REVIEW 4 major objections 5 minor
Holographic timelike complexity for de Sitter
T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash
Pith's one-line read Timelike subregion complexity in de Sitter is real and finite, grows exponentially at early durations, and diverges hyperfast near a maximal duration unless a black hole horizon intervenes.
desk verdict The pure dS calculation is solid and worth reading, but the SdS Case 2 headline is contradicted by the paper's own Appendix B, so the abstract and Section 5.2 need correction before this can be trusted. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The construction has three parts: (i) the stretched-horizon prescription, which places the holographic boundary just inside the cosmological horizon and regulates the maximum boundary duration as T_max = (1/2) ln((2-epsilon)/epsilon); (ii) the timelike-subregion complexity prescription, in which two spacelike extremal surfaces anchored at the past and future endpoints and one timelike extremal surface joining their turning points bound a bulk volume; and (iii) the reflection-symmetry assumption that the turning points lie on the t=0 slice (U=V in Kruskal coordinates), which fixes all boundary conditions. In pure dS_3 the extremal surfaces are known analytically; for higher dimensions and for
What would settle it
For the Schwarzschild–de Sitter Case-2 geometry, numerically integrate the extremal-surface equations without imposing U=V (or v(r*) = r̃(r*)) at the turning point; if the resulting late-time complexity growth differs from the claimed power-law/linear regime, the symmetry assumption is the point of failure. Separately, a late-time run with epsilon and T-max approached to high precision can settle whether C_s grows like r_min^{d-1} T or like a genuinely nonlinear power of T, since the paper currently asserts both.
Extended reading notes
Core claim
On the paper's own terms, the discovery is that timelike-subregion volume complexity is computable and well-defined in de Sitter static-patch holography. For a boundary subregion of duration 2T on the stretched horizon, the complexity is the bulk volume bounded by two spacelike extremal surfaces anchored at the subregion's endpoints and a timelike extremal surface joining their turning points. In pure dS_3 this volume is analytic, real and positive; in higher dimensions it is numerical, checked in Kruskal and Eddington–Finkelstein coordinates. The observable reproduces the early exponential and late hyperfast growth of spacelike volume complexity, with the divergence exponent lowered by one,
Load-bearing premise
The entire calculation assumes that the turning point of every spacelike extremal surface lies exactly on the t=0 slice (U=V), argued from the left-right symmetry of the extended Penrose diagram but not independently derived for finite boundary subregions; if the turning point is off t=0, every boundary condition, volume, and late-time scaling in the paper changes.
Editorial extensions
If this is right
- If the proposal is correct, timelike subregion complexity is a real, positive, finite holographic quantity in de Sitter, with no ultraviolet divergence, in contrast to timelike entanglement entropy which is complex.
- In pure de Sitter, the observable grows as e^{2T} at short durations and diverges as (T_max - T)^{-(d-2)} near the maximal duration, with T_max identical to the critical time of full-volume complexity.
- For Schwarzschild–de Sitter with the stretched horizon near the cosmological horizon, the same exponential-then-hyperfast pattern holds, with the black-hole mass increasing T_max and modifying the divergence exponent.
- With the stretched horizon near the black hole horizon, timelike complexity does not diverge at finite time; its long-duration growth is a power law (per the paper's abstract and discussion) or linear with coefficient r_min^{d-1} (per the appendix), indicating the black hole interior rather than the expanding cosmological region.
- The growth-rate pattern — e^{T} for full volume complexity and e^{2T} for timelike subregion complexity — suggests the rate tracks the codimension of the boundary anchoring surface, giving a new organizational rule for holographic complexity.
Reading between the lines
- A testable extension: compute timelike subregion complexity for rotating or charged de Sitter black holes; if the Case-2 softening persists, the black-hole-horizon effect is generic rather than a quirk of Schwarzschild–de Sitter.
- The e^{2T} early growth matches the behavior expected of Krylov-type operator growth, so timelike subregion complexity may be a more direct holographic proxy for operator growth in the static patch than full volume complexity.
- The paper's Case-2 late-time result is stated in two ways: the abstract and discussion describe nonlinear growth, while the appendix derives C_s ~ r_min^{d-1} T, linear in T. Since these are not the same statement, a high-precision late-time numerical check is needed to decide which behavior the extremal-surface construction actually produces.
