betaFeedback

WorksPhysics and Astronomy

Quantum Reference Frames, Locality, and Microcausality in Quantum Field Theory

Abstract

As a consequence of the relativistic spectrum condition, quantum reference frames for translations on Minkowski space are never described by projection-valued measures, and in many cases fail a weaker spacetime localisation property known as the norm-11 condition. This has several implications for the local and observable structure of quantum field theories described relative to even the best available frame: for a large class of frame states, strict locality is replaced by almost locality, and microcausality by commutators that decay rapidly with increasing separation, corresponding to an effective smearing of the light cone by the localisation density of the frame state. Operations built from finitely many almost local Kraus operators, for example nonselective measurements of relativised local observables, satisfy an approximate no-signalling bound, with a residue attributable to the localisation uncertainty of the frame. Whether causality in its strongest operational form – non-disturbance under all physically realisable operations – survives relativisation is left open. An upshot of our finding is that quantum field theories formulated relative to a quantum reference frame cannot rest on the traditional axioms, and some general relational axiomatisation must replace at least some of them by taking into account physically realistic properties of the frame.

Introduction

If a physical system possesses a symmetry, every preparation procedure is accompanied by an orbit of others, related by the symmetry. The same applies to every measurement procedure. Specifying the nature of the procedure only specifies the orbit, not a particular instance within it. Without a way to break the symmetry and identify which preparation procedure on the orbit was used, the experimenter is restricted to those measurement procedures that are invariant with respect to the symmetry. In other words, physical observables are invariants.

One way to extend the range of available observables is to introduce a reference frame - a physical system that shares the same symmetry as the original, for then the invariant observables of the joint system become available. For example, while the position of a single particle in Euclidean space is not translationally invariant, the relative position of the particle with respect to a reference particle is invariant under common translations of both particles. If the reference particle is in a sufficiently well localised state, the relative position behaves, to good approximation, like the position of the original particle; the relative position is the translation invariant object that is measured in practice, with the absolute position being a good proxy in the limit of a perfectly localised reference.

Let us describe how this programme is implemented within quantum theory. Here we consider a quantum system S\mathcal{S}, and a reference R\mathcal{R}, itself a quantum system. For this reason, R\mathcal{R} is called a quantum reference frame (QRF). We suppose that both S\mathcal{S} and R\mathcal{R} admit a symmetry described by a symmetry group GG, which is assumed to be locally compact. The mathematical description of S\mathcal{S} and R\mathcal{R}, and of S+R\mathcal{S}+\mathcal{R}, includes quantities that would be observable in the absence of the symmetry. For want of a better term, we temporarily describe these as ‘potential observables’. If S\mathcal{S} is a quantum-mechanical system, one may always find an ideal QRF R\mathcal{R} so that the algebra of potential observables of S\mathcal{S} is isomorphic to an algebra of invariant relative observables of S+R\mathcal{S}+\mathcal{R} (e.g. [1, 2, 3, 4, 5, 6]), and this algebra can be recovered arbitrarily precisely via a physically well motivated conditional expectation, which gives the description of the system contingent on the specification of the state of the frame. This indicates that for quantum mechanical systems, it is to some degree a matter of taste whether to include the frame in the description, or not: in this sense, all potential observables become truly observable. Reference frames that are described by quantum mechanical systems are now well studied (e.g., [1, 4, 6, 7, 8, 9, 10]).

The situation changes when considering quantum field theory, where the axiomatic basis for the theory puts more stringent conditions on the frame; here the ideality property may be lost, and the above-mentioned isomorphism is typically no longer tenable. More precisely, given a locally compact group GG with a strongly continuous unitary representation URU_{\mathcal{R}} in some Hilbert space HR\mathcal{H}_{\mathcal{R}}, a (principal) quantum reference frame is defined to be a system of covariance based on GG, that is, the triple (HR,UR,E)(\mathcal{H}_{\mathcal{R}}, U_{\mathcal{R}}, E), where EE is a (possibly subnormalised) positive operator valued measure satisfying the covariance condition

U(g)E(Δ)U(g)∗=E(g⋅Δ)for all g∈G,Δ∈Bor⁡(G).(1)U(g)E(\Delta)U(g)^* = E(g \cdot\Delta) \quad\text{for all } g \in G, \Delta\in\operatorname{Bor}(G). \tag*{(1)}

From here, the relativisation map (e.g. [2, 3]) Υ:B(HS)→B(HS⊗HR)\mathsf{\Upsilon}: \mathcal{B}(\mathcal{H}_{\mathcal{S}}) \to\mathcal{B}(\mathcal{H}_{\mathcal{S}} \otimes\mathcal{H}_{\mathcal{R}}) is given as

Υ(A)=∫GU(g)AU(g)∗⊗E(dg),(2)\mathsf{\Upsilon}(A) = \int_G U(g)AU(g)^* \otimes E(\mathrm{d}g), \tag*{(2)}

where the integral is understood in an appropriate manner ([3, 11, 12]).

Relativisation is understood as giving a description of S\mathcal{S} relative to the frame EE; Υ\mathsf{\Upsilon} is a normal completely positive subunital map whose image lies in the invariant algebra B(HS⊗HR)G\mathcal{B}(\mathcal{H}_{\mathcal{S}} \otimes\mathcal{H}_{\mathcal{R}})^G, and an injective ∗*-homomorphism if EE is projection-valued. Those operators in B(HS)\mathcal{B}(\mathcal{H}_{\mathcal{S}}) not invariant under GG are understood as not representing bona fide physical quantities, and acquire their physical meaning only relative to an appropriate frame. Given that one of the motivations for introducing quantum reference frames is based on the view that the frame should be described by the same theory as the system of interest, it is important to construct quantum reference frames appropriate to quantum field theory, ideally that are themselves QFTs.

In this paper, we study quantum reference frames in Minkowski space that are covariant under (4-)translation, hence with G=R4G = \mathbb{R}^4. It is known (due to a result of Toller, [13]; see also [14]) that no POVM satisfying the covariance requirement (I.1) plus the spectrum condition can be projection-valued, and therefore that multiplication is not preserved by Υ\mathsf{\Upsilon} in general, and some extra ‘fuzziness’ is present. Moreover, these POVMs typically do not satisfy a weaker ‘localisability’ condition, known as the norm-1 property, which allows the system algebra to be recovered up to arbitrarily good approximation under a natural conditional expectation. Our assumptions are weaker than those of Wightman/AQFT in Minkowski, so although our QRFs are not constructed from systems of fields, they would certainly apply in that setting.

Our main findings are as follows. Under the spectrum condition, no translation-covariant frame is projection-valued (Corollary II.2, following Toller [13]), and a quantitative bound rules out even the norm-1 property for a large class of frames (Theorem II.7). Relative to a frame whose localisation density is a Schwartz function, relativised local observables are almost local (Proposition III.4; see [15, 16]), their commutators at causal separation LL are O(L−∞)O(L^{-\infty}) (Lemma III.5), and the sharp light cone is replaced by its smearing with the autocorrelation of the localisation density (Remark III.7). Causality nonetheless survives in some form: relativisation preserves compatibility (Proposition IV.2), and operations with finitely many almost local Kraus operators, for instance the Lüders measurements of relativised local observables, disturb causally distant relativised observables only at order O(L−∞)O(L^{-\infty}) (Proposition IV.4, Corollary IV.7), locality of operations being, under Haag duality, equivalent to non-signalling (Remark IV.5). Finally, accurate recovery of local observables requires the frame to resolve small translations, and hence to carry correspondingly high energies (Proposition IV.8).

A complementary, axiomatic development has recently been given by Fedida and Glowacki [17], who formulate relational covariance and causality conditions for Poincaré-covariant frame observables. Their causality results are built on frame preparations with compactly supported localisation distributions; however, as follows from Proposition II.1 below, no such preparations exist once the frame itself satisfies the spectrum condition usually assumed in QFT [15, 18]. By contrast, in this work we enforce the spectrum condition at the expense of microcausality but, as we shall see, this does not necessarily entail acausal signalling.

