Science › Creation theories
Cosmic inflation
1980· Theory
Well supportedStrong evidence or math, but not fully tested.
The leading theory that in its first instant the universe grew at least 10²⁶-fold in under 10⁻³² seconds, before the hot Big Bang began.
See it on the Science map →How it works
Alexei Starobinsky (1980) found quantum effects of gravity could drive runaway expansion. Alan Guth (1980), independently, proposed it to fix the Big Bang's horizon and flatness puzzles, and named it. Guth's version could not end smoothly; Andrei Linde, and separately Andreas Albrecht and Paul Steinhardt, fixed that in 1982 with slow roll. The energy of a field called the inflaton made space double in size 87 times or more, roughly 10⁻³⁶ to 10⁻³² seconds in. Its decay, reheating, then filled space with hot particles.
Why it matters
It explains why space is flat and evenly hot, and where the seeds of galaxies came from. It means the hot Big Bang had a prelude whose physics we can partly test.
Think of it like
Like inflating a balloon to the size of Earth: an ant on it sees a flat surface, and every wrinkle is stretched far beyond its view.
How we know
Its forecasts came true: space is flat within about 0.2% (Planck, 2018), and the ripples have a slight tilt, about 0.965 against a perfectly even 1, predicted by Mukhanov and Chibisov in 1981.
Still a mystery
Unproven: the inflaton's identity, its energy hill, and how inflation began. Its gravitational waves have not been seen. In 2017 critics publicly called it unfalsifiable; 33 leading physicists signed a reply defending it.
Connections (14)
▸Heisenberg's uncertainty principle supplied the tiny jitters stretched by Cosmic inflationWell supported
The relationship
The uncertainty principle means no field can be perfectly smooth; it always jitters slightly. During inflation, a fraction of a second after the Big Bang, space expanded enormously and stretched these microscopic jitters to cosmic size. They became slightly denser regions that gravity later pulled together into galaxies and galaxy clusters.
How the link was found
Worked out in 1981–1982 by Viatcheslav Mukhanov and Gennady Chibisov, and by Hawking, Starobinsky, Guth, Pi, Bardeen, Steinhardt and Turner around a 1982 Cambridge workshop.
Is it proven?
The predicted pattern of tiny temperature variations in the cosmic microwave background, nearly but not exactly equal across scales, was measured by COBE, WMAP and Planck. Strong support, though inflation itself is not proven.
Open questions and mysteries
What drove inflation is unknown. A predicted imprint of primordial gravitational waves in the background radiation has not yet been detected.
▸Cosmic inflation seeded Galaxies and the cosmic webWell supported
The relationship
During inflation, tiny quantum jitters in the driving field were stretched to astronomical size. They left regions very slightly denser than others, by about 1 part in 100,000. Over billions of years, gravity pulled matter from thinner regions into denser ones, growing those ripples into galaxies, clusters, filaments and voids.
How the link was found
Viatcheslav Mukhanov and Gennady Chibisov in 1981; Hawking, Starobinsky, Guth and others at a 1982 Cambridge workshop.
Is it proven?
The ripple pattern measured in the microwave background, fed into computer simulations, reproduces the cosmic web seen in galaxy surveys. That the ripples came from inflation is well supported but unconfirmed.
Open questions and mysteries
Whether the seeds were exactly as inflation predicts, or show small departures hinting at different physics, is being tested now.
▸False vacuum: energy trapped in empty space powers Cosmic inflationWell supported
The relationship
False-vacuum energy stays the same in every cubic meter as space grows, so it never dilutes. General relativity says such energy has negative pressure, which makes gravity repel. The result is a steady expansion rate: space doubles, then doubles again, in equal tiny intervals. As long as the false vacuum lasts, the growth continues exponentially.
How the link was found
Alan Guth, 1979–80. Erast Gliner had suggested in 1965 that vacuum-like matter repels; Katsuhiko Sato in Japan and Demosthenes Kazanas proposed exponential growth from vacuum energy in 1980–81.
Is it proven?
Vacuum energy's push is observed today: dark energy, found in 1998, accelerates expansion the same way, though, in typical models, over 10¹⁰⁰ times more weakly. That inflation itself ran on this is inferred, not observed.
Open questions and mysteries
Modern models mimic a false vacuum with a slowly rolling field rather than a truly stuck one. Which, if either, powered our universe's inflation is unknown.
▸The inflaton field drives Cosmic inflationWell supported
The relationship
How high the inflaton sits on its hill sets how fast space inflates; the slope sets how long. A higher perch means faster expansion and stronger gravitational waves; a gentler slope means a longer run and more doublings. When the field drops off, inflation stops.
How the link was found
Guth's 1980 model used a field already present in particle theories. Andrei Linde's 1983 'chaotic inflation' showed any field starting high enough on a simple hill would do.
Is it proven?
Indirect. The ripple pattern fits a single field: Planck (2013–2018) found the ripples' statistics almost perfectly random and bell-shaped, with no sign of extra fields or exotic effects.
