The universe as we observe it is not eternal in the straightforward sense. It had a definite beginning roughly 13.8 billion years ago, and multiple lines of evidence point toward an eventual end, though physicists disagree sharply about which kind of end. What makes the question richer than a simple “no” is that several serious theoretical frameworks attempt to preserve a form of eternity, either by positing that something existed before the Big Bang or by arguing that the cosmos will cycle through deaths and rebirths without ever truly ceasing. Whether any of those frameworks hold up depends on unresolved questions in quantum gravity, dark energy, and the deepest structure of spacetime itself.
How We Know the Universe Had a Beginning
For most of human intellectual history, a static and eternal cosmos was the default assumption. Even Einstein initially tweaked his own equations to keep the universe from expanding or contracting. That changed in 1927 when Georges Lemaître proposed that the expanding universe began from a single primordial state, and it changed decisively in 1965 when Arno Penzias and Robert Wilson detected the cosmic microwave background (CMB), the faint afterglow of the hot, dense early universe.1Classical and Quantum Gravity. The cosmic microwave background: the history of its experimental investigation and its significance for cosmology Before that discovery, a rival model called the steady state theory, championed by Fred Hoyle and others, held that new matter was continuously created to keep the universe looking the same at all times. Radio surveys in the 1950s and 1960s had already shown that distant galaxies looked different from nearby ones, suggesting the cosmos had changed over time, but the CMB was the decisive blow.2Journal of Physics: Conference Series. The Big Bang versus the Steady State: Gamow, Hoyle and Ryle, rivals in cosmology
The CMB is not just evidence that a hot beginning occurred. Its tiny temperature fluctuations, mapped in extraordinary detail by satellites like COBE, WMAP, and Planck, have confirmed inflation (the idea that the universe underwent an extremely rapid expansion in its earliest moments) and enabled what cosmologists call precision cosmology, where the age, composition, and geometry of the universe can be pinned down to remarkable accuracy.1Classical and Quantum Gravity. The cosmic microwave background: the history of its experimental investigation and its significance for cosmology In short, the observational case for a beginning is as strong as anything in modern physics.
Why an Eternal Past Is Difficult to Sustain
Even if you set aside the CMB and just focus on the mathematics of an expanding universe, an eternal past runs into trouble. A result known as the Borde-Guth-Vilenkin (BGV) theorem, published in 2003, demonstrates that any region of spacetime with a net positive rate of expansion cannot be extended infinitely far into the past. It must have some kind of boundary or beginning.3PubMed. Inflationary spacetimes are incomplete in past directions The theorem does not assume any particular form of energy or exotic physics; it is a purely geometric argument about what happens when you trace the paths of particles backward through an expanding space. At some point those paths simply run out.
The theorem’s conclusion is sometimes stated more precisely: any spacetime with net positive expansion is geodesically incomplete to the past.4arXiv. The Borde-Guth-Vilenkin Theorem in extended de Sitter spaces That means the expanding region cannot be the whole story. Something else, some other physics, must describe what happens at or before that boundary.5Journal of Cosmology and Astroparticle Physics. Possible relationship between initial conditions for inflation and past geodesic incompleteness of the inflationary spacetime The BGV theorem does not say the universe was created out of nothing, and it does not say what that “something else” is. But it does close the door on the simplest version of an eternally inflating past.
Models That Try to Avoid a True Beginning
The BGV theorem motivates a lot of creative theorizing. If an expanding universe must have a past boundary, maybe the universe was not always expanding. Several proposals try to get around the problem by positing that conditions before the Big Bang were radically different from what came after.
One family of models involves a “bounce.” Instead of emerging from a singularity where density and temperature become infinite, the universe might have contracted from a previous phase, reached some minimum size, and then rebounded into expansion. Loop quantum gravity, a candidate theory that tries to describe gravity at the smallest scales using quantum mechanics, produces exactly this kind of picture. In loop quantum gravity calculations, the singularity at the heart of the Big Bang gets replaced by a Big Bounce, and the evolution of the universe becomes periodic: expansion, contraction, bounce, expansion again.6arXiv. Bouncing Universe in Loop Quantum Gravity: full theory calculation Other quantum approaches also generate bouncing solutions, where the bounce emerges from scattering processes in the quantum description of the cosmos.7Physics Letters B. Quantum Big-Bounce as a phenomenology of RQM in the Mini-superspace
A separate but related idea is Roger Penrose’s conformal cyclic cosmology (CCC). In this picture, the universe does not bounce in the usual sense. Instead, the infinitely distant future of one cosmic era, where everything has diluted into radiation, is mathematically identical to the Big Bang of the next era. Each cycle is called an “aeon,” and the sequence of aeons has no beginning and no end.8arXiv. The Physics of Conformal Cyclic Cosmology CCC does away with inflation entirely and requires that all massive particles eventually decay. It is elegant but remains speculative, and whether signals from a previous aeon could be detected in the CMB is hotly debated.