- If the turning-point-at-t=0 symmetry is relaxed, the entire Case-2 late-time scaling would shift; testing this assumption in the numerical integration would reveal whether the claimed black-hole-horizon softening is robust.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper applies the recently proposed holographic timelike subregion volume complexity to de Sitter spacetime in the static patch holography framework. For pure dS it computes analytic dS3 expressions for the spacelike and timelike extremal surfaces, the timelike entanglement entropy, and the complexity, then extends the computation numerically to d=4,5. For Schwarzschild–de Sitter, it treats two stretched-horizon positions: near the cosmological horizon (Case 1) and near the black hole horizon (Case 2). The headline claims are that pure dS timelike complexity grows as e^{2T} for short durations and diverges hyperfastly near a maximal duration, while SdS Case 2 replaces the hyperfast growth by a 'nonlinear growth regime' for long durations.
Significance. If established, this would be a useful extension of timelike complexity to cosmological horizons, showing that the construction yields a finite, real, positive measure in dS and that the late-time divergence is dimension-dependent. The dS3 analytic solution, the numerical cross-checks in Kruskal and Eddington–Finkelstein coordinates (Tables 3–6, Appendix A), and the absence of fitted constants in the extremization are genuine strengths. However, the central Case 2 claim is internally inconsistent with the paper's own Appendix B, and the pure-dS late-time divergence formulae contradict each other for d=3. These issues must be resolved before the paper's main conclusions can be accepted.
major comments (4)
- [Abstract; §5.2; Appendix B] The SdS Case 2 claim of 'nonlinear growth' is contradicted by Appendix B. Eq. (B.6) gives C_s ∼ r_min^{d−1} T, which is linear in T for fixed r_min, while the sentence immediately after (B.6) states that this 'does not show linear growth with time.' Section 5.2 and the Discussion instead describe 'nonlinear growth (mix of powers)' and a 'power law increase with the boundary subregion duration,' without any fitted exponent. Both statements cannot be correct. The authors must determine whether Eq. (B.6) or the numerical reading of Fig. 17 is right; if the former, the abstract and §5.2 must be revised to report linear growth; if the latter, the asymptotic reduction in (B.4)–(B.6) contains an error.
- [Eqs. (1.2), (1.3), (3.40), (4.13)] The pure-dS late-time divergence is stated inconsistently. Eq. (1.2) and Eq. (3.40) give C_late ∼ ln[(T_max−T)^{−1}] for d=3, while Eq. (1.3) and Eq. (4.13) give C_late ∼ (T_max−T)^{−(d−2)}, which for d=3 is (T_max−T)^{−1}, not logarithmic. Section 4 and Fig. 8 apply the power-law form to d=3,4,5. This is a direct contradiction in a central quantitative claim. If logarithmic divergence is special to d=3, the general formula must state d>3 and explain how the d=3 limit arises; if the power law is correct, the §3.2 analytic result needs re-examination.
- [§5.1, Eq. (5.16); §5.2, Eqs. (5.29), (5.35)] The SdS computation relies on the assumption that the turning point of the spacelike extremal surface lies on the t=0 slice, and that the timelike connecting curve is u(r)=−r̃(r) or v(r)=r̃(r). This is asserted via 'insight based on symmetry' (§5.1) and 'we can argue from the symmetry of SdS geometry' (§5.2), rather than derived from the extremization conditions with finite boundary subregions. For pure dS this condition follows from the U↔V reflection symmetry of the maximal-radius point; for the infinitely extended SdS Penrose diagram and a finite timelike boundary interval it is not automatic. Since this assumption fixes the boundary conditions for all computed surfaces and volumes in Cases 1 and 2, a proper justification is load-bearing. The authors should either prove the t=0 turning point from the equations of motion or state the precise symmetry and why it survives the choice of a
- [§5.2, Fig. 17] The evidence for 'nonlinear (polynomial)' growth is not quantified. Fig. 17 plots ln C_s versus T, not versus ln T, and shows only two mass values with no error bars, no fitted exponent, and no alternative linear fit. Given the conflict with Appendix B, a quantitative late-time analysis is necessary: for example, extract d ln C_s/d ln T at large T, or directly test C_s = a T + b and compare with a power law. The present plot does not by itself support a 'mix of powers' claim.
minor comments (5)
- [Eq. (3.36)] The expression contains 'ln ln' where a single logarithm is clearly intended; the formula should read C_s = (4c/3) Tϵ + (4c/3) ln(...).
- [Fig. 13 caption] The caption mentions 'AdS radius L' in a de Sitter context; this should be 'de Sitter radius L'.
- [Eq. (5.21)] The exponent is written as '(d−2)−#g(μ)' with the symbol '#' unexplained and the function g(μ) not defined. Please define g and specify the range of T_max−T over which this fit is valid.
- [§2] The notation V_{S^{d−1}} is used without definition; it should be identified as the volume of the unit (d−1)-sphere.