Quantum Reference Frames for Translations in Minkowski space

Here we present Toller’s construction [13] of translation-covariant POVMs on (here, 1 + 3-dimensional) Minkowski space MM, adding some further particulars where they may be helpful. Throughout, the Minkowski product is k⋅x=k0x0−k⃗⋅x⃗k \cdot x = k^0x^0 - \vec{k} \cdot\vec{x}, and Fourier transforms of vector-valued functions are normalised so that (F−1F)(x)=∫R4F(k)d4k/(2π)4(\mathcal{F}^{-1}\mathcal{F})(x) = \int_{\mathbb{R}^4} \mathcal{F}(k)\mathrm{d}^4 k/(2\pi)^4, unitary from L2(R4,d4k/(2π)4)L^2(\mathbb{R}^4,\mathrm{d}^4k/(2\pi)^4) to L2(M,d4x)L^2(M,\mathrm{d}^4x).

Let H\mathcal{H} be a complex, separable Hilbert space carrying a strongly continuous unitary representation of translations

U:R4→U(H),U(x)=eixμPμ,(3)U:\mathbb{R}^4 \to\mathcal{U}(\mathcal{H}), \qquad U(x) = e^{i x_\mu P^\mu}, \tag*{(3)}

whose generators satisfy the spectrum condition - that the joint spectrum of the PμP^\mu is contained in V+V^+, the closed forward causal cone. We seek a POVM E:Bor⁡(M)→B(H)+E:\operatorname{Bor}(M) \to\mathcal{B}(\mathcal{H})^+ satisfying:

U(x)E(Δ)U(x)∗=E(Δ+x),(4)U(x)E(\Delta)U(x)^* = E(\Delta+ x), \tag*{(4)}
0≠E(M)≤1H.(5)0 \ne E(M) \leq\mathbb{1}_{\mathcal{H}}. \tag*{(5)}

Note that the usual normalisation E(M)=1HE(M)=\mathbb{1}_{\mathcal{H}} is too strong and is replaced by subnormalisation (II.3). If E(Δ)E(\Delta) is interpreted as an effect associated with a certain event occurring in Δ\Delta (or a measurement returning a value in Δ\Delta), then E(M)E(M) is the effect corresponding to the event occurring at all (or that the measurement succeeds in returning a result). As will be seen below (Theorem II.4), EE vanishes on the singular subspace Hsing\mathcal{H}_{\mathrm{sing}} of the energy-momentum spectral measure, which contains the vacuum and any single-particle states. Vectors in Hsing\mathcal{H}_{\mathrm{sing}} therefore do not define events in this sense; more generally, writing ψ=ψac+ψsing\psi=\psi_{\mathrm{ac}}+\psi_{\mathrm{sing}}, one has (ψ,E(M)ψ)=(ψac,E(M)ψac)≤∥ψac∥2(\psi,E(M)\psi)=(\psi_{\mathrm{ac}},E(M)\psi_{\mathrm{ac}})\leq\lVert\psi_{\mathrm{ac}}\rVert^{2}, so the singular component of a state contributes only to the ‘non-occurrence event’.

Furthermore, in any QFT with a translation-invariant vacuum vector that is separating for the local algebras (as guaranteed by the Reeh–Schlieder theorem), no nonzero E(Δ)E(\Delta) can belong to a local algebra, and under the cluster property not even to the quasilocal algebra [19]. The next proposition collects two related facts which together are the foundation of much of what follows. Both are due to Toller [13]; for part (ii) we write the proof out in detail.

Proposition II.1. Let E:Bor⁡(M)→B(H)E:\operatorname{Bor}(M)\to\mathcal{B}(\mathcal{H}) be a POVM satisfying (II.2)–(II.3).

(i) For every nonempty open set Δ⊂M\Delta\subset M, the positive operator E(Δ)E(\Delta) is non-zero.

(ii) For every nonempty open set Δ⊂M\Delta\subset M and every ψ∈H\psi\in\mathcal{H},

(ψ,E(Δ)ψ)=0⟹(ψ,E(M)ψ)=0.(6)(\psi,E(\Delta)\psi)=0\quad\Longrightarrow\quad(\psi,E(M)\psi)=0. \tag*{(6)}

Proof. (i) Suppose E(Δ)=0E(\Delta)=0 for some nonempty open Δ\Delta. Choose ϵ>0\epsilon>0 such that Δ\Delta contains an open ball BϵB_{\epsilon} of radius ϵ\epsilon. Countably many translates of BϵB_{\epsilon} cover R4\mathbb{R}^{4}, and by (II.2),

E(Bϵ+x)=U(x)E(Bϵ)U(x)∗≤U(x)E(Δ)U(x)∗=0.E(B_{\epsilon}+x)=U(x)E(B_{\epsilon})U(x)^{*}\leq U(x)E(\Delta)U(x)^{*}=0.

Countable additivity of EE then gives E(M)=0E(M)=0, contradicting (II.3). (ii) Suppose (ψ,E(Δ)ψ)=0(\psi,E(\Delta)\psi)=0 for some nonempty open Δ\Delta, and choose nonempty open sets Δ′,Δ′′\Delta^{\prime},\Delta^{\prime\prime} with Δ′+Δ′′⊆Δ\Delta^{\prime}+\Delta^{\prime\prime}\subseteq\Delta (two balls of half the radius of a ball contained in Δ\Delta suffice). For x∈Δ′′x\in\Delta^{\prime\prime} we have Δ′+x⊆Δ\Delta^{\prime}+x\subseteq\Delta, and therefore monotonicity of the POVM together with (II.2) gives

0≤∥E(Δ′)1/2U(x)∗ψ∥2=(ψ,E(Δ′+x)ψ)≤(ψ,E(Δ)ψ)=0,(7)0\leq\lVert E(\Delta^{\prime})^{1/2}U(x)^{*}\psi\rVert^{2}=(\psi,E(\Delta^{\prime}+x)\psi)\leq(\psi,E(\Delta)\psi)=0, \tag*{(7)}

so E(Δ′)1/2U(x)∗ψ=0E(\Delta^{\prime})^{1/2}U(x)^{*}\psi=0 for all x∈Δ′′x\in\Delta^{\prime\prime}. Now, fix any φ∈H\varphi\in\mathcal{H} and define

f(z)=(E(Δ′)1/2φ,e−iz⋅Pψ),z∈C4.f(z)=(E(\Delta^{\prime})^{1/2}\varphi,e^{-iz\cdot P}\psi),\qquad z\in\mathbb{C}^{4}.

The spectrum condition σ(P)⊆V+\sigma(P)\subseteq V^{+} implies that e−iz⋅Pψe^{-iz\cdot P}\psi is well defined as an H\mathcal{H}-valued holomorphic function of zz in the lower tube T−=R4+i(−V)T_{-}=\mathbb{R}^{4}+i(-V), where VV denotes the interior of V+V^{+}, with continuous boundary values on R4\mathbb{R}^{4} (by strong continuity of UU and uniform boundedness of ∥e−iz⋅Pψ∥\lVert e^{-iz\cdot P}\psi\rVert on T−‾\overline{T_{-}}). Hence ff is holomorphic in T−T_{-}, continuous on the boundary, and by (II.5) vanishing on Δ′′\Delta^{\prime\prime}. Setting f∗(z)=f(z‾)‾f^{*}(z)=\overline{f(\overline{z})} gives a holomorphic function on the upper tube T+=R4+iVT_{+}=\mathbb{R}^{4}+iV whose boundary values agree with those of ff on Δ′′\Delta^{\prime\prime}. The edge-of-the-wedge theorem [18] then provides a holomorphic continuation across Δ′′\Delta^{\prime\prime}, giving a single holomorphic function on a connected open set containing T−∪Δ′′∪T+T_{-}\cup\Delta^{\prime\prime}\cup T_{+}. As this continuation vanishes on Δ′′\Delta^{\prime\prime} and the domain is connected, it vanishes identically. In particular, f(x)=0f(x)=0 for every x∈R4x\in\mathbb{R}^{4}, and since φ∈H\varphi\in\mathcal{H} was arbitrary, E(Δ′)1/2U(x)∗ψ=0E(\Delta^{\prime})^{1/2}U(x)^{*}\psi=0 for every x∈R4x\in\mathbb{R}^{4}. Equivalently,

(ψ,E(Δ′+x)ψ)=0for all x∈R4.(\psi,E(\Delta^{\prime}+x)\psi)=0\quad\text{for all }x\in\mathbb{R}^{4}.

