REVIEW 4 major objections 3 minor 1 cited by
Observables are glocal
T0 review · 4 major / 3 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read A complete set of physical observables exists for graph-based quantum gravity — and building it means solving graph isomorphism.
desk verdict A genuinely interesting construction for abstract graph theories, but the leap from label-permutation invariance to physical diffeomorphism invariance is argued, not shown. 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 central machinery is group averaging over the permutation group acting on graph labels. Starting from a spin network state associated with a labeled graph, one forms invariants by summing, or projecting, over all relabelings, so that each invariant is constant along orbits of the relabeling action and therefore probes the whole graph. The new structural ingredient is the construction of complete sets whose members are each supported on connected subgraphs; completeness of such a set is then shown to be equivalent to a graph isomorphism test. This equivalence is what makes the construction explicit and ties the existence of observables to computational complexity.
What would settle it
Compute the constructed complete observables for all spin network states on graphs with up to five vertices and compare the resulting values with the partition into isomorphism classes of the underlying graphs. Any two states from different isomorphism classes that receive identical values for every observable would falsify completeness, while any two states from the same class that receive different values would falsify the claim that the observables are permutation-invariant.
Extended reading notes
Core claim
The central claim is that the problem of observables is fully resolved for background-independent theories on graphs: for any state in the spin network state space, there exists a set of permutation-invariant, group-averaged functionals that completely separates inequivalent states. Each individual invariant is global in that its value depends on the entire labeled graph, yet the paper shows that a complete set can be organized so that every invariant is associated with a connected subgraph, encoding local correlations. The completeness of a candidate set is equivalent to deciding whether two graphs are isomorphic: the invariants fail to be complete precisely when distinct graph orbits are not distinguished, so checking completeness is solving a graph isomorphism problem. The paper concludes that this supplies physically meaningful complete observables for discrete general relativity and a permutation-invariant reformulation of the spin network state space.
Load-bearing premise
The load-bearing premise is that two discrete geometries should be considered physically identical when they differ only by a relabeling of graph vertices — that is, permutation invariance is the discrete analogue of coordinate independence, and if that identification is wrong, the constructed invariants are invariants of the wrong equivalence relation and are not physical observables.
Editorial extensions
If this is right
- Every permutation-invariant physical state of a graph-based background-independent theory is uniquely pinned down by a finite, explicit set of glocal observables; no observable information is lost by restricting to invariants.
- Discrete general relativity inherits a well-defined algebra of physical observables, giving the spin network formulation a concrete answer to the problem of observables.
- The spin network state space of loop quantum gravity admits a permutation-invariant reformulation, so one can work directly with unlabeled graphs plus complete invariant data.
- Constructing or verifying a complete set of observables is at least as hard as graph isomorphism, so there is no general efficient procedure unless graph isomorphism turns out to be tractable in polynomial time.
- Completeness comes in local pieces: because each member of the set is tied to a connected subgraph, physical geometry can be recovered from correlations among subgraph observables.
Reading between the lines
- If permutation invariance is the right gauge equivalence, then the physical content of a discrete spacetime is its isomorphism class as an unlabeled graph; observable quantities are isomorphism invariants, and known graph invariants such as spectra provide candidate checks or completions.
- The same glocal structure may appear in other discrete, label-based approaches to quantum gravity: any theory whose gauge group is a permutation of fundamental constituents will face the same isomorphism-completeness trade-off.
- The graph-isomorphism hardness suggests that even when observables exist in principle, evaluating a complete set on a large spin network is intractable; practical predictions may require observables that are only approximately complete.
- A concrete extension would be to implement the group averaging for all graphs up to a small number of vertices and verify that the number of independent complete invariants matches the number of isomorphism classes.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper claims to resolve the problem of observables for background-independent theories defined on graphs by explicitly constructing complete sets of observables. The proposed observables are formed by group averaging over graph-label permutations, and each invariant is said to probe the entire graph while also being tied to a connected subgraph structure, a combination the authors term "glocal." The paper further claims that checking completeness of such a set is equivalent to solving a graph isomorphism problem, implying computational hardness, and that these results provide complete physical observables for discrete general relativity and a permutation-invariant reformulation of the spin-network state space of loop quantum gravity.
Significance. If the technical claims are correct, the paper would supply an explicit, complete observable set for a discrete gauge theory, which is a long-standing goal in background-independent quantum gravity. The connection between observables and graph isomorphism is conceptually striking and could be of independent interest to the quantum-gravity and complexity-theory communities. The paper also makes a clear, falsifiable prediction: that constructing complete observables is computationally hard. These strengths are conditional, however, because the supplied full text is largely unreadable, preventing verification of the derivations, and because the central physical identification of label-permutation invariance with coordinate independence is argued rather than derived. The manuscript's contribution would be solid for abstract combinatorial graphs, but its relevance to loop quantum gravity as a physical theory requires additional justification.
major comments (4)
- [Abstract] The central identification, stated as "argued" in the abstract, is that coordinate independence in discrete gravity is the same as invariance under relabeling of graph vertices and edges. This premise is load-bearing: the completeness theorem, the graph-isomorphism connection, and the claim to resolve the LQG observable problem all inherit this equivalence. In loop quantum gravity, however, spin-network states are based on embedded graphs, and the diffeomorphism group acts by moving the embedding rather than by permuting labels of an abstract graph. Two states with identical abstract graph and identical SU(2) data, such as an unknotted and a knotted embedding of the same graph, can be non-diffeomorphic yet indistinguishable under label-permutation invariants. The authors should either restrict their claims to abstract combinatorial graphs and revise the title/abstract accordingly, or provide a concrete argument or test showing that permutation invariance captures the physical equivalence relevant to LQG. Without this, the construction is a genuine result for label-gauge fixing but not a resolution of the LQG observable problem.