- [References] Several references use incomplete entries, e.g., [1] is listed with only an arXiv number and no year, and [82] repeats the arXiv identifier. Please normalize the bibliography.
Circularity Check
No significant circularity: the central derivation is self-contained; the noted issues are internal consistency and unproven symmetry assumptions, not circular reasoning.
full rationale
The central derivation chain is self-contained. For pure dS, the d=3 case is solved analytically (Eq. 3.15 and Eq. 3.35), and the higher-dimensional cases use numerical solutions of Eq. 4.5, validated against the analytic d=3 results in Tables 3-5 and against independent Eddington-Finkelstein numerics in Appendix A. The SdS analysis uses the conserved charge from Eq. 5.12/5.26 with boundary conditions; no fitted parameter is later renamed as a prediction. The stretched-horizon cutoff epsilon and mass parameter mu are physical inputs, not parameters fitted to the output. Self-citations (e.g., [82], [103], [104], [106], [107], [109]) appear as comparisons or outlook examples; for instance, "Contrast this with volume complexity growth in global de Sitter space... [82]" is a side remark, not a load-bearing premise. The paper is benchmarked against external work [2], [90] and its own analytic/numeric cross-checks, so the computation does not reduce to its inputs by construction. Two issues are correctness risks rather than circularity. First, Appendix B's late-time analysis gives Eq. (B.6), "Cs ∼ rd−1 min T", i.e., linear growth, yet the next sentence says "the timelike complexity does not show linear growth with time," and Sections 5.2/6 claim "nonlinear growth (mix of powers)". Second, the pure-dS d=3 divergence is stated logarithmically in Eq. (1.2) and Eq. (3.40), but as a power law (Tmax−T)^{-(d−2)} in Eq. (1.3) and Section 4; for d=3 these cannot both hold. Also, the SdS turning-point condition is invoked "based on symmetry" (Secs. 5.1-5.2) without an independent derivation, an assumption that may affect the results but is not circular.
Assumptions & free parameters
free parameters (3)
- ϵ (stretched-horizon cutoff) =
10^{-7} (pure dS and SdS Case 1 plots), 10^{-5} (SdS Case 2), 10^{-3} (some comparisons)
- µ (SdS mass parameter) =
0.1, 0.11, 0.2, 0.3
- L (dS radius) =
1 (pure dS), 3 (SdS numerics)
assumptions (4)
- domain assumption Timelike subregion complexity is the volume of the bulk region bounded by the spacelike and timelike extremal surfaces anchored on the timelike boundary subregion (the Alishahiha prescription, adapted to dS by joining turning points with the U=V curve).
- domain assumption Static patch holography: the dual field theory lives on the stretched horizon at r=1−ϵ and the standard dictionaries S=A/4G_N and CV=V/G_N apply.
- domain assumption The classical extremization (Euler–Lagrange equations (3.14), (4.5), (5.12), (5.27)) selects the correct codimension-two surface, and the branch choices (spacelike vs timelike) are the ones dictated by the timelike-entanglement proposals [65, 66].
- domain assumption In SdS, the turning point of the spacelike extremal surface lies on the t=0 slice by reflection symmetry of the infinitely extended Penrose diagram (leading to Eq. (5.16) and Eq. (5.29)).
Cite this review
Pith. "Pith review of Holographic timelike complexity for de Sitter." pith.science (2026). https://pith.science/paper/V4NG7MYR
@misc{pith2026260729662,
author = {Pith},
title = {Pith review of: Holographic timelike complexity for de Sitter},
year = {2026},
howpublished = {\url{https://pith.science/paper/V4NG7MYR}},
note = {Machine review of arXiv:2607.29662}
}
read the original abstract
We investigate the recent proposal of holographic volume complexity for timelike subregions \cite{Alishahiha:2025xml} in the framework of static patch holography for de Sitter spacetime. Using the stretched-horizon prescription, we compute the timelike subregion complexity as a function of the subregion duration for pure de Sitter and Schwarzschild de Sitter geometries. In pure de Sitter spacetime, the timelike subregion complexity displays exponential growth for short durations, and hyperfast growth near a maximal duration, paralleling the features of spacelike volume complexity \cite{Jorstad:2022mls}. For Schwarzschild de Sitter, when the stretched horizon is near the cosmological horizon, the behavior broadly remains similar to pure de Sitter. However, when the stretched horizon is near the black hole horizon, the hyperfast growth for long durations is replaced by nonlinear growth regime. Along the way, we also compute the corresponding timelike holographic entanglement entropy for de Sitter and Schwarzschild de Sitter.
Reviewed August 3, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.