REVIEW 2 major objections 3 minor 25 references
Non-singular string cosmology via $\alpha^{\prime}$ corrections
T0 review · 2 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read This paper constructs non-perturbative solutions of the Hohm-Zwiebach equations, with all alpha-prime corrections included, in which the Hubble parameter stays finite for all time and the pre-big-bang and post-big-bang phases are smoothly…
desk verdict Explicit non-singular ansatz in the Hohm–Zwiebach framework, but the n=1 solution fails the single-valued f(H) consistency check, so the central claim is unsupported. 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 machinery is the Hohm-Zwiebach truncation of string effective actions: for FLRW backgrounds with vanishing Kalb-Ramond field, all orders of $\alpha'$ corrections can be written as $I=\int dt\,e^{-\Phi}\left(-\dot\Phi^2+\sum_{k\ge1}(\alpha')^{k-1}c_k\,\mathrm{tr}(\dot S^{2k})\right)$, where $S$ is the $O(d,d)$ matrix built from the scale factor. The equations of motion reduce to $\ddot\Phi+\tfrac12 H f(H)=0$, $\frac{d}{dt}(e^{-\Phi}f(H))=0$, and $\dot\Phi^2+g(H)=0$, with $f(H)=-2dH-2d\alpha'H^3+O(\alpha'^2)$ and $g(H)=-dH^2-\tfrac32 d\alpha'H^4+O(\alpha'^2)$, related by $g'(H)=H f'(H)$. The construction chooses a positive integer $n$, writes $f(t)$ and $g(t)$ directly as functions of time, solves for $H(t)$ and $\Phi(t)$, and then checks that the large-$|t|$ expansion reproduces the first two orders of the perturbative $f(H)$ and $g(H)$.
What would settle it
Compute the explicit $n=1$ solution and locate two times $t_1\neq t_2$ with $H(t_1)=H(t_2)$; such pairs exist because $H(t)$ rises and then falls. If $f(t_1)\neq f(t_2)$, then no function $f(H)$ can reproduce the ansatz, and the solution cannot be a solution of the Hohm-Zwiebach equations (3.19) with $f$ and $g$ as functions of $H$. Directly substituting the ansatz into (3.19) at those two times settles the question.
Extended reading notes
Core claim
The central claim is that the equations of motion (3.19) derived from the Hohm-Zwiebach action admit an explicit one-parameter family of non-perturbative, non-singular cosmological solutions. For the bosonic string, where $c_2=1/64$, the relevant member is the $n=1$ solution $$H_\pm(t)=\mp\frac{\sqrt{2}(\$\alpha$'-2d $t^{2}$)}{(\$\alpha$'+2d $t^{2}$)^{3/2}}, \qquad \Phi(t)=\log\left(\frac{1}{2}\frac{\sqrt{\$\alpha$'}}{d(\$\alpha$'+2d $t^{2}$)}\right),$$ with $f_\pm(t)=\mp 2\sqrt{2}d/\sqrt{\alpha'+2d t^2}$ and $g(t)=-4d^2t^2/(\alpha'+2d t^2)^2$. These functions are finite for all real $t$, and their expansion at large $|t|$ matches $H(t)=1/(\sqrt{d}\,t)-(5/4)\alpha'/(d^{3/2}t^3)+O(\alpha'^2)$ together with the corresponding dilaton series. The paper presents this as evidence that the big-bang singularity is smoothed out by higher-derivative $\alpha'$ corrections, with the singular point $t=0$ appearing only in the truncated perturbative solution.
Load-bearing premise
The construction assumes that the time-dependent coefficient functions $f(t)$ and $g(t)$ can be expressed as single-valued functions of the Hubble rate $H$, because the Hohm-Zwiebach action defines them through $H$, but this is never proved and already fails for the $n=1$ solution, where $H(t)$ is non-monotonic and the same $H$ occurs at two different times with different $f$ values.
Editorial extensions
If this is right
- If the central claim is correct, the big-bang singularity in the tree-level gravi-dilaton cosmology disappears once all $\alpha'$ corrections are included; the singular point is replaced by a regular evolution with finite $H(t)$.
- The pre-big-bang and post-big-bang branches become one continuous history: the universe contracts, passes through a regular phase near $t=0$, and then expands.
