{"id":"653fa0a9-87ff-4d8a-82e0-b35ef7cbf5ce","arxiv_id":"2505.16555","paper_version":2,"verdict":"REJECT","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Curl forces admit Darboux-style generalized potential representations, but the proposed conserved auxiliary Hamiltonian is not a constant of the original dynamics.","lead":"This paper classifies non-conservative, non-dissipative forces using a classical theorem about differential forms, writing any such force with two or three generalized potentials. It also claims that a constructed auxiliary Hamiltonian is a conserved quantity of the original motion, but that conservation claim does not survive differentiation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The auxiliary-Hamiltonian conservation claim fails at Eq. (2.49): the auxiliary momentum is evaluated on the original trajectory, so dH/dt is nonzero for genuine curl forces.","rationale":"I read the paper in good faith and separated the Darboux classification from the dynamical claim. The classification of curl forces via the work 1-form, the generalized potentials U, V, W, and the accessibility remarks are mostly standard consequences of Darboux's theorem and are not the main point of contention. The advertised novelty is the construction of a conservative auxiliary force whose Hamiltonian is conserved along the original curl-force motion. That claim is load-bearing and it fails. The derivation in Section 2.6 defines \\bar p and \\bar x through integrals of the original trajectory, but then uses these in a Hamiltonian that is only conserved for the auxiliary trajectory. Equation (2.49) incorrectly equates -∇U(\\bar x) with -∇U(x). This is not a subtle sign error: for any genuine curl force with V(x0) ≠ 1 and ∇U(x0) ≠ 0, the original and auxiliary accelerations differ immediately, so \\bar x(t) ≠ x(t) for small t > 0, and the derivative of H along the original trajectory does not vanish. The reader's strongest claim and weakest assumption identify exactly this point, and my independent reading agrees. Since the central advertised result is unsupported, the REJECT verdict stands; the paper may contain a correct exposition of classical material, but that alone does not justify acceptance as a new research result.","tokens_in":12674,"tokens_out":4659,"duration_ms":28989,"concrete_test":"Integrate Example 2.6 numerically: take F0/a^3 = 1, m = 1, x0 = (1, 1), v0 = (0, 1), and integrate x¨ = -(x y^2, x^3) from t = 0 to t = 1. Define \\bar p(t) = p0 - ∫_0^t ∇U(x(s)) ds and \\bar x(t) = x0 + v0 t - (1/m)∫_0^t∫_0^τ ∇U(x(ξ)) dξ dτ with U(x,y) = -1/x - 1/y. Then evaluate H(t) = |\\bar p(t)|^2/(2m) + U(\\bar x(t)) at several times. If H(t) changes by more than the integration tolerance, H is not conserved under the curl force. An equivalent analytical check is to verify that dH/dt = (1/m) \\bar p · [∇U(\\bar x) - ∇U(x)] is nonzero in this example.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central new claim is that H defined in (2.46), expressed as the functional (2.52) of the original motion, is conserved under the original curl force F = -V∇U. This is false. The error enters at Eq. (2.49) and the second equality of (2.50), where \\bar p is defined through \\dot{\\bar p} = (1/V(x)) \\dot p = -∇U(x). For the auxiliary particle, Newton's law is \\dot{\\bar p} = Fbar(\\bar x) = -∇U(\\bar x), not -∇U(x). Since \\bar x and x are different trajectories for a genuine curl force (they have different accelerations at t=0 whenever V(x0) ≠ 1 and ∇U(x0) ≠ 0), the identification is invalid. Differentiating H = |\\bar p|^2/(2m) + U(\\bar x) along the original flow gives dH/dt = (1/m) \\bar p · [∇U(\\bar x) - ∇U(x)], which does not vanish in general. Thus the conserved quantity advertised in the abstract and Section 2.6 does not exist; only the trivial statement that H is conserved along the auxiliary dynamics remains. The Darboux representation itself appears standard, but it does not rescue the conservation claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies particle dynamics under curl forces, i.e., position-dependent non-conservative and non-dissipative forces with nonzero curl. It uses Darboux's classification of 1-forms to represent any 2D force as F = -V∇U and any 3D force as F = -V∇U - ∇W, where U, V, W are called generalized potentials. The paper then constructs an auxiliary conservative force Fbar = -∇U and the associated Hamiltonian H, and claims that H, although not the physical energy, is a conserved quantity of motion under the original curl-force dynamics. Additional sections discuss work over closed paths, kinetic energy changes via Carathéodory's formulation of thermodynamics, and local accessibility of the Pfaffian equation F·dx = 0.","tokens_in":12899,"tokens_out":9819,"duration_ms":70632,"significance":"If the conservation claim were correct, the paper would establish a striking and potentially useful result: every