All-orders consistency of the SO(8192) unoriented open bosonic string: higher-loop dilaton-tadpole cancellation, the closed-string-tachyon regulator, and the origin of the special Chan–Paton choice

The 26-dimensional unoriented open bosonic string with Chan–Paton gauge group $SO(8192)=SO(2^{13})$ is the unique choice for which the tree-level dilaton tadpole cancels and the divergences of the one-loop vacuum amplitudes are absent (Douglas–Grinstein, 1987; Weinberg, 1987). Weinberg's analysis left three linked issues open: whether the cancellation persists at the next and higher orders of string perturbation theory, where the torus, cylinder, Möbius strip and Klein bottle with a single dilaton insertion contribute with Chan–Paton weights $1$, $n^{2}$, $n$, $1$ and no symmetry is known to enforce cancellation; whether the analytic-continuation prescription used to regulate the separately divergent closed-string-tachyon tadpole is physically correct or unique; and whether an underlying principle — perhaps a symmetry relating the Marcus–Sagnotti boundary fermions to the spacetime coordinates — enforces the special gauge-group choice. The problem is open: no computation of the next-order dilaton tadpole for SO(8192) is known, the tachyon-tadpole treatment remains tied to the unsolved closed-string tachyon problem, and no enforcing symmetry has been constructed.

Background

The open bosonic string in the critical dimension $D=26$ carries, at each of its two ends, internal matrix degrees of freedom — Chan–Paton factors, after the names of the originators of this device — whose products supply the group-theory factors of open-string scattering amplitudes. For an orthogonal gauge group $SO(n)$ these multiplicities weight every open-string world sheet by powers of the dimension $n$. In 1986–1987 it was discovered that this apparently arbitrary integer is sharply constrained in the unoriented theory, obtained by gauging world-sheet parity $\Omega$ (the operation exchanging left- and right-movers; the resulting quotient construction is now called an orientifold): consistent vacua then necessarily mix open and closed strings.

String perturbation theory organizes amplitudes by the Euler characteristic of the world sheet. A massless closed-string state such as the dilaton $\varphi$ — the massless scalar universally present in string spectra, whose vacuum expectation value controls the string coupling — can be emitted at tree level from the disk and from the projective plane $RP^2$ (a disk with a crosscap, i.e. a boundary identified after a half-twist, which is the world-sheet imprint of the gauged parity $\Omega$). At the next order four topologies of vanishing Euler characteristic contribute to the vacuum amplitude: the torus ${\cal T}$ (weight $1$), the annulus or cylinder ${\cal A}$ (weight $n^2$), the Möbius strip ${\cal M}$ (weight $\epsilon n$, with $\epsilon=\pm1$ a discrete sign fixed by the projection) and the Klein bottle ${\cal K}$ (weight $1$). The last three have ultraviolet-divergent regions of their modular-parameter space (the space of world-sheet shape parameters); rewriting them in the transverse channel (the modular $S$-transformed form, with $S$ the standard exchange of the two cycles of the world sheet, in which the surface depicts a closed string propagating between boundaries and/or crosscaps) turns these divergences into infrared singularities from the zero-momentum exchange of massless closed-string states. The singular terms are controlled by the dilaton tadpole — the one-point function of the dilaton on a disk or on a crosscap: a nonzero tadpole means the assumed flat background does not solve the field equations, and perturbation theory must be reorganized around a shifted background (the Fischler–Susskind mechanism).

For general $n$ the singular transverse-channel parts combine into

$\tilde{\cal K}+\tilde{\cal A}+\tilde{\cal M}\;\sim\;\frac{1}{2}\left(2^{13}+2^{-13}n^{2}-2\,\epsilon\,n\right)=\frac{2^{-13}}{2}\left(n-\epsilon\,2^{13}\right)^{2},$