- In the limits $|t|\to\infty$ or $\alpha'\to0$, the non-perturbative solutions reduce exactly to the known perturbative results to first order in $\alpha'$, so the new solutions are consistent with earlier results.
- The scale-factor dual obtained by $H\to -H$ is also a solution, so the construction respects the $O(d,d)$ symmetry underlying the Hohm-Zwiebach action.
- The Einstein-frame Hubble parameter $H_E(t)$ is regular for all $t$ as well, so the singularity resolution is not an artifact of working in the string frame.
Reading between the lines
- Because the higher-order coefficients $c_{k\ge3}$ are not fixed by the paper, the $\alpha'$ resolution is a phenomenological completion rather than a unique prediction; an independent string computation of $c_{k\ge3}$ would show whether the smoothing survives all consistent completions.
- A natural test is to add a time-dependent Kalb-Ramond field or matter sources; if regular non-perturbative solutions still exist, the mechanism is robust, and the dilaton might be stabilized at late times as in loop-corrected models.
- The brief contraction phase near $t=0$ suggests that, after perturbations are added, this class of backgrounds would produce a characteristic primordial spectrum; computing it would give an observational way to distinguish an $\alpha'$-driven bounce from a loop-correction bounce.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper constructs a one-parameter family of cosmological solutions to the Hohm-Zwiebach action for the bosonic gravi-dilaton system, with the goal of showing that α′ corrections can remove the big-bang singularity. For n=1, the solutions (3.24) give regular H±(t), Φ(t), f±(t), and g(t); the authors verify that the large-|t| expansion matches the known perturbative solution to first order in α′, and compare with loop-corrected bouncing solutions. The paper concludes that the big-bang singularity can indeed be smoothed out by higher-derivative corrections, while acknowledging in Section 4 that the construction is phenomenological.
Significance. If valid, the explicit non-singular solutions would be a useful demonstration within the Hohm-Zwiebach framework. The paper is clearly written, and it is genuinely valuable that the authors present a worked example whose first two α′ orders match the perturbative results in [15]; the comparison with loop-corrected bounces is also informative. However, the central consistency requirement—that f and g be single-valued functions of H—is not met by the explicit n=1 solution, so the construction as stated does not solve the equations of motion it claims to solve.
major comments (2)
- [Sec. 3, Eqs. (3.21)–(3.24)] The n=1 solution is not a solution of the EOM (3.19) because f(t) and g(t) cannot be expressed as single-valued functions of H(t). For H_+(t) = −√2(α′−2d t^2)/(α′+2d t^2)^{3/2}, the function is even and non-monotone: it has a negative local minimum at t=0, a zero at |t|=√(α′/(2d)), and a positive maximum at |t|=√(5α′/(2d)), after which it decays to zero. Hence each H in (0,H_max) is attained at two different values of |t|, while f_+(t) = −2√(2d)/√(α′+2d t^2) takes different values at those two times. No single-valued function f(H) reproduces this relation, and the identity g(H)=H f(H)−∫_0^H f(x)dx cannot hold globally. The consistency check in Eqs. (3.27)–(3.28) is performed only on the large-|t| monotone branch and therefore cannot detect this multivaluedness. A branch rule could in principle repair the argument, but none is supplied, so the trajectory is not a legitimate solution of the action (3.18), whose Lagrangian depends on H only through single-valued f(H) and g(H).
- [Abstract; Sec. 3, Eq. (3.20); Sec. 4] The Abstract's claim that the big-bang singularity 'indeed could be smoothed out' is stronger than what the construction supports. The coefficients c_k≥3 in (3.20) are free and are never determined from the bosonic string; the proposed solution implicitly postulates a specific all-order completion by choosing f(t), g(t), and H(t), but no independent derivation of c_k is given. The perturbative matching at large |t| fixes only c_1 and c_2 and cannot select the infinite set of higher-order coefficients. The paper's own characterization of the construction as phenomenological is appropriate, but in that case the result is a conditional existence statement for an unspecified completion, not a demonstration about the bosonic string's α′ corrections. To support the abstract, the authors would need either to derive the required c_k from string theory or to show that some known principle fixes them.
minor comments (3)
- [Sec. 3, after Eq. (3.25)] There is a typo: 'diaton' should be 'dilaton', and the same paragraph contains 'the the singularity', which should be 'the singularity'.
