CoChat

Is cosmological tuning fine or coarse?

Open access
Díaz-Pachón, Daniel Andrés, Hössjer, Ola, Marks II, Robert J. · 2021 · Journal of Cosmology and Astroparticle Physics

The DOI matches its Crossref record; 59 of the 64 references it cites were found in a registry or an index; 5 of the 25 sources its claims cite were checked for support.

Abstract

The fine-tuning of the universe for life, the idea that the constants of nature (or ratios between them) must belong to very small intervals in order for life to exist, has been debated by scientists for several decades. Several criticisms have emerged concerning probabilistic measurement of life-permitting intervals. Herein, a Bayesian statistical approach is used to assign an upper bound for the probability of tuning, which is invariant with respect to change of physical units, and under certain assumptions it is small whenever the life-permitting interval is small on a relative scale. The computation of the upper bound of the tuning probability is achieved by first assuming that the prior is chosen by the principle of maximum entropy (MaxEnt). The unknown parameters of this MaxEnt distribution are then handled in such a way that the weak anthropic principle is not violated. The MaxEnt assumption is “maximally noncommittal with regard to missing information.” This approach is sufficiently general to be applied to constants of current cosmological models, or to other constants possibly under different models. Application of the MaxEnt model reveals, for example, that the ratio of the universal gravitational constant to the square of the Hubble constant is finely tuned in some cases, whereas the amplitude of primordial fluctuations is not.

Overview

CoChat Summary

TL;DR

The paper proposes a Bayesian maximum-entropy (MaxEnt) method to compute an upper bound \(P_{\max}\) on the probability that a constant of nature, or a ratio of constants, lies in its life-permitting interval. Applying it to examples, the ratio of the gravitational constant to the square of the Hubble constant is finely tuned for very small relative intervals, whereas the amplitude of primordial fluctuations is coarsely tuned.

Key findings

  • The tuning probability upper bound \(P_{\max}\) is dimensionless and, under the paper’s assumptions, proportional to the relative half-width \(\varepsilon\) of the life-permitting interval; small \(\varepsilon\) implies fine tuning unless the prior distribution carries substantial information concentrated in that interval.
  • For \(x=G_{\rm obs}/d\), with \(d=H^2_{\rm obs}\), the paper cites \(\varepsilon=10^{-60}\); with \(d=\Lambda_{\rm vac}\), it cites \(\varepsilon=10^{-50}\) or \(10^{-100}\). Under MaxEnt priors, \(P_{\max}\) is of the same order as \(\varepsilon\) (or slightly smaller when MaxEnt is applied separately to \(G\) and \(D\)), so these cases are extremely finely tuned.
  • For a real-valued ratio with a normal MaxEnt prior, if \(\mu/\sigma\) is unrestricted, \(P_{\max}\) can reach 1, making tuning coarse; enforcing \(\mu=0\) gives an upper bound of \(e^{-1/2}/\sqrt{2\pi}\approx 0.242\).
  • For the amplitude of primordial fluctuations \(Q\), with sample space \(\mathbb{R}^+\), \(Q\sim{\rm Exp}(1/\mu)\), and life-permitting interval \(\ell_Q=(10^{-6},10^{-5})\), the maximum tuning probability is “rather large”; since \(Q_{\rm obs}\approx 2\times10^{-5}\) lies outside this generous interval, the paper concludes coarse tuning. Adams’s wider interval \(10^{-6}\le Q_{\rm obs}\le10^{-2}\) is said to make little difference to that conclusion.

Methods

  • Bayesian three-step procedure: (I) choose the sample space \(\Omega\) for the constant or ratio; (II) choose a finite-dimensional MaxEnt prior class \(F(\cdot;\theta)\) from Table 1—e.g., exponential on \(\mathbb{R}^+\), normal on \(\mathbb{R}\)—either for \(X\) directly or for the two constants \(G\) and \(D\) separately; (III) maximize the tuning probability over the hyperparameter \(\theta\) to obtain \(P_{\max}\).
  • The examples use literature values and intervals: \(G_{\rm obs}=6.67408\times10^{-11}\,{\rm m^3\,kg^{-1}\,s^{-2}}\), \(Q_{\rm obs}\approx2\times10^{-5}\), Rees’s and Adams’s life-permitting intervals for \(Q\), and cited \(\varepsilon\) values for gravitational ratios.
  • The analysis is analytic, solving for maximizing hyperparameters and deriving bounds such as \(P_{\max}\propto\varepsilon\) and the \(\approx0.242\) bound for the \(\mu=0\) normal case.

