{"id":"61592b8d-2ac1-4b47-80b3-cd1af5ecaed9","arxiv_id":"2608.05197","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A polynomial probability functional on complex amplitudes that is phase invariant, phase-averaged additive, and normalized must equal the squared modulus, the Born rule.","lead":"The paper proves that if a probability rule on complex path amplitudes stays the same when all phases are rotated and adds up correctly when paths are averaged over random relative phases, the only possible rule is the square of the amplitude's size. It is a local uniqueness theorem for the Born rule in a categorical framework with no Hilbert space assumed.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The derivation is internally sound, but the theorem's content is carried almost entirely by Axiom (iv), a uniform-phase averaging identity whose physical realization is deferred to an unpublished companion; this limits the result to a conditional characterization rather than an independent…","rationale":"I read the proof line by line. Steps 1 and 2 are standard: U(1) invariance forces a radial polynomial; the Fourier moments (3.3) are correct. Step 3's elimination of higher-degree coefficients is valid: the non-separable monomial r1^{2(m-1)}r2^2 appears only in the degree-2m homogeneous component of the m-th moment, so no cancellation from other coefficients is possible; the text could have made the homogeneity argument explicit, but the conclusion follows. Normalization fixes c1=1. I find no algebraic error. The load-bearing concern is therefore not internal consistency but the status of Axiom (iv). The identity is an averaging property that the Born rule satisfies by construction; the theorem proves uniqueness within the class of polynomials satisfying it, which is a characterization rather than an independent derivation. The paper honestly flags the deferred physical realization. I agree with the reader's weakest_assumption. The verdict ACCEPT with high confidence is appropriate because the theorem is precisely stated (modulo the wording issue on θ) and the limitations are acknowledged; I would only ask the authors to clarify in (iv) that θ is an independent uniform random variable. Thus no change to the verdict.","tokens_in":10335,"tokens_out":33107,"duration_ms":300216,"concrete_test":"In the companion categorical-refinement model (Zenodo 10.5281/zenodo.21443525), construct the two-slit category with two mutually exclusive paths whose morphism weights are w(f)=exp(i s(f)/κ) as specified in Remark 3.4', and compute the relative-phase distribution that the model actually assigns to such path pairs. Then evaluate whether E[|A1+e^{iθ}A2|^2] = |A1|^2+|A2|^2 holds for that distribution with P=|z|^2. If the model's phases are not uniform (or the required cosine moment does not vanish), Axiom (iv) fails in the motivating framework, and the theorem's physical relevance is unsupported. A second check: re-run the proof with axiom (iv) interpreted literally (θ fixed to arg(A2)-arg(A1)); |z|^2 will fail, confirming that the theorem requires the independent-uniform-phase reading.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a valid conditional theorem, but the weight of the argument rests on Axiom (iv): E[P(A1+e^{iθ}A2)] = P(A1)+P(A2). Step 3 of the proof eliminates every c_k for k≥2 solely through the non-separable cross-terms of the moments (3.3), and those cross-terms vanish under uniform phase averaging precisely because the Born rule is quadratic in the amplitude. Thus (iv) is not a classical-limit additivity condition in the usual sense (e.g., ℏ→0 with definite phases); it is an ensemble randomization that already encodes the absence of interference. The categorical mutual-exclusivity formalism and the existence of an admissible category make the axiom applicable to arbitrary magnitudes, but they do not reduce its strength; any P satisfying (iv) must have the interference-averaging behaviour of |z|^2. The paper itself, in Section 6.2(b), concedes that the physical realization of the randomized-phase hypothesis is not established and refers to a companion Zenodo work. In addition, the formal statement of (iv) defines θ as the fixed relative phase arg(A2)-arg(A1) and then averages over it; under that literal reading |z|^2 itself fails the identity, so the theorem depends on the charitable reading (made explicit in Remark 3.4') that θ is an independent uniform phase. Neither point invalidates the algebra, but together they show the result is a characterization under a strong, not-yet-grounded postulate.