- [Sec. 3, Eqs. (3.21)–(3.22)] The general formulas for H(t), f(t), and g(t) are difficult to reproduce as printed; I could not straightforwardly reduce (3.22) to the explicit n=1 expressions (3.24) without additional intermediate steps. Please display the simplified n=1 derivation or add a supplementary computation, since the main claim depends on these formulas.
- [Throughout Sec. 3] The notation alternates between f(t), g(t) and f(H(t)), g(H(t)); the paper should state explicitly that, on any chosen branch, f(t) is defined as f(H(t)) and explain how branch choices are made. This is related to the major issue above, but even for a corrected version a notational clarification would help.
Circularity Check
The n=1 ansatz (3.24) is selected to be regular and to match two known coefficients; the conclusion that α′ corrections smooth the singularity is therefore built into the ansatz, and the perturbative 'complete agreement' uses a local inversion that does not hold globally.
-
fitted input called prediction
[Section 3, Eq. (3.21)-(3.24) and the consistency check after Eq. (3.24)]
"To match c2 = 1/64 for the bosonic string theory we are concerned here, we set n = 1 and the solutions are ... We now check the consistency ... It is obvious that, for fixed finite α′, the solutions are regular everywhere ... The big-bang singularity is indeed smoothed out."
The family (3.21)-(3.22) is constructed by making an ansatz for H(t), f(t), and g(t) whose regularity is the desired conclusion. The only string-determined input is that the first two coefficients c1 and c2 are matched, which fixes the discrete parameter n by 'To match c2 = 1/64 ... we set n = 1.' The higher coefficients ck≥3 remain free, and the regular H(t) in (3.24) is chosen, not derived, to have no pole. Hence the statement that α′ corrections can smooth the singularity is an input assumption about the unknown completion, not a prediction extracted from known α′ corrections.
-
other
[Section 3, Eq. (3.24) and Eq. (3.27)-(3.28)]
"In eq. (3.27) and eq. (3.28), we used eq. (3.25) to replace t by the Hubble parameter H. Comparing with the perturbative results eq. (3.20), we find complete agreement."
This check assumes H(t) can be inverted into a single-valued function so that f(t)=f(H(t)) as required by the Hohm-Zwiebach EOM (3.19). For n=1, H+(t) is even and non-monotone: it has a negative local minimum at t=0, a zero at |t|=(α′/(2d))^{1/2}, a positive maximum at |t|=(5α′/(2d))^{1/2}, and then decays to 0, while f+(t) is an even function of |t|. Therefore the same H value occurs at different |t| with different f values, so no single-valued f(H) exists. The expansion (3.25) is valid only at large |t| on one monotone branch, so the 'complete agreement' is a local consistency check rather than a global proof that (3.24) solves (3.19); moreover the matching was selected by choosing the ansatz family to reproduce (3.20).
full rationale
The paper's own text frames the construction as phenomenological: 'one can assume values for ck≥3 and solve the EOM' and 'the constructed solution is anticipated to be regular everywhere.' The regular Hubble parameter (3.24) is therefore an input, not an output. The consistency check verifies only the first two orders (3.25)-(3.28), which were already imposed by the matching condition 'To match c2 = 1/64 ... we set n = 1'; the infinitely many higher-order coefficients are never fixed by independent string-theoretic data. Moreover, the check that f(t) and g(t) agree with the perturbative functions f(H),g(H) of (3.20) requires replacing t by H; for n=1, H+(t) is non-monotone, so f+(t) cannot be written as a single-valued function of H, and the relation g(H)=Hf(H)-∫f(x)dx cannot hold globally. The large-|t| expansions (3.27)-(3.28) sample only one monotone branch and cannot detect this. Because of these two issues, the central claim that 'the big-bang singularity indeed could be smoothed out by higher derivative α′ corrections' is not an independent derivation: the regular history was chosen, the two matching coefficients were used to fix the only free discrete parameter n, and the global solution status in the HZ action is unproven. This is partial circularity rather than a self-citation chain; the paper honestly labels the construction phenomenological.
