Papex

Open problem · Cosmology · Open

Low entropy of the early universe

Why did the universe begin in a state of such low entropy?

Summary

The second law of thermodynamics and the arrow of time require the early universe to have had very low entropy; why it began in such a special state, rather than a typical one, is unexplained.

Why it is open

In Ludwig Boltzmann's definition, entropy measures the volume of the set of microscopic states that look the same at the macroscopic level; the second law says that, roughly speaking, the entropy of the universe increases with time [4]. Tracing entropy back in time, Robert Wald concludes that "the reason why the entropy of the present universe is low is that the entropy of the universe at or near the 'big bang' was extremely low" [5]. In 2000 the philosopher David Albert called the posit that the universe began in a tiny part of its available phase space the "Past Hypothesis" [14].

Yet the early universe's matter was nearly uniform and in thermal equilibrium, which for an ordinary gas means maximum entropy [5]. The answer that the philosopher David Wallace calls conventional, set out by Roger Penrose, is that this refers to matter, not to gravity: gravity makes matter clump and eventually form black holes, so a smooth distribution has very low entropy once gravity is counted [4, 5, 9]. The low value is then a claim about gravity, for which "there does not presently exist a general notion of 'gravitational entropy'", Wald notes [5].

The problem is several questions of different kinds. Physicists ask whether a mechanism or a law produced the initial state [5]. Philosophers of physics also ask whether the Past Hypothesis needs explaining at all, and whether it can be stated precisely in general relativity [6, 14]. No body sets a bar for a resolution. Wald asks that an explanation make the observed universe "a prediction of the model", not merely "a (highly unlikely) possibility that is allowed in the model" [5].

What is known

How special (1979–2010). In 1979 Penrose compared the entropy of the early universe, measured by its radiation, with that of a black hole holding the mass of the observable universe, and found the first about 35 orders of magnitude smaller [1]. In a 2006 version he collapses the roughly 10⁸⁰ baryons within our past light cone into one black hole, obtains an entropy of order 10¹²³, and concludes that the volume of phase space exceeds that of the region representing the big bang by a factor of more than 10 raised to the power 10¹²³ [4]. In 2010 Chas Egan and Charles Lineweaver estimated the entropy of today's observable universe at 3.1 (+3.0/−1.7) × 10¹⁰⁴ times Boltzmann's constant, most of it in supermassive black holes, against 5.4 × 10⁸⁹ in the photons of the cosmic microwave background (CMB) [7].

Gravity and entropy (2006–2013). Penrose reads the early uniformity as "the gravitational degrees of freedom that are potentially available to the universe being not excited at all" [4]. His Weyl curvature hypothesis states that the Weyl curvature, the part of spacetime curvature that describes the gravitational degrees of freedom, must vanish at initial singularities, while final ones, such as those inside black holes, are unconstrained [4]. In 2013 Timothy Clifton, George Ellis and Reza Tavakol proposed a measure of gravitational entropy that reproduces the entropy of a non-rotating black hole, grows as cosmic structure forms and is "in keeping with" that hypothesis [10]. In 2010 Wallace argued that, black holes apart, gravitational degrees of freedom are "not relevant in known physical situations", and that the early universe was out of global thermal equilibrium both because it was uniform and because it was expanding [9].

Inflation, explanation or presupposition (2002–2010). Inflation is a proposed early period of exponential expansion that would leave each inflated region extremely uniform and flat [5]; see cosmic inflation. Penrose argues that any scenario that starts from a "random" state and drives it into a special one faces a conundrum, because "'random' means 'non-special' in the extreme" [4]. Wald argues that the conditions for a patch to inflate are themselves special, that predictions in such models lead "down an essentially circular path" through anthropic selection, and that the universe more plausibly "came into existence in a very special state" [5]. Andreas Albrecht, "definitely in the dynamical camp", agrees that inflation needs special initial conditions, but only for one field in a small region, which inflation "leverages" into large regions like ours [1]. Counting cosmological histories with the canonical measure, Sean Carroll and Heywood Tam found in a 2010 preprint that inflation occurs in fewer than 1 in 10 raised to the power 6.6 × 10⁷ of the histories that are realistic at late times [8].