curl-force trajectory would carry a nonlocal conserved functional. However, the conservation claim fails because the auxiliary momentum is evaluated along the original trajectory rather than the auxiliary trajectory. The Darboux classification itself is standard and, insofar as it is applied here, correctly derived; the examples illustrate the construction of generalized potentials. But the central advertised contribution—the auxiliary conserved Hamiltonian—does not exist, and the remaining classification is a known differential-geometric fact rather than a new dynamical theorem. The accessibility discussion also contains an internal contradiction in the statement and use of Carathéodory's theorem.","major_comments":[{"comment":"The identification ⌂p(t) = ṁp(t)/V(x(t)) is invalid. For the auxiliary dynamics, Newton's law reads ⌂p = Fbar(̄x) = -∇U(̄x), not -∇U(x). The paper replaces ṁp/V(x) by -∇U(x) using the original equation of motion, thereby evaluating the auxiliary force at the original position x(t) instead of the auxiliary position ̄x(t). Consequently, Eq. (2.52) defines a functional of the original trajectory that is not conserved. Differentiating this functional along the original flow gives dH/dt = (1/m)̄p(t)·[∇U(̄x(t)) - ∇U(x(t))], which is generally nonzero for genuine curl forces (when V is not identically 1). This invalidates the abstract's claim that the auxiliary Hamiltonian is a conserved quantity of motion under the curl force.","section":"§2.6.1, Eq. (2.49)"},{"comment":"The same error propagates into the three-dimensional construction. The auxiliary momentum is again defined through ṁp/V(x), but the auxiliary force must be evaluated at ̄x(t), not x(t). Therefore the expression in Eq. (2.58) is not a conserved quantity for the original curl-force dynamics. Only the trivial statement that H is conserved along the auxiliary dynamics (with ̄x and ̄p as a solution of ̄F) holds. The 3D generalization does not repair the 2D error; it relies on the same invalid identification.","section":"§2.6.2, Eqs. (2.57)-(2.58)"},{"comment":"The statement and use of Carathéodory's theorem are internally inconsistent. With the rank defined in the text as the integer r such that Ω∧(dΩ)^r ≠ 0 and Ω∧(dΩ)^{r+1} = 0, a 3D form with Ω∧dΩ ≠ 0 has rank r = 1. The text states that local accessibility holds if and only if r ≥ 2, which would imply that no 3D work 1-form is accessible. Yet §2.5.2 asserts that a 3D curl force with F·curl F ≠ 0 (i.e., Ω∧dΩ ≠ 0, rank r = 1) has the accessibility property. These two statements contradict each other. The correct statement for a single 1-form is that accessibility holds when Ω∧dΩ ≠ 0 (non-integrable distribution), so the theorem quotation and its application need correction.","section":"§2.5"}],"minor_comments":[{"comment":"The phrase 'without without loss of generality' contains a duplicated word.","section":"§2.1, Example 2.6"},{"comment":"The notation ̄v(x) is used for the auxiliary velocity, but ̄v is a function of time along a trajectory, not a field; this is confusing.","section":"§2.6.1, after Eq. (2.45)"},{"comment":"The term p0 · p0 should be written as |p0|^2 for notational consistency with Eq. (2.52).","section":"§2.6.2, Eq. (2.57)"},{"comment":"The equality chain 'dK = F·v dt = F·dx = F♭ = Ω' mixes differentials and differential forms; F·dx is not the same object as the 1-form F♭. This should be clarified.","section":"§2.4"},{"comment":"There is a typo: 'if an only if' should be 'if and only if'.","section":"Proposition 2.3"}],"recommendation":"reject","confidential_remarks":"The paper's central claim is false, and the error is load-bearing: the auxiliary-Hamiltonian conservation fails at Eq. (2.49). The Darboux classification part is standard and not novel in itself; the application to curl forces is straightforward but does not support the advertised conserved quantity. The accessibility section also contains a contradictory statement of Carathéodory's theorem. I recommend rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's advertised new result—that every curl force admits a conserved auxiliary Hamiltonian—is wrong. The error is at Eq. (2.49), where the auxiliary momentum is differentiated along the original trajectory instead of the auxiliary trajectory. Differentiating the H of Eq. (2.52) along the original flow gives a leftover term proportional to ∇U(xbar) − ∇U(x), which does not vanish for genuine curl forces. The claimed conserved quantity exists only when the original and auxiliary trajectories coincide, i.e., in the conservative case. That is a load-bearing flaw: the abstract and Section 2.6 promise a conserved quantity that does not exist.\n\nWhat the paper does well is the classification itself. Propositions 2.3 and 2.4 correctly transcribe Darboux's normal forms for rank-1 1-forms: any 2D curl force is −V∇U, and any 3D curl force is −V∇U − ∇W. The sections on helicity, accessibility, and the worked examples (2.6 and 2.9) are clear and could be useful for a mechanics audience. The citation pattern is honest, and the writing is careful, so the mistake does not look like carelessness—it looks like a genuine mathematical slip.