Applying the covering of part (i) to Δ′\Delta^{\prime} now yields (ψ,E(M)ψ)=0(\psi,E(M)\psi)=0.

Part (ii) of Proposition II.1 is enough to rule out projection-valued EE:

Corollary II.2. No POVM E:Bor⁡(M)→B(H)E:\operatorname{Bor}(M)\to\mathcal{B}(\mathcal{H}) satisfying (II.2)–(II.3) is projection-valued.

Proof. Suppose, for a contradiction, that EE is a PVM. For any nonempty open Δ⊂M\Delta\subset M and any ψ∈ker⁡E(Δ)\psi\in\ker E(\Delta), (ψ,E(Δ)ψ)=0(\psi,E(\Delta)\psi)=0, so by Proposition II.1(ii), E(M)ψ=0E(M)\psi=0. Thus ker⁡E(Δ)⊆ker⁡E(M)\ker E(\Delta)\subseteq\ker E(M), equivalently ran⁡E(M)⊆ran⁡E(Δ)\operatorname{ran}E(M)\subseteq\operatorname{ran}E(\Delta). Monotonicity gives the reverse inclusion. As both are projections (since EE is assumed PVM), E(Δ)=E(M)E(\Delta)=E(M) for every nonempty open Δ\Delta. Now choose any bounded open Δ\Delta; then M∖ΔM\setminus\Delta is open and nonempty, so Proposition II.1(i) gives E(M∖Δ)≠0E(M\setminus\Delta)\neq0. But E(M∖Δ)=E(M)−E(Δ)≤E(M)−E(Δ)=0E(M\setminus\Delta)=E(M)-E(\Delta)\leq E(M)-E(\Delta)=0, a contradiction.

We now turn to the structure of translation-covariant POVMs. The spectral theorem for the commuting self-adjoint operators PμP^{\mu} provides a direct integral decomposition

H≅∫R4⊕dμ(k)H(k),(8)\mathcal{H}\cong\int_{\mathbb{R}^{4}}^{\oplus}d\mu(k)\mathcal{H}(k), \tag*{(8)}

where μ\mu is supported in V+V^{+}. The decomposition μ=μac+μsing\mu=\mu_{\mathrm{ac}}+\mu_{\mathrm{sing}} induces a corresponding orthogonal decomposition H=Hac⊕Hsing\mathcal{H}=\mathcal{H}_{\mathrm{ac}}\oplus\mathcal{H}_{\mathrm{sing}}; the vacuum and any single-particle sectors are contained in Hsing\mathcal{H}_{\mathrm{sing}}, on which E(M)E(M) vanishes. After rescaling the inner products on H(k)\mathcal{H}(k) if necessary, the absolutely continuous part can be written as

Hac=∫Vac+⊕d4k(2π)4H(k),\mathcal{H}_{\mathrm{ac}}=\int_{V_{\mathrm{ac}}^{+}}^{\oplus}\frac{\mathrm{d}^{4}k}{(2\pi)^{4}}\mathcal{H}(k),

where Vac+=supp⁡μacV_{\mathrm{ac}}^{+}=\operatorname{supp}\mu_{\mathrm{ac}}. In a theory with mass gap, σ(PμPμ)⊂{0}∪[m2,∞)\sigma(P_{\mu}P^{\mu})\subset\{0\}\cup[m^{2},\infty) with m>0m>0, one has Vac+⊂{k∈V+:kμkμ≥4m2}V_{\mathrm{ac}}^{+}\subset\{k\in V^{+}:k_{\mu}k^{\mu}\ge4m^{2}\}.

Lemma II.3 (Toller, [13 Prop. 2]). For integers r≥0r\ge0 write ∥ψ∥r:=∥(1+P0)rψ∥H\lVert\psi\rVert_{r}:=\lVert(1+P^{0})^{r}\psi\rVert_{\mathcal{H}}, and let D:=⋂r≥0D((1+P0)r)D:=\bigcap_{r\ge0}D((1+P^{0})^{r}), equipped with the Fréchet topology generated by the seminorms {∥⋅∥r}r=0,1,2,…\{\lVert\cdot\rVert_{r}\}_{r=0,1,2,\ldots}; DD is dense and translation-invariant in H\mathcal{H}. For every φ,ψ∈D\varphi,\psi\in D, the complex measure Δ↦(φ,E(Δ)ψ)\Delta\mapsto(\varphi,E(\Delta)\psi) is absolutely continuous with respect to Lebesgue measure on MM, and is represented by a smooth density ρφ,ψ∈C∞(M)\rho_{\varphi,\psi}\in C^{\infty}(M):

(φ,E(Δ)ψ)=∫Δρφ,ψ(x)d4xfor all Borel Δ⊆M.(9)(\varphi,E(\Delta)\psi)=\int_{\Delta}\rho_{\varphi,\psi}(x)\mathrm{d}^{4}x\qquad\text{for all Borel }\Delta\subseteq M. \tag*{(9)}

The density is sesquilinear in (φ,ψ)(\varphi,\psi), positive in the sense that ρψ,ψ(x)≥0\rho_{\psi,\psi}(x)\ge0, and satisfies the covariance relation

ρU(y)φ,U(y)ψ(x+y)=ρφ,ψ(x),x,y∈M,(10)\rho_{U(y)\varphi,U(y)\psi}(x+y)=\rho_{\varphi,\psi}(x),\qquad x,y\in M, \tag*{(10)}

together with the bound

∣ρφ,ψ(x)∣≤24∥φ∥4∥ψ∥4,x∈M.(11)\lvert\rho_{\varphi,\psi}(x)\rvert\le2^{4}\lVert\varphi\rVert_{4}\lVert\psi\rVert_{4},\qquad x\in M. \tag*{(11)}

In particular, ∫Mρψ,ψ(x)d4x=(ψ,E(M)ψ)≤∥ψ∥2\int_{M}\rho_{\psi,\psi}(x)\mathrm{d}^{4}x=(\psi,E(M)\psi)\le\lVert\psi\rVert^{2}.

The next theorem gives a basis-free formulation of Toller’s representation of the density ρφ,ψ\rho_{\varphi,\psi} of translation-covariant POVMs, which can be cast as a Naimark dilation.

Theorem II.4. Let E:Bor⁡(M)→B(H)E:\operatorname{Bor}(M)\to\mathcal{B}(\mathcal{H}) be a POVM satisfying (II.2)–(II.3). Then EE vanishes on Hsing\mathcal{H}_{\mathrm{sing}}, and there exist a Hilbert space H~\widetilde{\mathcal{H}} and a measurable family of contractions

K(k):H(k)→H~,∥K(k)∥≤1 for a.e. k∈Vac+,K(k):\mathcal{H}(k)\to\widetilde{\mathcal{H}},\qquad\lVert K(k)\rVert\le1\ \text{for a.e. }k\in V_{\mathrm{ac}}^{+},

such that the L2L^{2} map

Wψ=F−1(K(⋅)ψ(⋅))(12)W\psi=\mathcal{F}^{-1}(K(\cdot)\psi(\cdot)) \tag*{(12)}

defines a contraction W:Hac→L2(M,H~)W:\mathcal{H}_{\mathrm{ac}}\to L^{2}(M,\widetilde{\mathcal{H}}), and the POVM EE is the (subnormalised, since WW is a contraction) compression of the canonical dilation on L2(M;H~)L^{2}(M;\widetilde{\mathcal{H}}):

E(Δ)=W∗MχΔW,Δ∈Bor⁡(M),(13)E(\Delta)=W^{*}M_{\chi_{\Delta}}W,\qquad\Delta\in\operatorname{Bor}(M), \tag*{(13)}

where MχΔM_{\chi_{\Delta}} denotes multiplication by the indicator function χΔ\chi_{\Delta} on L2(M;H~)L^{2}(M;\widetilde{\mathcal{H}}). Equivalently,

(φ,E(Δ)ψ)=∫Δ((Wφ)(x),(Wψ)(x))H~d4x(14)(\varphi,E(\Delta)\psi)=\int_{\Delta}((W\varphi)(x),(W\psi)(x))_{\widetilde{\mathcal{H}}}\mathrm{d}^{4}x \tag*{(14)}

for all φ,ψ∈H\varphi,\psi\in\mathcal{H} and Borel Δ⊆M\Delta\subseteq M, where WW is extended by zero on Hsing\mathcal{H}_{\mathrm{sing}}. Conversely, any such measurable family {K(k)}\{K(k)\} of contractions, not vanishing almost everywhere, defines a translation-covariant POVM (vanishing on Hsing\mathcal{H}_{\mathrm{sing}}) via (II.10)–(II.11). Moreover, E(M)=ΠacE(M)=\Pi_{\mathrm{ac}} (the projection onto Hac\mathcal{H}_{\mathrm{ac}}) if and only if almost every K(k)K(k) is an isometry.