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proves a conditional uniqueness theorem for the Born rule in the setting of small categories with complex-valued morphism weights. The author defines path amplitudes as products of weights, gives a categorical definition of mutually exclusive paths, and shows that any probability functional P on amplitudes satisfying (i) non-negativity, (ii) polynomiality, (iii) global U(1) invariance, (iv) a uniform-phase additivity identity, and (v) normalization must equal P(z)=|z|^2. The proof proceeds in four steps: U(1) invariance reduces P to a polynomial in |z|^2; binomial and Fourier moment identities compute the phase averages; a degree analysis forces all terms of degree at least 4 to vanish; normalization fixes the remaining coefficient. The paper also compares the result with Gleason's theorem, Hardy's reconstruction, Chiribella–D'Ariano–Perinotti, and other recent derivations, and formulates the global Chapman–Kolmogorov coherence problem for the family of local functionals.","tokens_in":10709,"tokens_out":12638,"duration_ms":117202,"significance":"The algebraic core of the paper is correct, and the theorem is a clean local characterization: within the stated axioms, no Hilbert space is needed, and the proof is elementary and fully explicit. The paper is unusually honest about its scope, explicitly acknowledging in Section 6.2 that the physical realization of the randomized-phase hypothesis is open and deferred to a companion work. My main reservation concerns interpretation: axiom (iv) is not a classical limit in the usual sense but a uniform-phase averaging condition that the Born rule itself satisfies, so the result is best read as a characterization under a strong averaging postulate rather than as an independent derivation from physically unproblematic principles. With that framing, the result is a useful contribution to the Born-rule reconstruction literature, and the explicit statement of the global coherence problem is a valuable opening for further work.","major_comments":[{"comment":"The formal statement of axiom (iv) is internally inconsistent as written: it defines theta as the deterministic relative phase arg(A2)-arg(A1) and then asserts that this quantity is drawn from U[0,2pi]. Under the literal reading, the theorem is false even for the claimed conclusion, since for A1=A2=1 the identity would require E[P(2)]=4 to equal P(1)+P(1)=2. The proof and Remark 3.4' use the charitable reading of theta as an independent uniform phase; the theorem must be restated with that reading made explicit.","section":"Theorem 3.1, Eq. (3.1)"},{"comment":"The term 'admissible' appears in the theorem statement before it is defined, and Remark 3.5 gives only an informal two-disjoint-cycles sketch rather than a construction proving that arbitrary independent magnitudes r1,r2>0 are realizable as total amplitudes of mutually exclusive paths. Since admissibility is part of the hypothesis and controls the quantification over r1,r2 in Step 3, the definition and an explicit existence argument (or a precise reference to the companion work) should be moved into Section 3.","section":"Theorem 3.1 / Remark 3.5"},{"comment":"The degree-by-degree elimination of c_k for k>=2 is stated as an inspection of Eq. (3.3), but the equation being analyzed couples all coefficients c_k through the sum over k. The argument should be written as an induction on the highest remaining degree, or as comparison of homogeneous components of the polynomial identity in r1,r2, since only then is it clear that the non-separable monomial r1^{2(m-1)} r2^2 cannot be cancelled by contributions from other k.","section":"Proof of Theorem 3.1, Step 3"}],"minor_comments":[{"comment":"The sentence 'Axiom (iv) is satisfied in any category where mutually exclusive path pairs exist with independent morphism weights' is too quick, because the identity constrains the functional P rather than the category alone; the wording may confuse the condition on (C,w) with the condition on P.","section":"Remark 3.4'"},{"comment":"The notation switches between k (the index in P) and m (the exponent in Eq. (3.3)); using a single index throughout would avoid the appearance that the m-th moment is being treated independently of the sum over k.","section":"Proof of Theorem 3.1, Step 3"},{"comment":"The claim that 'any choice of complex weights with nontrivial relative phase' produces a nonzero interference cross-term should specify nonzero path amplitudes; if one of the two intermediate amplitudes is zero, the cross-term vanishes trivially.","section":"Example 5.2"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the paper is likely acceptable after a revision that fixes the statement of axiom (iv), moves the admissibility definition into the main body with an explicit existence argument, and tightens the rigor of Step 3. The reliance on an unpublished Zenodo companion for the physical realization of the randomized-phase hypothesis is a scope concern, but it is disclosed and acknowledged by the author in Section 6.2. I do not see a need for additional external review."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The four-step proof of Theorem 3.1 is correct, and the categorical framing is doing real work, not just providing jargon. The main caveat is that Axiom (iv) carries nearly all the physical content: it is a uniform-phase averaging identity that the Born rule itself satisfies, so the theorem is best read as a characterization under a strong postulate, not an independent derivation. That still has value, and the paper deserves a serious referee.\n\nWhat is genuinely good: the algebraic core is clean. U(1) invariance reduces P to a polynomial in |z|^2, the uniform-phase moments are computed correctly, and the non-separability argument in Step 3—showing that cross-terms like r1^{2(m-1)} r2^2 cannot be cancelled by any separable sum—is sound. The paper is also unusually honest. Section 5 openly states the global coherence problem and gives a concrete three-object example where Chapman–Kolmogorov fails, which is real evidence of clear thinking. The mutual-exclusivity definition via absence of shared factorization is a reasonable categorical formalization, and the comparison with Gleason, Hardy, and Caticha is fair.