Shape of the universe
What is the global shape and topology of space?
Summary
Why it is open
The shape of space has two parts. Curvature is local: the equations of general relativity constrain the curvature at every point "but say nothing about the global topology of the Universe" [10]. Topology describes global properties, such as closed loops that cannot be shrunk to a point; spaces with the same curvature can have different topologies, and a flat space, for example, can wrap around like a three-dimensional torus [15]. In a 2026 review, a preprint, Craig Copi and colleagues call it a "long-standing misconception" that measuring curvature settles whether the universe is finite [15]. Complete Euclidean three-dimensional space has 18 topology classes, while positively and negatively curved geometries each allow infinitely many [15].
Curvature is measured by the parameter Ω_K: zero for flat space, negative for a closed (positively curved) universe and positive for an open one [13]. Even if space were exactly flat, the density fluctuations in our observable volume give any measurement of Ω_K confined to that volume a "cosmic variance" of roughly 2 × 10⁻⁵, a floor about a hundred times below Planck's best constraint [8]. Topology can be seen only if the distance around the universe is not much larger than the observable universe, and "whether cosmic topology is observable remains uncertain" [15].
The sources point to two possible outcomes: a curvature measured above the cosmic-variance floor [8], or a detected topological signature, for which null results only push the limits outwards [15].
What is known
Curvature from Planck (2018). From its maps of the cosmic microwave background (CMB), the Planck satellite's temperature and polarisation spectra alone gave Ω_K = −0.044 (+0.018/−0.015), "an apparent detection of curvature at well over 2σ"; adding Planck's lensing reconstruction pulled this to −0.0106 ± 0.0065, and adding baryon acoustic oscillation (BAO) data gave 0.0007 ± 0.0019, flat to 0.2% at 1σ [7].
The closed-universe debate (2019–2021). Eleonora Di Valentino, Alessandro Melchiorri and Joseph Silk argued that the Planck spectra prefer a closed universe at 3.4 standard deviations, which would also explain an anomalously strong lensing signal (the distortion of the CMB by intervening matter) in them [5]. With curvature free, they found strong disagreement between Planck and BAO, and a Hubble constant of 54.4 km/s/Mpc in 5.2σ tension with local measurements, worsening the Hubble tension; "the assumption of a flat universe could, therefore, mask a cosmological crisis" [5]. Will Handley put the tension between Planck, CMB lensing and BAO at 2.5 to 3σ and wrote that curvature conclusions drawn from combining them "should therefore be viewed with suspicion" [9]. George Efstathiou and Steven Gratton, using a new Planck likelihood, found the spectra "consistent with a spatially flat Universe, in contrast to recent claims in the literature", and Ω_K = 0.0004 ± 0.0018 with BAO [6]. In their preprint they attribute the high significance to the choice of likelihood and of a uniform prior on Ω_K, and find the pull towards closed models at about 2.1σ [6].
Later data (2024–2026). Analysing Planck's final maps, Matthieu Tristram and colleagues found Ω_K = −0.012 ± 0.010, "complete consistency with a flat universe" [12]. The Atacama Cosmology Telescope (ACT) found its own spectra consistent with flatness and, combined with Planck, lensing and BAO, Ω_K = 0.0019 ± 0.0015; the radius of curvature exceeds 105 gigaparsecs (Gpc), or 343 billion light years, if space is closed and 66 Gpc if open, at 95% confidence [13]. In a paper published in 2026, the SPT-3G team found each of their CMB combinations compatible with flatness on its own, but adding BAO from the Dark Energy Spectroscopic Instrument (DESI) moved them away: combining SPT, ACT and Planck with DESI gave Ω_K = 0.0027 ± 0.0012, 2.5σ from flatness on the open side [14]. They note that this conflicts with the excess lensing in Planck's high-resolution temperature data, so using Planck's spectrum in place of ACT's could weaken it, and that CMB plus BAO gives similar 2–3σ shifts when the lensing amplitude or the equation of state of dark energy is freed instead [14].