Proof. Translating Toller’s component formula [13 §3] into our notation, the density of any translation-covariant POVM satisfying (II.2)–(II.3) vanishes when either argument lies in Hsing\mathcal{H}_{\mathrm{sing}}, and on Hac\mathcal{H}_{\mathrm{ac}} takes the form

ρφ,ψ(x)=⟨∫Vac+e−ik⋅xK(k)φ(k)d4k(2π)4,∫Vac+e−ik⋅xK(k)ψ(k)d4k(2π)4⟩H~\rho_{\varphi,\psi}(x)=\left\langle\int_{V_{\mathrm{ac}}^{+}}e^{-i k\cdot x}K(k)\varphi(k)\frac{\mathrm{d}^{4}k}{(2\pi)^{4}},\int_{V_{\mathrm{ac}}^{+}}e^{-i k\cdot x}K(k)\psi(k)\frac{\mathrm{d}^{4}k}{(2\pi)^{4}}\right\rangle_{\widetilde{\mathcal{H}}}

for some Hilbert space H~\widetilde{\mathcal{H}} and measurable field of contractions K(k):H(k)→H~K(k):\mathcal{H}(k)\to\widetilde{\mathcal{H}} in the usual direct-integral sense, with ∥K(k)∥≤1\lVert K(k)\rVert\le1 a.e. on V+acV_{+}^{\mathrm{ac}}. Defining W:Hac→L2(M;H~)W:\mathcal{H}_{\mathrm{ac}}\to L^{2}(M;\widetilde{\mathcal{H}}) by (II.10), this reads ρφ,ψ(x)=((Wφ)(x),(Wψ)(x))H~\rho_{\varphi,\psi}(x)=((W\varphi)(x),(W\psi)(x))_{\widetilde{\mathcal{H}}}. Plancherel’s theorem gives

∥Wψ∥L2(M;H~)2=∫V+ac∥K(k)ψ(k)∥H~2d4k(2π)4≤∫V+ac∥ψ(k)∥H(k)2d4k(2π)4=∥ψ∥H2.\lVert W\psi\rVert_{L^{2}(M;\widetilde{\mathcal{H}})}^{2}=\int_{V_{+}^{\mathrm{ac}}}\lVert K(k)\psi(k)\rVert_{\widetilde{\mathcal{H}}}^{2}\frac{\mathrm{d}^{4}k}{(2\pi)^{4}}\leq\int_{V_{+}^{\mathrm{ac}}}\lVert\psi(k)\rVert_{\mathcal{H}(k)}^{2}\frac{\mathrm{d}^{4}k}{(2\pi)^{4}}=\lVert\psi\rVert_{\mathcal{H}}^{2}.

so WW is a contraction, and integration of ρφ,ψ\rho_{\varphi,\psi} over Δ\Delta gives (II.12), equivalently the Naimark-dilation formula (II.11). In particular E(M)=W∗WE(M)=W^{*}W; hence E(M)=ΠacE(M)=\Pi_{\mathrm{ac}} iff ∥Wψ∥=∥ψ∥\lVert W\psi\rVert=\lVert\psi\rVert for every ψ∈Hac\psi\in\mathcal{H}_{\mathrm{ac}}, which by the Plancherel identity above is equivalent to K(k)∗K(k)=1H(k)K(k)^{*}K(k)=\mathbf{1}_{\mathcal{H}(k)} a.e., i.e., to almost every K(k)K(k) being an isometry. The converse direction - that any measurable field K(k)K(k) of contractions, in the usual direct-integral sense, defines a translation-covariant POVM via (II.10)–(II.11) - is a direct computation.

If H~\widetilde{\mathcal{H}} is chosen minimal, in the sense that the ranges of the K(k)K(k) together span a dense subspace, the pair (H~,K)(\widetilde{\mathcal{H}},K) is unique up to a unitary on H~\widetilde{\mathcal{H}}.

Remark II.5. On DD, the L2L^{2} equivalence class WψW\psi has a smooth pointwise representative given by

(Wψ)(x)=hU(x)∗ψ,hψ=∫V+acK(k)ψ(k)d4k(2π)4,(W\psi)(x)=hU(x)^{*}\psi,\qquad h\psi=\int_{V_{+}^{\mathrm{ac}}}K(k)\psi(k)\frac{\mathrm{d}^{4}k}{(2\pi)^{4}},

to be understood weakly, or as the value at x=0x=0 of the Fourier transform on DD. In general hh does not extend to a bounded operator Hac→H~\mathcal{H}_{\mathrm{ac}}\to\widetilde{\mathcal{H}}. The boundedness or otherwise of hh will play a structural role in the next subsection.

Localisability

Theorem II.4 allows us to address whether translation-covariant POVMs admit the norm-1 property.

Definition II.6. A POVM E:Bor⁡(M)→B(H)E:\operatorname{Bor}(M)\to\mathcal{B}(\mathcal{H}) satisfies the norm-1 property if ∥E(Δ)∥op=1\lVert E(\Delta)\rVert_{\mathrm{op}}=1 for every nonempty open Δ⊆M\Delta\subseteq M.

This has two equivalent formulations (i) for every nonempty open Δ⊆M\Delta\subseteq M there is a sequence of unit vectors (φn)⊂H(\varphi_n)\subset\mathcal{H} for which lim⁡n→∞(φn,E(Δ)φn)=1\lim_{n\to\infty}(\varphi_n,E(\Delta)\varphi_n)=1 (ii) for every nonempty open Δ⊆M\Delta\subseteq M and any ϵ>0\epsilon>0 there is a unit vector φϵ∈H\varphi_\epsilon\in\mathcal{H} for which (φϵ,E(Δ)φϵ)>1−ϵ(\varphi_\epsilon,E(\Delta)\varphi_\epsilon)>1-\epsilon. The norm-1 property is strictly weaker than being projection-valued, but characterises those observables for which arbitrarily sharp localisation in any open region is possible. By Corollary II.2, no translation-covariant POVM on MM is projection-valued, so it is natural to ask whether they may still satisfy the norm-1 condition.

Theorem II.7. Let E:Bor⁡(M)→B(H)E:\operatorname{Bor}(M)\to\mathcal{B}(\mathcal{H}) be a translation-covariant POVM satisfying 0≠E(M)≤1H0\ne E(M)\leq\mathbf{1}_{\mathcal{H}}, and let W:Hac→L2(M,H~)W:\mathcal{H}_{\mathrm{ac}}\to L^{2}(M,\widetilde{\mathcal{H}}) be the contraction as given above, so that E(Δ)=W∗MχΔWE(\Delta)=W^{*}M_{\chi_{\Delta}}W. Suppose that the map

ψ↦(Wψ)(0),\psi\mapsto(W\psi)(0),

initially defined on the domain DD, extends to a bounded operator h:Hac→H~h:\mathcal{H}_{\mathrm{ac}}\to\widetilde{\mathcal{H}}. By translation covariance, this is equivalent to asserting that the pointwise representative of WψW\psi is given by

(Wψ)(x)=hU(x)∗ψ(W\psi)(x)=hU(x)^{*}\psi

for all ψ∈Hac\psi\in\mathcal{H}_{\mathrm{ac}} and x∈Mx\in M, with hh bounded on Hac\mathcal{H}_{\mathrm{ac}}. Then, for every Borel set Δ⊆M\Delta\subseteq M of finite Lebesgue measure,

∥E(Δ)∥op≤∥h∥op2∣Δ∣.(15)\lVert E(\Delta)\rVert_{\mathrm{op}}\leq\lVert h\rVert_{\mathrm{op}}^{2}|\Delta|. \tag*{(15)}

Consequently,

sup⁡∣Δ∣≤ϵ∥E(Δ)∥op≤∥h∥op2ϵ,\sup_{|\Delta|\leq\epsilon}\lVert E(\Delta)\rVert_{\mathrm{op}}\leq\lVert h\rVert_{\mathrm{op}}^{2}\epsilon,

and EE does not have the norm-1 property.

