The epsilon-to-zero ideal limit and composite-measurement semantics of causal continuous-outcome measurements of quantum field operators

Oeckl (J. Math. Phys. 66, 042302 (2025)) constructed, for free bosonic quantum field theory, a one-parameter family of Gaussian-regularized positive-operator-valued measures and associated quantum operations that measure a field observable with continuous outcomes; the family inherits the field's locality, is causally transparent (it permits no superluminal signalling and is a possible measurement in Sorkin's sense), and converges weakly to the spectral measure of the field operator. Later work by the same author completed the covariant spacetime extension and showed that the epsilon-to-zero limit does not exist at the level of quantum operations, that the physically meaningful limits are quotient-type probability expressions, and that measurement-induced correlations between later measurements diverge as the regulator is removed. What remains open is the rigorous epsilon-to-zero limit theory of composite measurements together with the ideality question: precisely which composite measurement expressions admit a limit as the regularization is removed and in what sense; whether that limit, where it exists, defines an ideal (projective, repeatable) measurement of the observable, as the heuristic composition identities of the source suggest; and the operational comparison, for joint and sequential measurements, between extracting the full value of a linear observable and measuring its directly quantized function, in relation to the modulus-square construction, its renormalization scheme, and the sharpness lower bound on physically implementable measurements.

Background

In relativistic quantum field theory (QFT), the naive extension of the quantum measurement postulates fails: as Sorkin showed, applying the collapse (Lüders) update associated with a projector localised in some spacetime region can transmit information faster than light between spacelike separated regions, even though all local operators commute at spacelike separation. These are the so-called impossible measurements. A positive-operator-valued measure (POVM) is a family of positive operators summing (or integrating) to the identity whose expectation values give the statistics of a general quantum measurement; a selective quantum operation (or instrument) is the accompanying completely positive map on trace-class operators (finite-trace-norm operators, which represent quantum states) that updates the state given an outcome, and its non-selective version is the outcome-averaged, trace-preserving map. Several routes to a consistent local measurement theory for QFT exist: the Fewster-Verch framework models the apparatus as a probe quantum field locally coupled to the system field; detector (Unruh-DeWitt-type) models describe measurements along worldlines; and Jubb identified a necessary and sufficient causality condition on state updates in real scalar QFT (the past-support-non-increasing property, also called causal transparency by Oeckl), under which Gaussian measurements of smeared fields are causal.

Oeckl's construction starts from free bosonic QFT described through the holomorphic (Bargmann) representation: the Hilbert space is a space of square-integrable holomorphic functions on the dual of a complex linear phase space, with a reproducing family of coherent states (Gaussian pure states labelled by phase-space points). Field operators arise by Weyl quantization (the symmetric ordering prescription assigning operators to classical phase-space functions) of linear phase-space functions $D_\xi$ (smeared fields). Because the spectral measure of a field operator is singular at points, its decomposition is obtained as a limit of a strongly continuous one-parameter family of regularized POVMs built from bounded positive operators $\Pi_\xi^\epsilon(q)$ (essentially Gaussian kernels of width $\epsilon$ in the value of $D_\xi$). The continuous-outcome measurement of the observable is then encoded by selective quantum operations $A_\xi^\epsilon(q)(\sigma)=\sqrt{2\pi}\epsilon\,\Pi_\xi^\epsilon(q)\,\sigma\,\Pi_\xi^\epsilon(q)$ on trace-class states $\sigma$, and their weighted integrals $A_\xi^\epsilon[f](\sigma)=\int\mathrm{d}q\,f(q)A_\xi^\epsilon(q)(\sigma)$. These inherit the locality (microcausality) properties of the field operator, and Oeckl proved (his Theorem 5.2) that the non-selective operations are causally transparent: inserting such a measurement cannot be detected through the outcome statistics of measurements outside its causal future. This is surprising because Albertini and Jubb showed that ideal (projective) measurements of smeared field operators generically do violate causality in Sorkin-type scenarios, and Oeckl's discrete-outcome (spectral-bin) variants do as well.