a perfect square that vanishes for $\epsilon=+1$ and $n=2^{13}=8192$, fixing the gauge group $SO(8192)$. Douglas and Grinstein computed the disk dilaton tadpole and found that, combined with the projective-plane contribution, the tree-level dilaton tadpole vanishes precisely for $SO(8192)$. Weinberg then showed that the usual one-loop divergence is absent in $SO(8192)$ open bosonic string theory — but to exhibit the cancellation he had to regulate by analytic continuation the separately divergent contribution from the closed-string tachyon (the bosonic string's unavoidable mode of negative mass-squared), and his results agreed only, in his word, '(nearly)' with those obtained by other methods, reflecting subtleties of regulator and modular-parameter conventions. In modern D-brane language the condition equates the total tension of $2^{13}$ spacetime-filling extended objects on which open strings end (D25-branes, counted with their mirror images) with the negative tension of the orientifold plane (the O25-plane, the spacetime object sourced by the gauged parity operation) that they must cancel.

The result was striking because it mimics, in a non-supersymmetric and tachyonic theory, the famous infinity cancellations of the supersymmetric $SO(32)$ type-I superstring (the unoriented open superstring) found by Green and Schwarz. Marcus and Sagnotti showed that the $SO(2^{13})$ structure is what emerges from 26 world-sheet boundary fermions — fermionic variables living at the string endpoints, one for each spacetime coordinate and with matching signature — and Bianchi and Sagnotti constructed the full partition function (the vacuum amplitude summed over world sheets), showing that the involution structure (the action of the world-sheet parity — a symmetry operation that squares to the identity — on the double covers of the world sheets) and the multiplicity factors persist at arbitrary genus. Yet no symmetry of the $SO(8192)$ model is known that would require the cancellations, and the model retains both a tachyon and a nonzero cosmological constant, so flat 26-dimensional Minkowski space is at best a formal starting point for its perturbation theory. Whether the one-loop miracle reflects genuine all-order perturbative consistency — and what principle, if any, stands behind it — is the unresolved target.

Problem Statement

Decide whether the SO(8192) unoriented open bosonic string in 26 dimensions is a consistently divergence-free perturbative string theory beyond the cancellations already established at tree level and one loop. The question, as left open by Weinberg's 1987 one-loop analysis, has three linked facets that jointly settle this single objective:

1. Higher orders. The cancellations established so far concern the tree-level dilaton tadpole and the one-loop vacuum amplitudes: the tree-level tadpole — the one-point function of the dilaton on the disk plus that on the projective plane — vanishes precisely for $n=2^{13}$ (Douglas–Grinstein), and the divergences of the one-loop vacuum amplitudes on the four world sheets of vanishing Euler characteristic — torus, cylinder, Möbius strip, Klein bottle, carrying relative Chan–Paton weights $1$, $n^{2}$, $n$, $1$ — are correspondingly absent for $n=2^{13}$ (Weinberg). At the next order the dilaton tadpole itself receives contributions from these same four surfaces, each with a single dilaton insertion and again carrying relative weights $1$, $n^{2}$, $n$, $1$; by the Fischler–Susskind mechanism these one-loop one-point functions control the divergences that appear in the vacuum amplitudes at subsequent orders. It is unknown whether the divergent parts of these next-order tadpole contributions cancel for $n=2^{13}$, and no known symmetry mandates such a cancellation. Determine whether they cancel, and whether cancellation persists order by order (all genera and all numbers of boundaries and crosscaps).

2. Regulator. The one-loop cancellation is obtained only after regulating the separately divergent closed-string tachyon tadpole by an analytic continuation (as reported, inserting a factor $q^{\alpha' k^{2}/2}$, integrating at large loop momentum $k^{2}$, and continuing back to $k^{2}=0$), and the individually divergent contributions depend on the regulator and on modular-parameter conventions. Establish whether this prescription is the unique or physically correct treatment of the closed-string tachyon tadpole, or replace it by a well-defined alternative.

3. Principle. Identify the mechanism, if any, that enforces the special Chan–Paton choices: the Marcus-Sagnotti boundary-fermion construction realizes $SO(2^{13})$ with 26 fermions matching the 26 spacetime coordinates in number and signature, and Weinberg conjectured a possible new supersymmetry-like relation between these boundary fermions and the spacetime coordinates. Establish whether such a symmetry (or any other principle) exists, and whether it enforces the cancellations at one loop and protects them at all orders.