Proof. For the asserted equivalence: given ψ∈Hac\psi\in\mathcal{H}_{\mathrm{ac}}, choose ψn∈D\psi_n \in\mathcal{D} with ψn→ψ\psi_n \to\psi; then hU(⋅)∗ψn→hU(⋅)∗ψhU(\cdot)^*\psi_n \to hU(\cdot)^*\psi uniformly, while Wψn→WψW\psi_n \to W\psi in L2L^2, so by Remark II.5 the continuous function x↦hU(x)∗ψx \mapsto hU(x)^*\psi represents WψW\psi. Now let ψ∈Hac\psi\in\mathcal{H}_{\mathrm{ac}} with ∥ψ∥≤1\lVert\psi\rVert\le1. By the hypothesis and unitarity of U(x)U(x),

∥(Wψ)(x)∥H~2=∥hU(x)∗ψ∥H~2≤∥h∥op2∥ψ∥2≤∥h∥op2\lVert(W\psi)(x)\rVert_{\widetilde{\mathcal{H}}}^{2}=\lVert hU(x)^*\psi\rVert_{\widetilde{\mathcal{H}}}^{2}\le\lVert h\rVert_{\mathrm{op}}^{2}\lVert\psi\rVert^{2}\le\lVert h\rVert_{\mathrm{op}}^{2}

for all x∈Mx \in M. Hence

(ψ,E(Δ)ψ)=∫Δ∥(Wψ)(x)∥H~2 d4x≤∥h∥op2∣Δ∣.(\psi,E(\Delta)\psi)=\int_{\Delta}\lVert(W\psi)(x)\rVert_{\widetilde{\mathcal{H}}}^{2}\,d^{4}x\le\lVert h\rVert_{\mathrm{op}}^{2}|\Delta|.

Taking the supremum over ∥ψ∥≤1\lVert\psi\rVert\le1 (and using that EE vanishes on Hsing\mathcal{H}_{\mathrm{sing}}) gives (II.13). If EE had the norm-1 property, every nonempty open Δ\Delta would satisfy ∥E(Δ)∥=1\lVert E(\Delta)\rVert=1; choosing a nonempty bounded open Δ\Delta with ∣Δ∣<∥h∥op−2|\Delta|<\lVert h\rVert_{\mathrm{op}}^{-2} contradicts the estimate (and if h=0h=0 the conclusion is immediate), so EE does not have the norm-1 property.

A simple, explicit sufficient condition for the hypothesis of Theorem II.7 is square-integrability of the Hilbert–Schmidt norm of KK in momentum space.

Corollary II.8. Suppose

∫Vac+∥K(k)∥HS2d4k(2π)4=:C2<∞.(16)\int_{V_{\mathrm{ac}}^{+}}\lVert K(k)\rVert_{\mathrm{HS}}^{2}\frac{d^{4}k}{(2\pi)^{4}}=:C^{2}<\infty. \tag*{(16)}

Then the evaluation map ψ↦(Wψ)(0)\psi\mapsto(W\psi)(0) extends to a bounded operator h:Hac→H~h:\mathcal{H}_{\mathrm{ac}}\to\widetilde{\mathcal{H}} with ∥h∥op≤C\lVert h\rVert_{\mathrm{op}}\le C. Consequently ∥E(Δ)∥op≤C2∣Δ∣\lVert E(\Delta)\rVert_{\mathrm{op}}\le C^{2}|\Delta| for every Borel Δ⊆M\Delta\subseteq M of finite Lebesgue measure and EE does not have the norm-1 property.

Proof. Choose an orthonormal basis {en}\{e_n\} of H~\widetilde{\mathcal{H}} and let φn∈Hac\varphi_n\in\mathcal{H}_{\mathrm{ac}} be the section k↦K(k)∗en∈H(k)k\mapsto K(k)^*e_n\in\mathcal{H}(k). By Fubini and the definition of the Hilbert–Schmidt norm, we have

∑n∥φn∥H2=∫Vac+∑n∥K(k)∗en∥H(k)2d4k(2π)4=∫Vac+∥K(k)∥HS2d4k(2π)4=C2.\sum_n\lVert\varphi_n\rVert_{\mathcal{H}}^{2}=\int_{V_{\mathrm{ac}}^{+}}\sum_n\lVert K(k)^*e_n\rVert_{\mathcal{H}(k)}^{2}\frac{d^{4}k}{(2\pi)^{4}}=\int_{V_{\mathrm{ac}}^{+}}\lVert K(k)\rVert_{\mathrm{HS}}^{2}\frac{d^{4}k}{(2\pi)^{4}}=C^{2}.

For ψ∈Hac\psi\in\mathcal{H}_{\mathrm{ac}} and x∈Mx\in M,

(en,(Wψ)(x))H~=∫Vac+e−ik⋅x(K(k)∗en,ψ(k))H(k)d4k(2π)4=(φn,e−ix⋅Pψ)H,(e_n,(W\psi)(x))_{\widetilde{\mathcal{H}}}=\int_{V_{\mathrm{ac}}^{+}}e^{-ik\cdot x}(K(k)^*e_n,\psi(k))_{\mathcal{H}(k)}\frac{d^{4}k}{(2\pi)^{4}}=(\varphi_n,e^{-ix\cdot P}\psi)_{\mathcal{H}},

and Cauchy–Schwarz gives ∣(φn,e−ix⋅Pψ)∣≤∥φn∥∥ψ∥\lvert(\varphi_n,e^{-ix\cdot P}\psi)\rvert\le\lVert\varphi_n\rVert\lVert\psi\rVert. Summing,

∥(Wψ)(x)∥H~2=∑n∣(φn,e−ix⋅Pψ)∣2≤∑n∥φn∥H2∥ψ∥H2=C2∥ψ∥H2,\lVert(W\psi)(x)\rVert_{\widetilde{\mathcal{H}}}^{2}=\sum_n\lvert(\varphi_n,e^{-ix\cdot P}\psi)\rvert^{2}\le\sum_n\lVert\varphi_n\rVert_{\mathcal{H}}^{2}\lVert\psi\rVert_{\mathcal{H}}^{2}=C^{2}\lVert\psi\rVert_{\mathcal{H}}^{2},

so the hypothesis of Theorem II.7 holds with ∥h∥op≤C\lVert h\rVert_{\mathrm{op}}\le C.

Corollary II.9. If H~=C\widetilde{\mathcal{H}}=\mathbb{C} and K(k)=⟨φ(k)∣K(k)=\langle\varphi(k)\rvert for some φ∈H\varphi\in\mathcal{H} with ∥φ(k)∥H(k)≤1\lVert\varphi(k)\rVert_{\mathcal{H}(k)}\le1 a.e. and ∥φ∥H≤1\lVert\varphi\rVert_{\mathcal{H}}\le1, then (Wψ)(x)=(φ,e−ix⋅Pψ)(W\psi)(x)=(\varphi,e^{-ix\cdot P}\psi) is bounded in absolute value by ∥ψ∥H\lVert\psi\rVert_{\mathcal{H}}, and

∥E(Δ)∥op≤∣Δ∣\lVert E(\Delta)\rVert_{\mathrm{op}}\le|\Delta|

for every Borel Δ⊆M\Delta\subseteq M with ∣Δ∣<∞|\Delta|<\infty. Thus, EE does not have the norm-1 property.

