{"id":"070b9046-b59f-40d0-8dec-0fb0b02fa3db","arxiv_id":"1908.05273","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The Mandelstam-Tamm and Heisenberg time-energy uncertainty relations are not universally valid because their derivation divides by a vanishing time derivative for eigenstates of energy or the observable.","lead":"This paper argues that the standard time-energy uncertainty relations, unlike position-momentum uncertainty, are not universally valid in quantum mechanics. It shows that the usual derivation fails for stationary states and questions common applications in physics and cosmology.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The non-universality claim rests on an invalid denseness inference: eigenvectors are not a dense set of violations, and Mandelstam–Tamm explicitly excludes ΔE=0.","rationale":"In good faith, the paper performs a service by identifying a genuine gap in many textbook derivations that divide by d⟨A⟩/dt without checking that it is nonzero. The reader's CONDITIONAL verdict captures this. The most load-bearing part of the paper is the proof of the headline claim that the time–energy relations 'cannot hold on linearly dense sets.' That proof has a real soft spot: it conflates the algebraic property that eigenvectors span the Hilbert space with the analytic claim that the counterexamples form a dense set of violations. In the two-level model, (29) holds with equality for all non-eigenstate superpositions, which form a linearly dense set; the only singularities are the eigenstates themselves, where the relation is undefined. The paper itself concedes that the original Mandelstam–Tamm paper contains the ΔH ≠ 0 reservation (Sec. 4), so the counterexamples fall outside the standard domain. For continuous-spectrum Hamiltonians, the eigenvector set is empty, further eroding the claim. The Planck–Einstein discussion is heuristic and does not rescue the proof. The concern is therefore not that the paper's qualitative conclusion is false, but that its central mathematical argument is unsupported as written; the conclusion should be weakened to 'not universally valid without domain restrictions' in line with the reader's conditional verdict. No change to the reader's verdict is needed.","tokens_in":9772,"tokens_out":11004,"duration_ms":110999,"concrete_test":"Perform the two-level calculation explicitly. Let H = E0|0⟩⟨0| + E1|1⟩⟨1|, A = |0⟩⟨1| + |1⟩⟨0|, and |ψ(θ)⟩ = cosθ|0⟩ + i sinθ|1⟩. Compute ΔE = (E1−E0) sin(2θ)/2, |d⟨A⟩/dt| = (E1−E0) sin(2θ)/ℏ, and ΔA = 1, so τ_A ΔE = ℏ/2 for every θ ∈ (0,π/2). As θ → 0 the state tends to the eigenstate |0⟩, ΔE → 0, τ_A → ∞, and the product remains ℏ/2; (29) is not violated on any sequence approaching the eigenstate. This directly tests the paper's claim that (29) cannot hold on a linearly dense set, since the non-eigenstate family is linearly dense in the two-level Hilbert space and satisfies the inequality with equality.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The decisive weakness is the inference from undefinedness on eigenvectors to the headline that (29) 'cannot hold on linearly dense sets' (Sec. 3). For eigenstates of H, Eq. (26) reads 0·0 ≥ 0; (27)/(29) is obtained by dividing by |d⟨A⟩/dt| = 0, so the relation is undefined there, not contradicted. This alone does not show non-universality, since the Mandelstam–Tamm derivation standardly and in the original paper (as the author concedes in Sec. 4) is restricted to ΔE > 0. The denseness step is doubly flawed. First, eigenvectors are not themselves a dense set of counterexamples: the relation can hold on a dense set of nearby states. Explicitly, for a two-level system H = E0|0⟩⟨0| + E1|1⟩⟨1|, A = |0⟩⟨1| + |1⟩⟨0| and |ψ(θ)⟩ = cosθ|0⟩ + i sinθ|1⟩, a direct calculation gives τ_A ΔE = ℏ/2 for every θ ∈ (0,π/2); these states are linearly dense in the two-level Hilbert space, so (29) in fact holds on a linearly dense set. Second, for continuous-spectrum Hamiltonians such as the free particle, Σ_H contains no normalizable eigenvectors, so the counterexample class is empty. The argument establishes a caveat about division by zero, not the non-universality of time–energy uncertainty relations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript critically examines the Heisenberg and Mandelstam–Tamm (MT) time–energy uncertainty relations. It first reviews the standard derivation of the Robertson inequality (5) and then considers the MT step that replaces the commutator average with ℏ|d⟨A⟩_φ/dt|/2 and divides by this derivative to define τ_A. The author observes that for eigenstates of H or of A the derivative and the relevant variance vanish, so Eq. (26) becomes 0·0 ≥ 0 and the quotient defining τ_A is undefined. From this, Section 3 concludes that relations (27) and (29) cannot hold on linearly dense sets and are therefore not universally valid. Section 4 discusses stationary states and the Planck–Einstein relation, and Section 5 repeats the non-universality conclusion, also invoking the author's previous preprint [35].","tokens_in":9998,"tokens_out":7944,"duration_ms":76663,"significance":"If the