A complete answer settles this single yes/no objective. The three facets are independently refutable sub-claims, and a refutation of each does not resolve the objective in the same way. A refutation of facet 1 (the cancellations demonstrably fail at some order for $n=2^{13}$) or of facet 2 (no consistent regulator treatment exists) resolves the overarching question negatively: in particular, a demonstration that the next-order divergences fail to cancel for $n=2^{13}$, or that no consistent regulator treatment exists, refutes the model's all-orders consistency. A negative resolution of facet 3 (no enforcing symmetry or principle exists), by contrast, does not refute the model's consistency and leaves the objective open, since the absence of a protective symmetry is compatible with the cancellations holding at the orders where they are established. Only a positive facet-3 resolution — an enforcing symmetry, strictly stronger than and subsuming the next-order cancellation of facet 1 — or an all-orders facet-1 computation settles the objective affirmatively; the next-order computation alone does not. The question concerns the perturbative consistency of the tadpole-cancellation mechanism itself, not phenomenological viability; the model's open and closed tachyons and its nonzero one-loop cosmological constant are part of the given setting and may be handled by any consistent prescription the answer proposes and justifies.

References

1. S. Weinberg, 'Cancellation of one-loop divergences in SO(8192) string theory', Phys. Lett. B 187 (1987) 278-282, doi:10.1016/0370-2693(87)91096-3, https://doi.org/10.1016/0370-2693(87)91096-3
2. M. B. Green and J. H. Schwarz, 'Infinity cancellations in SO(32) superstring theory', Phys. Lett. B 151 (1985) 21-25, doi:10.1016/0370-2693(85)90816-0, https://doi.org/10.1016/0370-2693(85)90816-0
3. W. Fischler and L. Susskind, 'Dilaton tadpoles, string condensates and scale invariance', Phys. Lett. B 171 (1986) 383-389, doi:10.1016/0370-2693(86)91425-5, https://doi.org/10.1016/0370-2693(86)91425-5
4. M. R. Douglas and B. Grinstein, 'Dilaton tadpole for the open bosonic string', Phys. Lett. B 183 (1987) 52-58, erratum: Phys. Lett. B 187 (1987) 442, doi:10.1016/0370-2693(87)91416-X, https://doi.org/10.1016/0370-2693(87)91416-X
5. Nobuyoshi Ohta, 'Cancellation of dilaton tadpoles and two-loop finiteness in SO(32) type-I superstring theory', Phys. Rev. Lett. 59 (1987) 176-179, doi:10.1103/PhysRevLett.59.176, https://doi.org/10.1103/PhysRevLett.59.176
6. N. Marcus and A. Sagnotti, 'Group theory from "quarks" at the ends of strings', Phys. Lett. B 188 (1987) 58-64, doi:10.1016/0370-2693(87)90705-2, https://doi.org/10.1016/0370-2693(87)90705-2
7. Jun Liu and Joseph Polchinski, 'Renormalization of the Möbius volume', Phys. Lett. B 203 (1988) 39-43, doi:10.1016/0370-2693(88)91566-3, https://doi.org/10.1016/0370-2693(88)91566-3
8. Soo-Jong Rey, 'Off-shell prescription for dilaton tadpoles', Phys. Lett. B 203 (1988) 393-399, doi:10.1016/0370-2693(88)90189-X, https://doi.org/10.1016/0370-2693(88)90189-X
9. Yoshiaki Tanii, 'One loop amplitudes and divergences in unoriented string theories', Phys. Lett. B 203 (1988) 99-104, doi:10.1016/0370-2693(88)91578-X, https://doi.org/10.1016/0370-2693(88)91578-X