- [Full Text (group averaging)] The paper does not demonstrate, in the available text, that group averaging over the permutation group is well defined and yields nonzero invariants on the spin-network state space. For a finite graph, the permutation group is finite and compact, so averaging is formally straightforward, but for infinite or arbitrary graphs the averaging map may diverge or be ill-defined, and the paper does not state the precise class of graphs and Hilbert-space domains. Moreover, the claim that the invariants "probe the entire graph" while being "tied to connected subgraph structures" needs a precise definition of the subgraph structure used and a proof that the resulting set separates orbits. Because no definitions or proofs are visible in the supplied text, this technical core cannot be assessed.
- [Abstract / Full Text] The claimed equivalence between checking completeness and solving a graph isomorphism problem is not stated with enough precision to be evaluated. The authors should specify the exact decision problem: is the equivalence polynomial-time, polynomial-time with oracles, or something weaker? Does it concern labeled or unlabeled graphs, and are graph colors (e.g., spin labels) included? The complexity claim "computationally costly" is also imprecise, since graph isomorphism is not known to be NP-complete and admits quasi-polynomial algorithms. A precise formal statement, ideally with a proof sketch or theorem, is necessary for the advertised connection to complexity theory.
- [Full Text (entire manuscript)] The supplied manuscript body consists of an unreadable character stream, so no equation, definition, lemma, or proof could be inspected. This is a fundamental obstacle to refereeing: the central claims of existence, completeness, and the glocal property are entirely unverifiable from the submitted text. I cannot determine whether the derivations are sound or whether there are internal inconsistencies. The authors should provide a readable version so that the technical content can be evaluated.
minor comments (3)
- [Abstract] The term "glocal" is introduced without a formal definition in the abstract; a precise mathematical definition should appear early in the paper.
- [Abstract] The phrase "permutation invariant reformulation of the spin networks state space" needs a clear statement of how this reformulation relates to the standard group-averaging techniques used in loop quantum gravity; relevant literature should be cited.
- [Full Text] The text contains repeated and fragmented passages that appear to be a formatting or encoding corruption; even aside from the unreadable equations, the narrative structure is not recoverable, so a carefully typeset manuscript is needed.
Circularity Check
No significant circularity beyond a definitional reduction: the graph-isomorphism connection is a direct consequence of defining physical equivalence as relabeling invariance.
-
self definitional
[Abstract (physical-equivalence identification and computational-complexity claim)]
"The appropriate analogue of coordinate independence is argued to be the invariance under changes of graph labels, a kind of permutation invariance. ... the construction of a complete set of observables for discrete spacetime theories is computationally costly, as it corresponds to solving a graph isomorphism problem."
The paper defines physical equivalence as invariance under graph relabeling and defines complete observables as invariants separating those equivalence classes. Under that identification, a complete set of observables is definitionally a set of functions separating graph-isomorphism orbits, so checking completeness is graph isomorphism by construction. The 'deep connection' to computational complexity is therefore a restatement of the chosen definition rather than an independent derivation. The glocal construction itself is not circular; the circularity is limited to the framing of the complexity result as a discovered connection.
full rationale
The paper's core construction -- group-averaged invariants tied to connected subgraphs -- is a self-contained invariant-theoretic construction for finite permutation groups, and no fitted parameters or predictions are involved. The load-bearing physical premise (relabeling invariance as the discrete analogue of coordinate independence) is explicitly labeled as 'argued' rather than derived, which is an assumption, not a circular step. The only mild circularity is definitional: once physical states are taken to be graph-isomorphism classes, the equivalence between completeness and graph isomorphism follows from the meaning of 'complete', so the complexity connection does not add independent content beyond the construction. No self-citation chain or uniqueness theorem appears in the abstract and no such argument can be identified from the supplied text. Overall the paper is not circular in any damaging sense.
Assumptions & free parameters
assumptions (3)
- ad hoc to paper Graph-label permutation invariance is the discrete analogue of coordinate or diffeomorphism independence.
- domain assumption Group averaging over label permutations produces well-defined, nonzero invariants on the spin network state space.
- domain assumption Geometrical information is fully encoded in graphs together with their spin network data.
Cite this review
Pith. "Pith review of Observables are glocal." pith.science (2026). https://pith.science/paper/EXHKTCNK
@misc{pith2026250802346,
author = {Pith},
title = {Pith review of: Observables are glocal},
year = {2026},
howpublished = {\url{https://pith.science/paper/EXHKTCNK}},
note = {Machine review of arXiv:2508.02346}
}
read the original abstract
We show that the problem of observables can be fully resolved for background independent theories defined on graphs, through the explicit construction of complete observables. The appropriate analogue of coordinate independence is argued to be the invariance under changes of graph labels, a kind of permutation invariance. Invariants are formed by group averaging and they probe the entire graph -- they are global. Strikingly, sets of complete observables can be constructed so that each of the invariants comprising them seeks a connected subgraph structure -- local correlations. Geometrical information is fully encoded through this subtle interplay of global and local graph notions, a behavior we term glocal. This provides physically meaningful complete sets of observables for discrete general relativity, and a permutation invariant reformulation of the spin networks state space of loop quantum gravity. Our analysis reveals an important new aspect of the problem of observables, demonstrating a deep connection between the theory of spacetime and computational complexity theory: the construction of a complete set of observables for discrete spacetime theories is computationally costly, as it corresponds to solving a graph isomorphism problem.
Forward citations
Cited by 1 Pith paper
-
Bosonic and fermionic statistics in nonperturbative quantum gravity
In loop quantum gravity, enforcing invariance under graph automorphisms produces fermionic and mixed-statistics sectors for the quanta of volume, not only bosonic ones.
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.