non-universality claim were established, the paper would be significant: it would challenge the textbook status of time–energy uncertainty as a rigorous counterpart to position–momentum uncertainty. The paper has the merit of highlighting a genuine domain issue in the textbook derivation of the MT relation, namely that τ_A is not defined when d⟨A⟩/dt = 0, and the algebraic derivation up to Eq. (26) is standard. However, the paper's decisive inference from undefinedness to non-universality is not sound, as detailed below; the result, as it stands, is a caveat about the domain of the derivation rather than a proof that the relations are not universally valid. The paper contains no fitted parameters or numerical benchmarks; its argument is analytic.","major_comments":[{"comment":"The inference from the vanishing of d⟨A⟩_φ/dt on eigenstates of H or A to the conclusion that (27)/(29) 'cannot hold on linearly dense sets' is invalid. For an eigenstate |φ_β⟩ of H, Eq. (26) gives the true statement 0 · Δ_{φ_β}E ≥ 0; what fails is only the subsequent division by |d⟨A⟩_φ/dt| = 0, which makes τ_A undefined. An undefined expression is not a counterexample to an inequality. Furthermore, states on which the relation does hold can be linearly dense: for the two-level system H = E0|0⟩⟨0| + E1|1⟩⟨1|, A = |0⟩⟨1| + |1⟩⟨0|, and |ψ(θ)⟩ = cosθ|0⟩ + i sinθ|1⟩, a direct calculation gives τ_A ΔE = ℏ/2 for every θ ∈ (0,π/2). Hence the proof establishes only a domain caveat about division by zero, not non-universality.","section":"Sec. 3, Eqs. (34)–(37)"},{"comment":"The paper concedes in Sec. 4 that Mandelstam and Tamm themselves restricted the derivation to states with Δ_φH ≠ 0 and to eigenvectors corresponding to the continuous spectrum. This concession means the eigenvector counterexamples of Sec. 3 fall outside the domain of the original conditional relation. Moreover, for Hamiltonians with purely continuous spectrum, such as a free particle, Σ_H contains no normalizable eigenvectors, so the counterexample class can be empty. To sustain the headline claim, the manuscript must show that the conditional form τ_A Δ_φE ≥ ℏ/2 for states with Δ_φH > 0 and d⟨A⟩_φ/dt ≠ 0 fails or is unjustified; the present argument does not address that form.","section":"Sec. 4, paragraph on Mandelstam–Tamm reservation"}],"minor_comments":[{"comment":"There are several typographical errors, including 'conlusion' in the abstract, 'Schroödinger' in the introduction, 'Mandelstm', 'operstor', 'Hesienberg's', and 'nad'; these should be corrected.","section":"Abstract and Introduction"},{"comment":"The statement that Eqs. (38) and (41) are 'mathematically identical' is misleading: (38) is an inequality about uncertainties Δ_φt and Δ_φE, while (41) is an exact relation between the period T_φ and energy E_φ. Please rephrase this as a scaling analogy or a dimensional observation.","section":"Sec. 4, Eqs. (38) and (41)"},{"comment":"The conclusion states that the analysis 'together with the conclusions presented in [35]' supports the result; since [35] is the author's own preprint, the reader needs either a summary of its argument or a published reference.","section":"Sec. 5"},{"comment":"The phrase 'linearly dense set' is used without a definition. Please define it explicitly and distinguish between the set itself being dense and the property (29) holding on all elements of a dense set; the current usage is ambiguous.","section":"Sec. 3, 'linearly dense' terminology"}],"recommendation":"major_revision","confidential_remarks":"To the editor: this is a short proceedings-style contribution, and the central claim is not supported by the proof it contains. A revised version that restricts the claim to the domain of definition of τ_A (states with ΔH > 0 and d⟨A⟩/dt ≠ 0) could be a useful pedagogical note, but as submitted the 'not universally valid' conclusion overreaches. I would also ask the author to engage with the existing review literature, particularly Busch's chapter (ref. [20]), which already distinguishes several time–energy relations and their validity conditions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is my take. The paper correctly identifies a real defect in the standard textbook route to the Mandelstam–Tamm relation: after Eq. (26), dividing by |d⟨A⟩/dt| is not allowed when that derivative vanishes, which happens for eigenstates of H or A. That observation is sound, and the paper explains it clearly. But it is not new. It is in Gray and Vogt (2005), which the paper cites, and Mandelstam and Tamm themselves explicitly restricted the relation to states with ΔE≠0. The paper then pushes the point too far. It claims relations (27)/(29) 'cannot hold on linearly dense sets' and therefore are not universally valid. That inference is invalid. Undefinedness on a set is not falsity on that set, and eigenvectors are not a dense set of counterexamples anyway. The stress-test example nails it: for H and A as given, τ_A ΔE = ℏ/2 for all θ in (0,π/2), so (29) actually holds on a linearly dense set. And for continuous-spectrum Hamiltonians, the eigenvector set contains no normalizable states, so the counterexample class is empty. The denseness argument is the load-bearing part of the paper, and it collapses.