Skip to content
Saturday, September 26, 202668 days to the Dice Letter centennialNo physics degree or shared belief required.
GOD PLAYS DICE™The magazine of big questions

THE QUESTION LIBRARY ORIGINS

Is the universe infinite?

The part we can see has a limit, space looks flat, and flat can still mean finite, so here is what the evidence says and why the final answer may stay out of reach.

A black line drawing of a ring-shaped surface, like a doughnut, seen at an angle. On its front face, a small circled dot stands for an observer, and a small curved square holds a large question mark. A straight black arrow labeled B runs from the question mark to the dot. A red arrow labeled A leaves the question mark, runs down and around the back of the ring as a dashed line, and comes over the top to reach the dot from the other side.
A two-dimensional picture of a finite space with no edge. Light from the object marked “?” can reach the observer by a short path (B) or by going all the way around (A), so the same object could appear twice. Astronomers have searched the sky for this kind of repeat on a cosmic scale and, for the shapes tested so far, have not found one.Boud, via Wikimedia Commons · CC BY-SA 4.0Image source ↗

THE SHORT ANSWER

Nobody knows yet. The part we can see, called the observable universe, is finite. Its edge is now about 46 billion light-years away in every direction. The whole universe is much bigger than that. Careful measurements show that space is flat, or very close to flat. A flat universe could go on forever. But it could also be finite and wrap around on itself, like a video game screen. We may never be able to tell which.

  • The observable universe is a bubble around us whose edge is now about 46 billion light-years away.
  • Its size is set by how far light has had time to travel since the Big Bang, not by where the universe stops.
  • The Planck satellite, together with maps of galaxies, found that space is flat to within about 0.2%.
  • Flat does not have to mean infinite, because a flat space can loop back on itself and still have a finite size.
  • Searches have found no sign of such a loop smaller than the part of the universe we can see, for the shapes tested so far.

WHERE THE EVIDENCE STANDS

How sure are we?

  • EstablishedThe part of the universe we can observe is finite, reaching about 46 billion light-years in every direction.Astronomers Tamara Davis and Charles Lineweaver (2004) give about 46 billion light-years as the present distance to the edge of what we can see; a NASA expert gives about 92 billion light-years across.
  • EstablishedOn the largest scales, space is flat or very close to flat.Planck 2018 results combined with galaxy maps (BAO): flat to within about 0.2%. NASA’s WMAP team had found flatness within 0.4%. Planck data taken alone leaned slightly toward a closed shape, which is still discussed.
  • Not supportedBecause space is flat, the universe must be infinite.Einstein’s theory describes the local shape of space, not how it connects overall. Researchers list flat shapes of finite size, such as the three-torus. NASA’s Ask an Astrophysicist answered “No” to whether WMAP proved an infinite universe.
  • Open questionThe whole universe goes on forever.A NASA expert says we do not know if the universe is finite or infinite. A 2009 study found the odds of an infinite universe fall between 67% and 98%, depending on assumptions, and argued curvature below a tiny level may never be measurable.
  • Not supportedThe universe wraps around on a scale smaller than the part we can see.Planck searched the microwave background for repeating patterns and found no sign of such a loop for the shapes it tested. Researchers note that many other possible shapes have not been fully tested yet.
What the labels mean

WHAT WE DON’T KNOW YET

Nobody knows whether space goes on forever, curves back on itself very gently, or wraps around in a loop larger than what we can see. Searches for wrap-around shapes have not yet tested every possible shape.

WHAT WOULD CHANGE THIS ANSWER

A clear sign of curvature would change this answer: a slight closed curve would mean a finite universe. Matching patterns in the oldest light in the sky would show a wrap-around universe with a definite size.

Nobody knows yet. That is the honest answer, and it is a good one. What we do know is where our view ends. The part of the universe we can see, called the observable universe, is a giant bubble around us. Its edge is now about 46 billion light-years away in every direction. A light-year is the distance light travels in one year, about 6 trillion miles. So the bubble we can see is finite. It is huge, but it has a size.

That edge is not a wall. It is a horizon, like the line where the sea seems to end. It marks how far light has had time to travel to us since the Big Bang, about 13.8 billion years ago. Space kept stretching while that light was on its way, so its starting points are now much farther than 13.8 billion light-years. The universe keeps going well beyond that horizon. We cannot see that part.

So is the whole thing infinite? Here scientists turn to the shape of space. Spacecraft have measured it with great care. The best result, from the Planck satellite together with maps of galaxies, is that space is flat to within about 0.2%. In flat space, the angles of a huge triangle add up to 180 degrees, just like on paper.

Many people hear “flat” and think “infinite.” The simplest flat universe does go on forever. But flat does not have to mean endless. A flat space can also wrap around, like an old video game where you walk off the right side of the screen and come back on the left. That kind of universe would be finite, with no edge at all.

So far, no one has seen signs of such wrapping. And if space curves only very gently, the curve may be too small to ever measure. We label the whole-universe question an Open question. The size of the part we can see is Established.

