Inflation is a wildfire that will inevitably take over the forest, as long as there is some chance that it will start.
Could the Big Bang have been caused by a colossal stick of TNT, or perhaps a thermonuclear explosion? Or maybe a gigantic ball of matter collided with a gigantic ball of antimatter, releasing an untold amount of energy in a powerful cosmic blast.
In fact, none of these scenarios can plausibly account for the Big Bang that started our Universe, which had two very special features distinguishing it from any typical explosion.
First, the Big Bang was far more homogeneous, on large scales, than can be explained by an ordinary explosion. In discussing homogeneity, however, I must first clarify that the Universe is in many ways conspicuously inhomogeneous. Palo Alto is very different from San Francisco, and the stars, galaxies, and clusters of galaxies are scattered through space in a lumpy, complex pattern. Cosmologically speaking, however, all this structure is small-scale. We can focus on the large scales, for example, by dividing space into cubes of 300 million light-years or more on a side. We would find that each such cube closely resembles the others in all its average properties, such as mass density, galaxy density, and light output. This large-scale uniformity can be seen in galaxy surveys, but the most dramatic evidence comes from the cosmic background radiation. Data from the COBE satellite, confirmed by subsequent ground-based observations, show that this radiation has the same temperature in all directions (after correcting for the motion of the Earth) to an accuracy of one part in 100,000.
To see how difficult it is to explain this uniformity as the result of an ordinary explosion, we need to know a little about the history of the cosmic background radiation. The early Universe was so hot that electrons would have been ripped away from atoms, resulting in a plasma that filled space. Such a plasma is very opaque, so the photons that now make up the cosmic background radiation were constantly absorbed and re-emitted. After about 300,000 years, however, the Universe cooled enough for the plasma to form a gas of neutral atoms, which is very transparent. The photons of the cosmic background radiation have traveled on straight lines ever since, so they provide today an image of the Universe at an age of 300,000 years, just as the photons reaching your eye at this moment provide an image of the page in front of you. Thus, the observations of the cosmic background radiation show that the Universe was uniform in temperature, to one part in 100,000, at an age of several hundred thousand years.
Under many circumstances such uniformity would be easy to understand, since anything will come to a uniform temperature if left undisturbed for a long enough time. In the standard Big Bang theory, however, the Universe evolves so quickly that there is no time for the uniformity to be established. One can pretend, for the sake of discussion, that the Universe is populated by little purple creatures, each equipped with a furnace and a refrigerator, and each dedicated to the cause of creating a uniform temperature. Those little creatures, however, would have to communicate at roughly 100 times the speed of light if they are to achieve their goal of creating a uniform temperature across the visible Universe by 300,000 years after the Big Bang. Since neither sticks of dynamite nor balls of matter and antimatter can transmit their energy faster than light, they cannot account for the uniformity. The classical form of the Big Bang theory requires us to postulate, without explanation, that the primordial fireball filled space from the beginning. The temperature was the same everywhere by assumption, not as a consequence of any physical process. This shortcoming is known as the "horizon problem," since cosmologists use the word "horizon" to indicate the largest distance that information or energy could have traversed since the instant of the Big Bang, given the restriction of the speed of light.
The second special feature of the Big Bang, which is very difficult to imagine arising from a standard explosion, is a remarkable coincidence called the "flatness problem." This problem concerns the pinpoint precision with which the mass density of the early Universe must be specified for the Big Bang theory to agree with reality.
First, we need to review a little vocabulary. If the mass density of the Universe exceeds a value called the critical density, then gravity will be strong enough to reverse the expansion eventually, causing the Universe to recollapse into what is sometimes called the big crunch. If the mass density is less than the critical value, the Universe will go on expanding forever. The ratio of the actual mass density to the critical value is known to cosmologists by the Greek letter omega (). General relativity implies that the geometry of the Universe is Euclidean only if omega is one, so an = 1 universe is called "flat" (see box on the right).
Omega is very difficult to determine, but it is safe to say that its present value lies somewhere in the range of 0.1 to 2. That seems like a broad range, but consideration of the time development of the Universe leads to a spectacularly different point of view. = 1 is an unstable equilibrium point of cosmological evolution, which means that it resembles the situation of a pencil balancing on its sharpened tip. The phrase equilibrium point implies that if omega is ever exactly equal to one, it will remain exactly equal to one forever - just as a pencil balanced precisely on end will, according to the laws of classical physics, remain forever vertical. The word unstable means that any deviation from the equilibrium point, in either direction, will rapidly grow. If the value of omega in the early Universe was just a little above one, it would have rapidly risen toward infinity; if it was just a smidgen below one, it would have rapidly fallen toward zero. For omega to be anywhere near one today, it must have been extraordinarily close to one at early times. For example, consider one second after the Big Bang, the time at which the processes related to Big Bang nucleosynthesis were just beginning. For omega to be anywhere in the allowed range today, at that time omega must have equaled one to an accuracy of 15 decimal places!
A simple explosion gives no explanation for this razor-sharp finetuning, and indeed no explanation can he found in the traditional version of the Big Bang theory. The initial values of the mass density and expansion rate are not predicted by the theory, but must be postulated. Unless we postulate that the mass density at one second just happened to have a value between 0.999999999999999 and 1.000000000000001 times the critical density, however, the theory will not describe a universe that resembles the one in which we live.