{"id":"9d53248c-3ef8-4b7e-961b-27272747f978","arxiv_id":"1908.01973","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A pAQFT-based framework constructs classical and quantum, free and interacting real scalar field models on finite causal sets.","lead":"This paper builds quantum field theory on causal sets, discrete space-times where the only structure is an ordering of points. It shows how to define free and interacting fields on any finite causal set and quantize them with the algebraic methods of perturbative algebraic quantum field theory.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"On-shell quotient Eq. (90) removes only the kernel of E, not the discretized field equations; the interacting algebras are off-shell models unless that gap is closed.","rationale":"The reader's weakest_assumption identifies the same structural point: the replacement of on-shell reduction by the kernel-of-E quotient in Eq. (90) is only approximate because EP ≠ 0. My reading of Secs. III.B, III.D, and IV.D confirms that this is the single most load-bearing assumption. The algebraic construction of the star products, Weyl algebra, states, and interacting Møller map is largely self-contained and checkable, and the paper's own caveat citing Sorkin [16] is honest. However, the abstract's central claim—'construct interacting models for arbitrary interactions'—does not carry the caveat. The quotient by ker E does not impose the bulk equations of motion on a finite causal set, because P is invertible and hence im P fills the configuration space, whereas ker E is only the degenerate part of the Peierls bracket. Consequently the free fields and their interacting deformations live in an off-shell algebra whose physical interpretation as a theory on causal sets is not established. This does not invalidate the algebraic construction itself; it means the physical claim is conditional on solving the on-shell problem. Since the paper already flags the issue and the construction is otherwise coherent, the conditional verdict remains appropriate. I found no additional independent fatal flaw: the S-matrix and Møller formulas follow the standard pAQFT pattern, the continuum limit of the free Green function is demonstrated, and the finite-dimensional setting removes UV issues. A concrete numerical test on a small diamond lattice would confirm or refute whether the quotient captures dynamics, and would make the caveat quantitative.","tokens_in":31102,"tokens_out":16121,"duration_ms":176355,"concrete_test":"On a small finite causal set, e.g. a 6x6 regular diamond lattice with PΛ and k=2, compute numerically: (a) the kernel of E = (E^+)^T - E^+, (b) the solution space Sol = E^+ E(C^-_2), and (c) the SJ two-point matrix W from Eq. (113). Then check whether dim(Sol) = N - dim ker E and whether P W = 0 or W P^T = 0 (equivalently whether the smeared free field obeys the bulk equation). The concern is settled if dim(Sol) differs from N - dim ker E and/or W does not satisfy the discretized field equation; either result would show that the quotient (90) does not encode the dynamics.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the quotient by Eq. (90), i.e. by functionals F with E_ij F,_j = 0, implements the free dynamics. In the continuum this works because of the exact sequence (91), where ker E = im P and EP = PE = 0. On a finite causal set, P is lower-triangular with nonzero diagonal and hence invertible, so im P is all of R^N, while ker E is generically much smaller; moreover the paper itself states EP ≠ 0 (Sec. III.D). Thus the quotient by ker E removes only the degenerate directions of the Peierls bracket, not the bulk equations of motion (Pφ)_p = 0 for p outside C^-_k. The free algebra and its interacting deformation in Secs. III.D and IV.D are therefore deformation quantizations of an off-shell configuration space with a nondegenerate bracket, not a field theory satisfying the discretized wave equation. The authors explicitly flag this as an approximation via Sorkin [16], but the abstract's unqualified 'first construction of interacting QFT models on causal sets' depends on this unproven step; the models are only physical if the on-shell content of the theory is restored by some additional mechanism, such as a state condition, and no such mechanism is supplied.