THE LONG ANSWER

How big is the part we can see?

The universe is about 13.8 billion years old, so you might expect to see 13.8 billion light-years in each direction and no farther. But space has been expanding the whole time. The light from the most distant places left them long ago, and since then the space between us has stretched. In a 2004 paper, astronomers Tamara Davis and Charles Lineweaver worked out where those places are now: about 46 billion light-years away. That distance is the radius of the observable universe. A NASA expert, speaking in NASA’s “We Asked a NASA Expert” series, rounds the full width to about 92 billion light-years.

Two things surprise people here. First, the bubble is centered on us only because we are the ones looking. Someone in a galaxy billions of light-years away would have their own bubble, centered on them. Second, the edge is not a place where anything stops. NASA’s Ask an Astrophysicist team explains that the universe’s age limits how much of it we can see. It does not tell us whether space itself is infinite. For a fuller tour of the sizes involved, see our story How Big Is the Universe, and Does It Have an Edge?

What does “flat” mean, and how was it measured?

“Flat” does not mean shaped like a pancake. It is about geometry. Draw a giant triangle in space. In flat space, its three angles add up to 180 degrees, as on paper. In a closed universe, curved like the surface of a ball, they add up to more. In an open universe, curved like a saddle, they add up to less.

Astronomers can draw such a triangle using the cosmic microwave background, the faint glow left from about 380,000 years after the Big Bang. The glow has hot and cold spots. NASA explains that if space is flat, the brightest spots should look about one degree across on the sky. Several experiments found about one degree. NASA’s WMAP spacecraft then pinned it down: by 2013, its team reported that space is flat with only a 0.4% margin of error.

The European Space Agency’s Planck satellite did even better. Its final 2018 results, combined with maps of how galaxies are spread out, show space is flat to within about 0.2%.

There is one honest wrinkle. Planck’s own microwave data, taken alone, leaned toward a slightly closed, ball-like shape. Adding other measurements pulled the result back to flat. In 2019, three researchers argued in Nature Astronomy that the closed-shape hint deserved serious attention. They wrote that future measurements are needed to settle whether it comes from hidden errors, new physics, or chance. Planck’s standard model of the universe still assumes flat space, but the question is being tested, not ignored.

Does flat mean infinite?

Not by itself. NASA’s WMAP team wrote that a flat universe “suggests” an infinite one. But they added that, since we can only see a finite part, all we can truly conclude is that the universe is much larger than what we can observe.

Here is why flat is not enough. Einstein’s theory of gravity describes the local shape of space: how it bends in each small region. It says nothing about how the pieces connect overall. That overall connection is called topology. A 2024 paper in Physical Review Letters by a team called the COMPACT Collaboration makes this point. It lists 18 possible flat shapes. One is ordinary endless space. Others are finite, or finite in some directions. The simplest finite one is the three-torus, a box where leaving through one wall brings you back through the opposite wall.

Analogy: think of that old video game screen, or the surface of a doughnut. You can travel forever without hitting an edge, yet the total area is finite. Where the analogy breaks: a doughnut’s surface curves in a bigger space around it, and a game screen sits inside a machine. A three-torus universe needs no outside space at all. Its flat space simply connects back to itself.

If our universe wrapped around on a scale smaller than our view, we would see repeats: the same pattern of hot and cold spots in two places on the sky. Planck searched for these and found none. For the simple shapes tested, any loop must be about as large as the whole observable universe, or larger. The COMPACT team adds a caution. Past searches, they write, have “far from exhausted” the possible shapes. The hunt goes on.

What is the strongest case each way?

The case for an infinite universe is simple. The data say flat, and the simplest flat space has no end. A 2009 study in Monthly Notices of the Royal Astronomical Society by Mihran Vardanyan, Roberto Trotta and Joe Silk weighed the evidence. They found the chance that the universe is spatially infinite lies between 67% and 98%, depending on the starting assumptions.

The strongest objection is that we may never be able to know. Space could curve so gently that no measurement could ever tell it from flat. Analogy: stand in a wide field. Earth is round, but your patch looks flat. (Where it breaks: on Earth you can travel far enough to see the curve, but we cannot travel beyond what we can see of the universe.) The same 2009 study argued that if the curvature is below a very tiny level, the geometry of the universe is “not knowable.” A gently closed universe would be finite. And a wrap-around shape larger than our view might leave no mark we can read. So even a perfect measurement of “flat” could leave the question open.

What do people often get wrong?

“The universe is 13.8 billion light-years wide.” That mixes up age and distance. Because space expanded while light traveled, the edge of what we can see is now about 46 billion light-years away.

“The universe has an edge out there.” The limit of our view is a horizon, not a wall. As far as anyone can tell, space has no edge, whether it is finite or infinite.

“The Big Bang happened at one spot, so the universe must be finite.” NASA’s Ask an Astrophysicist team explains that the Big Bang happened everywhere at once. It was the start of an expansion, not an explosion from a center.