Limitations

  • Stated: the paper does not determine the life-permitting interval bounds itself; it relies on existing physics literature, and those bounds may change, constants may be removed, or new constants may be identified. It also notes that without a theory of everything the first step is incomplete.
  • Stated: results are conditional on the chosen sample space and MaxEnt prior class; the choice of \(\Omega\) and prior is debated, and additional knowledge can shrink the hyperparameter space. If a “known” tuning probability \(P^*\) differs from \(P_{\max}\), the paper proposes active information \(I^+=\log(P^*/P_{\max})\) to account for bias.
  • Evident: the examples are simplified, assuming no additional knowledge beyond the sample space and one or two moments; only two main examples are analyzed, and joint tuning of several constants is left for future work.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1088/1475-7516/2021/07/020

Cites

59 of 59 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 When the variance σ 2 of the prior distribution of X approaches 0, the normal distribution approaches a Dirac delta measure at µ ; thus µ ∈ ℓ X , implies P max = 1 (left). On the other hand, when µ / ∈ ℓ X , where ℓ X is the life-permitting interval (LPI), P [ ℓ G ] will go to zero either when σ → 0 or when σ →∞ . Therefore P max is strictly less than 1 (right). For the figure at the left, the tuning is coarse, whereas for the figure at the right it is fine.
Figure 1 When the variance σ 2 of the prior distribution of X approaches 0, the normal distribution approaches a Dirac delta measure at µ ; thus µ ∈ ℓ X , implies P max = 1 (left). On the other hand, when µ / ∈ ℓ X , where ℓ X is the life-permitting interval (LPI), P [ ℓ G ] will go to zero either when σ → 0 or when σ →∞ . Therefore P max is strictly less than 1 (right). For the figure at the left, the tuning is coarse, whereas for the figure at the right it is fine.From the PDF

Verification

Claims the paper makes

3 Valid · 2 Not valid · 15 Not checked · 44 Own results and interpretations

29 more claims

Create a free account to see all 64 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 23
  • It may state two separate claims as one. 14

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Research works printed but not linked: 1

  1. [23] R. Penrose. Time-Assymetry and Quantum Gravity. In C. J. Isham, R. Penrose, and D. W. Sciama, editors, Quantum Gravity 2: A Second Oxford Symposium , pages 244-272. Oxford Univeristy Press, 1981. -

Cited by

See all 12

Metadata

Authors
Díaz-Pachón, D. A., Hössjer, O., & Marks II, R. J.
Affiliations
Stockholm University (SE)
Corresponding author
Hössjer, Ola
Venue
Journal of Cosmology and Astroparticle Physics
Published
2021-07-01
Volume
2021
Issue
07
Pages
020
ISSN
  • 1475-7508 (ISSN-L)
  • 1475-7516
OpenAlex
W3153095692
Cited by
10
References
64
Citations by year
2026: 1 · 2025: 3 · 2024: 1 · 2023: 2 · 2022: 3
FWCI
0.81
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Cosmology and Astroparticle Physics (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
  • DESY Publication Database (PUBDB) (Deutsches Elektronen-Synchrotron)
Funders
  • Deutsche Forschungsgemeinschaft (EXC2121, 390833306)
  • Universidade Federal de Minas Gerais
  • Science and Technology Facilities Council (ST/T00102X/1, Consolidated Grant ST/T00102X/1)
Topics
  • Physical Sciences › Physics and Astronomy › Astronomy and Astrophysics › Cosmology and Gravitation Theories
  • Physical Sciences › Physics and Astronomy › Astronomy and Astrophysics › Space Science and Extraterrestrial Life
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Advanced Mathematical Theories and Applications
Keywords
cosmological fine-tuning, anthropic principle, constants of nature, maximum entropy prior, life-permitting universe

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.