Assumptions & free parameters
free parameters (2)
- ck>=3 (higher-order alpha-prime coefficients) =
not specified; implicitly fixed by the ansatz
- n (solution family label) =
n=1 for bosonic string (c2=1/64); n>=2 for type II
assumptions (3)
- domain assumption The O(d,d) matrix S keeps the standard form (3.11) to all orders in alpha-prime via field redefinitions.
- domain assumption The complete alpha-prime corrections for FLRW cosmology are given by the Hohm-Zwiebach action (3.18) with c1=-1/8, c2=1/64 and unknown ck>=3.
- ad hoc to paper For the proposed solutions, f(t) and g(t) are single-valued functions of H(t) and satisfy g'(H)=H f'(H).
invented entities (1)
-
Implicit non-perturbative completion of the alpha-prime expansion
Cite this review
Pith. "Pith review of Non-singular string cosmology via $\alpha^{\prime}$ corrections." pith.science (2026). https://pith.science/paper/BQVHK5HT
@misc{pith2026190900830,
author = {Pith},
title = {Pith review of: Non-singular string cosmology via $\alpha^\prime$ corrections},
year = {2026},
howpublished = {\url{https://pith.science/paper/BQVHK5HT}},
note = {Machine review of arXiv:1909.00830}
}
abstract
In string theory, an important challenge is to show if the big-bang singularity could be resolved by the higher derivative $\alpha'$ corrections. In this work, based on the Hohm-Zwiebach formula, we construct a series of non-singular non-perturbative cosmological solutions with the complete $\alpha^{\prime}$ corrections, for the bosonic gravi-dilaton system. In the perturbative regime, these solutions exactly match the perturbative results given in literature. Our results show that the big-bang singularity indeed could be smoothed out by the higher derivative $\alpha'$ corrections.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[15]
Duality Invariant Cosmology to all Orders inα′,
O. Hohm and B. Zwiebach, “Duality Invariant Cosmology to all Orders inα′,” arXiv:1905.06963 [hep-th]
arXiv 1905
-
[1]
Scale factor duality for classical and quantum strings,
G. Veneziano, “Scale factor duality for classical and quantum strings,” Phys. Lett. B265, 287 (1991). doi:10.1016/0370-2693(91)90055-U
-
[2]
A. Sen, “O(d) x O(d) symmetry of the space of cosmological solutions in string theory, scale factor duality and two-dimensional black holes,” Phys. Lett. B271, 295 (1991). doi:10.1016/0370-2693(91)90090-D
-
[3]
Twisted black p-brane solutions in string theory,
A. Sen, “Twisted black p-brane solutions in string theory,” Phys. Lett. B274, 34 (1992) doi:10.1016/0370- 2693(92)90300-S [hep-th/9108011]
arXiv 1992
-
[4]
A. A. Tseytlin, “Duality and dilaton,” Mod. Phys. Lett. A6, 1721 (1991). doi:10.1142/S021773239100186X 11
-
[5]
A.A.TseytlinandC.Vafa, “Elementsofstringcosmology,” Nucl.Phys.B 372, 443(1992)doi:10.1016/0550- 3213(92)90327-8 [hep-th/9109048]
arXiv 1992
-
[6]
String cosmology: The Pre - big bang scenario,
G. Veneziano, “String cosmology: The Pre - big bang scenario,” doi:10.1007/3-540-45334-2-12 [hep- th/0002094]
-
[7]
The Pre - big bang scenario in string cosmology,
M. Gasperini and G. Veneziano, “The Pre - big bang scenario in string cosmology,” Phys. Rept.373, 1 (2003) doi:10.1016/S0370-1573(02)00389-7 [hep-th/0207130]
arXiv 2003
Show all 25 references
-
[8]
String Theory and Pre-big bang Cosmology,
M. Gasperini and G. Veneziano, “String Theory and Pre-big bang Cosmology,” Nuovo Cim. C38, no. 5, 160 (2016) doi:10.1393/ncc/i2015-15160-8 [hep-th/0703055]
2016 arXiv
-
[9]
Pre - big bang in string cosmology,