\n\nThe soft spots are proportionate. The classification is not new mathematics; it is a known theorem applied to force fields, and the authors do not claim otherwise. The real problem is the auxiliary-Hamiltonian construction. Once you read (2.49) carefully, the flaw is obvious: pbar-dot is set equal to −∇U(x), not −∇U(xbar). Everything built on that identification collapses.\n\nWho should read this? Researchers interested in the Darboux representation of work 1-forms might read the first half, but they should skip Section 2.6. As is, the paper cannot be accepted. A serious referee would spot the error in minutes, so it deserves referee time only to confirm the rejection. If the authors strip out the conservation claim and present the classification as an expository contribution, it could become a reasonable short note. In current form, my recommendation is reject.","headline":"A clean Darboux classification of curl forces is undermined by a false auxiliary-Hamiltonian conservation claim that fails at Eq. (2.49).","tokens_in":13482,"tokens_out":2568,"would_cite":false,"duration_ms":22709,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37J05","58A10","70F99"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that Darboux classification of the work 1-form gives every curl force at most two generalized potentials in 2D and three in 3D, and that a rescaled conservative auxiliary force has a Hamiltonian conserved along the motion…","keywords":["curl forces","Darboux classification","work 1-form","generalized potentials","auxiliary Hamiltonian","nonconservative forces","helicity","Pfaffian systems"],"falsifier":"Take a genuine two-dimensional curl force such as $F = -(xy^2, x^3)$, integrate the equations of motion numerically, and evaluate the claimed conserved quantity from Eq. (2.52) along the orbit. Direct differentiation of that expression gives $dH/dt = (\\bar p/m)\\cdot(\\nabla U(\\bar x) - \\nabla U(x))$, which is nonzero whenever the auxiliary trajectory $\\bar x$ differs from the physical trajectory $x$; a converged numerical integration will show $H$ drifting rather than staying constant.","tokens_in":12422,"feed_emoji":"🌀","tokens_out":8536,"duration_ms":65962,"temperature":0.7,"pith_summary":"This paper aims to show that curl forces — position-dependent forces that cannot be gradients of a potential and are neither conservative nor dissipative — have a hidden structure when viewed through their work 1-form $\\Omega = F\\cdot dx$. Using Darboux's classification of 1-forms, it claims that any two-dimensional curl force can be written as $F = -V\\,\\nabla U$ with two generalized potentials $U,V$, and any three-dimensional one as $F = -V\\,\\nabla U - \\nabla W$ with three. The paper further claims that rescaling the force by $1/V$ produces a conservative auxiliary force, and that the Hamiltonian of that auxiliary force is a conserved quantity along the original curl-force motion, even though it is not the physical energy. If these claims hold, the classification gives curl forces an analogue of potential structure and a conserved quantity that, being nonlocal, does not by itself make the dynamics integrable.","feed_headline":"Curl forces hide a conserved Hamiltonian that is not energy","feed_subtitle":"Darboux's classification yields two or three generalized potentials for any curl force and an extra conserved quantity.","key_machinery":"The machinery is the work 1-form $\\Omega = F\\cdot dx$ together with Darboux's normal-form classification of rank-one 1-forms. Darboux's theorem gives local coordinates in which $\\Omega$ is either $y_1\\,dz_1$ or $y_1\\,dz_1 + dy_2$; reading the coefficients as generalized potentials turns the force into $F = -V\\,\\nabla U$ in 2D and $F = -V\\,\\nabla U - \\nabla W$ in 3D. The auxiliary step rescales by the potential $V$ so that $\\bar F = -\\nabla U$, making the auxiliary force conservative and endowing it with a Hamiltonian. That Hamiltonian, pulled back to the original trajectory as a double time integral, is the claimed conserved quantity.","core_discovery":"The central discovery, stated on the paper's own terms, is that the natural object for curl-force dynamics is not the vector field but its work 1-form $\\Omega = F^\\flat$. Darboux's theorem applies because $\\Omega$ has rank one: on $\\mathbb{R}^2$ the canonical form is $\\Omega = \\phi\\,d\\psi$, giving $F = -V\\,\\nabla U$; on $\\mathbb{R}^3$ the alternative $\\Omega = \\phi\\,d\\psi + d\\zeta$ gives $F = -V\\,\\nabla U - \\nabla W$. The paper calls $U,V,W$ generalized potentials, with the number of required potentials controlled by the helicity $F\\cdot \\operatorname{curl} F$. It then defines an