“Flat means infinite.” As shown above, flat space can still be finite if it wraps around.

How can you check this yourself?

  • Read NASA’s own words. The WMAP “Shape of the Universe” page (in the Internet Archive copy linked above) explains closed, open and flat in a few paragraphs.
  • Find the Planck number. The Planck 2018 paper’s abstract gives the curvature result. Look for the words “consistent with a flat universe.”
  • Watch for “observable.” When an article gives a size for “the universe,” check whether it means the part we can see. The two are very different.
  • Ask what was tested. When you read “no sign of a wrap-around universe,” ask which shapes were searched. The COMPACT team’s paper shows many remain.

The answer stays the same after all this detail. The universe we can see is finite, and very large. Space is flat or very close to it. Whether the whole universe goes on forever is still an open question, and it may stay that way.

THREE THINGS TO REMEMBER

  1. The observable universe is finite, with its edge now about 46 billion light-years away.
  2. Space is flat or very close to flat, but flat space can still be finite.
  3. Whether the whole universe is infinite is unknown, and it may never be knowable.

WORDS WORTH KNOWING

Observable universe
The part of the universe whose light has had time to reach us since the Big Bang. It is a sphere centered on the observer.
Light-year
The distance light travels in one year, about 6 trillion miles.
Flat space
Space where the angles of a triangle add up to 180 degrees and parallel lines never meet, as on a sheet of paper.
Topology
How a space connects to itself overall, for example whether it goes on forever or loops back around.
Three-torus
A finite flat space in which leaving through one side brings you back through the opposite side, in all three directions.
Cosmic microwave background
The faint glow left from about 380,000 years after the Big Bang. Its hot and cold spots help astronomers measure the shape of space.

Sources & further reading

  1. Expanding Confusion: common misconceptions of cosmological horizons and the superluminal expansion of the Universe (Davis and Lineweaver) ↗Figure 1 discussion: the past light cone reaches about 46 Glyr, the current distance to the particle horizon; light traveling since the beginning was emitted from positions now 46 Glyr away. Text read from the arXiv PDF.
  2. How Big is Space? We Asked a NASA Expert: Episode 61 ↗Observable universe about 92 billion light-years across; the universe is probably bigger than what we can see; “we don’t actually know if the universe is finite or infinite.” Also: one light-year is about 6 trillion miles.
  3. Planck 2018 results. VI. Cosmological parameters ↗Abstract: joint constraint with BAO consistent with a flat universe, curvature 0.001 ± 0.002. Section 7.3: Planck TT,TE,EE+lowE alone gives an apparent detection of (closed) curvature; adding lensing and BAO gives 0.0007 ± 0.0019, flat to a 1-sigma accuracy of 0.2%. Conclusions: the model describes the Universe from 380 000 years after the Big Bang (the last-scattering surface seen in the CMB) to the present day at an age of 13.8 billion years. Section 7.3 also states that the base model assumes flat space.
  4. WMAP: Shape of the Universe (Internet Archive copy, May 31, 2025) ↗Closed, open and flat geometries; spots about one degree across if flat; as of 2013, flat with a 0.4% margin of error; flatness “suggests” an infinite universe, but all we can truly conclude is that the universe is much larger than the volume we can observe. The live URL now redirects to the WMAP overview.
  5. What is the shape of the universe? (WMAP) ↗Spherical, saddle and flat geometries explained with triangles; a one-degree angle between hot and cold spots points to flat space.
  6. Ask an Astrophysicist: Cosmology questions ↗“Did WMAP results prove an infinite universe?” Answer: No; there could be much more universe beyond what we see, which we cannot know about; the finite age has nothing to do with whether the universe is spatially infinite. Also: the Big Bang occurred everywhere all at once, the start of an expansion rather than an explosion.
  7. Planck 2015 results. XVIII. Background geometry and topology of the Universe ↗No detection of a compact topology at a scale below the diameter of the last-scattering surface; for the cubic torus the inscribed radius must exceed 0.97 of the distance to the last-scattering surface.
  8. Promise of Future Searches for Cosmic Topology (COMPACT Collaboration) ↗General relativity concerns only the local geometry, not topology; 18 Euclidean (flat) topologies, 17 of them nontrivial; the three-torus has finite volume; prior searches “have far from exhausted” the possibilities.
  9. How flat can you get? A model comparison perspective on the curvature of the Universe (Vardanyan, Trotta and Silk) ↗Probability that the Universe is spatially infinite between 67% and 98% depending on priors; geometry “not knowable” if the curvature parameter is below about 10^-4.
  10. Planck evidence for a closed Universe and a possible crisis for cosmology (Di Valentino, Melchiorri and Silk) ↗Planck spectra alone prefer positive (closed) curvature; the authors say future measurements are needed to tell whether the discordances come from systematics, new physics, or a statistical fluctuation.

KEEP ASKING

See a mistake? Report a problem. Corrections are made openly.