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper adapts the framework of perturbative algebraic quantum field theory (pAQFT) to scalar fields on finite causal sets. It introduces a preferred past structure, a new discretized d'Alembertian P_Λ, retarded Green functions, a Peierls bracket, and a classical free algebra, then quantizes via deformation quantization, constructs Weyl algebras, discusses the Sorkin-Johnston state, and defines interacting algebras through time-ordered products, S-matrices, and quantum Møller operators. The abstract claims that this yields the first construction of interacting quantum field theory models for arbitrary smooth interactions on causal sets.","tokens_in":31386,"tokens_out":16560,"duration_ms":183858,"significance":"If the central claims are substantiated, the paper would be a genuinely useful bridge between causal set theory and pAQFT, providing explicit algebraic tools for interacting scalar fields on discrete causal structures. The paper contains several concrete strengths: a new discretized d'Alembertian with a clear Taylor-expansion continuum limit on the 2D diamond lattice, an exact computation of its Green function, a careful proof of positivity of the evaluation state, a self-contained construction of the Weyl algebra and SJ state, and the import of graph-expansion formulas for interacting star products from [37]. However, the physical interpretation of the free and interacting algebras is currently incomplete, because the on-shell reduction is replaced by a quotient by the kernel of the Peierls bracket matrix, which does not impose the discretized equations of motion. The significance of the claimed 'first construction of interacting QFT models on causal sets' therefore depends on closing or clearly qualifying this gap.","major_comments":[{"comment":"","section":"Sec. III.D, Eq. (90); Abstract; Sec. IV.D"},{"comment":"","section":"Secs. III.B-III.C, Eqs. (34)-(36) and (72)-(83)"}],"minor_comments":[{"comment":"","section":"Sec. III.D, Eq. (90)"},{"comment":"","section":"Sec. III.B.2, Definition III.4; Sec. V"},{"comment":"","section":"Lemma III.5, Eq. (55)"},{"comment":"","section":"Sec. IV.D.2, Eq. (154)"}],"recommendation":"major_revision","confidential_remarks":"The main gap is between the abstract's strong claim and the paper's own caveat in Sec. III.D. I would suggest asking the authors to either prove that the quotient by ker E implements the dynamics in a suitable sense, or to reframe the paper as a construction of off-shell algebraic models. The graph-expansion results are properly credited to [37], and the paper is otherwise carefully written."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is a serious, largely checkable step: it gives the first pAQFT-based construction of interacting scalar QFT models on finite causal sets, including a new preferred-past d'Alembertian whose retarded Green function on the diamond lattice is computed exactly and shown to have the correct continuum limit. That part is genuinely new and well executed. The algebraic machinery—Peierls bracket, Moyal/Wick products, S-matrix and quantum Møller operator—is imported from continuum pAQFT and transfers cleanly to finite-dimensional configuration space; no renormalization is needed, and the graph expansions from [37] apply unchanged. I would trust the algebra; the continuum limit for P_Λ is demonstrated by explicit Taylor expansion, not fitted.\n\nThe soft spot is the one the stress-test names, and it is real. The free theory is quantized off-shell: because EP ≠ 0 on causal sets, the paper does not quotient by the equations of motion (Pφ = φ^−) but by the smaller ideal of functionals with E_ij F,_j = 0 (Eq. (90)). On a finite causal set this kills only the degeneracy of the Peierls bracket, not the bulk dynamics. The interacting algebras built on top are therefore deformation quantizations of an off-shell configuration space, not of the solution space of a discretized wave equation. The authors say this explicitly in Sec. III.D and cite Sorkin's argument that dynamics can be implemented only approximately, so the limitation is not hidden. But the abstract's unqualified \"first construction of interacting QFT models on causal sets\" overstates what is established: the construction is a well-defined algebraic framework, with the physical on-shell interpretation left as an open problem. Whether that gap matters depends on whether one is content with an off-shell algebraic QFT on causal sets; the paper does not settle it.\n\nMinor issues: the nonlinear Peierls response formula (100) is stated without derivation, and Eq. (154) has a copy-paste typo (Φ_{g1} repeated instead of Φ_{gn}). The preferred past structure is admittedly ad hoc, but the authors flag this and suggest averaging over choices. These are minor.