Emergent Symmetries in Ensemble Averages

Two things to know before you read arXiv:2609.36542. First, it is a review, explicitly so: the author states it is based on four of his own papers. The Siegel–Weil ensemble average, the duality-origami mechanism, and the SWSSB conjecture all live in the cited literature ([22, 25, 32, 35]). What the review adds is synthesis, and the synthesis is good. Second, the paper's genuinely new element — the holographic SWSSB demonstration and the Swampland SWSSB conjecture (Sections 5.5–5.6) — pivots on a single assertion the paper does not establish.

What it does well: the Siegel–Weil case study (Section 2) is a theorem with a credible two-step proof sketch (cusp behavior plus a differential equation), and the duality-origami identity (3.5) is a clean change-of-variables argument. The author is also honest about gaps: the complete bulk dual is unknown (b85), torus-level equality does not establish a symmetry of every observable (b97, b117), and the lattice SWSSB conjecture needs clustering at 2β and a volume-uniform bound checked per system (b191–192). That transparency is real and should count.

The soft spot is Eq. (5.22) and b215. The Rényi-2 correlator's large-distance limit becomes an integral of |G_µ(β,0)|² weighted by w(µ), and the paper asserts this is "bounded from below by a positive constant independent of the system size," citing [35]. That assertion carries the whole SWSSB-in-holography claim, and it is quoted, not proved. Worse, w(µ) is a quasi-probability — complex, not non-negative — so positivity is not automatic; sign oscillations could cancel against |G_µ|². And the source, [35], is an unpublished companion preprint. Nothing in this text rules out G_µ vanishing or decaying in the thermodynamic limit for µ in the support of w, in which case the diagnostic (5.13) fails and the conjecture loses its only concrete gravitational evidence. The author names local ensemble equivalence and the lower bound as things to check (b192), but that just underscores the point.

On Section 3 I side with your stress-test: I do not press the torus claim. Eq. (3.2) is sound because Aut(D,θ) preserves spin and braiding, hence commutes with S and T and the whole Weil representation. The full-observable gap is acknowledged and largely discharged, since g is a genuine automorphism of the bulk TQFT.

Who this is for: readers wanting a compact survey of the ensemble-average program, and referees of the companion [35]. It deserves peer review — the review format is serviceable and the open question is pointed — but the referee should insist the bound (5.22) be proved or sourced to a published computation before the SWSSB claim is taken as established. I would cite the primary papers rather than this review.

The review's only genuinely new physical element is the holographic SWSSB demonstration (Section 5.5) and the Swampland SWSSB Conjecture (Section 5.6). The demonstration's pivot is Eq. (5.22), lim R(2)(x,y) -> ∫ dµ w(µ)|G_µ(β,0)|^2 as |x−y|→∞, together with the assertion (b215) that this is 'bounded from below by a positive constant independent of the system size,' leading to 'This demonstrates SWSSB in this holographic model' (b216). This bound is asserted only by quotation from the author's companion preprint [35] (arXiv 2607.12022); the computation is not reproduced here. The bound is not automatic. The weight w(µ) ∝ e^{−iµQ} Tr[e^{iµQ}e^{−2βH}] is a quasi-probability: it need not be non-negative, so (5.22) is not literally a probability average, and positivity of the integral requires that sign oscillations not cancel against |G_µ|^2. Moreover |G_µ(β,0)|^2 is a charge-twisted thermal two-point function at the shifted Matsubara frequencies of (5.21); if G_µ vanishes for any µ in the support of w, or if its magnitude is not uniformly bounded away from zero as the boundary spatial volume V grows, the integral need not stay O(1) in the thermodynamic limit, and the SWSSB diagnostic (5.13) fails. Since the conjecture's stated motivation is exactly this model, failure of the bound would strip the conjecture of its only concrete gravitational evidence. I do not press Section 3's torus claim: Eq. (3.2) is sound because Aut(D,θ) preserves the spin and braiding, so the induced permutation commutes with U(S) and U(T), hence with the whole (Weil) representation, giving E_{Q,g·α}=E_{Q,α}. The residual full-observable gap is acknowledged by the authors (b97, b117) and is largely discharged, since g is a genuine automorphism of the bulk TQFT. The pointed, checkable, un-discharged risk is therefore b215.