Two features of the regularized family remain only partially controlled. First, the limit $\epsilon\to0$ that would recover the 'true' sharp measurement does not exist in the trace norm at the level of quantum operations; only limits 'under the trace' of composite probability and expectation expressions are guaranteed (Lemma 4.8 of the source), and the source's precise existence criteria and domains were left open. Second, a measurement is called ideal (or projective) if repeating it, or coarse-graining its outcomes, reproduces the same statistics; equivalently the corresponding non-selective operation is idempotent. For every $\epsilon>0$ the regularized operations are not ideal, but the composition formula

$A_\xi^\epsilon(q')\circ A_\xi^\epsilon(q)=\frac{1}{\sqrt{\pi}\epsilon}\exp\left(-\frac{(q-q')^2}{\epsilon^2}\right)A_\xi^{\epsilon/\sqrt{2}}\left(\frac{q+q'}{2}\right)$

suggests that in the limit $\epsilon\to0$ the operations collapse onto a delta-function behaviour, implying the ideality identities (equations (85)-(86) of the source) such as $\lim_{\epsilon\to0}A_\xi^\epsilon[\chi_A]\circ A_\xi^\epsilon[\chi_A]=\lim_{\epsilon\to0}A_\xi^\epsilon[\chi_A]$ for characteristic functions $\chi_A$; the rigorous justification was explicitly left to future work.

Subsequent developments define the current frontier. Oeckl and Zampeli proposed the modulus-square construction: spacetime probes (spacetime-localized generalizations of quantum operations in the positive formalism, a local path-integral framework) that measure expectation values of quadratic and similar observables, with a renormalization scheme that is compositional ('renormalization commutes with composition'). Oeckl's follow-up (Phys. Rev. D 113, 085002 (2026)) completed the covariant spacetime development deferred in the source: it constructs regularized measurements of spacetime-extended observables as probes in the positive formalism, proves causal transparency in that general setting, and quantifies how a temporally extended measurement back-reacts on itself and induces correlations between measurements in its causal future. The same paper shows explicitly that the limit $\epsilon\to0$ does not exist as a quantum operation (different boundary conditions yield different limits), identifies quotient-type probability expressions (probabilities written as ratios of quantities that individually vanish in the limit, such as an outcome density divided by the total outcome density) as the correct limiting objects, and finds that the measurement-induced correlations diverge as $\epsilon\to0$; it does not treat ideality, and defers the comparison with the modulus-square construction. In parallel, Mandrysch and Navascués proved that Gaussian-modulated measurements of locally smeared fields are realizable in the Fewster-Verch framework (with a movable Heisenberg cut), and Simmons, Papageorgiou, Christodoulou and Brukner derived, from an explicit pointer model based on local S-matrices, that the sharpness of any physical measurement of a smeared field $\phi(f)$ is bounded below by $\Delta_R(f,f)^{1/2}$, the square root of the retarded propagator smeared against $f$ - a constraint absent from the abstractly constructed measurement maps, which their channels otherwise reproduce once taken into account.

Problem Statement

In Oeckl's regularized causal measurement framework for free bosonic quantum field theory - with field observables $D_\xi$ quantized in the holomorphic representation, regularized positive operators $\Pi_\xi^\epsilon(q)$ forming the continuous-outcome POVM, and selective quantum operations $A_\xi^\epsilon(q)(\sigma)=\sqrt{2\pi}\epsilon\,\Pi_\xi^\epsilon(q)\sigma\Pi_\xi^\epsilon(q)$ with their weighted integrals $A_\xi^\epsilon[f]$ - establish the rigorous $\epsilon\to0$ limit theory for composite measurements and settle the ideality question, specifically:

(1) Existence criteria and domain. Characterize precisely the class of composite measurement expressions (obtained by composing selective and non-selective operations $A_\xi^\epsilon[f]$ with one another, with time evolution, and with initial states and final traces or post-selections, including the multi-observable operations of the source's equation (91) and their spacetime probe counterparts) for which the $\epsilon\to0$ limit exists, and specify the topology (trace norm, strong, weak, or quotient-type) and domain in which it exists. It is known that the limit does not exist at the level of individual quantum operations, that single-measurement traces converge (Lemma 4.8 of the source), that quotient-type probability expressions have well-defined limits in the spacetime setting for non-degenerate covariance, and that measurement-induced correlations between measurements in the causal future diverge as $\epsilon\to0$; determine where the boundary between existence and divergence lies, including degenerate cases and unbounded weight functions.