\n\nBottom line: this is a worthwhile paper. The central algebraic construction holds; the on-shell gap is real, honestly disclosed, and left open. It deserves a serious referee—one who can push the authors to soften the abstract and clarify the physical status of the quotient. I would cite it for the preferred-past d'Alembertian and for showing how pAQFT works on causal sets.","headline":"A largely checkable pAQFT framework for interacting fields on causal sets, with a real but openly acknowledged gap between the off-shell algebra and the discretized dynamics.","tokens_in":31906,"tokens_out":3061,"would_cite":true,"duration_ms":33752,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T20","83C27","81T05","81S10"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs interacting scalar quantum field theories on any fixed finite causal set, for arbitrary smooth interactions, by quantizing a classical theory built from discretised wave operators and retarded/advanced Green functions.","keywords":["causal sets","interacting quantum field theory","perturbative algebraic quantum field theory","discrete d'Alembertian","retarded Green function","deformation quantization","preferred past structure","relative Cauchy evolution"],"falsifier":"On a regular diamond lattice approximating two-dimensional flat spacetime, take the preferred-past operator $P_\\Lambda$ and a compactly supported smooth source $f$, compute $E_{+,\\Lambda}f$ with the paper's scaling, and compare pointwise with the continuum retarded solution; if the difference does not vanish as the lattice spacing tends to zero, the proposed Green function fails its continuum-limit claim. A second test: on a small random causal set, find a source $f$ for which $E P f \\neq 0$; because the equation-of-motion ideal is generated by functionals of the form $\\Phi_{Pf}$, such an $f$ shows directly that the quotient by the kernel of $E$ is not the same as imposing the wave equation.","tokens_in":30879,"feed_emoji":"⚛️","tokens_out":18386,"duration_ms":170817,"temperature":0.7,"pith_summary":"This paper uses the algebraic framework of perturbative quantum field theory to build quantum field theories on causal sets — discrete collections of points ordered by causality, proposed as the fundamental structure of spacetime. It constructs the classical theory on a fixed finite causal set: a discretised wave operator, retarded and advanced Green functions, a classical bracket, and an interacting classical theory. It then deforms this structure into a quantum theory with a distinguished state and, for any interaction that is a smooth functional of the field configuration, an explicit interacting algebra with S-matrix and correlation functions as formal power series. If the construction holds, it gives the first general framework for interacting quantum field theory on causal sets, with ultraviolet divergences absent because everything is finite on a finite causal set. It also introduces a new 'preferred past' discretised d'Alembertian that reproduces the continuum wave operator on a regular diamond lattice.","feed_headline":"Interacting quantum fields now exist on causal sets","feed_subtitle":"A finite causal set hosts a full interacting scalar theory, with no renormalization needed at the discrete level.","key_machinery":"The machinery is finite-dimensional linear algebra carried by the causal order. The retarded Green operator $E_+ = P^{-1}K$ is lower triangular because $P$ is retarded, so its transpose $E_-$ is advanced; everything else is built from their difference $E$. The paper's new discretised wave operator is defined through a preferred past structure: a choice $\\Lambda(p)$ of a rank-2 point in the past of $p$, giving $P_\\Lambda\\varphi_p = \\varphi_p - 2\\,\\mathrm{mean}_{\\Lambda(p)\\prec q\\prec p}\\varphi_q + \\varphi_{\\Lambda(p)}$ for points outside the two-layer past infinity, and $\\varphi_p$ on the boundary. On a regular diamond lattice in two-dimensional flat spacetime this has continuum limit $\\frac{1}{2}\\square$ and retarded Green function $\\frac{1}{2}(1+C)$, where $C$ is the causal chain matrix, and the paper proves this by direct matrix identities. The quantum deformation is carried by the star product $F \\star G = m \\circ \\exp(\\frac{1}{2}i\\hbar\\, \\mathcal{D}_E)(F\\otimes G)$, the normally ordered product with covariance $W$, the time-ordered product with covariance $\\Delta_F$, and the retarded quantum map that turns the free algebra into the interacting one.","core_discovery":"The central claim is that a single finite causal set $C$ of $N$ points can carry a full interacting