The strongest_claim rightly centers on emergent symmetries after ensemble averaging. Reading in good faith, I first checked the most visible item, Eq. (3.2): it holds, because any g in Aut(D,θ) preserves both the topological spin θ and the bilinear form B, so the permutation of the anyon basis commutes with the modular S- and T-matrices and therefore with the full SL(2,Z) (Weil) representation; the Eisenstein components then satisfy E_{Q,g·α}=E_{Q,α}. The residual 'torus-level only' gap is stated honestly by the authors (b97, b117), and it is largely discharged by the observation that g is an automorphism of the bulk Abelian Chern-Simons theory itself, i.e., a symmetry of the actual averaged bulk theory. For the SWSSB half, the lattice argument's assumptions (clustering at 2β, local equivalence of ensembles, volume-uniform lower bound) are explicitly acknowledged (b191, b192). The genuinely un-discharged, load-bearing step is the holographic computation: the bound (b215) is the sole concrete evidence offered for the paper's new Swampland conjecture, yet it is quoted from the author's companion preprint [35] rather than derived. Whether ∫ dµ w(µ)|G_µ(β,0)|^2 remains O(1) in the thermodynamic limit is a real, checkable question: w is a quasi-weight that can sign-oscillate, and the twisted thermal two-point function can in principle vanish or decay. This sharpens the reader's concern: the reader placed full-observable covariance as the weakest point, whereas I read that item as acknowledged and defensible, and the quoted holographic bound as the sharper, un-discharged pivot. [35] is an accessible arXiv preprint, so the check is feasible; nothing here calls into question the solidity of the review's arranged, previously published content, and I am not attacking a published theorem. On this basis I retain the reader's CONDITIONAL verdict: the review is transparent and the synthesis is plausible, but the new conjecture's gravitational evidence should be confirmed before assigning higher weight. Hence verdict_should_be is UNCHANGED, with partial agreement with the reader's weakest_assumption.

If the Section 3 mechanism lifts to the full observable set, 'duality origami' provides a concrete construction in which T-dualities between ensemble members become symmetries of an averaged theory, and the Section 5 recasting would unify SWSSB with ensemble-average emergence, with wormhole connectivity as the bulk mechanism. The review is unusually candid about the scope of its own results: b97 and b117 explicitly flag that torus-level covariance does not establish a symmetry of the complete boundary dual, and b191–b192 flag that the Lattice SWSSB argument requires clustering at 2β and a volume-uniform lower bound to be checked per system. The measure-theoretic identity (3.5) is clean, and the Siegel–Weil review is a credible account of a proven theorem, with the convergence condition p+q>4 stated explicitly. These strengths do not, however, carry the load-bearing positivity bound of Section 5.5, which is deferred to an unpublished companion preprint and is not automatic given that the weight in Eq. (5.22) is a quasi-probability.

This review argues that a global symmetry absent from every individual theory in an ensemble can reappear after the ensemble is averaged, a loophole in the quantum-gravity dictum that no exact global symmetries exist. In the precision case of two-dimensional Narain-type conformal field theories, the moduli-space average of the torus partition function is computed exactly by the Siegel–Weil formula, and the averaged theory inherits the zero-form symmetries of the bulk anyons. The paper names the general mechanism 'duality origami': T-dualities that relate different theories in the ensemble are folded into symmetries of a single averaged theory. It then shows the same logic recasts strong-to-weak spontaneous symmetry breaking (SWSSB) of mixed states as an emergence of symmetries, proposes a wormhole bulk realization, and states a new Swampland conjecture restricting strong symmetries in quantum gravity.

What carries the argument. The load-bearing identity is the Siegel–Weil formula, which equates the moduli-space average of a Siegel–Narain $\theta$ function with the Siegel–Eisenstein series, $\langle\vartheta_{Q,\alpha}\rangle = E_{Q,\alpha}$, of the quadratic form $Q$. The second mechanism is the general notion of an 'ensemble symmetry': a group $G$ acting on both the sector labels and the moduli space such that the measure is preserved, integration domains are mapped into each other, and the observable is covariant, $O_{g\cdot\alpha}(g\cdot m) = O_{\alpha}(m)$, so that a single change of variables gives $\langle O_{\alpha}\rangle = \langle O_{g\cdot\alpha}\rangle$ (Eq. 3.5). The anyon data of the Chern–Simons bulk, the discriminant group $D$ and topological spin $\theta$, fix the Gauss sums $\gamma_{Q,\alpha}(c,d)$ that build the Eisenstein series, and these are the invariant quantities. For SWSSB the machinery is the Schwinger–Keldysh doubled-Hilbert-space formulation with the Rényi-2 correlator as diagnostic, and the projector onto a fixed charge sector is implemented as an integral over chemical potential $\mu$ in the holographic computation.

