matter.

The inflationary era lasted from about 10
-37
second to 10
-33
second after the big bang. During that relatively brief stretch of time, the fabric of space and time expanded faster than light, growing in a billionth of a trillionth of a trillionth of a second from one hundred billion billion times smaller than the size of a proton to about 4 inches. Yes, the observable universe once fit within a grapefruit. But what caused the cosmic inflation? Cosmologists have named the culprit: a “phase transition” that left behind a specific and observable signature in the cosmic background radiation.

Phase transitions are hardly unique to cosmology; they often occur in the privacy of your home. We freeze water to make ice cubes, and boil water to produce steam. Sugary water grows sugar crystals on a string dangling within the liquid. And wet, gooey batter turns into cake when baked. There’s a pattern here. In every case, things look very different on the two sides of a phase transition. The inflationary model of the universe asserts that when the universe was young, the prevailing energy field went through a phase transition, one of several that would have occurred during these early times. This particular episode not only catapulted the early, rapid expansion but also imbued the cosmos with a specific fluctuating pattern of high- and low-density regions. These fluctuations then froze into the expanding fabric of space, creating a kind of blueprint for where galaxies would ultimately form. Thus in the spirit of Pooh-Bah, the character in Gilbert and Sullivan’s
Mikado
who proudly traced his ancestry back to a “primordial atomic globule,” we can assign our origins, and the beginnings of all structure, to the fluctuations on a sub-nuclear scale that arose during the inflationary era.