quantum field theory. The field is a real function on $C$; the configuration space is $\\mathbb{R}^N$. Choosing a lower-triangular discretised wave operator $P$ and a source-averaging operator $K$ gives a retarded Green matrix $E_+ = P^{-1}K$, with advanced Green matrix $E_- = E_+^T$. Their antisymmetric difference $E = E_- - E_+$ defines the classical bracket $F_i E_{ij} G_j$ and, after deformation quantization, the quantum commutation relations. Since $E P$ is not zero on a causal set, the paper does not impose the equations of motion directly; instead it quotients the algebra by functionals whose functional derivatives are annihilated by $E$, noting that this is the causal-set replacement for going on shell and is only approximate. On this base it defines a two-point function $W = \\frac{1}{2}(iE + \\sqrt{-E^2})$, a propagator $\\Delta_F = \\frac{1}{2}(E_+ + E_-) + H$, the time-ordered product $F \\cdot_T G = m \\circ \\exp(\\hbar/2\\, \\mathcal{D}_{\\Delta_F})(F\\otimes G)$, the formal S-matrix $S(\\lambda V) = \\exp(i\\lambda V/\\hbar)_T$, and the retarded quantum map $R_{\\lambda V}(F) = S(\\lambda V)^{-1} \\star_H\\bigl(S(\\lambda V) \\cdot_T F\\bigr)$. The interacting algebra $A_{\\mathrm{int}}(C)$ is then obtained by deforming the free product; $n$-point functions of interacting fields are explicit formal power series in $\\hbar$ and $\\lambda$. This is the construction the paper claims as the first of its breadth in quantum field theory on causal sets.","pith_inferences":["The paper leaves open whether the quotient by the kernel of $E$ becomes a faithful on-shell condition for dense sprinklings; an editor's test is to compute the spectrum of $E$ on growing causal sets and check that the kernel collapses to the image of the wave operator in the continuum limit.","Averaging the preferred-past operator over all admissible choices of $\\Lambda$ would remove the choice of preferred past; this could yield a dimension-agnostic discrete wave operator, a program the paper names but does not carry out.","Because the propagator $\\Delta_F$ depends on the chosen covariance $H$, the scheme actually defines a family of interacting theories; selecting $H$ so that the continuum limit is Hadamard, as the paper suggests for the free state, is a reasonable criterion for choosing among them.","A numerical check on small sprinklings — computing interacting $n$-point functions for a $\\varphi^4$-type local interaction and comparing with continuum perturbation theory on a lattice — would show whether the formal construction has the expected physical limit."],"forward_implications":["Every fixed finite causal set now carries a definite family of interacting scalar field theories, one for each smooth interaction functional, with explicit algebraic definitions and no renormalization step.","Correlation functions of interacting fields are computable as formal power series, and the paper's graphical expansions give explicit Feynman-like diagram rules on finite causal sets.","On a regular diamond lattice the preferred-past wave operator and its Green function reproduce the continuum retarded Green function as the spacing goes to zero, so the free theory has a concrete continuum-limit check.","The distinguished state is pure, and because the configuration space is finite, all sufficiently regular representations of the algebra are unitarily equivalent; changing the auxiliary inner product produces alternative states, including precursors of Hadamard states in the continuum.","The relative Cauchy evolution provides a way to measure how observables respond when the background causal set is altered, a step toward connecting discrete geometry to the dynamics of observables."],"supporting_citations":[{"why":"Defines the retarded and advanced responses whose difference gives the classical bracket used throughout the paper.","marker":"[15]"},{"why":"Provides the original causal-set d'Alembertian and the layered past/future infinity used for boundary data and the retarded Green operator.","marker":"[1]"},{"why":"Supplies the retarded Green function starting point and the observation that causal-set dynamics hold only approximately, which motivates the quotient by the kernel of $E$.","marker":"[16]"},{"why":"Gives the classical retarded map framework used to introduce interactions in the classical theory.","marker":"[12]"},{"why":"Contains the graph expansions for the interacting star product and the retarded quantum map that the paper transplants to causal sets.","marker":"[37]"},{"why":"Supplies the deformation quantization and state