Papex

Open problem · Cosmology · Open

Size of the universe

Is the universe finite or infinite in extent?

Summary

The observable universe is about 93 billion light-years across, but space beyond it cannot be observed; whether the universe as a whole is finite or infinite is unknown.

Why it is open

The observable universe is the region from which light has had time to reach us; in the standard cosmological model, Tamara Davis and Charles Lineweaver put the present distance to its edge, the particle horizon, at about 46 billion light years [2]. About what lies beyond it "we can obtain no reliable information whatever", George Ellis writes [5].

Whether space is finite depends on two of its properties. Curvature can be positive, zero or negative [10]. Topology, the way space is connected on the largest scales, can make even a flat space finite, as in a three-dimensional torus [11], and spatial infinities "can be avoided by compact spatial sections, resulting either from positive spatial curvature, or from a choice of compact topologies" [5]. Both properties are discussed on the shape of the universe page; this page asks what they imply for size.

The two answers are not equally testable. We could see right around a finite universe only if it were small, with a closure scale within reach of light since the cosmic microwave background (CMB) was emitted, Ellis notes [5]; otherwise finiteness could be inferred only from positive curvature, which makes space closed, or from a detected topology [5, 10]. Of an infinite space, "we can never prove that any realised 3-space in the real universe continues in this way"; it is "an untestable concept", in his words [5]. No body sets a bar for a resolution; observations can set a minimum size or detect one of those signatures [5, 6].

What is known

Curvature and a minimum size (2009–2025). Curvature is measured by the parameter Ω_K, which is zero for flat space and negative for a positively curved, closed universe [10]. In 2009 Mihran Vardanyan, Roberto Trotta and Joe Silk compared flat, closed and open models and found that, given the data then available, the probability that the universe is spatially infinite lay between 67% and 98%, depending on the priors; they assumed the simplest topology [6]. For closed models they set a "robust, prior-independent" lower limit: the universe contains at least about five Hubble spheres, their name for the volume inside the apparent particle horizon (99% confidence) [6]. Averaging over models in 2011, under their more conservative prior they put the radius of curvature above 42 gigaparsecs (Gpc) and the number of Hubble spheres above 251, both at 99%, noting that the limits are stronger because flat models are favoured [7]. In 2025 the Atacama Cosmology Telescope (ACT), combining its data with Planck, lensing and baryon acoustic oscillations, measured Ω_K = 0.0019 ± 0.0015, "fully consistent with spatial flatness", and put the radius of curvature above 105 Gpc, or 343 billion light years, for a closed universe and above 66 Gpc for an open one (95% confidence) [10]. The dispute over hints of slight positive or negative curvature since 2019 is covered on the shape page.

Topology (to 2026). Searches have yielded "no definitive evidence for non-trivial topology", but in a 2026 review, a preprint, Craig Copi and colleagues report that "detectable signals may persist even when the topology scale exceeds the size of the visible Universe" [11]. They also write that "there is no established first-principles selection rule that singles out any particular spatial topology" [11].

What cannot be measured (2006–2012). Even if space is flat, the curvature measured within one Hubble volume receives contributions from long-wavelength density perturbations and will not in general be zero, Matthew Kleban and Marjorie Schillo point out [8]. Vardanyan and colleagues concluded that "the geometry of the Universe is not knowable" if |Ω_K| is below about 10⁻⁴, an order of magnitude above the size of curvature perturbations, about 10⁻⁵ [6]. Ellis adds that speculations about the structure of space beyond the horizon, such as chaotic inflation, are "not observationally testable" [5].

Curvature as a test of eternal inflation (2012). Kleban and Schillo argued that a measured |Ω_K| above 10⁻⁴ would rule out slow-roll eternal inflation in our past, that a value below −10⁻⁴ (positive curvature) would also rule out eternal inflation through false-vacuum decay, and that an open universe is consistent with the latter [8]. Alan Guth and Yasunori Nomura reached a similar conclusion: positive curvature beyond that level would exclude the eternally inflating multiverse, while negative curvature above it would mean that our universe was born by bubble nucleation; they found the 2012 limits consistent with multiverse expectations except in certain cases [9].

Main approaches

Infinite, from eternal inflation. Inflation, a brief period of exponential expansion in the early universe, generically never ends: in most models some regions are always still inflating [1]. Jaume Garriga and Alexander Vilenkin argued in 2001 that "a generic prediction of inflation is that the thermalized region we inhabit is spatially infinite", with infinitely many regions the size of our observable universe [1]. If our universe formed by tunnelling from a neighbouring vacuum, as Ben Freivogel and colleagues argued in 2006 that a rich "landscape" of vacua implies, it is open, with negative curvature [4]. Strongest argument: the conclusion follows from mechanisms that its proponents consider generic [1, 4]. Main difficulty: the regions beyond our horizon are untestable [5], and a measured Ω_K below −10⁻⁴ would rule out eternal inflation [8, 9].

Flat and infinite. The simplest reading of the data: flat space with the simplest topology [6, 10]. Strongest evidence: curvature consistent with zero [10] and, in 2009, model comparison favouring flat models [6]. Main difficulty: that comparison assumed the simplest topology [6], a flat space can be finite [5], and curvature below about 10⁻⁴ may never be known [6].