\n\nCredit where due: the paper is honest about the Mandelstam–Tamm caveat, the stationary-state discussion is intuitive, and the Planck–Einstein rescaling in Sec. 4 is clearly labeled as heuristic. The citation pattern is fine, though leaning on the author's own preprint [35] in the conclusion is weak but minor.\n\nNet: the paper is a well-written restatement of a known caveat with an overblown conclusion. It will not change the literature, and the novel part is wrong. A fresh reader might find the derivation walkthrough useful, but a knowledgeable referee would reject the dense-set claim. I would not bring it to the reading group, would not cite it, and would not send it to peer review. The right outcome is desk reject, with a note pointing to Gray & Vogt and the two-level counterexample.","headline":"The division-by-zero caveat is real but already known; the paper's dense-set claim is false, and the right outcome is desk reject.","tokens_in":10592,"tokens_out":4305,"would_cite":false,"duration_ms":41306,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["03.65.-w","03.65.Ta","01.55.+b"],"model":"deepseek-v4-flash","headline":"The paper argues that time–energy uncertainty relations cannot be universal quantum laws.","keywords":["time-energy uncertainty relation","Mandelstam-Tamm inequality","universal validity","stationary states","Pauli theorem","division by zero","quantum foundations","uncertainty principle"],"falsifier":"Check the Mandelstam–Tamm inequality on the free particle, whose Hamiltonian has a purely continuous spectrum and no normalizable energy eigenvectors; if the inequality holds for every normalizable state in that system, the paper's dense-set argument does not apply, because the set of eigenvectors is empty rather than dense.","tokens_in":9487,"feed_emoji":"⚛️","tokens_out":7715,"duration_ms":68608,"temperature":0.7,"pith_summary":"This paper challenges the common view that time–energy uncertainty relations have the same rigorous status as the position–momentum uncertainty relation. It argues that the standard derivation of the Mandelstam–Tamm relation divides both sides by a quantity that is zero for stationary states, making the relation undefined on a linearly dense set of states. The author concludes that the Heisenberg and Mandelstam–Tamm time–energy relations cannot be considered universally valid and should be treated as conditional estimates for non-stationary systems. If correct, this narrows what can be inferred from time–energy uncertainty in quantum mechanics and in applications such as cosmology.","feed_headline":"Time–energy uncertainty is not a universal law","feed_subtitle":"A division-by-zero step in the standard derivation means the relation fails for stationary states.","key_machinery":"The load-bearing object is the set $\\Sigma_H$ of Hamiltonian eigenvectors, which the paper takes to be linearly dense in the state space—meaning finite linear combinations of these vectors can approximate every state. The mechanism that carries the argument is the division step in the Mandelstam–Tamm derivation: one divides by $|d\\langle A\\rangle_\\varphi/dt|$ to define $\\tau_A$. For every eigenstate of $H$, this derivative is zero and $\\Delta_\\varphi E$ is also zero, so the division is undefined and the inequality reduces to $0 = 0$. Since the same happens for eigenstates of $A$, the relation fails on a dense set. A supporting mechanism is the no-self-adjoint-time-operator theorem, which keeps time from being treated as an observable like position.","core_discovery":"The central claim is a negative result: within standard quantum mechanics, the time–energy uncertainty relations of type $\\Delta_\\varphi t \\cdot \\Delta_\\varphi E \\geq \\hbar/2$ and $\\tau_A \\cdot \\Delta_\\varphi E \\geq \\hbar/2$ fail to be universally valid. The proof rests on the observation that in the Mandelstam–Tamm derivation, after using $\\langle [A,H]\\rangle_\\varphi = i\\hbar\\, d\\langle A\\rangle_\\varphi/dt$, one divides by $|d\\langle A\\rangle_\\varphi/dt|$ to define the time scale $\\tau_A$. For any eigenvector of $H$ (or of $A$), this derivative is zero, so the division is undefined; at the same time $\\Delta_\\varphi E$ is also zero, reducing the inequality to the trivial equality $0 = 0$. Because the eigenvectors of $H$ form a linearly dense set in the state space, the relation cannot hold on a dense set and therefore cannot be called a universal principle. The paper also notes that a well-known theorem rules out a self-adjoint time operator for a Hamiltonian bounded from below, further separating time from position.","pith_inferences":["If the linear-denseness argument generalizes, any derived inequality obtained by dividing by a quantity that can vanish on a dense set becomes a conditional estimate rather than a universal law; this may apply to other 'uncertainty' relations in the literature.","The paper's conclusion leaves open the possibility that the Mandelstam–Tamm bound is valid on