Core claim. The paper's central claim is that the ensemble-averaged Narain partition functions are invariant under the zero-form symmetry group $\mathrm{Aut}(D,\theta)$ of the bulk anyons: $\langle\vartheta_{Q,g\cdot\alpha}\rangle = E_{Q,g\cdot\alpha} = E_{Q,\alpha} = \langle\vartheta_{Q,\alpha}\rangle$ for every $g$ preserving the anyon labels' topological spin (Eq. 3.2). Because this group includes 'quantum' symmetries that do not leave any fixed $\theta$ function invariant, $\vartheta_{Q,g\cdot\alpha}(\tau,\bar\tau;m) \neq \vartheta_{Q,\alpha}(\tau,\bar\tau;m)$, the invariance is genuinely emergent: it holds only after the average, not in any member theory. The review further claims that the emergent-symmetry building blocks, the moduli-independent Gauss sums $\gamma_{Q,\alpha}(c,d)$ identified with lens-space partition functions of the Abelian Chern–Simons theory, are exactly the objects governing the infinite-distance limit, so the two emergence mechanisms coincide. Finally, it recasts SWSSB of mixed states as the same phenomenon: twirling a pure state over charge phases produces a weakly symmetric density matrix, and in the holographic BTZ model the Rényi-2 correlator stays positive because a wormhole connects the two sides of the thermal circle. On this basis it proposes the Swampland SWSSB conjecture: in weakly coupled Einstein gravity, any strong global symmetry of a mixed state must break at least down to a weak symmetry.

The premise it rests on. The load-bearing premise is that equality of the averaged torus partition functions, together with covariance of the labeled family of observables, is enough to attribute a genuine global symmetry to the complete averaged theory; the paper itself states that this requires the covariance to extend to the full set of observables of a complete boundary dual, and that local equivalence of the fixed-charge and grand canonical ensembles plus clustering at inverse temperature $2\beta$ must be checked for each system in the SWSSB argument.

If the average is the right description, the averaged Narain theory carries the zero-form symmetries $\mathrm{Aut}(D,\theta)$: the averaged modular data is invariant under anyon-label permutations that preserve topological spin.

The ensemble-average and infinite-distance mechanisms for emergent global symmetries coincide, since both are controlled by the same moduli-independent Gauss sums identified with lens-space partition functions.

SWSSB of mixed states is an emergence of symmetries in an ensemble average, with wormhole-connected bulk geometries producing a positive Rényi-2 correlator in the holographic model.

The Roelcke–Selberg spectral decomposition of the deviation from the average carries the moduli-dependent terms that break the emergent symmetries in individual theories, playing the role of the tower of states in the Distance Conjecture.

If the Swampland SWSSB conjecture holds, no exactly strongly symmetric mixed state can exist in weakly coupled Einstein gravity: any strong global symmetry must break at least to a weak one.

A natural next test is to push the covariance condition beyond the torus: if higher-genus ensemble averages of Narain theories fail $\langle O_{\alpha}\rangle = \langle O_{g\cdot\alpha}\rangle$ for some $g$, the emergent symmetry is a statement about modular data, not about the full theory; the paper leaves this gap open in Section 3.2.

The magnitudes of the moduli-dependent spectral terms in Eq. 4.1 suggest a quantitative measure of how strongly the emergent symmetry is broken in a given member theory, an analogue of stating how heavy the would-be symmetry-restoring tower is.

In condensed-matter open systems where SWSSB is already studied, the ensemble viewpoint suggests the twirling or averaging step could be engineered experimentally, providing a laboratory analogue of the holographic wormhole mechanism.

If the Swampland SWSSB conjecture is correct, it becomes a selection rule on holographic constructions: any proposed bulk state with a strong symmetry must exhibit SWSSB or conventional SSB, ruling out exact strong symmetries in weakly-coupled gravity.

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review