Robustness of the no-boundary predictions for inflation and cosmic isotropy beyond the Hawking–Luttrell minisuperspace treatment

Hawking and Luttrell (1984) argued, within a spatially homogeneous but anisotropic (Bianchi IX) minisuperspace model containing a single massive scalar field, that the quantum state of the universe defined by a path integral over compact four-geometries — the no-boundary proposal — predicts, with high probability, an early inflationary epoch that dilutes anisotropy far below observational limits. The prediction rests on three linked semiclassical conclusions: a weighting of the initial scalar amplitude that favors large amplitudes, a scalar-dominance condition that identifies inflationary initial data, and a quantitative bound on the present anisotropy ratio. Whether these conclusions survive beyond the minisuperspace truncation and the leading semiclassical approximation — under inhomogeneous perturbations, a fully specified path-integral measure or factor ordering, higher-order quantum corrections, and additional matter content — remains contested. Later work has qualitatively confirmed the suppression of anisotropy in restricted extensions and has sharpened the measure and contour issues, but the probability weighting of the initial scalar amplitude (equivalently, the predicted duration of inflation) is explicitly unsettled in the current literature, and the definition of the gravitational path integral on which all three conclusions rest is under active revision.

Background

The cosmic microwave background is observed to be isotropic to roughly one part in $10^{4}$ (apart from a dipole due to our motion). This extreme isotropy is not explained by classical dynamics alone: dissipative processes (such as neutrino viscosity or particle creation) and accelerated (inflationary) expansion can attenuate anisotropy during the evolution, but none of them explains why the initial data were of the kind that lets these mechanisms operate in the first place — one can always choose initial conditions that avoid inflation or retain large anisotropy. A proposed way to fix the initial data is quantum cosmology. There, the quantum state of a spatially closed universe is described by a wave function $\Psi$ of the three-geometry and matter fields on a space-like three-surface, obeying the Wheeler–DeWitt equation — the quantum version of the Hamiltonian constraint of general relativity, a functional differential equation on superspace, the infinite-dimensional space of all three-geometries and matter configurations. Since this equation is intractable in full, analyses truncate it to minisuperspace, a finite-dimensional configuration space retaining only a few homogeneous degrees of freedom. For a closed, homogeneous but anisotropic universe the standard truncation is the Bianchi IX model, parametrized by Misner variables: a logarithmic scale factor $\alpha$ and two anisotropy (shear) parameters $\beta_{+},\beta_{-}$ describing the squashing of the three-sphere; the anisotropy potential $V(\beta_{+},\beta_{-})$ grows exponentially in the $\beta$ directions, which is also why Bianchi IX is believed to capture the generic local form of spacetime near a space-like big-bang-type singularity.

The boundary condition selecting $\Psi$ was taken to be the no-boundary proposal of Hartle and Hawking: $\Psi$ is defined by a path integral over compact (Euclidean, positive-definite) four-geometries and matter histories whose only boundary is the three-surface on which the arguments of $\Psi$ are specified. In a minisuperspace truncation this path integral is evaluated semiclassically through saddle points (compact 'instantons'), producing a WKB-type wave function whose oscillatory regions correspond to ensembles of classical Lorentzian histories and whose amplitude defines relative probabilities.

Hawking and Luttrell studied exactly this setting: the configuration space spanned by $\alpha$ , $\beta_{+}$ , $\beta_{-}$ and one spatially constant massive scalar field $\phi$ of mass $m$ (in Planck units). Their analysis yields three linked conclusions. First, the Wheeler–DeWitt operator contains a factor-ordering (measure) ambiguity parametrized by $p$ , and the effective 'mass-squared' of the wave equation is $M^{2}=\mathrm{e}^{6\alpha}m^{2}\phi^{2}+\mathrm{e}^{4\alpha}V(\beta_{+},\beta_{-})$ ; classical (oscillatory) behaviour begins where the hypersurface $M^{2}=1$ is space-like, and a sufficient condition for this is the scalar-dominance inequality

$m^{2}\phi^{2}>\mathrm{e}^{-2\alpha}V(\beta_{+},\beta_{-}),$

which selects initial data with small anisotropy velocities that enter a slow-roll inflationary phase. Second, if the scalar amplitude at the start of that phase is $|\phi_{1}|\gg1$ , the scale factor grows by of order $\exp(\tfrac{9}{2}\phi_{1}^{2})$ before $\phi$ rolls through zero, so that the anisotropy energy density $\rho_{\beta}$ is redshifted relative to the scalar energy density $\rho_{\mathrm{m}}$ ; propagating to the present epoch gives the bound

$\frac{\rho_{\beta}}{\rho_{\mathrm{m}}}\Big|_{\text{present}}<10^{61}\,m\,\exp(-9\phi_{1}^{2}),$

which is far below observational limits on present anisotropy once $|\phi_{1}|$ is sufficiently large. Third, the semiclassical weighting of the compact path integral, reported as $|\Psi|^{2}\sim\exp(\tfrac{2}{3}m^{2}\phi_{1}^{2})$ in the relevant region, was argued to make such large amplitudes probable, so that the no-boundary state predicts both negligible present anisotropy and a present density very near critical.