10. Massimo Bianchi and Augusto Sagnotti, 'The partition function of the SO(8192) bosonic string', Phys. Lett. B 211 (1988) 407-416, doi:10.1016/0370-2693(88)91884-9, https://doi.org/10.1016/0370-2693(88)91884-9
11. Greg Nagao, 'Involution dependent constants and the cancellation of divergences in the one loop open string amplitude', Phys. Rev. D 38 (1988) 569-577, doi:10.1103/PhysRevD.38.569, https://doi.org/10.1103/PhysRevD.38.569
12. Neil Marcus, 'Unitarity and regularized divergences in string amplitudes', Phys. Lett. B 219 (1989) 265-272, doi:10.1016/0370-2693(89)90389-4, https://doi.org/10.1016/0370-2693(89)90389-4
13. Jan Govaerts, 'Quantum consistency of open string theories', Phys. Lett. B 220 (1989) 77-86, doi:10.1016/0370-2693(89)90017-8, https://doi.org/10.1016/0370-2693(89)90017-8
14. Greg Nagao, 'Divergences and involution dependent constants', Mod. Phys. Lett. A 4 (1989) 445-450, doi:10.1142/S021773238900054X, https://doi.org/10.1142/S021773238900054X
15. Ram Brustein and S. P. de Alwis, 'Nonperturbative divergence in critical string theory', Phys. Lett. B 247 (1990) 31-35, doi:10.1016/0370-2693(90)91044-C, https://doi.org/10.1016/0370-2693(90)91044-C
16. Hitoshi Konno and Takashi Suzuki, 'A secondary structure of general multiloop amplitude in open bosonic string theory', Phys. Rev. D 41 (1990) 3159-3164, doi:10.1103/PhysRevD.41.3159, https://doi.org/10.1103/PhysRevD.41.3159
17. Greg Nagao, 'Open strings as "space-time" Z(2) orbifolds of closed strings', Mod. Phys. Lett. A 6 (1991) 2483-2496, doi:10.1142/S021773239100292X, https://doi.org/10.1142/S021773239100292X
18. Z. Bern and D. C. Dunbar, 'Solution of the dilaton problem in open bosonic string theories', Gen. Rel. Grav. 23 (1991) 107-113, doi:10.1007/BF00766518, https://doi.org/10.1007/BF00766518
19. Paul H. Frampton and Thomas W. Kephart, 'Mersenne primes, polygonal anomalies and string theory classification', arXiv:hep-th/9904212, https://arxiv.org/abs/hep-th/9904212
20. Tsuguhiko Asakawa, Taichiro Kugo and Tomohiko Takahashi, 'One loop tachyon amplitude in unoriented open-closed string field theory', arXiv:hep-th/9905043, https://arxiv.org/abs/hep-th/9905043
21. Yutaka Matsuo, 'Fate of unoriented bosonic string after tachyon condensation', arXiv:hep-th/9905044, https://arxiv.org/abs/hep-th/9905044
22. Shin Nakamura, 'Closed string tachyon condensation and on-shell effective action of open string tachyons', arXiv:hep-th/0105054, https://arxiv.org/abs/hep-th/0105054
23. Francois Englert, Laurent Houart and Anne Taormina, 'The bosonic ancestor of closed and open fermionic strings', arXiv:hep-th/0203098, https://arxiv.org/abs/hep-th/0203098
24. Shyamoli Chaudhuri, 'Bosonic matrix theory and matrix D-branes', arXiv:hep-th/0205306, https://arxiv.org/abs/hep-th/0205306
25. Noriaki Kitazawa, 'Tadpole resummations in string theory', arXiv:0801.1702, https://arxiv.org/abs/0801.1702
26. Carlo Angelantonj and Augusto Sagnotti, 'Open strings', Phys. Rept. 371 (2002) 1-150, erratum: Phys. Rept. 376 (2003) 407, doi:10.1016/S0370-1573(02)00273-9, https://doi.org/10.1016/S0370-1573(02)00273-9 (arXiv:hep-th/0204089)
27. Emilian Dudas, Jean Mourad and Augusto Sagnotti, 'Supersymmetry breaking with fields, strings and branes', Phys. Rept. 1175 (2026) 1-256, doi:10.1016/j.physrep.2026.02.005, https://doi.org/10.1016/j.physrep.2026.02.005 (arXiv:2511.04367)