REVIEW 3 major objections 2 minor
Uniqueness of the Canonical Reciprocal Cost
T0 review · 3 major / 2 minor · reviewed 2026-08-29 · grok-4.5
Pith's one-line read A d'Alembert composition law plus one quadratic calibration uniquely fix the canonical reciprocal cost as the AM–GM gap of a positive ratio and its reciprocal.
desk verdict Clean uniqueness packaging for the AM-GM reciprocal cost via d’Alembert; classical route, limited scope, still worth a serious referee. 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 change of variables $H(t)=F(e^t)+1$ converts the multiplicative composition law on ratios into d'Alembert's functional equation $H(t+s)+H(t-s)=2H(t)H(s)$ on the additive group $\mathbb{R}$. Classical continuous solutions of that equation are hyperbolic cosine, ordinary cosine, or linear; the quadratic calibration at the origin eliminates the oscillatory and linear branches and pins the curvature of the hyperbolic-cosine solution, recovering $F$ as the AM–GM gap.
What would settle it
Exhibit a nonnegative $F$ on positive reals that satisfies the stated composition law and the quadratic calibration yet differs from $(x+1/x)/2-1$ at even one point $x\neq 1$, or produce a continuous one-parameter family that still meets both assumptions.
Extended reading notes
Core claim
Any function $F:\mathbb{R}_{>0}\to\mathbb{R}_{\ge 0}$ that obeys a d'Alembert-type composition law on positive reals and a single quadratic calibration at the identity in logarithmic coordinates is uniquely the canonical reciprocal cost $$F(x)=\frac{x+1/x}{2}-1,$$ equivalently the arithmetic–geometric-mean gap of $x$ and $1/x$. In the logarithmic coordinates $H(t)=F(e^t)+1$ the composition law becomes d'Alembert's functional equation on $\mathbb{R}$; the calibration supplies the minimal regularity that selects the continuous hyperbolic-cosine branch and fixes the remaining scaling constant.
Load-bearing premise
The uniqueness proof leans on whatever minimal regularity (continuity, measurability, or local boundedness) is needed to invoke the classical continuous classification of d'Alembert solutions; without that gate the composition law alone admits wild non-measurable solutions.
Editorial extensions
If this is right
- Any ratio-deviation penalty that is required to compose under multiplication of ratios and to open quadratically at equilibrium must be exactly the AM–GM gap of the ratio and its reciprocal.
- Dropping either the composition law or the calibration immediately destroys uniqueness, so both axioms are necessary as well as sufficient.
- Approximate solutions that obey the composition law only up to a bounded defect remain quantitatively close to the canonical cost.
- The same uniqueness supplies a canonical, parameter-free cost for any modelling setting whose natural variable is a positive ratio.
Reading between the lines
- The result supplies a first-principles justification for preferring the hyperbolic cosine (or cosh-minus-one) over other common ratio penalties whenever multiplicative composition is required.
- The stability estimate suggests that numerical or empirical approximate costs that nearly obey the composition law can be projected onto the canonical form with controlled error.
- Because the argument is purely functional-equation theoretic, the same uniqueness should transfer to any domain whose positive elements form a multiplicative group isomorphic to $(\mathbb{R}_{>0},\cdot)$.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript studies rigidity for nonnegative functions F on the positive reals that penalize deviation of a positive ratio from the equilibrium x=1. Under a d'Alembert-type composition law on R>0 and a single quadratic calibration at the identity in logarithmic coordinates, it claims F is uniquely the canonical reciprocal cost F(x)=(x+1/x)/2−1 (the AM–GM gap of x and 1/x). The argument passes to H(t)=F(e^t)+1, converting the composition law into d'Alembert’s equation on R; the calibration supplies the regularity needed for the classical continuous classification and selects the hyperbolic-cosine branch among continuous solutions. Necessity of each hypothesis is asserted (one-parameter continuous family without calibration; failure of global determination without composition; pathological non-measurable solutions without regularity), together with a stability estimate under bounded defect and some further properties of the canonical cost.
Significance. If the uniqueness theorem holds under explicitly stated and essentially minimal regularity, the paper supplies a clean axiomatic characterization of a standard reciprocal/AM–GM cost via a classical functional equation plus a local quadratic calibration. That is a useful rigidity result for anyone employing such costs in analysis or applied mathematics: the composition law forces F(1)=0, continuous (or measurable/locally bounded) d'Alembert solutions are known, and a second-order condition at the identity kills scaling and the oscillatory/linear branches. The necessity and stability addenda, if correctly proved, strengthen the claim beyond a pure existence-uniqueness statement. Credit is due for framing the problem so that the target form is selected by classification plus calibration rather than built into the axioms by definition.
major comments (3)
- [Abstract (composition law and calibration)] The full text is unavailable, so the precise statement of the d'Alembert-type composition law on R>0, the exact quadratic calibration in logarithmic coordinates, and the regularity hypothesis actually used cannot be audited. The uniqueness claim stands or falls on those details: without a clearly stated continuity, measurability, or local-boundedness gate, the composition law admits pathological solutions, as the abstract itself notes. The central claim is therefore not yet verifiable from the manuscript as supplied.