All of this is computed in a homogeneous truncation, at leading semiclassical order, for one matter field, with an essentially unspecified integration measure. These limitations were acknowledged in the original work and define the open question: are the three conclusions artifacts of the truncation and of the approximation scheme, or robust predictions of the no-boundary state?

Problem Statement

Determine whether the conclusions that Hawking and Luttrell drew from the no-boundary state in the Bianchi IX minisuperspace with a massive scalar field survive beyond that truncation and beyond the leading semiclassical approximation. Specifically, within a well-defined extension of the original model — the homogeneous single-field Bianchi IX minisuperspace treated at leading semiclassical order — that genuinely departs from it by at least one of: inhomogeneous (and further anisotropic) degrees of freedom with their backreaction; a fully specified path-integral measure, integration contour, or factor ordering (the ambiguity parameter $p$ ), fixed by an independent argument or supplied together with the evaluation it produces; higher-order quantum or higher-derivative corrections to the action; or additional matter fields and non-minimal couplings extending the original single-field matter content — establish whether and how the following are altered:

(i) the semiclassical probability weighting of the initial scalar amplitude $\phi_{1}$ , and hence the predicted duration of inflation, i.e. whether the no-boundary state assigns high probability to histories with sufficiently long inflationary phases;

(ii) the condition identifying inflationary initial data, expressed in the minisuperspace treatment as the scalar-dominance inequality $m^{2}\phi^{2}>\mathrm{e}^{-2\alpha}V(\beta_{+},\beta_{-})$ on the hypersurface where the Wheeler–DeWitt effective mass equals unity;

(iii) the quantitative prediction for the present anisotropy ratio, of the form $\rho_{\beta}/\rho_{\mathrm{m}}<10^{61}\,m\,\exp(-9\phi_{1}^{2})$ in the original treatment.

The formulas displayed in (i)–(iii) are stated in the 1984 paper's conventions and units. Conclusions (i)–(iii) are to be compared against a solution's own re-derivation of the original model in those conventions.

Within the declared extension, the question is whether and how each of conclusions (i)–(iii) is altered, and which definition of the path integral (contour, saddle-point selection, or regularity condition) fixes the probabilities entering (i) and (iii).

For the branches that add physical content — inhomogeneous or further anisotropic degrees of freedom, higher-order or higher-derivative corrections, and additional matter fields or non-minimal couplings — the declared extension must be non-degenerate: its added degrees of freedom, couplings, or corrections must produce a non-negligible backreaction, a modification of order the leading homogeneous single-field dynamics on the effective-mass-equals-unity hypersurface, so that an arbitrarily weak perturbation, a decoupled or infinitely weakly coupled extra field, or any other vanishing-backreaction limit does not qualify.