Topology searches (1998–2026). In 1998 Neil Cornish, David Spergel and Glenn Starkman showed that if the universe is smaller than the distance to the surface from which the CMB was emitted, copies of that surface intersect in circles, so pairs of circles on the sky would show matching patterns [1]. In 2003 Jean-Pierre Luminet and colleagues proposed the Poincaré dodecahedral space, finite and positively curved, to explain the vanishing temperature correlations on angles wider than 60° seen by WMAP [2]. A 2004 search of WMAP data found no matched circles with radius above 25°, ruling out topology scales below 24 Gpc for a wide class of models [3]; the preprint adds that this excludes the dodecahedral model, which predicts circles of radius 35° [3]. Planck found no evidence for the tested topologies below the diameter of the last scattering surface; for a cubic torus, the largest sphere that fits inside the repeating cell must have a radius at least 0.97 times the distance to that surface [4]. In 2023 Pip Petersen and colleagues noted that the absence of circles makes the shortest loop through our position longer than 98.5% of that diameter, but that in all but the simplest flat topologies the shortest loop can avoid us and be shorter than the diameter by a factor of 2 to at least 6 [10]. The 2026 review by Copi and colleagues concludes that current constraints "exclude only some topologies, parameter ranges, and observer positions" [15].
Main approaches
Flat, or too large to tell. One possibility is that curvature lies below the cosmic-variance floor and topological signatures beyond our observable region [8, 15]. Strongest evidence: combined CMB and BAO constraints centred on zero [7, 13] and the absence of matched circles [3, 4, 10]. Main difficulty: a curvature below the floor cannot be measured from our volume [8], and null circle searches leave many topologies allowed [10, 15].
A closed universe. Space is positively curved and therefore finite in the simplest topology [15]. Strongest evidence: the preference in Planck's 2018 spectra, which also accounts for their lensing excess [5, 9]. Main difficulty: lensing and BAO pull the result back to flat [7], newer likelihoods and ACT find flatness [6, 12, 13], and positive curvature "could pose a problem for the inflationary paradigm" because closed inflationary models are hard to produce [8]; see cosmic inflation.
An open universe. Space could be negatively curved; a measurement above the floor would test open-inflation models [8]. Strongest evidence: the 2.5σ result from CMB data combined with DESI [14]. Main difficulty: it depends on which CMB temperature data are used, and the same data can be fitted instead by changes in lensing or dark energy [14].
A multiply connected space. Space wraps around, as in a torus or the dodecahedral space [2, 15]. Strongest evidence: if large-angle CMB anomalies are due to topology, the shortest distance around the universe through us is unlikely to be much larger than the horizon diameter, and the circle limits leave "many possibilities" open [11]. Main difficulty: no detection so far [4, 10, 15]; searching the large space of topologies is "a formidable computational challenge" [11]; and in inflationary cosmology a visible topology requires inflation to have lasted "just right" [15].
Open sub-questions
- What drives the open hint. Does the 2.5σ preference for Ω_K > 0 from CMB data plus DESI survive when Planck's high-resolution temperature spectrum replaces ACT's [14]?
- Curvature or dark energy. Which BAO measurements, at which redshifts and precision, would separate a small positive Ω_K from evolving dark energy as the cause of the CMB–BAO shifts [14]?
- Loops that avoid us. For flat topologies whose shortest loop avoids us and is up to six times shorter than the diameter of the last scattering surface, which off-diagonal correlations of CMB harmonic coefficients would reveal them in Planck data [10]?
- Anomalies without circles. Does any topology allowed by the circle-search limits reproduce the large-angle CMB anomalies [10, 11, 15]?
- Polarisation reach. Can LiteBIRD polarisation data extend the reach of topology searches, measured by the distance to our nearest copy, from a projected Planck reach of about 1.05 to about 1.2 times the diameter of the last scattering surface, as the 2026 review estimates [15]?