\n\nThe soft spots, in order. First, Axiom (iv) is the whole show. The identity E[P(A1 + e^{i\\theta} A2)] = P(A1)+P(A2) with uniform theta is exactly what a quadratic functional produces because all interference cross-terms vanish under phase averaging. Calling it 'classical-limit additivity' is a bit misleading; it is an ensemble randomization, not a semiclassical limit with definite phases. The physical realization is deferred to a companion Zenodo work, so the theorem is conditional: if a particular weighted category realizes this averaging, then |z|^2 is the unique local law. Don't oversell it as Born rule from phase invariance alone. Second, the literal statement of (iv) defines theta as arg(A2)-arg(A1) and then averages over it; under that fixed-relative-phase reading, |z|^2 itself fails the identity. The intended reading—theta as an independent uniform phase—appears only in Remark 3.4'. The main theorem needs a rewrite of Axiom (iv) to remove the ambiguity. That is a genuine bug in presentation, not a hidden flaw in the argument. Third, admissibility is sketched informally rather than formalized; Remark 3.5 gestures at disjoint cycles but does not prove the category exists. Minor, but worth tightening.\n\nWho is this for? People working on categorical derivations of quantum probability, or on Born-rule reconstructions generally. It will not end the debate, but it is a precise local result with a valid proof and honest limitations. I would send it to peer review, asking for revision of Axiom (iv) and a slightly deeper engagement with Caticha's consistency derivation, which is the closest antecedent.","headline":"A clean local Born-rule uniqueness theorem whose proof is sound but whose work is done by a phase-averaging axiom that already encodes |z|^2; worth refereeing after a wording fix.","tokens_in":11131,"tokens_out":2961,"would_cite":true,"duration_ms":29035,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The Born rule is the unique local probability law on complex amplitudes.","keywords":["Born rule","categorical quantum mechanics","complex-weighted categories","path amplitudes","U(1) invariance","classical limit","mutually exclusive paths","Markov categories"],"falsifier":"Compute the left and right sides of the averaging axiom for the one-parameter family $P(z) = |z|^2 + \\varepsilon |z|^4$ with $A_1 = A_2 = 1$: the left side equals $2 + 6\\varepsilon$ while the right side equals $2 + 2\\varepsilon$, so the identity fails by $4\\varepsilon$ for any $\\varepsilon \\neq 0$. The theorem is confirmed insofar as no perturbation of the same kind passes the test; it would be refuted by exhibiting a polynomial or real-analytic functional satisfying all five axioms that is not $|z|^2$.","tokens_in":10143,"feed_emoji":"⚛️","tokens_out":8768,"duration_ms":75859,"temperature":0.7,"pith_summary":"This paper proves that, in a setting where physical processes are morphisms of a small category carrying complex weights and probabilities are assigned to path amplitudes, the only possible probability functional satisfying five local axioms is the Born rule, $P(z) = |z|^2$. The axioms are non-negativity, polynomiality, global phase invariance, classical-limit additivity over mutually exclusive paths, and normalization. The main theorem is local: it fixes the probability law on a single collection of paths between two objects, without assuming a Hilbert space or deriving the full measurement structure. If the result stands, the Born rule is not an additional postulate in path-amplitude frameworks but a consequence of phase symmetry and a classical-limit averaging condition. The paper leaves open how these local laws compose across morphisms, and pins that gap to an explicit open problem.","feed_headline":"Five axioms force probability to square the amplitude","feed_subtitle":"No Hilbert space needed: phase symmetry plus classical-limit averaging single out $|z|^2$.","key_machinery":"The central object is the averaging identity of axiom (iv), $\\mathbb{E}_\\theta[P(A_1 + e^{i\\theta}A_2)] = P(A_1) + P(A_2)$, applied to pairs of mutually exclusive paths, where mutual exclusivity means the two paths share no common interior factorization through a non-identity morphism. The identity is combined with global $U(1)$ invariance to reduce $P$ to a polynomial in $|z|^2$, and its content is exhibited by the Fourier moments $\\mathbb{E}[\\cos^j\\theta] = 0$ for odd $j$ and $\\mathbb{E}[\\cos^{2j}\\theta] = \\binom{2j}{j}/2^{2j}$ for even $j$. These moments make the degree-$2k$ contributions produce non-separable monomials such as $r_1^{2m-2}r_2^2$, which cannot be reproduced by any function of the form $f(r_1) + f(r_2)$, so all coefficients $c_k$ with $k \\geq 2$ vanish. The same machinery turns the global composition question into a classification problem about which weight structures make interference cross-terms vanish.","core_discovery":"Theorem 3.1 establishes that any probability functional $P$ on complex amplitudes, assumed to be a polynomial of bounded total degree in $z$ and $\\bar z$, globally $U(1)$-invariant, non-negative, normalized, and satisfying the expectation identity $\\mathbb{E}_\\theta[P(A_1 + e^{i\\theta}A_2)] = P(A_1) + P(A_2)$ for mutually