- [Abstract (stability estimate)] The stability estimate for approximate solutions under bounded defect is asserted but not quantified in the abstract. Load-bearing constants, the precise defect norm, and the sense in which approximate solutions are close to the canonical cost must appear explicitly; without them the stability claim cannot support the rigidity narrative.
- [Abstract (necessity)] Necessity of each assumption is claimed (continuous one-parameter family without calibration; non-uniqueness without composition; non-measurable pathologies without regularity). Each necessity direction needs a concrete counter-example or explicit family in the body of the paper; an abstract assertion alone does not establish that the hypotheses are sharp.
minor comments (2)
- [Abstract] The abstract uses both “canonical reciprocal cost” and the AM–GM description; the published version should fix a single primary name and state the equivalent formulae (including H(t)=cosh t or the normalized cosh branch) in one display for ease of reference.
- Once the full text is supplied, the classical references for the continuous/measurable classification of d'Alembert solutions on R should be cited at the point the classification is invoked, so the regularity gate is traceable.
Simulated Author's Rebuttal
We thank the referee for a careful reading of the abstract and for framing the contribution accurately. The present review is abstract-only: the full manuscript text was not available to the referee. Consequently several load-bearing details (precise statement of the composition law, the exact calibration, the regularity gate, explicit counter-examples, and the quantified stability estimate) cannot be audited from the abstract alone. Below we answer each major comment as far as the abstract and the intended theorem statements permit, and we record honestly what must await the full text. We will ensure that the next version states hypotheses, counter-examples, and constants with full explicitness so that the uniqueness, necessity, and stability claims are directly verifiable.
read point-by-point responses
-
Referee: Abstract (composition law and calibration): The full text is unavailable, so the precise statement of the d'Alembert-type composition law on R>0, the exact quadratic calibration in logarithmic coordinates, and the regularity hypothesis actually used cannot be audited. The uniqueness claim stands or falls on those details: without a clearly stated continuity, measurability, or local-boundedness gate, the composition law admits pathological solutions, as the abstract itself notes. The central claim is therefore not yet verifiable from the manuscript as supplied.
Authors: We agree that the uniqueness theorem is only as strong as the precise hypotheses, and that an abstract-only review cannot audit them. The intended statements are as follows. The composition law is the d'Alembert-type identity on ratios: F(xy)+F(x/y)=2F(x)F(y)+2F(x)+2F(y) (equivalently, after H(t)=F(e^t)+1, the standard d'Alembert equation H(t+s)+H(t-s)=2H(t)H(s) on R). The quadratic calibration is the second-order condition H(t)=1+t^2/2+o(t^2) as t o0 (i.e., F(e^t)=t^2/2+o(t^2)). Regularity is continuity of F (or of H), which is the classical gate that yields H(t)=cosh(ct) or the trigonometric/linear branches; the calibration forces c=1 and selects the cosh branch, giving F(x)=(x+1/x)/2-1. Pathological solutions are excluded precisely by this regularity, as the abstract already notes. Because the referee could not see the body, we will make every hypothesis, the reduction to d'Alembert, and the invocation of the continuous classification fully explicit in the revised manuscript so the claim is directly verifiable. revision: yes
-
Referee: Abstract (stability estimate): The stability estimate for approximate solutions under bounded defect is asserted but not quantified in the abstract. Load-bearing constants, the precise defect norm, and the sense in which approximate solutions are close to the canonical cost must appear explicitly; without them the stability claim cannot support the rigidity narrative.
Authors: The referee is correct that the abstract only asserts existence of a stability estimate and does not display constants, the defect norm, or the topology of closeness. In the manuscript the intended result is a Hyers–Ulam-type bound: if a nonnegative F satisfies the composition identity up to a uniformly bounded defect and satisfies the same quadratic calibration at the identity, then F remains within an explicit multiple of that defect (in the sup norm on compact log-intervals, or an equivalent weighted norm on R>0) of the canonical reciprocal cost. Load-bearing constants will be written out from the stability theory for d'Alembert’s equation under the calibration. We will add the precise statement, the defect norm, and the constants to the abstract or to a prominently placed theorem so the stability claim can support the rigidity narrative without requiring the reader to reconstruct them. revision: yes
-
Referee: Abstract (necessity): Necessity of each assumption is claimed (continuous one-parameter family without calibration; non-uniqueness without composition; non-measurable pathologies without regularity). Each necessity direction needs a concrete counter-example or explicit family in the body of the paper; an abstract assertion alone does not establish that the hypotheses are sharp.