Recent claims. In 2020 Daniel An, Krzysztof Meissner, Paweł Nurowski and Penrose reported anomalous circular spots in Planck and WMAP maps of the CMB and identified them as "Hawking points", which Penrose's conformal cyclic cosmology predicts from black holes evaporating in a universe before ours [11]. Dylan Jow and Douglas Scott repeated the analysis and, after marginalising over the size of the rings, found an excess at only 87% confidence; they concluded that "there is no statistically significant evidence" for Hawking points [12]. In 2024 Latham Boyle and Neil Turok computed a gravitational entropy for realistic cosmologies that "favors universes like our own", flat, homogeneous and isotropic, and in which inhomogeneities are "entropically disfavored" [13].

Main approaches

A dynamical origin. A physical proposal: inflation or another process turns less special conditions into ours [1, 2]. Strongest argument: inflation needs special conditions only for one field in a small region [1], and in a 2004 semiclassical calculation by Albrecht and Lorenzo Sorbo it is exponentially favoured over routes to our universe without it [2]. Main difficulty: the most concrete scheme then available disfavoured it [2]; with reversible evolution, inflation "does not affect the number of initial conditions that evolve into a late universe like our own" [8]; and, as Albrecht concludes, it "will necessarily require some special initial conditions itself" [1].

A law for the initial state. A proposal about laws: a principle, such as the Weyl curvature hypothesis, fixes the special state [4, 5]. Strongest argument: if the Past Hypothesis is a law, the probability of the initial state is one, not astronomically small [14]. Main difficulty: Wald concedes that this "begs the question" of which law governed the creation [5]; John Earman argues that the initial Boltzmann entropy is likely ill-defined [6], and that the Past Hypothesis cannot be stated in general relativity [14].

Unbounded entropy. In a physical model from a 2004 preprint, not peer reviewed, Carroll and Jennifer Chen propose that entropy has no maximum: inflation starts spontaneously from nearly empty space and creates universes like ours in both directions of time [3]. Strongest argument: in such models "one can do without a past Hypothesis" [14]. Main difficulty: everything rests on entropy being unbounded [3], and Wald finds the proposal implausible [5].

A universe before ours. A physical model with an observational test: in conformal cyclic cosmology the remote future of a previous universe becomes our big bang, with the Weyl curvature hypothesis as its smooth boundary [4]. Strongest evidence: the reported Hawking points [11]. Main difficulty: the re-analysis found no significant excess [12]; the details "have yet to be worked out", Penrose writes, and stray electrons must lose charge or mass by routes "none of which is part of conventional particle physics" [4].

No puzzle as posed. A philosophical position: Craig Callender doubts that the Past Hypothesis demands explanation [14], and Earman calls Boltzmann entropy's role in inferences about the past "vastly overrated" [6]. Strongest argument: an initial condition occurs only once, and whether it can be lawlike depends on the theory of laws [14]. Main difficulty: Huw Price holds that it does [14], and Wald considers seeking that answer "more fruitful" than dynamical explanations [5].

Open sub-questions

  1. Hawking points elsewhere. With Jow and Scott's marginalisation over ring size, does the ring-gradient statistic of An and colleagues find an excess in CMB data they did not use, such as polarisation maps [11, 12]?
  2. A measure that tracks structure. Does the Clifton–Ellis–Tavakol gravitational entropy grow monotonically in simulations of structure formation, and how does its present value compare with Egan and Lineweaver's black-hole entropy [7, 10]?
  3. How rare is inflation. How sensitive is Carroll and Tam's fraction of inflating histories to their assumption that every perturbation mode must be smaller than unity at the grand unification scale [8]?
  4. Two ways of counting. Which assumption makes the scheme that Albrecht and Sorbo describe disfavour inflation while their semiclassical calculation favours it, and can one calculation include both [2]?
  5. Entropy with structure. In the statistical ensemble that Boyle and Turok defer to later work, does their gravitational entropy still disfavour inhomogeneity once black holes, today's largest entropy term, are included [7, 13]?
  6. Stating the Past Hypothesis. Can a condition on the initial singularity, such as Penrose's vanishing Weyl curvature, give the Past Hypothesis a precise form in general relativity that answers Earman's objection [4, 6, 14]? See also the arrow of time.

