Authors: Carl Sagan
“It is a great ball resting on the flat back of the world turtle.”
“Ah yes, but what does the world turtle stand on?”
“On the back of a still larger turtle.”
“Yes, but what does he stand on?”
“A very perceptive question. But it’s no use, mister; it’s turtles all the way down.”
We now know that we live on a tiny dust mote in an immense and humbling universe. The gods, if they exist, no longer intervene daily in human affairs. We do not live in an anthropocentric universe. And the nature, origin and fate of the cosmos seem to be mysteries far more profound than they were perceived to be by our remote ancestors.
But the situation is once again changing. Cosmology, the study of the universe as a whole, is becoming an experimental science. Information obtained by optical and radio telescopes on the ground, by ultraviolet and X-ray telescopes in Earth orbit, by the measurement of nuclear reactions in laboratories, and by determinations of the abundance of chemical elements in meteorites, is shrinking the arena of permissible cosmological hypotheses; and it is not too much to expect that we will soon have firm observational answers to questions
once considered the exclusive preserve of philosophical and theological speculation.
This observational revolution began from an unlikely source. In the second decade of this century there was—as there still is—in Flagstaff, Arizona, an astronomical facility called the Lowell Observatory, established by none other than Percival Lowell, for whom the search for life on other planets was a consuming passion. It was he who popularized and promoted the idea that Mars was crisscrossed with canals, which he believed to be the artifacts of a race of beings enamoured of hydraulic engineering. We now know that the canals do not exist at all. They apparently were the product of wishful thinking and the limitations of observing through the Earth’s murky atmosphere.
Among his other interests, Lowell was concerned with the spiral nebulae—exquisite pinwheel-shaped luminous objects in the sky, which we now know to be distant collections of hundreds of billions of individual stars, like the Milky Way Galaxy of which our Sun is a part. But at that time there was no way to determine the distance to these nebulae, and Lowell was interested in an alternative hypothesis—that the spiral nebulae were not enormous, distant, multistellar entities, but rather smaller, closer objects which were the early stages of the condensation of an individual star out of the interstellar gas and dust. As such gas clouds contract under their self-gravitation, the conservation of angular momentum requires that they speed up to rapid rotation and shrink to a thin disc. Rapid rotation can be detected astronomically by spectroscopy, letting light from a distant object pass consecutively through a telescope, a narrow slit and a glass prism or other device which spreads white light out into a rainbow of colors. The spectrum of starlight contains bright and dark lines here and there in the rainbow, images of the slit of the spectrometer. An example is the bright yellow lines emitted by sodium, apparent as we throw a small piece of sodium into a flame. Material made of many different chemical elements will show many different spectral lines. The displacement of these spectral lines from their
usual wavelengths when the light source is at rest gives us information on the velocity of the source toward and away from us—a phenomenon called the Doppler effect and familiar to us, in the physics of sound, as the increase or decrease in the pitch of an automobile horn as the car rapidly approaches or recedes.
Lowell is thought to have asked a young assistant, V. M. Slipher, to check the larger spiral nebulae to determine whether one side showed spectral lines shifted toward the red and the other toward the blue, from which it would be possible to deduce the speed of rotation of the nebula. Slipher investigated the spectra of the nearby spiral nebulae but found to his amazement that almost all of them showed a red shift, with virtually no sign of blue shifts anywhere in them. He had found not rotation, but recession. It was as if all the spiral nebulae were retreating from
us.
A much more extensive set of observations was obtained in the 1920s at the Mount Wilson Observatory by Edwin Hubbell and Milton Humason. Hubbell and Humason developed a method of determining the distance to the spiral nebulae; it became apparent that they were not condensing gas clouds relatively nearby in the Milky Way Galaxy, but themselves great galaxies millions or more light-years away. To their amazement, they also found that the more distant the galaxy, the faster it was receding from us. Since it is unlikely that there is anything special about our position in the cosmos, this is best understood in terms of a general expansion of the universe; all galaxies recede from all others so that an astronomer on any galaxy would observe all other galaxies apparently retreating.
If we extrapolate such a mutual recession back into the past, we find that there was a time—perhaps 15 billion or 20 billion years ago—when all of the galaxies must have been “touching”; that is, confined to an extremely small volume of space. Matter in its present form could not survive such astonishing compressions. The very earliest stages of that expanding universe must have been dominated by radiation rather than
matter. It is now conventional to talk of this time as the Big Bang.
Three kinds of explanation have been offered for this expansion of the universe: the Steady State, Big Bang and Oscillating Universe cosmologies. In the Steady State hypothesis, the galaxies recede from one another, the more distant galaxies moving with very high apparent velocities, their light shifted by the Doppler effect to longer and longer wavelengths. There will be a distance at which a galaxy will be moving so fast that it passes over what is called its event horizon and, from our vantage point, disappears. There is a distance so great that, in an expanding universe, there is no chance of getting information from beyond it. As time goes on, if nothing else intervenes, more and more galaxies will disappear over the edge. But in the Steady State cosmology, the matter lost over the edge is exactly compensated for by new matter continuously created everywhere, matter that eventually condenses into new galaxies. With the rate of disappearance of galaxies over the event horizon just balanced by the creation of new galaxies, the universe looks more or less identical from every place and in every epoch. In the Steady State cosmology there is no Big Bang; one hundred billion years ago the universe would have looked just the same, and one hundred billion years from now, likewise. But where does the new matter come from? How can matter be created from nothing? Proponents of the Steady State cosmology answer that it comes from whatever place proponents of the Big Bang get their Bang from. If we can imagine all the matter in the universe discontinuously created from nothing 15 billion to 20 billion years ago, why are we unable to imagine it being created in a tenuous trickle everywhere, continuously and forever? If the Steady State hypothesis is true, there was never a time when the galaxies were much closer. The universe in its largest structures is then unchanging and infinitely old.