exclusive path pairs with $\\theta$ drawn uniformly from $[0,2\\pi]$, must equal $|z|^2$. The proof reduces $P$ to a polynomial in $|z|^2$ using phase invariance, computes the uniform-phase averages of powers of $|A_1 + e^{i\\theta}A_2|^2$ explicitly, and shows that every term of degree $2k$ with $k \\neq 1$ generates a non-separable cross-term in the magnitudes $r_1, r_2$ that cannot be canceled by the separable right-hand side. Only the quadratic term survives; normalization sets its coefficient to one.","pith_inferences":["A natural testable extension is to check whether the uniform-phase condition can be replaced by a deterministic phase-decorrelation property of the weights; the theorem suggests any category with effectively incommensurate phases between exclusive paths will satisfy the averaging identity.","If the coherence classification has a positive answer for weight structures factoring through a commutative monoid, the combined local-global result would supply a Hilbert-free derivation of quantum probability as a Markov category, placing interference under explicit structural control.","The separability argument is independent of the particular coefficients, so it hints that the same elimination works for any $U(1)$-invariant functional, analytic or not, once a suitable phase-averaging identity is imposed; weakening analyticity is the paper's own open direction."],"forward_implications":["Any complex-weighted small category that can host mutually exclusive path pairs with arbitrary independent magnitudes inherits the Born rule locally, with no Hilbert-space or projection-lattice input.","The polynomiality axiom is a convenience: the proof also works for real-analytic $U(1)$-invariant functionals, so the uniqueness conclusion extends beyond polynomials.","The local theorem does not make probabilities compose: the paper gives a four-object category where interference cross-terms make the Chapman–Kolmogorov identity fail, so only a restricted class of weight structures can be globally coherent.","The global coherence problem becomes a precise classification question: characterize the complex-weighted categories for which the family of local Born rules extends to a Markov-category-compatible probability assignment."],"supporting_citations":[{"why":"Defines the path-amplitude structure (products of weights along composable sequences) that the theorem's amplitudes are built from.","marker":"[6]"},{"why":"Supplies the small-category morphism setting and the categorical quantum mechanics program the theorem operates within.","marker":"[1, 2]"},{"why":"Gives the Markov-category and Chapman–Kolmogorov formalism that frames the open global coherence problem.","marker":"[5, 7, 12]"},{"why":"Provides the Hilbert-space measure-theoretic result whose projection-lattice assumptions the theorem strips away.","marker":"[8]"}],"fun_headline_variants":["Five axioms force the Born rule without Hilbert space","Phase invariance plus classical limit uniquely yields the Born rule","Born rule unique from five categorical axioms, no Hilbert space","Uniqueness of Born rule from phase symmetry and classical limit","Five assumptions single out the quantum probability rule"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rises or falls on the assumption that for any two mutually exclusive paths, averaging over a uniformly random relative phase makes the probability functional additive, and that the category can realize such path pairs with arbitrary independent magnitudes.","fun_headline_variants_meta":{"raw":{"variants":["Five axioms force the Born rule without Hilbert space","Phase invariance plus classical limit uniquely yields the Born rule","Born rule unique from five categorical axioms, no Hilbert space","Uniqueness of Born rule from phase symmetry and classical limit","Five assumptions single out the quantum probability rule"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000806,"raw_usage":{"total_tokens":3542,"prompt_tokens":953,"completion_tokens":2589,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":569,"completion_tokens_details":{"reasoning_tokens":2513}},"tokens_in":569,"tokens_out":2589,"duration_ms":15949,"temperature":1.0,"reasoning_tokens":2513,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:47:48.564196+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the left and right sides of the averaging axiom for the one-parameter family $P(z) = |z|^2 + \\varepsilon |z|^4$ with $A_1 = A_2 = 1$: the left side equals $2 + 6\\varepsilon$ while the right side equals $2 + 2\\varepsilon$, so the identity fails by $4\\varepsilon$ for any $\\varepsilon \\neq 0$. The theorem is confirmed insofar as no perturbation of the same kind passes the test; it would be refuted by exhibiting a polynomial or real-analytic functional satisfying all five axioms that is not $|z|^2$.","supporting_citations":[{"cited_title":"Space-time approach to non-relativistic quantum mechanics,","cited_arxiv_id":null,"evidence_quote":"Defines the path-amplitude structure (products of weights along composable sequences) that the theorem's amplitudes are built from."},{"cited_title":"Measures on the closed subspaces of a Hilbert space,","cited_arxiv_id":null,"evidence_quote":"Provides the Hilbert-space measure-theoretic result whose projection-lattice assumptions the theorem strips away."}],"review_version":2}