formalism, including the normally ordered product and the state construction that the quantum half relies on.","marker":"[14]"},{"why":"Defines the distinguished pure state that the paper uses as the two-point function on the free theory.","marker":"[27]"},{"why":"Shows that altering the auxiliary inner product yields softened versions of the distinguished state with good continuum properties, informing the paper's treatment of state choices.","marker":"[30]"}],"fun_headline_variants":["Interacting QFT on causal sets: first broad construction","Causal sets now support full interacting quantum fields","Finite causal sets host interacting scalar theory","First interacting quantum field theory on causal sets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that quotienting the algebra by functionals whose derivative is annihilated by the antisymmetric Green matrix $E$ is an adequate way to impose the field equation, even though $E$ does not invert the wave operator on a causal set; if that replacement misrepresents the dynamics, the free and interacting quantum theories built on it inherit the error.","fun_headline_variants_meta":{"raw":{"variants":["Interacting QFT on causal sets: first broad construction","Causal sets now support full interacting quantum fields","Finite causal sets host interacting scalar theory","First interacting quantum field theory on causal sets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000448,"raw_usage":{"total_tokens":2342,"prompt_tokens":1106,"completion_tokens":1236,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":722,"completion_tokens_details":{"reasoning_tokens":1177}},"tokens_in":722,"tokens_out":1236,"duration_ms":9773,"temperature":1.0,"reasoning_tokens":1177,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:58:32.746096+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"On a regular diamond lattice approximating two-dimensional flat spacetime, take the preferred-past operator $P_\\Lambda$ and a compactly supported smooth source $f$, compute $E_{+,\\Lambda}f$ with the paper's scaling, and compare pointwise with the continuum retarded solution; if the difference does not vanish as the lattice spacing tends to zero, the proposed Green function fails its continuum-limit claim. A second test: on a small random causal set, find a source $f$ for which $E P f \\neq 0$; because the equation-of-motion ideal is generated by functionals of the form $\\Phi_{Pf}$, such an $f$ shows directly that the quotient by the kernel of $E$ is not the same as imposing the wave equation.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the retarded and advanced responses whose difference gives the classical bracket used throughout the paper."},{"cited_title":"The starting-point is therefore a suitable discretization of the continuum field equation □φ =f (32) to a causal set","cited_arxiv_id":null,"evidence_quote":"Provides the original causal-set d'Alembertian and the layered past/future infinity used for boundary data and the retarded Green operator."},{"cited_title":"Henson, The causal set approach to quantum gravity, inApproaches to Quantum Gravity: Toward a New Understanding of Space, Time and Matter, edited by D","cited_arxiv_id":null,"evidence_quote":"Supplies the retarded Green function starting point and the observation that causal-set dynamics hold only approximately, which motivates the quotient by the kernel of $E$."},{"cited_title":"We use the framework of perturbative AQFT [11, 14, 51], where the interacting fields are constructed with the use of quantum Møller operators","cited_arxiv_id":null,"evidence_quote":"Gives the classical retarded map framework used to introduce interactions in the classical theory."},{"cited_title":"Cortês and L","cited_arxiv_id":null,"evidence_quote":"Contains the graph expansions for the interacting star product and the retarded quantum map that the paper transplants to causal sets."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the deformation quantization and state formalism, including the normally ordered product and the state construction that the quantum half relies on."},{"cited_title":"Rejzner, Perturbative Algebraic Quantum Field Theory","cited_arxiv_id":null,"evidence_quote":"Defines the distinguished pure state that the paper uses as the two-point function on the free theory."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that altering the auxiliary inner product yields softened versions of the distinguished state with good continuum properties, informing the paper's treatment of state choices."}],"review_version":1}