the dense set of states with $\\Delta_\\varphi H \\neq 0$; whether that counts as 'universal' is a definitional question, and the paper's choice to include stationary states is a substantive one.","A numerical study of the relation for a free particle or other continuous-spectrum Hamiltonian could determine whether the failure on stationary states is the only obstruction to universal validity.","The Planck–Einstein exact relation $T_\\varphi E_\\varphi = h$ suggests that many textbook 'time–energy uncertainty' applications may be re-readable as exact relations, which would change how those results are interpreted."],"forward_implications":["Any use of the Mandelstam–Tamm relation outside the domain where both $\\Delta_\\varphi H$ and $d\\langle A\\rangle_\\varphi/dt$ are nonzero needs an explicit justification.","Time–energy uncertainty should not be cited as a universal bound comparable to position–momentum uncertainty in foundational arguments.","Physicists using time–energy relations in cosmology or astrophysics should check that the states involved are non-stationary and that the relation is being read as a conditional estimate.","For unstable states, the paper says the relevant relations (such as lifetime–width) are exact-value relations, not uncertainty relations, so they should not be used to infer a fundamental time–energy trade-off."],"supporting_citations":[{"why":"Proposes the original time–energy uncertainty relation that the paper argues is not universal.","marker":"[1]"},{"why":"Gives the Mandelstam–Tamm derivation whose division step is the target of the critique.","marker":"[14]"},{"why":"Textbook reproduction of the Mandelstam–Tamm derivation where the division by $|d\\langle A\\rangle/dt|$ is performed without the nonzero caveat.","marker":"[5]"},{"why":"States the theorem that no self-adjoint time operator exists when the Hamiltonian is bounded from below, used to separate time from position.","marker":"[19]"},{"why":"Mathematical analysis of the Mandelstam–Tamm principle that notes cases where $d\\langle A\\rangle/dt$ vanishes for eigenvectors.","marker":"[25]"},{"why":"The author's earlier analysis of the universal validity of time–energy relations, whose conclusions this paper continues.","marker":"[35]"}],"fun_headline_variants":["Time–energy uncertainty fails for stationary states","Division by zero breaks time–energy uncertainty","Time–energy uncertainty not universal in QM","Stationary states void time–energy uncertainty","Zero derivative kills time–energy uncertainty"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the premise that the eigenvectors of the Hamiltonian form a linearly dense set in the state space, and that a relation called 'universally valid' must hold even for stationary states, where the expectation value of any observable does not change in time.","fun_headline_variants_meta":{"raw":{"variants":["Time–energy uncertainty fails for stationary states","Division by zero breaks time–energy uncertainty","Time–energy uncertainty not universal in QM","Stationary states void time–energy uncertainty","Zero derivative kills time–energy uncertainty"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000169,"raw_usage":{"total_tokens":1207,"prompt_tokens":834,"completion_tokens":373,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":450,"completion_tokens_details":{"reasoning_tokens":307}},"tokens_in":450,"tokens_out":373,"duration_ms":3722,"temperature":1.0,"reasoning_tokens":307,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:17:24.635442+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check the Mandelstam–Tamm inequality on the free particle, whose Hamiltonian has a purely continuous spectrum and no normalizable energy eigenvectors; if the inequality holds for every normalizable state in that system, the paper's dense-set argument does not apply, because the set of eigenvectors is empty rather than dense.","supporting_citations":[{"cited_title":"Heisenberg, Uber den anschaulichen Inhalt der quantentheoretischen Kinematik und Mechanik , Zeitschrift for Physik, 43, 172 – 98 (1927)","cited_arxiv_id":null,"evidence_quote":"Proposes the original time–energy uncertainty relation that the paper argues is not universal."},{"cited_title":"Mandelstam and Ig","cited_arxiv_id":null,"evidence_quote":"Gives the Mandelstam–Tamm derivation whose division step is the target of the critique."},{"cited_title":"Pauli, General Principles of Quantum Mechanics (Spr inger- Verlag, New York 1980), p","cited_arxiv_id":null,"evidence_quote":"States the theorem that no self-adjoint time operator exists when the Hamiltonian is bounded from below, used to separate time from position."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Mathematical analysis of the Mandelstam–Tamm principle that notes cases where $d\\langle A\\rangle/dt$ vanishes for eigenvectors."},{"cited_title":"Remarks on the uncertainty relations","cited_arxiv_id":"1810.11462","evidence_quote":"The author's earlier analysis of the universal validity of time–energy relations, whose conclusions this paper continues."}],"review_version":1}