Then there is the no-boundary proposal, associated with Stephen Hawking and James Hartle. Rather than asking what came before the Big Bang, it redefines the question: the universe’s wave function is calculated by summing over all possible geometries that have no boundary to the past. In this framework, asking what came “before” is like asking what is south of the South Pole. The universe is finite and self-contained, and the Big Bang singularity is avoided entirely.9arXiv. Review of the No-Boundary Wave Function The no-boundary proposal does not quite say the universe is eternal, but it removes the need for a sharp creation event.
There is also the more radical suggestion that Einstein’s equations can be continued mathematically through the Big Bang singularity itself, connecting our universe to a mirror image on the other side where the orientation of space is reversed.10Physics Letters B. Through the big bang: Continuing Einstein’s equations beyond a cosmological singularity These are all different flavors of the same impulse: to find some sense in which the universe did not simply pop into existence from nothing.
Eternal Inflation and the Multiverse Wrinkle
Even if our observable universe had a definite beginning, the larger structure it sits inside might not have. In the eternal inflation scenario, the rapid expansion that occurred in the first fraction of a second after the Big Bang never completely stops. Different regions stop inflating at different times, each one becoming a vast, causally disconnected pocket universe with its own set of physical constants and low-energy laws. Our universe would be just one such pocket in an endlessly spawning multiverse.11IOP Publishing (Reports on Progress in Physics). A brief history of the multiverse
Eternal inflation preserves a kind of eternity for the multiverse as a whole, even if each pocket universe within it has its own finite beginning. But this is where the BGV theorem bites again: even the eternally inflating multiverse, taken as a whole, still has a net positive expansion rate and therefore still requires a past boundary. The multiverse might generate new pocket universes forever into the future, but it cannot have been doing so forever into the past. So eternal inflation shifts the “beginning” problem rather than solving it. Something had to get inflation started.
How the Universe Might End
If the universe’s past is not eternal, what about its future? Here the answer depends almost entirely on the behavior of dark energy, the mysterious component that makes up roughly 68 percent of the universe’s total energy budget and is currently driving space to expand at an accelerating rate. The character of dark energy is described by a number called its equation of state parameter, usually written as w. Different values of w lead to dramatically different fates.
If dark energy behaves like a cosmological constant (w equals exactly −1), the universe keeps expanding forever, gradually cooling down until stars burn out, black holes evaporate, and everything approaches a state of maximal disorder. This is the heat death scenario, and it is the default expectation under the current standard model of cosmology. In this future, the universe is eternal in the forward direction but in the bleakest possible way: an infinite stretch of cold, empty, featureless space.
If w dips below −1, dark energy is “phantom” energy, and the outcome is much more violent. Phantom energy’s density grows over time rather than staying constant. Eventually it overwhelms every other force in the universe, tearing apart galaxy clusters, then galaxies, then solar systems, then planets, then atoms. This scenario is called the Big Rip, and the universe ends in a finite amount of time with everything shredded to pieces.12PubMed. Phantom energy: dark energy with w <–1 causes a cosmic doomsday Whether the equation of state can actually cross the phantom divide (from above −1 to below −1) is an active research question. Some modified gravity theories permit this crossing, and observational data have been used to argue it may already be happening.13IOP Publishing. Equation of state for dark energy in f(T) gravity On the other hand, quantum gravity effects could potentially produce phantom-like behavior without an actual Big Rip: the universe might mimic phantom energy for a time and then settle into steady exponential expansion, avoiding the catastrophic singularity.14Journal of Cosmology and Astroparticle Physics. Phantom-like dark energy from quantum gravity
A third possibility has gained fresh attention. If dark energy is not constant but weakens over time and eventually reverses sign, the expansion of the universe could slow down, stop, and then reverse into contraction, ultimately collapsing back to a point in a Big Crunch. Recent dark energy data have led at least one analysis to suggest the cosmos could be roughly halfway through a total lifespan of about 33 billion years, with expansion continuing for approximately another 11 billion years before the turnaround begins. That would make the universe’s future emphatically not eternal.15Journal of Cosmology and Astroparticle Physics. Universe may end in a “big crunch,” new dark energy data suggests This is a single analysis and far from settled, but it shows how sensitive the universe’s fate is to the precise properties of dark energy.
The Vacuum Decay Wildcard
Even if dark energy cooperates and the universe should, in principle, last forever, there is a separate threat lurking in fundamental particle physics. Based on our current measurements of the Higgs boson mass and the top quark mass, the electromagnetic vacuum that permeates the universe appears to be metastable rather than absolutely stable. That means the vacuum is sitting in a kind of valley that is not the lowest possible valley. It could, in principle, tunnel to a lower energy state.16Frontiers in Astronomy and Space Sciences. Cosmological Aspects of Higgs Vacuum Metastability
If that tunneling event happened anywhere in the universe, a bubble of “true vacuum” would expand outward at the speed of light, rewriting the laws of physics in its wake. Everything inside the bubble, every atom, every force, every structure, would be obliterated and replaced by something entirely different. The transition is not just inevitable in the sense that it could happen; under current Standard Model parameters, it is the favored outcome over absolute stability.17Journal of Physics: Conference Series. On the cosmological implications of the electroweak vacuum instability: constraining the non-minimal coupling with inflation The reassuring caveat is that the expected timescale for such a transition is absurdly long, far longer than the current age of the universe, and new physics beyond the Standard Model could easily stabilize the vacuum. But it is a genuine theoretical possibility that the universe could simply cease to exist as we know it, with no warning, at any point.