(2) Ideality. Prove or disprove that, at the level where the limits exist, the $\epsilon\to0$ limit defines an ideal (projective, repeatable) measurement of the observable $D_\xi$; that is, establish rigorously or refute the composition identities suggested in the source, such as $\lim_{\epsilon\to0}A_\xi^\epsilon[\chi_A]\circ A_\xi^\epsilon[\chi_A]=\lim_{\epsilon\to0}A_\xi^\epsilon[\chi_A]$ and the delta-function collapse of $A_\xi^\epsilon(q')\circ A_\xi^\epsilon(q)$, specifying exactly the states, composites, and topology for which they hold or fail. A positive answer must be consistent with causal transparency of all composites; a negative answer should exhibit a concrete composite scenario with non-ideal or non-existent limiting statistics.

(3) Operational comparison of composite schemes. For joint and sequential measurements built from the regularized family, characterize the operational difference between (i) extracting the full value of a linear observable (or several linear observables) and then applying a function $f$ to the extracted values, and (ii) measuring the directly quantized operator (for example $\widehat{D^2}$ versus extracting $D$ and squaring), where single-measurement expectation values agree in the $\epsilon\to0$ limit but joint statistics are expected to differ: derive the joint probability distributions of the two schemes, identify in what measurable quantities they differ, and compare both with the modulus-square construction and its compositional renormalization scheme, taking into account the sharpness lower bound $\Delta_R(f,f)^{1/2}$ on physically implementable measurements.

References

1. R. Oeckl, Spectral decomposition of field operators and causal measurement in quantum field theory, J. Math. Phys. 66, 042302 (2025). DOI: 10.1063/5.0245368. arXiv:2409.08748. https://arxiv.org/abs/2409.08748
2. R. Oeckl, Causal measurement in quantum field theory: Spacetime, Phys. Rev. D 113, 085002 (2026). DOI: 10.1103/5gl8-3j6p. arXiv:2511.06566. https://arxiv.org/abs/2511.06566
3. R. Oeckl, A. Zampeli, Towards local and compositional measurements in quantum field theory, arXiv:2505.10968. https://arxiv.org/abs/2505.10968
4. J. Mandrysch, M. Navascués, Quantum field measurements in the Fewster-Verch framework, Lett. Math. Phys. 115, 115 (2025). DOI: 10.1007/s11005-025-02001-3. arXiv:2411.13605. https://arxiv.org/abs/2411.13605
5. R. Simmons, M. Papageorgiou, M. Christodoulou, C. Brukner, Factorisation conditions and causality for local measurements in QFT, arXiv:2511.21644. https://arxiv.org/abs/2511.21644
6. J. Mandrysch, R. Simmons, M. Navascués, Causality and realizability of local operations in quantum field theory, arXiv:2607.12976. https://arxiv.org/abs/2607.12976
7. I. Jubb, Causal state updates in real scalar quantum field theory, Phys. Rev. D 105, 025003 (2022). DOI: 10.1103/PhysRevD.105.025003. arXiv:2106.09027. https://arxiv.org/abs/2106.09027
8. E. Albertini, I. Jubb, Are Ideal Measurements of Real Scalar Fields Causal?, arXiv:2306.12980. https://arxiv.org/abs/2306.12980
9. C. J. Fewster, R. Verch, Quantum fields and local measurements, Commun. Math. Phys. 378, 851-889 (2020). DOI: 10.1007/s00220-020-03800-6. arXiv:1810.06512. https://arxiv.org/abs/1810.06512
10. M. Papageorgiou, D. Fraser, Eliminating the 'impossible': Recent progress on local measurement theory for quantum field theory, Found. Phys. 54, 26 (2024). DOI: 10.1007/s10701-024-00756-8. arXiv:2307.08524. https://arxiv.org/abs/2307.08524
11. R. D. Sorkin, Impossible Measurements on Quantum Fields, in Directions in General Relativity (eds. B. L. Hu, T. A. Jacobson), Cambridge University Press, Cambridge, 1993, pp. 293-305. arXiv:gr-qc/9302018. https://arxiv.org/abs/gr-qc/9302018
12. R. Simmons, Causality is rare: some topological properties of causal quantum channels, arXiv:2603.25315. https://arxiv.org/abs/2603.25315
13. J. Polo-Gómez, T. R. Perche, E. Martín-Martínez, State updates and useful qubits in relativistic quantum information, arXiv:2506.18906. https://arxiv.org/abs/2506.18906
14. F. Daem, L. Ballesteros Ferraz, A. Zampeli, A. Matzkin, Sensing relativistic quantum fields with minimally perturbing local measurements, arXiv:2607.17920. https://arxiv.org/abs/2607.17920