M. Gasperini and G. Veneziano, “Pre - big bang in string cosmology,” Astropart. Phys.1, 317 (1993) doi:10.1016/0927-6505(93)90017-8 [hep-th/9211021]
1993 arXiv
-
[10]
Perturbations in a nonsingular bouncing universe,
M. Gasperini, M. Giovannini and G. Veneziano, “Perturbations in a nonsingular bouncing universe,” Phys. Lett. B 569, 113 (2003) doi:10.1016/j.physletb.2003.07.028 [hep-th/0306113]
2003 arXiv
-
[11]
Elements of String Cosmology
Maurizio Gasperini. Elements of String Cosmology. Cambridge University Press, 2007
2007
- [12]
-
[13]
Towards a nonsingular pre - big bang cosmology,
M. Gasperini, M. Maggiore and G. Veneziano, “Towards a nonsingular pre - big bang cosmology,” Nucl. Phys. B 494, 315 (1997) doi:10.1016/S0550-3213(97)00149-1 [hep-th/9611039]
1997 arXiv
-
[14]
Towards a stringy resolution of the cosmological singularity,
D. A. Easson, “Towards a stringy resolution of the cosmological singularity,” Phys. Rev. D68, 043514 (2003) doi:10.1103/PhysRevD.68.043514 [hep-th/0304168]
2003 arXiv
-
[16]
Bouncing universes in string-inspired gravity,
T. Biswas, A. Mazumdar and W. Siegel, “Bouncing universes in string-inspired gravity,” JCAP0603, 009 (2006) doi:10.1088/1475-7516/2006/03/009 [hep-th/0508194]
2006 arXiv
-
[17]
Towards a resolution of the cosmological singularity in non- local higher derivative theories of gravity,
T. Biswas, T. Koivisto and A. Mazumdar, “Towards a resolution of the cosmological singularity in non- local higher derivative theories of gravity,” JCAP1011, 008 (2010) doi:10.1088/1475-7516/2010/11/008 [arXiv:1005.0590 [hep-th]]
2010 arXiv
-
[18]
Towards singularity and ghost free theories of gravity,
T. Biswas, E. Gerwick, T. Koivisto and A. Mazumdar, “Towards singularity and ghost free theories of gravity,” Phys. Rev. Lett.108, 031101 (2012) doi:10.1103/PhysRevLett.108.031101 [arXiv:1110.5249 [gr- qc]]
2012 arXiv
-
[19]
Symmetries of higher order string gravity actions,
K. A. Meissner, “Symmetries of higher order string gravity actions,” Phys. Lett. B392, 298 (1997) doi:10.1016/S0370-2693(96)01556-0 [hep-th/9610131]
1997 arXiv
-
[20]
Symmetries of cosmological superstring vacua,
K. A. Meissner and G. Veneziano, “Symmetries of cosmological superstring vacua,” Phys. Lett. B267, 33 (1991). doi:10.1016/0370-2693(91)90520-Z 12
1991 doi
-
[21]
Are nonperturbative AdS vacua possible in bosonic string theory?,
P. Wang, H. Wu and H. Yang, “Are nonperturbative AdS vacua possible in bosonic string theory?,” Phys. Rev. D 100, no. 4, 046016 (2019) doi:10.1103/PhysRevD.100.046016 [arXiv:1906.09650 [hep-th]]
2019 arXiv
-
[22]
T-duality Constraints on Higher Derivatives Revisited,
O. Hohm and B. Zwiebach, “T-duality Constraints on Higher Derivatives Revisited,” JHEP1604, 101 (2016) doi:10.1007/JHEP04(2016)101 [arXiv:1510.00005 [hep-th]]
2016 arXiv
-
[23]
Non-perturbativedeSittervacuavia α′ corrections,
O.HohmandB.Zwiebach, “Non-perturbativedeSittervacuavia α′ corrections,” arXiv:1905.06583[hep-th]
1905 arXiv
-
[24]
de Sitter,α′-Corrections and Duality Invariant Cosmology,
C. Krishnan, “de Sitter,α′-Corrections and Duality Invariant Cosmology,” arXiv:1906.09257 [hep-th]
1906 arXiv
-
[25]
From trivial to nontrivial conformal string backgrounds via O(d,d) transformations,
M. Gasperini, J. Maharana and G. Veneziano, “From trivial to nontrivial conformal string backgrounds via O(d,d) transformations,” Phys. Lett. B272, 277 (1991). doi:10.1016/0370-2693(91)91831-F 13
1991 doi
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.