Remark II.10. Theorem II.7 and Toller’s [13 Prop. 5] are complementary. Theorem II.7 shows that boundedness of hh, which is guaranteed, for instance, by condition (II.14), obstructs the norm-1 property being satisfied, with the quantitative bound (II.13). Toller gives a sufficient condition for the norm-1 property to hold, given by the asymptotic constancy of K(k)K(k) as ∣k∣→∞\lvert k\rvert\to\infty, under which a translation-covariant POVM is in Toller’s terminology definite, which coincides with the norm-1 property. In particular, any definite translation-covariant POVM has unbounded hh, by Theorem II.7. These results thus correspond to opposite regimes of the asymptotic behaviour of KK—rapid decay excludes arbitrarily sharp localisation, and asymptotic constancy guarantees it. The intermediate regime, as well as the question of whether definite POVMs arise from physically natural QRFs in QFT, appears to be open.

Relative Observables

Let EE be a covariant POVM of the form described above, and let A\mathcal{A} be a von Neumann algebra with a σ\sigma-weakly continuous action α:R4→Aut⁡(A)\alpha:\mathbb{R}^{4}\to\operatorname{Aut}(\mathcal{A}). For A∈AA\in\mathcal{A} set

Ψ(A)=∫R4αx(A)⊗dE(x),(17)\mathsf{\Psi}(A)=\int_{\mathbb{R}^{4}}\alpha_x(A)\otimes dE(x), \tag*{(17)}

where the integral is understood weakly; the matrix elements (φ⊗ξ,Ψ(A)(ψ⊗η))=∫R4(φ,αx(A)ψ)(ξ,E(dx)η)(\varphi\otimes\xi,\mathsf{\Psi}(A)(\psi\otimes\eta))=\int_{\mathbb{R}^{4}}(\varphi,\alpha_x(A)\psi)(\xi,E(dx)\eta) determine Ψ(A)\mathsf{\Psi}(A) uniquely. Existence, normality, complete positivity and subunitality of Ψ\mathsf{\Psi} are established in [11] for an arbitrary von Neumann algebra A\mathcal{A} and subnormalised EE (the generality needed in QFT, where local algebras are typically not of type I) provided EE extends to a normal map on L∞(R4)L^\infty(\mathbb{R}^{4}), as it does here by Theorem II.4; explicitly, ΦE(f)=W∗MfW\Phi_E(f)=W^*M_fW with MfM_f denoting multiplication by ff.

Since Ψ(1)=1⊗E(M)\mathsf{\Psi}(\mathbb{1})=\mathbb{1}\otimes E(M) and E(M)≤1E(M)\leq\mathbb{1}, the map Ψ\mathsf{\Psi} is subunital in general. For φ,ψ∈H\varphi,\psi\in\mathcal{H}, put ηφ,ψ=id⁡A⊗ωφ,ψ\eta_{\varphi,\psi}=\operatorname{id}_{\mathcal{A}}\otimes\omega_{\varphi,\psi}, where ωφ,ψ(T)=(φ,Tψ)\omega_{\varphi,\psi}(T)=(\varphi,T\psi). Then

ηφ,ψ(Ψ(A))=∫R4αx(A)(φ,dE(x)ψ)=∫R4ρφ,ψ(x)αx(A)d4x.(18)\eta_{\varphi,\psi}(\mathsf{\Psi}(A))=\int_{\mathbb{R}^{4}}\alpha_x(A)(\varphi,dE(x)\psi)=\int_{\mathbb{R}^{4}}\rho_{\varphi,\psi}(x)\alpha_x(A)d^{4}x. \tag*{(18)}

Here

ρφ,ψ(x)=((Wφ)(x),(Wψ)(x))H~(19)\rho_{\varphi,\psi}(x)=((W\varphi)(x),(W\psi)(x))_{\widetilde{\mathcal{H}}} \tag*{(19)}

is an L1L^{1}-density for arbitrary φ,ψ∈H\varphi,\psi\in\mathcal{H}, and agrees with Toller’s smooth density when φ,ψ∈D\varphi,\psi\in D. For φ=ψ\varphi=\psi and ∥ψ∥=1\lVert\psi\rVert=1, we write ρψ=ρψ,ψ\rho_\psi=\rho_{\psi,\psi}; this is a subprobability density, and we refer to ηψ=ηψ,ψ(Ψ(A))\eta_\psi=\eta_{\psi,\psi}(\mathsf{\Psi}(A)) as a restricted relative observable.

Lemma III.1. If (ψ,E(M)ψ)>0(\psi,E(M)\psi)>0, then the essential support of ρψ\rho_\psi is noncompact.

Proof. Suppose that ρψ\rho_\psi has compact essential support. Then there is a compact set C⊂MC\subset M such that ρψ(x)=0\rho_\psi(x)=0 for almost every x∉Cx\notin C. Since

ρψ(x)=∥(Wψ)(x)∥H~2,(20)\rho_\psi(x)=\lVert(W\psi)(x)\rVert_{\widetilde{\mathcal{H}}}^{2}, \tag*{(20)}

it follows that (Wψ)(x)=0(W\psi)(x)=0 for almost every x∉Cx\notin C. Thus WψW\psi is a compactly supported element of L2(M;H~)L^{2}(M;\widetilde{\mathcal{H}}). By Paley-Wiener, the Fourier transform of a compactly supported L2L^{2}-function has an entire H~\widetilde{\mathcal{H}}-valued extension to C4\mathbb{C}^{4}. On the other hand, by the Toller representation,

Wψ^(k)=K(k)ψ(k)(21)\widehat{W\psi}(k)=K(k)\psi(k) \tag*{(21)}

as an element of L2(R4;H~)L^{2}(\mathbb{R}^{4};\widetilde{\mathcal{H}}), with K(k)ψ(k)K(k)\psi(k) extended by zero outside Vac+V_{\mathrm{ac}}^{+}. Hence Wψ^\widehat{W\psi} vanishes on the nonempty open set R4∖V+\mathbb{R}^{4}\setminus V^{+}. Analyticity forces Wψ^\widehat{W\psi} to vanish identically, so Wψ=0W\psi=0. Consequently

(ψ,E(M)ψ)=∥Wψ∥L2(M;H~)2=0,(22)(\psi,E(M)\psi)=\lVert W\psi\rVert_{L^{2}(M;\widetilde{\mathcal{H}})}^{2}=0, \tag*{(22)}

which proves the result.

The decay of ρψ\rho_\psi at infinity controls the localisation properties of the restricted relative observables. The next proposition gives a simple sufficient condition for rapid decay.

Proposition III.2. Let Fψ:R4→H~F_\psi:\mathbb{R}^{4}\to\widetilde{\mathcal{H}} be given by Fψ(k)=K(k)ψ(k)F_\psi(k)=K(k)\psi(k) for k∈Vac+k\in V_{\mathrm{ac}}^{+} and by Fψ(k)=0F_\psi(k)=0 otherwise. If Fψ∈S(R4;H~)F_\psi\in\mathcal{S}(\mathbb{R}^{4};\widetilde{\mathcal{H}}), then ρψ∈S(R4)\rho_\psi\in\mathcal{S}(\mathbb{R}^{4}).

