Run the film backward
In the guides just before this one you learned the central fact of modern cosmology: space itself is stretching, carrying the galaxies apart, so the more distant a galaxy is the faster it recedes. You also learned the crucial subtlety — that cosmological redshift is not the galaxies flying through space like shrapnel, but [[expansion-of-space|space expanding]] underneath them, lengthening the light in flight. Now we do something simple and audacious with that fact: we run the film backward. If everything is drifting apart today, then yesterday it was a little closer together, and a billion years ago closer still.
Keep winding back and the consequences pile up. Squeeze the same matter and radiation into ever less room and it gets denser; and just as a bicycle pump warms when you compress the air inside it, a compressed universe gets hotter. So the early universe was not just crowded — it was scorching. Far enough back, it was too hot for atoms to hold together, then too hot even for atomic nuclei, a glowing fog of bare particles and light. That hot, dense early state, and the expansion that has been cooling and thinning it ever since, is what we mean by the [[big-bang-theory|Big Bang]].
Notice what kind of claim this is. The Big Bang is not, in the first instance, a story about a moment of creation; it is a story about a hot, dense beginning that the universe has been expanding away from for about 13.8 billion years (we will see where that [[age-of-the-universe|age]] comes from below). A natural worry is that this is just clever extrapolation — winding a graph back past the data and trusting it. That worry is fair, and it is exactly why the rest of this guide matters. A backward extrapolation only becomes science when it predicts something specific that we can then go and check. The Big Bang made two such predictions, decades before either was found.
Pillar one: the expansion itself
The first pillar is the one you have already met: the expansion. A universe that is expanding today is the universe asking to have a denser yesterday. This is more than a vibe — the expansion comes with a clock. If two galaxies are moving apart at a steady rate, then running them back to where they touched takes a definite amount of time, and that time is roughly the inverse of how fast space is stretching per unit distance. Plug in the measured stretching rate and you get a number close to 14 billion years. That this homemade estimate lands so near the carefully computed age of about 13.8 billion years is the first quiet reassurance that the whole picture hangs together.
Here we must kill the most stubborn misconception in all of cosmology, because the name actively misleads. The Big Bang was not an explosion in space, and it had no center. An explosion is debris flung outward from one spot into surrounding empty room; you could, in principle, walk back to the spot it came from. The Big Bang was nothing like that. It was not stuff expanding into a pre-existing void — it was space itself expanding, everywhere at once. There was no special point that was 'the Bang'; the early hot dense state filled all of space, all of it equally, and it has been thinning out uniformly ever since.
Pillar two: the relic glow
The expansion is suggestive but, on its own, it could fit other stories too. The second pillar is far harder to explain away, and it is a genuine prediction. Wind back to when the universe was a few thousand times smaller and correspondingly hotter — a glowing plasma at thousands of degrees. In that fog, light could not travel freely: it scattered off loose electrons at every turn, the way light cannot pass cleanly through dense mist. The universe was opaque, like the inside of a star.
Then, about 380,000 years after the beginning, the expansion had cooled everything to around 3,000 kelvin — cool enough that electrons and protons could finally stick together into neutral hydrogen atoms. This moment is called [[recombination|recombination]], and at a stroke it cleared the fog: with the loose electrons mopped up, light no longer scattered, and the universe became transparent. The light that was flying free at that instant has been travelling ever since, across all of cosmic history, with nothing to stop it. The sphere from which it set out is the [[surface-of-last-scattering|surface of last scattering]] — the most distant thing we can ever see with light, a wall of glow surrounding us on every side.
That glow set out at about 3,000 kelvin, the color of a dull-red ember. But it has been climbing toward us through expanding space for 13.8 billion years, and that expansion has stretched its wavelengths by a factor of about a thousand — the same stretching of light you met with the galaxies, only far more extreme. Stretched a thousandfold, the ember-glow is no longer visible light but faint microwaves, and its temperature has dropped to just 2.7 kelvin above absolute zero. This is the [[cmb-relic-radiation|cosmic microwave background]] — the CMB — and it really is the relic heat of the Big Bang, arriving from every direction in the sky.
Pillar three: the first elements
The third pillar reaches back further still — to the first few minutes, long before any atoms or stars. Recall from the stellar-physics rungs that stars forge heavy elements in their cores by fusion, and scatter them when they die. That neatly explains the carbon, oxygen, and iron around us. But it does not explain the helium. The universe is about 24% helium by mass almost everywhere we look — far more than stars have had time to make. Worse, even the oldest, most pristine gas clouds, which have never been polluted by any star, already carry this same large helium fraction. Where did it come from, if not from stars?
The hot Big Bang answers cleanly. In the first second, the universe was hotter than the core of any star — a soup of bare protons and neutrons. As it expanded and cooled, for a brief window between roughly one second and a few minutes old, it was at just the right temperature for these to fuse, exactly the way a stellar core fuses, but everywhere at once. This is [[big-bang-nucleosynthesis|Big Bang nucleosynthesis]]. Then the relentless expansion cooled and thinned the soup below the threshold for fusion, and the cooking stopped — after only a few minutes. There was no time to build heavy elements; the furnace shut off too soon.
The triumph is in the numbers. From a single knob — the density of ordinary matter — the theory predicts the primordial recipe: about three-quarters hydrogen and one-quarter helium by mass, plus a trace of deuterium (heavy hydrogen) and a whisper of lithium, and essentially nothing heavier. Those predictions match what we measure in the oldest, least-polluted gas with striking precision. The same matter density that fits the helium also fits the deuterium and the lithium — one number, several independent checks, all agreeing. That is the signature of a theory describing something real, not a fudge.
TIMELINE OF THE HOT EARLY UNIVERSE
~1 sec to ~3 min : Big Bang nucleosynthesis
protons + neutrons fuse -> H, He, traces of D, Li
primordial mix ~ 75% hydrogen + 25% helium (by mass)
~380,000 years : Recombination
electrons + protons -> neutral atoms; fog clears
released light -> (stretched x1000) -> the CMB today
NOW (~13.8 Gyr) : CMB observed at ~2.7 K from every directionWhat the Big Bang does and does not claim
Three independent lines — the expansion's clock, the relic glow at 2.7 kelvin, and the primordial helium and deuterium — all point back to the same hot, dense beginning about 13.8 billion years ago, each predicted in advance and each confirmed. No rival theory explains all three together. That convergence is why the Big Bang is not a guess but the well-tested backbone of cosmology. The point is not that any single observation is unanswerable, but that they triangulate: three different clocks, set by three different bits of physics, read the same age and the same story.
Now the honest limits, because they matter as much as the evidence. The Big Bang theory describes the universe expanding and cooling from a hot dense state; it does not claim to describe the very first instant, and it certainly does not claim 'something came from nothing'. As you wind the clock toward time zero, the known physics breaks down before you reach it, and what happened in that first sliver of a second is at the frontier — the territory of cosmic inflation and ideas you will meet in the early-universe rung ahead. 'The Big Bang' names the hot, dense, expanding era we can test; it does not name a proven moment of creation. Honesty about that boundary is part of the science, not a weakness in it.
And the story is not finished. The precise expansion rate today is genuinely in dispute: two careful ways of measuring it disagree by more than their stated errors, a live puzzle called the [[hubble-tension|Hubble tension]] that may yet force a tweak to the model — or may turn out to be a subtle measurement error. That is a quarrel about the fine print, not about the framework: nobody serious doubts the expansion, the relic glow, or the primordial elements. With the hot beginning and its evidence in hand, the remaining guides in this rung can ask the next questions — what *shape* the universe has, what it is made of, and how the whole thing will end.