The phase of the gravitational path integral
Xiaoyi Shi and Gustavo J. Turiaci · 2025
Read The phase of the gravitational path integral by Xiaoyi Shi and Gustavo J. Turiaci (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-00900.txt.
Introduction
Xiaoyi Shi and Gustavo Turiaci ask whether a key quantity in quantum gravity is real or imaginary. Their twenty twenty-five paper studies the gravitational path integral on a space made of two spheres. The answer matters for how the result is read. One reading treats it as part of the norm of the Hartle-Hawking wave function, which needs a real value. Another treats it as a black hole creation rate in de Sitter space, which needs an imaginary one.
The pair builds a clean method to compute the phase of such integrals on Einstein spaces. On the spaces they test, the phase comes out real and positive. They extend parts of the work to charge and rotation.
Catalog record
| CREATOR | Xiaoyi Shi and Gustavo J. Turiaci |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 99 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-00900
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
The phase of the gravitational path integral Xiaoyi Shi and Gustavo J. Turiaci Abstract The gravitational path integral on S 2 × S 2 S^{2}\times S^{2} can be interpreted either as evaluating a contribution to the norm of the Hartle-Hawking wavefunction conditional on spatial S 1 × S 2 S^{1}\times S^{2} topology, or the pair creation rate of black holes in de Sitter. Both interpretations are distinguished at the quantum level. The former requires the path integral to be real and the latter to be imaginary. We develop a formalism to efficiently compute the phase of the gravitational path integral on Einstein spaces. We apply it to a broad class of spacetimes and in particular S 2 × S D − 2 S^{2}\times S^{D-2} , finding it to be real and positive. We generalize some of the analysis to cases with charge and rotation. 1 Introduction The Hartle and Hawking proposal Hartle:1983ai uses the gravitational path integral on spaces with a positive cosmological constant to prepare the wavefunction of the universe. It instructs us to sum over all smooth geometries that approach a given expanding spatial future geometry. In the case that the spatial slices are 3-spheres, the geometry is complex, combining an early period of Euclidean evolution glued onto an expanding Lorentzian de Sitter solution. In Euclidean signature, this geometry is a 4-sphere. In this context, a reasonable interpretation of the gravitational path integral over the sphere is that it evaluates the norm of the Hartle-Hawking wavefunction. This identification certainly works at the classical level, as already noticed in Hartle:1983ai . The situation at the quantum level is less clear. For pure gravity with a positive cosmological constant, it was realized in Polchinski:1988ua that the gravitational path integral on the sphere has a dimension-dependent phase and an overall sign that was recently corrected in Maldacena:2024spf . This is incompatible with the claim that the path integral evaluates a norm on a unitary dS Hilbert space, which is expected to be real and positive Gibbons:1977mu . A recent proposal by Maldacena, also in Maldacena:2024spf , claims that when an observer is added to dS beyond pure gravity, the resulting phase of the path integral on Euclidean dS disappears. This is an interesting observation, although we believe several open questions remain. The motivation for the work in this paper is the following. It is interesting to consider the physics of the Hartle-Hawking wavefunction on other spatial topologies. The simplest case is S^{1}\times S^{2} . Surprisingly, the norm of the wavefunction with spatial topology S^{1}\times S^{2} is infinite at the classical level. Quantum effects in the gravitational path integral, that arise when the size of S^{1} is much larger than the size of the S^{2} , are crucial to render the norm finite Maldacena:2019cbz ; Turiaci:2025xwi . The contribution to the norm of the Hartle-Hawking wavefunction conditional on a S^{1}\times S^{2} spatial topology matches with the small G_{N}\Lambda approximation of the path integral on S^{2}\times S^{2} . This geometry is the Euclidean section of a black hole in dS in its Nariai limit Nariai ; GINSPARG1983245 ; Bousso:1996au . Can we extend this match beyond the leading order in G_{N}\Lambda ? This problem was emphasized recently in Turiaci:2025xwi . Motivated by the results of Maldacena:2024spf we will calculate the phase of the gravitational path integral on S^{2}\times S^{2} . In this case, there are two possible interpretations of the S^{2}\times S^{2} path integral. The first we already described, as a correction to the norm of the Hartle-Hawking wavefunction. The second interpretation of the S^{2}\times S^{2} path integral is that it computes the decay rate of de Sitter. This was originally proposed by Ginsparg and Perry GINSPARG1983245 . Following up on the analysis of the instability of gravity on thermal flat space in Gross:1982cv , they found that S^{2}\times S^{2} has one unstable mode, naively suggesting the path integral is imaginary. This would be the correct result if the path integral is to be interpreted as capturing dS decay and black hole pair creation Bousso:1995cc ; Bousso:1996au . Perhaps surprisingly, the phase of the path integral on S^{2}\times S^{2} has not been carefully evaluated. It was guessed in GINSPARG1983245 that it is imaginary based on the presence of one unstable mode, but we learned from Polchinski:1988ua that even in the absence of instabilities, the path integral can be complex. The goal of this paper is to develop a streamlined formalism to evaluate the phase of the gravitational path integral in pure gravity. We will work in D dimensions and mostly focus on the case with a positive cosmological constant. The metric fluctuations around a given geometry can be decomposed into tensor (physical) components, that we take to be transverse and traceless (TT), a vector, and two scalars. We will show that using some classic results from Riemannian geometry one can give a completely general account of the contribution to the phase from the vector and scalar sector. Importantly, the latter contribution is non-trivial due to the presence of negative modes Polchinski:1988ua . Therefore, the nontrivial step in the evaluation of the phase is reduced to finding unstable physical TT modes. For example, we find that the gravitational path integral on S^{2}\times S^{D-2} is always real and positive regardless of dimension. This is consistent with its interpretation as a correction to the norm of the Hartle-Hawking state, but incompatible with its interpretation as a rate of black hole pair production! Path integrals over Euclidean configurations are required to be imaginary in order to compute vacuum decay in quantum field theory, as illustrated by the conversation between Sagredo and Salviati in Coleman:1985rnk . The same feature is presumably true in quantum gravity as well as field theory. The fact that the path integral on S^{2}\times S^{D-2} is real and positive seems to favor its state-counting interpretation. In a sense, it is reasonable that it does not require introducing an observer since, in Lorentzian signature, this geometry corresponds to a black hole in dS. We present evidence that the path integrals for charged and rotating black holes are also real and positive. Regardless of these results, it is interesting to study the effect an observer can have in these geometries, and we will make some preliminary remarks on this at the end of the paper. This also raises the following question. What is the real rate of black hole pair creation in dS? The Ginsparg-Perry result was used throughout in the literature Bousso:1995cc ; Bousso:1996au but as far as we know, there is no rigorous derivation of this rate due to the puzzles on the phase mentioned earlier. Settling this puzzle might have interesting implications for the black hole creation rate during inflation, as well as determining the lifetime of dS. The organization of the paper is the following. In section 2 we review the derivation of the Einstein action to quadratic order and put it in a form that will be optimal to evaluate quantum corrections. A convenient choice of gauge decouples the scalar modes. In section 3 we give a general derivation of the phase of the gravitational path integral in terms of the number of unstable tensor modes, accounting for the vector and