Proof. By Theorem II.4, Wψ=F−1FψW\psi=\mathcal{F}^{-1}F_\psi as an L2L^{2}-Fourier transform. Since the inverse Fourier transform is an automorphism of S(R4;H~)\mathcal{S}(\mathbb{R}^{4};\widetilde{\mathcal{H}}), one has Wψ∈S(R4;H~)W\psi\in\mathcal{S}(\mathbb{R}^{4};\widetilde{\mathcal{H}}). Moreover

ρψ(x)=∥(Wψ)(x)∥H~2.(23)\rho_\psi(x)=\lVert(W\psi)(x)\rVert_{\widetilde{\mathcal{H}}}^{2}. \tag*{(23)}

For every multi-index α\alpha,

∂αρψ=∑β≤αcα,β(∂βWψ,∂α−βWψ)H~,(24)\partial^\alpha\rho_\psi=\sum_{\beta\leq\alpha}c_{\alpha,\beta}(\partial^\beta W\psi,\partial^{\alpha-\beta}W\psi)_{\widetilde{\mathcal{H}}}, \tag*{(24)}

where cα,βc_{\alpha,\beta} is a multinomial coefficient. The Schwartz estimates for WψW\psi, together with Cauchy–Schwarz, imply the corresponding estimates for ρψ\rho_\psi. For example,

∥ρψ∥2γ,α≤∑β≤αcα,β∥Wψ∥γ,β∥Wψ∥γ,α−β,(25)\|\rho_\psi\|_{2\gamma,\alpha} \le\sum_{\beta\le\alpha} c_{\alpha,\beta}\|W\psi\|_{\gamma,\beta}\|W\psi\|_{\gamma,\alpha-\beta}, \tag*{(25)}

where Schwartz seminorms are defined as usual, ∥f∥β,α=sup⁡x∈R4(1+∣x∣β)∣∂αf∣\|f\|_{\beta,\alpha}=\sup_{x\in\mathbb{R}^{4}}(1+|x|^\beta)|\partial^\alpha f| for f∈S(R4)f\in\mathcal{S}(\mathbb{R}^{4}), replacing ∣∂αf∣|\partial^\alpha f| by ∥∂αf∥H~\|\partial^\alpha f\|_{\widetilde{\mathcal{H}}} for f∈S(R4;H~)f\in\mathcal{S}(\mathbb{R}^{4};\widetilde{\mathcal{H}}).

Definition III.3. Let O↦A(O)O\mapsto\mathcal{A}(O) be a local net of von Neumann algebras with quasilocal algebra A\mathcal{A} on Minkowski space MM. We assume throughout that the local algebras possess a common unit id\mathrm{id}, (see [20 p. 849]). An element B∈AB\in\mathcal{A} is almost local if, for some z∈Mz\in M and future-pointing unit timelike vector uu, there is a family of local operators

BR∈A(CR(z,u)),R>0,(26)B_R\in\mathcal{A}(C_R(z,u)),\qquad R>0, \tag*{(26)}

where CR(z,u)=I+(z−Ru)∩I−(z+Ru)C_R(z,u)=I^+(z-Ru)\cap I^-(z+Ru) is the open double-cone with tips at z±Ruz\pm Ru, such that

sup⁡R>0RN∥B−BR∥<∞for all N∈N(27)\sup_{R>0}R^N\|B-B_R\|<\infty\qquad\text{for all }N\in\mathbb{N} \tag*{(27)}

See [16], and cf. Haag’s field version [15 §II.4.1].

The local approximant BRB_R belongs to the local algebra of a double-cone about zz in the rest frame of uu. If BB is almost local then there is a family of local approximants obeying (III.10) and (III.11) for every z∈Mz\in M and future-pointing unit timelike uu. Note that the double-cones have the property CL(z,u)±CL′(z′,u)=CL+L′(z±z′,u)C_L(z,u)\pm C_{L'}(z',u)=C_{L+L'}(z\pm z',u).

Proposition III.4. Let A∈A(O)A\in\mathcal{A}(O) be bounded and local, with O⊂MO\subset M bounded. If ρψ∈S(R4)\rho_\psi\in\mathcal{S}(\mathbb{R}^{4}), then

Bψ(A):=ηψ(Ψ(A))=∫R4ρψ(x)αx(A) d4x(28)B_\psi(A):=\eta_\psi(\Psi(A))=\int_{\mathbb{R}^{4}}\rho_\psi(x)\alpha_x(A)\,\mathrm{d}^{4}x \tag*{(28)}

is almost local.

Proof. Let z∈Oz\in O, fix any unit future-pointing timelike vector uu and suppose that RR is sufficiently large that O⊂CR/2(z,u)O\subset C_{R/2}(z,u), setting BR=0B_R=0 for all smaller values of R>0R>0. Choose χR∈C∞(R4)\chi_R\in C^\infty(\mathbb{R}^{4}) with 0≤χR≤10\le\chi_R\le1, χR(x)=1\chi_R(x)=1 for x∈CR/4(0,u)x\in C_{R/4}(0,u), and χR(x)=0\chi_R(x)=0 for x∉CR/2(0,u)x\notin C_{R/2}(0,u), and set

BR=∫R4χR(x)ρψ(x)αx(A) d4x,(29)B_R=\int_{\mathbb{R}^{4}}\chi_R(x)\rho_\psi(x)\alpha_x(A)\,\mathrm{d}^{4}x, \tag*{(29)}

understood in the weak sense. Since αx(A)∈A(O+x)⊂A(CR/2(z+x,u))\alpha_x(A)\in\mathcal{A}(O+x)\subset\mathcal{A}(C_{R/2}(z+x,u)) and χRρψ\chi_R\rho_\psi is supported in CR/2(0,u)C_{R/2}(0,u), isotony gives BR∈A(CR(z,u))B_R\in\mathcal{A}(C_R(z,u)). Moreover,

∥Bψ(A)−BR∥≤∥A∥∫R4∣1−χR(x)∣∣ρψ(x)∣ d4x≤∥A∥∫R4∖CR/4(0,u)∣ρψ(x)∣ d4x.(30)\|B_\psi(A)-B_R\|\le\|A\|\int_{\mathbb{R}^{4}}|1-\chi_R(x)||\rho_\psi(x)|\,\mathrm{d}^{4}x\le\|A\|\int_{\mathbb{R}^{4}\setminus C_{R/4}(0,u)}|\rho_\psi(x)|\,\mathrm{d}^{4}x. \tag*{(30)}

Since ρψ\rho_\psi is Schwartz, the last integral is O(R−N)O(R^{-N}) for every NN as R→∞R\to\infty (independently of the chosen χR\chi_R).

Lemma III.5. Let B1,B2B_1,B_2 be almost local and fix any future-pointing timelike unit vector uu. Then, for every N∈NN\in\mathbb{N}, there is CN>0C_N>0 such that, for all L>0L>0 and all x1,x2∈Mx_1,x_2\in M for which

x1−x2⊥C2L(0,u),(31)x_1-x_2\perp C_{2L}(0,u), \tag*{(31)}

i.e., x1−x2x_1-x_2 belongs to the causal complement of C2L(0,u)C_{2L}(0,u), one has

∥[αx1(B1),αx2(B2)]∥≤CNL−N.(32)\|[\alpha_{x_1}(B_1),\alpha_{x_2}(B_2)]\|\le C_NL^{-N}. \tag*{(32)}

In particular, ∥[αx1(B1),αx2(B2)]∥=O(L−∞)\|[\alpha_{x_1}(B_1),\alpha_{x_2}(B_2)]\|=O(L^{-\infty}) as L→∞L\to\infty, uniformly in x1,x2x_1,x_2 satisfying (III.15).

Proof. Fixing any z∈Mz \in M and any future-pointing unit timelike vector uu, let Bi,R∈A(CR(z,u))B_{i,R} \in\mathcal{A}(C_R(z,u)) be local approximating families for the BiB_i. Now let L>0L > 0 and let x1,x2∈Mx_1,x_2 \in M with x1−x2x_1-x_2 spacelike to C2L(0,u)C_{2L}(0,u). By covariance, αxi(Bi,L)∈A(CL(z+xi,u))\alpha_{x_i}(B_{i,L}) \in\mathcal{A}(C_L(z+x_i,u)), so locality gives

[αx1(B1,L),αx2(B2,L)]=0,(33)[\alpha_{x_1}(B_{1,L}),\alpha_{x_2}(B_{2,L})]=0, \tag*{(33)}

because every vector in CL(z+x1,u)−CL(z+x2,u)=C2L(x1−x2,u)=x1−x2−C2L(0,u)C_L(z+x_1,u)-C_L(z+x_2,u)=C_{2L}(x_1-x_2,u)=x_1-x_2-C_{2L}(0,u) is spacelike. Hence, writing