Authors: We agree that sharpness requires concrete counter-examples, not merely an abstract assertion. The three directions are intended to be witnessed as follows. (1) Without calibration: continuous solutions of the composition law are F_c(x)=(x^c+x^{-c})/2-1 for c>0 (the one-parameter family), all of which satisfy F(1)=0 and nonnegativity; only c=1 meets the quadratic calibration. (2) Without the composition law: many functions satisfy the local quadratic calibration (e.g., F(x)=(log x)^2/2, or compactly supported smooth bumps added away from 1 in log scale) but fail to be the global reciprocal cost. (3) Without regularity: using a Hamel basis for R over Q one obtains wild solutions of d'Alembert’s equation that are non-measurable and unbounded on every interval; the corresponding F are pathological and nonnegative only after additional truncation arguments that still violate any reasonable regularity. These families will be written explicitly in the body (and briefly signposted in the abstract) so that necessity is demonstrated rather than merely claimed. revision: yes
- Full manuscript text was unavailable to the referee; uniqueness, quantified stability constants, and the explicit necessity counter-examples therefore could not be audited from the supplied abstract alone. Those items can be confirmed only once the full text is provided.
Circularity Check
No significant circularity: uniqueness follows from classical d'Alembert classification plus local calibration, not from defining the target into the axioms.
full rationale
The abstract-only record presents a standard rigidity argument: a d'Alembert-type composition law on R>0 is transported via H(t)=F(e^t)+1 to d'Alembert's equation on R; classical continuous (or measurable/locally bounded) solutions are cos, cosh, or linear; a single quadratic calibration at the identity kills the free scale and selects the cosh branch, yielding F(x)=(x+1/x)/2−1. Composition forces F(1)=0 by the equation itself; the calibration is an independent local second-order condition, not a global fit of the target form; the cosh solution class is external classical mathematics rather than an author-fitted ansatz or a self-cited uniqueness theorem that already assumes the result. Necessity and stability add-ons are likewise standard for this equation and do not redefine the conclusion as an input. No quoted step reduces a claimed prediction to a fitted parameter, a self-definition, or a load-bearing unverified self-citation. Residual risk is only the precise regularity gate (acknowledged in the abstract), which is a correctness/assumption issue, not circularity. Score 0 with empty steps is the honest finding.
Assumptions & free parameters
assumptions (4)
- ad hoc to paper d'Alembert-type composition law for F on R>0 (implies F(1)=0)
- ad hoc to paper Single quadratic calibration of F at the identity in logarithmic coordinates
- standard math Classical classification of continuous (or similarly regular) solutions of d'Alembert's equation on R
- domain assumption Minimal regularity (continuity, measurability, or local boundedness) sufficient to exclude pathological solutions
Cite this review
Pith. "Pith review of Uniqueness of the Canonical Reciprocal Cost." pith.science (2026). https://pith.science/paper/ETY4VLBO
@misc{pith2026260205753,
author = {Pith},
title = {Pith review of: Uniqueness of the Canonical Reciprocal Cost},
year = {2026},
howpublished = {\url{https://pith.science/paper/ETY4VLBO}},
note = {Machine review of arXiv:2602.05753}
}
abstract
We study a rigidity problem for functions \(F:\R_{>0}\to\R_{\ge 0}\) that penalize deviation of a positive ratio from equilibrium \(x=1\). Assuming (i) a d'Alembert-type composition law on \(\R_{>0}\), and (ii) a single quadratic calibration at the identity (in logarithmic coordinates), we prove that \(F\) is uniquely determined. The composition law implies the normalization $F(1)=0.$ The unique solution is called the canonical reciprocal cost, namely the difference between the arithmetic and geometric means of \(x\) and its reciprocal. Our proof uses the logarithmic coordinates \(H(t)=F(e^t)+1\), where the composition law becomes d'Alembert's functional equation on \(\R\). The calibration provides the minimal regularity needed to invoke the classical classification of continuous solutions and fixes the remaining scaling freedom, selecting the hyperbolic-cosine branch. We also establish necessity of each assumption: without calibration the composition law admits a continuous one-parameter family, without the composition law the calibration does not determine the global form, and without regularity the composition law admits pathological non-measurable solutions. Finally, we establish a stability estimate for approximate solutions under bounded defect and characterize some properties of the canonical cost.
Reviewed August 29, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.