Open questions and mysteries
Why did the field start high on its hill, in a patch smooth enough to begin inflating? How the very first inflating patch arose is unknown.
▸Slow roll: energy rolling down a hill sustains and smoothly ends Cosmic inflationWell supported
The relationship
To flatten space and even out the sky, inflation must last at least about 60 'e-folds', each multiplying size by 2.7, for 10²⁶-fold growth. Slow roll provides that and ends smoothly. Because the field speeds up slightly, later ripples, now on smaller scales, are a bit weaker. The tilt depends on how long inflation ran after today's visible scales were stretched; in many models, 60 e-folds gives about 0.965.
How the link was found
Linde and Albrecht–Steinhardt showed in 1982 that slow roll lasts long enough. Viatcheslav Mukhanov and Gennady Chibisov predicted the slight tilt in 1981.
Is it proven?
Planck (2013) measured the tilt below 1 at over five standard deviations, as predicted three decades earlier. It is one of inflation's strongest successes.
Open questions and mysteries
Inflation could have lasted far longer than 60 e-folds; if so, every trace of what came before is stretched beyond view. Why it lasted at least that long is unexplained.
▸Alan Guth proposed Cosmic inflationWell supported
The relationship
Guth realized that a universe stuck in a false vacuum would grow exponentially, and that this growth would flatten space and let distant regions share one temperature. He named the idea 'inflation' and laid out its problem-solving power in one paper. His 1980 lectures spread it fast, starting a worldwide effort to make it work.
How the link was found
Alan Guth at SLAC, Stanford. His notebook entry of December 1979 is headed 'SPECTACULAR REALIZATION'. The paper, 'Inflationary universe: A possible solution to the horizon and flatness problems', appeared in Physical Review D in January 1981.
Is it proven?
His specific model failed, as he admitted in the paper, but the core idea passed later tests: flat space and the ripple pattern measured by COBE, WMAP and Planck.
Open questions and mysteries
Did inflation happen at all? It is well supported but awaits a signature no rival can mimic, such as its gravitational waves. Which of hundreds of models is right is open.
▸Cosmic inflation may never fully stop, giving Eternal inflation and bubble universesSpeculative
The relationship
While the inflaton rolls, it keeps getting random quantum kicks. In each e-fold of time, every region grows into about 20 regions. If more than about one in 20 is kicked uphill rather than rolling down, the volume still inflating grows rather than shrinks. In many models that happens high on the hill, so inflation ends only in patches, never everywhere.
How the link was found
Alexei Starobinsky built the standard math for these random kicks, 'stochastic inflation', in 1986. Guth later gave the simple false-vacuum version: if the half-life exceeds the doubling time, inflation is eternal.
Is it proven?
It uses the same quantum-jitter physics that correctly predicted the microwave background's ripples, but applies it to field values far beyond anything observable. Whether it occurs depends on the hill's shape.
Open questions and mysteries
In their final joint paper (2018), Stephen Hawking and Thomas Hertog argued that eternal inflation yields a smoother, finite universe rather than an endless fractal of pockets. Unresolved.
▸Cosmic inflation pushes back the mystery of What came before the Big Bang?Speculative
The relationship
If inflation happened, the hot Big Bang did not start at t = 0 but at reheating, when inflation ended. The question 'what came before the hot Big Bang?' gets an answer, inflation, and is replaced by a harder one: what came before inflation? Inflation also erases clues, since anything earlier was stretched far beyond view.
How the link was found
Arvind Borde, Alan Guth and Alexander Vilenkin proved in 2003 that inflating spacetimes cannot extend forever into the past, so inflation alone cannot remove the need for a beginning.
Is it proven?
The 2003 theorem is a mathematical result that does not even depend on Einstein's equations, only on space expanding on average. What it implies, a true beginning or just the edge of our description, is debated.
Open questions and mysteries
What started inflation: a quantum birth from nothing, the no-boundary state, a bounce from an earlier universe, or something else? Inflation hides the answer behind its own stretching.
▸The horizon and flatness puzzles found their leading answer in Cosmic inflationWell supported
The relationship
Inflation treats all three puzzles by stretching. Any curvature is flattened, so flatness becomes the expected outcome rather than a coincidence. Regions now far apart were in contact before inflation pulled them out of reach. Any magnetic monopoles made earlier were diluted until essentially none remain in view. Fine-tuning is traded for one requirement: enough inflation.
How the link was found
Guth set out in 1979, with Henry Tye, to solve the monopole puzzle; John Preskill had shown in 1979 that grand unified theories make far too many. Guth then saw the same mechanism fixed flatness and horizons.
Is it proven?
Inflation predicted more than it was built for: slightly tilted, random-looking ripples, and correlations on scales larger than the early horizon (WMAP, 2003). Those successes suggest the answer is real, not just possible.
Open questions and mysteries
Bouncing and cyclic universes, and theories in which light was once faster, also claim to solve the puzzles. Whether any can match inflation's full record is debated.