What Flatness Tells Us
One of the more practical ways to constrain the universe’s fate is to measure its geometry. If the universe has positive curvature (like the surface of a sphere), it could eventually re-collapse. If it has zero or negative curvature, it tends to expand forever. Observations from the Planck satellite, combined with measurements of galaxy clustering, constrain the curvature parameter to be essentially zero, consistent with a perfectly flat universe to better than one percent precision.18Physics of the Dark Universe. The galaxy power spectrum take on spatial curvature and cosmic concordance Separate analyses combining CMB temperature and polarization data with baryon acoustic oscillation measurements reach similar conclusions.19Monthly Notices of the Royal Astronomical Society: Letters. The evidence for a spatially flat Universe
A flat universe, on its own, would expand forever. But flatness only tells you about the geometry, not about whether dark energy changes character in the future. A flat universe driven by a cosmological constant expands eternally. A flat universe in which dark energy weakens and reverses could still collapse. So the flatness measurements are necessary but not sufficient for predicting the end of the story.
The Thermodynamic Puzzle
There is a subtler reason the question of the universe’s eternity is philosophically tricky, and it has to do with entropy and the arrow of time. The second law of thermodynamics says entropy (roughly, disorder) increases over time. That gives time a direction: the past is low-entropy and the future is high-entropy. But cosmologists face a puzzle: the early universe was extremely hot and dense, which sounds like it should be high-entropy, yet the gravitational state of the early universe was extraordinarily uniform and low-entropy. This low-entropy starting point is what physicists call the “past hypothesis,” and without it, the second law has no foundation.20Foundations of Physics. Modeling the Past Hypothesis: A Mechanical Cosmology
Why does this matter for eternity? Because an eternal universe in either direction creates problems for thermodynamics. If the universe had existed forever into the past, it should have reached thermodynamic equilibrium long ago, and nothing interesting (stars, galaxies, life) would exist. The fact that we clearly live in a universe far from equilibrium is itself evidence against a simple eternal past. And if the universe lasts forever into the future, it will eventually reach that equilibrium, the heat death. The existence of a low-entropy beginning is not just an observation; it is a deep theoretical constraint that any model of an eternal cosmos has to explain.
The Boltzmann Brain Problem
An eternally expanding universe also runs into one of the strangest paradoxes in modern cosmology. In a universe that lasts forever and asymptotically approaches a low-energy state, random quantum or thermal fluctuations will occasionally produce organized structures purely by chance. Given enough time, those fluctuations will produce a functioning human brain, complete with false memories, far more often than the ordinary evolutionary process produces actual observers. These hypothetical entities are called Boltzmann brains, and in many scenarios involving infinite duration, they vastly outnumber real observers.21arXiv. Are there Boltzmann brains in the vacuum
This is not just a curiosity. If Boltzmann brains are the dominant type of observer in the universe’s history, then a randomly selected observer (which, under certain reasoning, you should expect yourself to be) is overwhelmingly likely to be a Boltzmann brain rather than a product of biological evolution. Since you have consistent, non-random experiences, this creates a contradiction. An eternal future where the universe settles into a near-empty state seems to predict something that we can directly rule out by looking around.
The Boltzmann brain problem plagues many cosmological models that involve infinite duration or infinite spatial extent. The inflationary multiverse is a prime offender: in a landscape of infinitely many pocket universes, defining what counts as a “typical” observer becomes enormously difficult. This is called the measure problem in eternal inflation.22Zenodo. The Measure Problem in Eternal Inflation: Foundations, Proposals, and Conceptual Challenges Some cosmologists have argued that the seriousness of the Boltzmann brain problem is itself evidence that the universe must eventually stop expanding or transition to a state where such fluctuations cannot occur, effectively requiring the universe’s future to be finite.23Journal of Cosmology and Astroparticle Physics. Sinks in the landscape, Boltzmann brains and the cosmological constant problem
Why the Answer Keeps Shifting
A century ago, the reigning view was that the universe was eternal and unchanging. Then the Big Bang model established that it had a beginning. For decades after that, the main question was whether the universe would expand forever or eventually re-collapse, and the answer seemed to hinge on how much matter it contained. The discovery of accelerating expansion in 1998 appeared to settle the question in favor of eternal expansion toward a cold, empty heat death. Now, recent dark energy survey data are reopening the possibility of a Big Crunch, and quantum gravity researchers continue to explore models in which the Big Bang was not a true beginning but a transition from a prior state.
The honest answer in 2025 is that the universe almost certainly had something like a beginning, in the sense that our expanding spacetime cannot be traced back indefinitely. Whether that beginning was an absolute beginning of everything, or a transition from some prior phase, depends on physics we do not yet understand. As for the end, dark energy holds the key, and we are still measuring its properties. The universe’s claim to eternity remains unproven in both directions, and the evidence keeps getting more interesting rather than more settled.