Finite through topology. Andrei Linde argued in 2004 that if inflation happens far below the Planck density, quantum creation of a closed or an infinite open universe is exponentially suppressed, while under certain conditions compact flat or open universes are not, so that they "should be considered a rule rather than an exception" [3]. Strongest argument: Copi and colleagues judge that the universe "very plausibly starts out with a non-trivial topology" [11]. Main difficulty: no topology has been detected, and a topology scale near the observable one requires inflation of a duration that is "just right" [11].

Finite by principle. A philosophical argument: Ellis holds that "the often claimed physical existence of infinities is questionable", citing David Hilbert, and that Machian considerations make compact space "highly preferable" [5]. Strongest argument: an infinite universe can never be verified, and implies "the infinite duplication of life and all events" [1, 5]. Main difficulty: it predicts no size, and the data that Ellis's review cited as suggesting "k = +1", a closed universe [5], have given way to a 2025 combination consistent with flatness [10].

Open sub-questions

  1. An updated minimum size. What lower limit on the number of Hubble spheres does the 2025 ACT combination give, with and without model averaging, under the definitions of Vardanyan and colleagues [7, 10]?
  2. Reaching 10⁻⁴. Which combination of planned CMB and baryon acoustic oscillation data would reach the |Ω_K| ≈ 10⁻⁴ level at which the sign of curvature tests eternal inflation, given the 2009 forecast of about 4.5 × 10⁻⁴ for an ideal CMB experiment with an SKA-like survey [6, 8, 9]?
  3. Topology beyond the horizon. For which topologies, and up to what topology scale beyond the observable universe, do the signals described by Copi and colleagues stay detectable with planned data [11]?
  4. Suppression in current models. Does Linde's result, that compact flat universes escape exponential suppression, hold for the inflation models favoured by current CMB data [3]? See cosmic inflation.
  5. The probability of infinity. How does the 2009 estimate of a 67–98% probability that space is infinite change with current data, and when compact topologies are included in the model comparison [6, 10]?

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Garriga, J. & Vilenkin, A. (2001). Many worlds in one. Physical Review D 64(4), 043511. 10.1103/PhysRevD.64.043511 · Checked 15 Oct 2026
  2. Davis, T. M. & Lineweaver, C. H. (2004). Expanding confusion: common misconceptions of cosmological horizons and the superluminal expansion of the Universe. Publications of the Astronomical Society of Australia 21(1), 97–109. 10.1071/AS03040 · Checked 15 Oct 2026
  3. Linde, A. (2004). Creation of a compact topologically nontrivial inflationary universe. Journal of Cosmology and Astroparticle Physics 2004(10), 004. 10.1088/1475-7516/2004/10/004 · Checked 15 Oct 2026
  4. Freivogel, B., Kleban, M., Rodríguez Martínez, M. & Susskind, L. (2006). Observational consequences of a landscape. Journal of High Energy Physics 2006(03), 039. 10.1088/1126-6708/2006/03/039 · Checked 15 Oct 2026
  5. Ellis, G. F. R. (2007). Issues in the philosophy of cosmology. In Philosophy of Physics, pp. 1183–1285. Elsevier. 10.1016/B978-044451560-5/50014-2 · Checked 15 Oct 2026
  6. Vardanyan, M., Trotta, R. & Silk, J. (2009). How flat can you get? A model comparison perspective on the curvature of the Universe. Monthly Notices of the Royal Astronomical Society 397(1), 431–444. 10.1111/j.1365-2966.2009.14938.x · Checked 15 Oct 2026
  7. Vardanyan, M., Trotta, R. & Silk, J. (2011). Applications of Bayesian model averaging to the curvature and size of the Universe. Monthly Notices of the Royal Astronomical Society: Letters 413(1), L91–L95. 10.1111/j.1745-3933.2011.01040.x · Checked 15 Oct 2026
  8. Kleban, M. & Schillo, M. (2012). Spatial curvature falsifies eternal inflation. Journal of Cosmology and Astroparticle Physics 2012(06), 029. 10.1088/1475-7516/2012/06/029 · Checked 15 Oct 2026
  9. Guth, A. H. & Nomura, Y. (2012). What can the observation of nonzero curvature tell us? Physical Review D 86(2), 023534. 10.1103/PhysRevD.86.023534 · Checked 15 Oct 2026
  10. Calabrese, E., Hill, J. C., Jense, H. T., La Posta, A., Abril-Cabezas, I., Addison, G. E. et al. (2025). The Atacama Cosmology Telescope: DR6 constraints on extended cosmological models. Journal of Cosmology and Astroparticle Physics 2025(11), 063. 10.1088/1475-7516/2025/11/063 · Checked 15 Oct 2026
  11. Copi, C. J., Mihaylov, D. P., Negro, A., Samandar, A., Starkman, G. D., Akrami, Y. et al. (2026). The topology of the Universe. Preprint, arXiv:2606.24886 (version 1, 23 June 2026), invited review for Nature Astronomy, not peer reviewed. Source · Checked 15 Oct 2026

Page reviewed 15 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in physics. Checked 08 Oct 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Cosmology & astronomy