auxiliary conservative force $\\bar F = F/V$ (or $(F+\\nabla W)/V$ in 3D), with Hamiltonian $H = |\\bar p|^2/(2m) + U(\\bar x)$. The paper claims that when $H$ is expressed as a nonlocal functional of the original trajectory — Eqs. (2.52) and (2.58) — it is a conserved quantity of motion under the curl force, although it is not the physical energy and does not partition phase space into invariant regions.","pith_inferences":["Beyond the paper: the nonlocal conserved functional could be tested numerically as a check on integration accuracy for curl-force orbits; if it drifts, the discrepancy measures how far the auxiliary trajectory has separated from the original one.","Beyond the paper: because the three-potential representation is non-unique up to conservative additions, one could look for a canonical gauge fixing, for instance by minimizing the $L^2$ norm of $V$, to make the decomposition computationally convenient.","Beyond the paper: the same Darboux-based three-potential representation may transfer to continuum mechanics, where a stress work 1-form plays the analogue of $\\Omega$; the paper notes the analogy but does not develop the conservation consequences in that setting."],"forward_implications":["In two dimensions every curl force is locally of the form $-V\\,\\nabla U$, so the two generalized potentials play the role that a single potential plays for conservative forces.","In three dimensions a curl force needs a third potential $W$ exactly when $F\\cdot \\operatorname{curl} F \\neq 0$; when this helicity vanishes, two potentials suffice.","Curl forces can do nonzero work around closed loops, but a closed motion followed by its reverse does zero net work, in contrast to dissipative forces.","The auxiliary Hamiltonian supplies a conserved quantity of motion for every curl force, but because it is defined through integrals of the motion it is nonlocal and cannot serve as a standard first integral for reducing dimension."],"supporting_citations":[{"why":"Supplies the Darboux classification theorem that yields the canonical forms of the work 1-form.","marker":"[Darboux, 1882]"},{"why":"Provides a modern formulation of Darboux normal forms used in the paper's proof of the potential representations.","marker":"[Slebodzinski, 1970]"},{"why":"Supplies the exterior differential systems framework and Carathéodory's theorem used for the accessibility analysis.","marker":"[Bryant et al., 2013]"},{"why":"Introduces the notion of curl forces and their non-Hamiltonian, non-Noetherian behavior that motivates the paper's classification.","marker":"[Berry and Shukla, 2012]"},{"why":"Gives a subset of curl forces that admit anisotropic-kinetic-energy Hamiltonians, the result the paper generalizes with auxiliary Hamiltonians.","marker":"[Berry and Shukla, 2015]"},{"why":"Provides the flat and sharp operators and Hodge star identities used to relate the work 1-form to the curl of the force.","marker":"[Abraham et al., 2012]"}],"fun_headline_variants":["Darboux classification yields hidden conserved Hamiltonian","Curl forces: Darboux yields conserved Hamiltonian, not energy","Generalized potentials from Darboux classify curl forces","Conserved hidden Hamiltonian for curl forces via Darboux","Work 1-form yields conserved Hamiltonian for curl forces"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the rescaled force acting on the auxiliary particle can be evaluated at the real particle's position; once the two trajectories diverge, which they do for any genuine curl force, the auxiliary Hamiltonian is no longer conserved.","fun_headline_variants_meta":{"raw":{"variants":["Darboux classification yields hidden conserved Hamiltonian","Curl forces: Darboux yields conserved Hamiltonian, not energy","Generalized potentials from Darboux classify curl forces","Conserved hidden Hamiltonian for curl forces via Darboux","Work 1-form yields conserved Hamiltonian for curl forces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000772,"raw_usage":{"total_tokens":3418,"prompt_tokens":944,"completion_tokens":2474,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":560,"completion_tokens_details":{"reasoning_tokens":2396}},"tokens_in":560,"tokens_out":2474,"duration_ms":16241,"temperature":1.0,"reasoning_tokens":2396,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:58:03.870349+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a genuine two-dimensional curl force such as $F = -(xy^2, x^3)$, integrate the equations of motion numerically, and evaluate the claimed conserved quantity from Eq. (2.52) along the orbit. Direct differentiation of that expression gives $dH/dt = (\\bar p/m)\\cdot(\\nabla U(\\bar x) - \\nabla U(x))$, which is nonzero whenever the auxiliary trajectory $\\bar x$ differs from the physical trajectory $x$; a converged numerical integration will show $H$ drifting rather than staying constant.","supporting_citations":[],"review_version":1}