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
- Cornish, N. J., Spergel, D. N. & Starkman, G. D. (1998). Circles in the sky: finding topology with the microwave background radiation. Classical and Quantum Gravity 15(9), 2657–2670. 10.1088/0264-9381/15/9/013 · Checked 15 Oct 2026
- Luminet, J.-P., Weeks, J. R., Riazuelo, A., Lehoucq, R. & Uzan, J.-P. (2003). Dodecahedral space topology as an explanation for weak wide-angle temperature correlations in the cosmic microwave background. Nature 425(6958), 593–595. 10.1038/nature01944 · Checked 15 Oct 2026
- Cornish, N. J., Spergel, D. N., Starkman, G. D. & Komatsu, E. (2004). Constraining the topology of the universe. Physical Review Letters 92(20), 201302. 10.1103/PhysRevLett.92.201302 · Checked 15 Oct 2026
- Planck Collaboration: Ade, P. A. R. et al. (2016). Planck 2015 results. XVIII. Background geometry and topology of the Universe. Astronomy & Astrophysics 594, A18. 10.1051/0004-6361/201525829 · Checked 15 Oct 2026
- Di Valentino, E., Melchiorri, A. & Silk, J. (2020). Planck evidence for a closed Universe and a possible crisis for cosmology. Nature Astronomy 4(2), 196–203 (published online 4 November 2019). 10.1038/s41550-019-0906-9 · Checked 15 Oct 2026
- Efstathiou, G. & Gratton, S. (2020). The evidence for a spatially flat Universe. Monthly Notices of the Royal Astronomical Society: Letters 496(1), L91–L95. 10.1093/mnrasl/slaa093 · Checked 15 Oct 2026
- Planck Collaboration: Aghanim, N. et al. (2020). Planck 2018 results. VI. Cosmological parameters. Astronomy & Astrophysics 641, A6. 10.1051/0004-6361/201833910 · Checked 15 Oct 2026
- Planck Collaboration: Akrami, Y. et al. (2020). Planck 2018 results. X. Constraints on inflation. Astronomy & Astrophysics 641, A10. 10.1051/0004-6361/201833887 · Checked 15 Oct 2026
- Handley, W. (2021). Curvature tension: evidence for a closed universe. Physical Review D 103(4), L041301. 10.1103/PhysRevD.103.L041301 · Checked 15 Oct 2026
- Petersen, P., Akrami, Y., Copi, C. J., Jaffe, A. H., Kosowsky, A., Starkman, G. D. et al. (2023). Cosmic topology. Part I. Limits on orientable Euclidean manifolds from circle searches. Journal of Cosmology and Astroparticle Physics 2023(01), 030; erratum 2024(04), E01, DOI 10.1088/1475-7516/2024/04/E01. 10.1088/1475-7516/2023/01/030 · Checked 15 Oct 2026
- Akrami, Y., Anselmi, S., Copi, C. J., Eskilt, J. R., Jaffe, A. H., Kosowsky, A. et al. (2024). Promise of future searches for cosmic topology. Physical Review Letters 132(17), 171501. 10.1103/PhysRevLett.132.171501 · Checked 15 Oct 2026
- Tristram, M., Banday, A. J., Douspis, M., Garrido, X., Górski, K. M., Henrot-Versillé, S. et al. (2024). Cosmological parameters derived from the final Planck data release (PR4). Astronomy & Astrophysics 682, A37. 10.1051/0004-6361/202348015 · Checked 15 Oct 2026
- 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
- Camphuis, E., Quan, W., Balkenhol, L., Khalife, A. R., Ge, F., Guidi, F. et al. (2026). SPT-3G D1: CMB temperature and polarization power spectra and cosmology from 2019 and 2020 observations of the SPT-3G main field. Physical Review D 113(8), 083504. 10.1103/7wt3-9v2y · Checked 15 Oct 2026
- 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
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.