But as placid and, in a strange way, as satisfying as the Steady State cosmology is, there is strong evidence against it. Whenever a sensitive radio telescope is
pointed anywhere in the sky, the constant chatter of a kind of cosmic static can be detected. The characteristics of this radio noise match almost exactly what we would expect if the early universe was hot and filled with radiation in addition to matter. The cosmic blackbody radiation is very nearly the same everywhere in the sky and looks very much to be the distant rumblings of the Big Bang, cooled and enfeebled by the expansion of the universe but coursing still down the corridors of time. The primeval fireball, the explosive event that initiated the expanding universe, can be observed. Supporters of the Steady State cosmology must now be reduced to positing a large number of special sources of radiation which together somehow mimic exactly the cooled primeval fireball, or proposing that the universe far beyond the event horizon is steady state but, by a peculiar accident, we live in a kind of expanding bubble, a violent pimple in a much vaster but more placid universe. This idea has the advantage or flaw, depending on your point of view, of being impossible to disprove by any conceivable experiment, and virtually all cosmologists have abandoned the Steady State hypothesis.
If the universe is not in a steady state, then it is changing, and such changing universes are described by evolutionary cosmologies. They begin in one state, and they end in another. What are the possible fates of the universe in evolutionary cosmologies? If the universe continues to expand at its present rate and galaxies continue to disappear over the event horizon, there will eventually be less and less matter in the visible universe. The distances between galaxies will increase, and there will be fewer and fewer of the spiral nebulae for the successors of Slipher, Hubbell and Humason to view. Eventually the distance from our Galaxy to the nearest galaxy will exceed the distance to the event horizon, and astronomers will no longer be able to see even the nearest galaxy except in (very) old books and photographs. Because of the gravity that holds the stars in our Galaxy together, the expanding universe will not dissipate our Galaxy, but even here a strange
and desolate fate awaits us. For one thing, the stars are evolving, and in tens or hundreds of billions of years most present stars will have become small and dark dwarf stars. The remainder will have collapsed to neutron stars or black holes. No new matter will be available for a vigorous younger generation of stars. The Sun, the stars, the entire Milky Way Galaxy, will slowly turn off. The lights in the night sky will go out.
But in such a universe there is a further evolution still. We are used to the idea of radioactive elements, certain kinds of atoms that spontaneously decay or fall to pieces. Ordinary uranium is one example. But we are less familiar with the idea that every atom except iron is radioactive, given a long enough period of time. Even the most stable atoms will radioactively decay, emit alpha and other particles, and fall to pieces, leaving only iron, if we wait long enough. How long? The American physicist Freeman Dyson of the Institute for Advanced Study calculates that the half-life of iron is about 10
500
years, a one followed by five hundred zeros—a number so large that it would take a dedicated numerologist the better part of ten minutes just to write it down. So if we wait just a little longer—10
600
years would do just fine—not only would the stars have gone out, but all the matter in the universe not in neutron stars or black holes would have decayed into the ultimate nuclear dust. Eventually, galaxies will have vanished altogether. Suns will have blackened, matter disintegrated, and no conceivable possibility will remain for the survival of life or intelligence or civilizations—a cold and dark and desolate death of the universe.
But need the universe expand forever? If I stand on a small asteroid and throw a rock up, it will leave the asteroid, there being on such a worldlet not enough gravity to drag the rock back. If I throw the same rock at the same speed from the surface of the Earth, it will of course turn around and fall down because of the substantial gravity of our planet. But the same sort of physics applies to the universe as a whole. If there is less than a certain amount of matter, each galaxy will feel an insufficient tug from the gravitational attraction of
the others to be slowed down appreciably, and the expansion of the universe will continue forever. On the other hand, if there is more than a certain critical mass, the expansion will eventually slow, and we will be saved from the desolation teleology of a universe that expands forever.
What, then, would be the fate of the universe? Why, then an observer would see expansion eventually replaced by contraction, the galaxies slowly and then at an ever-increasing pace approaching one another, a careening, devastating smashing together of galaxies, worlds, life, civilizations and matter until every structure in the universe is utterly destroyed and all the matter in the cosmos converted into energy: instead of a universe ending in cold and tenuous desolation, a universe finishing in a hot and dense fireball. It is very likely that such a fireball would rebound, leading to a new expansion of the universe and, if the laws of nature remain the same, a new incarnation of matter, a new set of condensations of galaxies and stars and planets, a new evolution of life and intelligence. But information from our universe would not trickle into that next one and, from our vantage point, such an oscillating cosmology is as definitive and depressing an end as the expansion that never stops.
The distinction between a Big Bang with expansion forever and an Oscillating Universe clearly turns on the amount of matter there is. If the critical amount of matter is exceeded, we live in an Oscillating Universe. Otherwise we live in one that expands forever. The expansion times—measured in tens of billions of years—are so long that these cosmological issues do not affect any immediate human concerns. But they are of the most profound import for our view of the nature and fate of the universe and—only a little more remotely—of ourselves.
In a remarkable scientific paper published in the December 15, 1974 issue of the
Astrophysical Journal
, a wide range of observational evidence is brought to bear on the question of whether the universe will continue to expand forever (an “open” universe) or whether it will
gradually slow down and recontract (a “closed” universe), perhaps as part of an infinite series of oscillations. The work is by J. Richard Gott III and James E. Gunn, then both of the California Institute of Technology, and David N. Schramm and Beatrice M. Tinsely, then of the University of Texas. In one of their arguments they review calculations of the amount of mass in and between galaxies in “nearby” well-observed regions of space and extrapolate to the rest of the universe: they find that there is not enough matter to slow the expansion down.