▸Cosmic inflation predicts gravitational waves sought as B-modes: hunting inflation's gravitational wavesOpen question
The relationship
Inflation stretched quantum jitters of spacetime itself, not just of the inflaton, into gravitational waves spanning the cosmos. Their strength depends directly on inflation's energy: the higher the energy, the stronger the waves. They would twist the afterglow's polarization into B-modes on scales of a degree and larger, so measuring them would read inflation's energy off the sky.
How the link was found
Alexei Starobinsky predicted gravitational waves from the early universe in 1979. Valery Rubakov and colleagues (1982) and Larry Abbott and Mark Wise (1984) showed their strength measures inflation's energy.
Is it proven?
Not yet seen. BICEP/Keck (2021) limits the waves to under about 3.6% of the density ripples, ruling out some of the simplest models. Starobinsky's model predicts about 0.3–0.4%, a target for coming experiments.
Open questions and mysteries
Some workable inflation models make waves far too faint ever to detect, so a non-detection cannot rule inflation out. Critics see that flexibility as a weakness.
▸Grand unification epoch may have triggered Cosmic inflationSpeculative
The relationship
In Guth's original idea, the field that broke the grand unified force got stuck as the universe cooled, like water supercooled below freezing without turning to ice. Trapped in that false vacuum, its energy drove runaway expansion. That would also dilute the magnetic monopoles a unification transition should make in huge numbers, explaining why none are found.
How the link was found
Alan Guth, working with Henry Tye on the overproduction of monopoles in grand unified theories, realized in December 1979 that supercooling would inflate space. His paper appeared in 1981.
Is it proven?
Untested. Grand unification is unconfirmed, and Guth's unification-based model failed because its bubbles could not merge smoothly. Most current models use a separate inflaton field. Monopole dilution is an argument, not evidence.
Open questions and mysteries
Searches for inflation's gravitational waves cap its energy below about 10¹⁶ GeV, close to the unification scale. Coincidence, or a real connection? Unknown.
▸Cosmic inflation ends with Reheating: inflation's energy becomes hot matterWell supported
The relationship
Inflation by itself only makes space: huge, smooth, flat and empty, with all the energy held in the field driving it. To give a universe like ours, inflation must end and that energy must turn into matter almost evenly everywhere. Quantum jitter made it end a hair earlier in some places than others; those tiny timing differences became the 1-in-100,000 density ripples later seen in the microwave background.
How the link was found
Even reheating became possible with 1982's slow-roll models. How its uneven timing seeds the ripples was worked out that year, around a Cambridge workshop, by Hawking, Starobinsky, Guth and So-Young Pi, and Bardeen, Steinhardt and Turner.
Is it proven?
The ripple pattern these timing differences predict matches Planck's maps. Reheating itself is inferred rather than seen: if inflation occurred and the universe was hot afterward, it must have happened.
Open questions and mysteries
Was reheating quick or gradual? Did other regions keep inflating while ours reheated, as eternal inflation suggests?
▸Cosmic inflation might have made Primordial black holes as dark matterSpeculative
The relationship
Simple inflation leaves ripples of about 1 part in 100,000, far too gentle to make black holes. But if the inflaton crossed a nearly flat ledge on its energy hill, it would slow down and amplify ripples on small scales, perhaps a thousandfold. After inflation, those strongly overdense patches collapse directly into black holes as soon as light can cross them.
How the link was found
Pavel Ivanov, Pavel Naselsky and Igor Novikov (1994) and Juan García-Bellido, Andrei Linde and David Wands (1996) linked inflation models to primordial black holes as dark matter.
Is it proven?
Not proven. The afterglow measures ripples only on large scales; small scales are barely constrained. Such spikes would also leave a background of gravitational waves, a target for pulsar-timing arrays.
Open questions and mysteries
The gravitational-wave hum pulsar-timing arrays reported in 2023 most likely comes from pairs of giant black holes, but an early-universe origin is not excluded. Building such a ledge needs careful tuning.
▸ΛCDM: the standard model of cosmology takes its starting ripples from Cosmic inflationWell supported
The relationship
ΛCDM, the standard model of cosmology, fits the afterglow and galaxy surveys with six numbers. Two describe the starting ripples: their overall strength, about 1 part in 100,000, and their tilt, about 0.965. ΛCDM uses these as inputs it cannot explain. Inflation is the leading account of where they came from and why they look as they do.
How the link was found
The six-number model was settled with WMAP's first results in 2003 (David Spergel and colleagues). Its ripple inputs take the simple, nearly even form inflation predicted in the early 1980s.
Is it proven?
Planck (2018) fits thousands of measured ripple sizes with these six numbers. Adding extras, such as curvature, gravitational waves or a changing tilt, does not improve the fit.
Open questions and mysteries
The fit does not prove inflation; any process making the same ripples would serve. Strains in ΛCDM, like the Hubble tension, may need changes, though probably not to the ripple inputs.