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. Albrecht, A. (2004). Cosmic inflation and the arrow of time. In Science and Ultimate Reality: Quantum Theory, Cosmology, and Complexity, ed. J. D. Barrow, P. C. W. Davies & C. L. Harper, Jr, pp. 363–401. Cambridge University Press. 10.1017/CBO9780511814990.021 · Checked 15 Oct 2026
  2. Albrecht, A. & Sorbo, L. (2004). Can the universe afford inflation? Physical Review D 70(6), 063528. 10.1103/PhysRevD.70.063528 · Checked 15 Oct 2026
  3. Carroll, S. M. & Chen, J. (2004). Spontaneous inflation and the origin of the arrow of time. Preprint, arXiv:hep-th/0410270 (version 1, 27 October 2004), not peer reviewed. Source · Checked 15 Oct 2026
  4. Penrose, R. (2006). Before the big bang: an outrageous new perspective and its implications for particle physics. In Proceedings of EPAC 2006, Edinburgh, Scotland, pp. 2759–2762 (paper THESPA01). Source · Checked 15 Oct 2026
  5. Wald, R. M. (2006). The arrow of time and the initial conditions of the universe. Studies in History and Philosophy of Modern Physics 37(3), 394–398. 10.1016/j.shpsb.2006.03.005 · Checked 15 Oct 2026
  6. Earman, J. (2006). The "Past Hypothesis": not even false. Studies in History and Philosophy of Modern Physics 37(3), 399–430. 10.1016/j.shpsb.2006.03.002 · Checked 15 Oct 2026
  7. Egan, C. A. & Lineweaver, C. H. (2010). A larger estimate of the entropy of the universe. The Astrophysical Journal 710(2), 1825–1834. 10.1088/0004-637X/710/2/1825 · Checked 15 Oct 2026
  8. Carroll, S. M. & Tam, H. (2010). Unitary evolution and cosmological fine-tuning. Preprint, arXiv:1007.1417 (version 1, 8 July 2010), not peer reviewed. Source · Checked 15 Oct 2026
  9. Wallace, D. (2010). Gravity, entropy, and cosmology: in search of clarity. The British Journal for the Philosophy of Science 61(3), 513–540. 10.1093/bjps/axp048 · Checked 15 Oct 2026
  10. Clifton, T., Ellis, G. F. R. & Tavakol, R. (2013). A gravitational entropy proposal. Classical and Quantum Gravity 30(12), 125009. 10.1088/0264-9381/30/12/125009 · Checked 15 Oct 2026
  11. An, D., Meissner, K. A., Nurowski, P. & Penrose, R. (2020). Apparent evidence for Hawking points in the CMB sky. Monthly Notices of the Royal Astronomical Society 495(3), 3403–3408. 10.1093/mnras/staa1343 · Checked 15 Oct 2026
  12. Jow, D. L. & Scott, D. (2020). Re-evaluating evidence for Hawking points in the CMB. Journal of Cosmology and Astroparticle Physics 2020(03), 021. 10.1088/1475-7516/2020/03/021 · Checked 15 Oct 2026
  13. Boyle, L. & Turok, N. (2024). Thermodynamic solution of the homogeneity, isotropy and flatness puzzles (and a clue to the cosmological constant). Physics Letters B 849, 138442. 10.1016/j.physletb.2024.138442 · Checked 15 Oct 2026
  14. Callender, C. (2001, substantively revised 12 June 2026). Thermodynamic asymmetry in time. Stanford Encyclopedia of Philosophy. 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