References

S. W. Hawking and J. C. Luttrell, The isotropy of the universe, Physics Letters B 143 (1984) 83-86, doi:10.1016/0370-2693(84)90809-8, https://doi.org/10.1016/0370-2693(84)90809-8
J. B. Hartle and S. W. Hawking, Wave function of the universe, Physical Review D 28 (1983) 2960, doi:10.1103/PhysRevD.28.2960, https://doi.org/10.1103/PhysRevD.28.2960
S. W. Hawking, The quantum state of the universe, Nuclear Physics B 239 (1984) 257, doi:10.1016/0550-3213(84)90093-2, https://doi.org/10.1016/0550-3213(84)90093-2
C. W. Misner, Mixmaster universe, Physical Review Letters 22 (1969) 1071, doi:10.1103/PhysRevLett.22.1071, https://doi.org/10.1103/PhysRevLett.22.1071
S. W. Hawking and J. C. Luttrell, Higher derivatives in quantum cosmology. 1. The isotropic case, Nuclear Physics B 247 (1984) 250, doi:10.1016/0550-3213(84)90380-8, https://doi.org/10.1016/0550-3213(84)90380-8
J. J. Halliwell and S. W. Hawking, The origin of structure in the universe, Physical Review D 31 (1985) 1777, doi:10.1103/PhysRevD.31.1777, https://doi.org/10.1103/PhysRevD.31.1777
W. A. Wright and I. G. Moss, The anisotropy of the universe, Physics Letters B 154 (1985) 115-119, doi:10.1016/0370-2693(85)90569-6, https://doi.org/10.1016/0370-2693(85)90569-6
P. Amsterdamski, Wave function of an anisotropic universe, Physical Review D 31 (1985) 3073-3078, doi:10.1103/PhysRevD.31.3073, https://doi.org/10.1103/PhysRevD.31.3073
S. W. Hawking and D. N. Page, Operator ordering and the flatness of the universe, Nuclear Physics B 264 (1986) 185, doi:10.1016/0550-3213(86)90478-5, https://doi.org/10.1016/0550-3213(86)90478-5
P. Chmielowski and D. N. Page, Probability of Bianchi type I inflation, Physical Review D 38 (1988) 2392, doi:10.1103/PhysRevD.38.2392, https://doi.org/10.1103/PhysRevD.38.2392
J. Louko, Semiclassical path measure and factor ordering in quantum cosmology, Annals of Physics 181 (1988) 318, doi:10.1016/0003-4916(88)90170-4, https://doi.org/10.1016/0003-4916(88)90170-4
M. J. Duncan and L. G. Jensen, The quantum cosmology of an anisotropic universe, Nuclear Physics B 312 (1989) 662-672, doi:10.1016/0550-3213(89)90576-2, https://doi.org/10.1016/0550-3213(89)90576-2
S. del Campo and A. Vilenkin, Tunneling wavefunction for an anisotropic universe, Physics Letters B 224 (1989) 45-48, doi:10.1016/0370-2693(89)91047-2, https://doi.org/10.1016/0370-2693(89)91047-2
J. J. Halliwell and J. Louko, Steepest-descent contours in the path-integral approach to quantum cosmology. III. A general method with applications to anisotropic minisuperspace models, Physical Review D 42 (1990) 3997-4031, doi:10.1103/PhysRevD.42.3997, https://doi.org/10.1103/PhysRevD.42.3997
K. Fujio and T. Futamase, Appearance of classical Mixmaster universe from the no-boundary quantum state, Physical Review D 80 (2009) 023504, doi:10.1103/PhysRevD.80.023504, arXiv:0906.2616, https://arxiv.org/abs/0906.2616
G. W. Gibbons and N. Turok, The measure problem in cosmology, Physical Review D 77 (2008) 063516, doi:10.1103/PhysRevD.77.063516, arXiv:hep-th/0609095, https://arxiv.org/abs/hep-th/0609095
J. Feldbrugge, J.-L. Lehners and N. Turok, No smooth beginning for spacetime, Physical Review Letters 119 (2017) 171301, doi:10.1103/PhysRevLett.119.171301, arXiv:1705.00192, https://arxiv.org/abs/1705.00192
J. Feldbrugge, J.-L. Lehners and N. Turok, Lorentzian quantum cosmology, Physical Review D 95 (2017) 103508, doi:10.1103/PhysRevD.95.103508, arXiv:1703.02076, https://arxiv.org/abs/1703.02076
S. F. Bramberger, S. Farnsworth and J.-L. Lehners, Wavefunction of anisotropic inflationary universes with no-boundary conditions, Physical Review D 95 (2017) 083513, doi:10.1103/PhysRevD.95.083513, arXiv:1701.05753, https://arxiv.org/abs/1701.05753
J. Diaz Dorronsoro, J. J. Halliwell, J. B. Hartle, T. Hertog, O. Janssen and Y. Vreys, Damped perturbations in the no-boundary state, Physical Review Letters 121 (2018) 081302, doi:10.1103/PhysRevLett.121.081302, arXiv:1804.01102, https://arxiv.org/abs/1804.01102
O. Janssen, J. J. Halliwell and T. Hertog, No-boundary proposal in biaxial Bianchi IX minisuperspace, Physical Review D 99 (2019) 123531, doi:10.1103/PhysRevD.99.123531, arXiv:1904.11602, https://arxiv.org/abs/1904.11602
J.-L. Lehners, Review of the no-boundary wave function, Physics Reports 1022 (2023) 1-82, doi:10.1016/j.physrep.2023.06.002, arXiv:2303.08802, https://arxiv.org/abs/2303.08802
J.-L. Lehners and K. S. Stelle, Higher-order gravity, finite action, and a safe beginning for the universe, European Physical Journal Plus 139 (2024) 380, doi:10.1140/epjp/s13360-024-05125-y, arXiv:2312.14048, https://arxiv.org/abs/2312.14048
T. Hertog, O. Janssen and J. Karlsson, Kontsevich-Segal criterion in the no-boundary state constrains anisotropy, Physical Review D 111 (2025) 046008, doi:10.1103/PhysRevD.111.046008, arXiv:2408.02652, https://arxiv.org/abs/2408.02652
V. Mondal and S. Mukohyama, Anisotropic quantum universe in Hořava-Lifshitz gravity, Physical Review D 112 (2025) 083542, doi:10.1103/6wp2-b17k, arXiv:2505.16266, https://arxiv.org/abs/2505.16266
G. Lavrelashvili and J.-L. Lehners, Birth of inflationary universes via wineglass wormholes and their no-boundary relatives, Physical Review D 114 (2026) 046009, doi:10.1103/jpwp-2nzv, arXiv:2605.10548, https://arxiv.org/abs/2605.10548