[αx1(B1),αx2(B2)]=[αx1(B1−B1,L),αx2(B2)]+[αx1(B1,L),αx2(B2−B2,L)]+[αx1(B1,L),αx2(B2,L)](34)\begin{aligned} [\alpha_{x_1}(B_1),\alpha_{x_2}(B_2)]={}&[\alpha_{x_1}(B_1-B_{1,L}),\alpha_{x_2}(B_2)]\\ &+[\alpha_{x_1}(B_{1,L}),\alpha_{x_2}(B_2-B_{2,L})]\\ &+[\alpha_{x_1}(B_{1,L}),\alpha_{x_2}(B_{2,L})] \tag*{(34)} \end{aligned}

and using that αx1,αx2\alpha_{x_1},\alpha_{x_2} are isometric, we have

∥[αx1(B1),αx2(B2)]∥≤2∥B1−B1,L∥∥B2∥+2∥B1,L∥∥B2−B2,L∥,(35)\lVert[\alpha_{x_1}(B_1),\alpha_{x_2}(B_2)]\rVert\le2\lVert B_1-B_{1,L}\rVert\lVert B_2\rVert+2\lVert B_{1,L}\rVert\lVert B_2-B_{2,L}\rVert, \tag*{(35)}

from which, by (III.11) and ∥B1,L∥≤∥B1∥+∥B1−B1,L∥\lVert B_{1,L}\rVert\le\lVert B_1\rVert+\lVert B_1-B_{1,L}\rVert, we arrive at the claim for L≥1L\ge1; for L<1L<1 it holds trivially, since the left-hand side is at most 2∥B1∥∥B2∥2\lVert B_1\rVert\lVert B_2\rVert.

Corollary III.6. Let Ai∈A(Oi)A_i\in\mathcal{A}(O_i) be bounded local observables, with Oi⊂MO_i\subset M bounded, and suppose that ρψ∈S(R4)\rho_\psi\in\mathcal{S}(\mathbb{R}^4). Put

Bi=ηψ(Ψ(Ai)),i=1,2.(36)B_i=\eta_\psi(\Psi(A_i)),\qquad i=1,2. \tag*{(36)}

Then the conclusion of Lemma III.5 holds for B1,B2B_1,B_2: for every N∈NN\in\mathbb{N} there is CN>0C_N>0 such that

∥[αx1(B1),αx2(B2)]∥≤CNL−N(37)\lVert[\alpha_{x_1}(B_1),\alpha_{x_2}(B_2)]\rVert\le C_NL^{-N} \tag*{(37)}

for all L>0L>0 and x1,x2∈Mx_1,x_2\in M satisfying (III.15).

Proof. This follows from Prop. III.4 and Lemma III.5.

At the formal level, we observe the following. If Φ(f)\Phi(f) is a field, f∈Cc∞(M)f\in C^\infty_c(M), and αx(Φ(f))=Φ(fx)\alpha_x(\Phi(f))=\Phi(f_x) with fx(y)=f(y−x)f_x(y)=f(y-x), then

ηψ(Ψ(Φ(f)))=∫Mρψ(x)Φ(fx) d4x=Φ(ρψ∗f),(38)\eta_\psi(\Psi(\Phi(f)))=\int_M\rho_\psi(x)\Phi(f_x)\,\mathrm{d}^4x=\Phi(\rho_\psi*f), \tag*{(38)}

where

(ρψ∗f)(y)=∫Mρψ(x)f(y−x) d4x.(39)(\rho_\psi*f)(y)=\int_M\rho_\psi(x)f(y-x)\,\mathrm{d}^4x. \tag*{(39)}

Thus a compactly supported test function is replaced by a smeared test function. If ρψ\rho_\psi is Schwartz, then so is ρψ∗f\rho_\psi*f, and the resulting field is almost local in Haag’s sense [15]. This explains formally why exact microcausality is replaced by rapidly decaying commutators. To avoid domain questions for unbounded fields, we now pass to the Weyl algebra. For real-valued f∈Cc∞(M)f\in C^\infty_c(M), let W(f)W(f) denote Weyl operators satisfying

W(f)W(g)=e−iE(f,g)/2W(f+g),(40)W(f)W(g)=e^{-\mathrm{i}E(f,g)/2}W(f+g), \tag*{(40)}

where E(f,g)=∫f(E−g−E+g) dvolE(f,g)=\int f(E^-g-E^+g)\,\mathrm{dvol} and E±E^\pm are retarded/advanced Green operators, and suppose translations act by αx(W(f))=W(fx)\alpha_x(W(f))=W(f_x). Define

Bfψ:=ηψ(Ψ(W(f)))=∫Mρψ(x)W(fx) d4x.(41)B_f^\psi:=\eta_\psi(\Psi(W(f)))=\int_M\rho_\psi(x)W(f_x)\,\mathrm{d}^4x. \tag*{(41)}

Then ∥Bfψ∥≤∫Mρψ(x) d4x≤1\lVert B_f^\psi\rVert\le\int_M\rho_\psi(x)\,\mathrm{d}^4x\le1. The Weyl relations give

[W(fx),W(gy)]=−2isin⁡(12E(fx,gy))W(fx+gy),(42)[W(f_x),W(g_y)]=-2\mathrm{i}\sin\left(\frac{1}{2}E(f_x,g_y)\right)W(f_x+g_y), \tag*{(42)}

and hence

[Bfψ,Bgψ]=−2i∫M×Mρψ(x)ρψ(y)sin⁡(12E(fx,gy))W(fx+gy) d4x d4y.(43)[B_f^\psi,B_g^\psi]=-2\mathrm{i}\int_{M\times M}\rho_\psi(x)\rho_\psi(y)\sin\left(\frac{1}{2}E(f_x,g_y)\right)W(f_x+g_y)\,\mathrm{d}^4x\,\mathrm{d}^4y. \tag*{(43)}

Note that E(fx,gy)E(f_x,g_y) depends only on y−xy-x. Even if ff and gg have causally disjoint supports, the integrand will be nonvanishing for suitable x,yx,y, so microcausality fails. Let

Δf,g:={(x,y)∈M×M:supp⁡fx and supp⁡gy are causally disjoint}.(44)\Delta_{f,g}:=\{(x,y)\in M\times M:\operatorname{supp} f_x\text{ and }\operatorname{supp} g_y\text{ are causally disjoint}\}. \tag*{(44)}

On Δf,g\Delta_{f,g} the integrand vanishes. Therefore

∥[Bfψ,Bgψ]∥≤2(ρψ×ρψ)((M×M)∖Δf,g).(45)\left\lVert[B_f^\psi,B_g^\psi]\right\rVert\le2(\rho_\psi\times\rho_\psi)((M\times M)\setminus\Delta_{f,g}). \tag*{(45)}

This gives a bounded, quantitative version of the preceding formal picture: exact microcausality is replaced by a bound controlled by the probability that two independent frame translations fail to preserve causal disjointness.

Remark III.7. The bound (III.29) entails a statement about the operationally accessible causal structure. For z∈Mz\in M let χf,g(z)∈{0,1}\chi_{f,g}(z)\in\{0,1\} be the set indicator function corresponding to the situation that (supp⁡f)+z(\operatorname{supp} f)+z and supp⁡g\operatorname{supp} g fail to be causally disjoint, and let

κψ(z):=∫Mρψ(x)ρψ(x−z) d4x(46)\kappa_\psi(z):=\int_M \rho_\psi(x)\rho_\psi(x-z)\,\mathrm{d}^{4}x \tag*{(46)}

denote the autocorrelation of the localisation density of the frame state ψ\psi. Then, for translated faf_a, a∈Ma\in M, a change of variables in (III.29) gives

∥[Bfaψ,Bgψ]∥≤2(κψ∗χf,g)(a).(47)\left\lVert[B_{f_a}^\psi,B_g^\psi]\right\rVert\le2(\kappa_\psi*\chi_{f,g})(a). \tag*{(47)}

Relative to the frame, the sharp indicator of causal connectedness is replaced by its convolution with κψ\kappa_\psi: the boundary of the light cone is smeared by the localisation uncertainty of the frame state. By Lemma III.1, ρψ\rho_\psi, and hence κψ\kappa_\psi, has non-compact support, so the smearing cannot be removed even in principle: no frame state restores a sharp causal boundary (although for ρψ∈S(R4)\rho_\psi\in\mathcal{S}(\mathbb{R}^{4}) there is rapid decay).

Commutativity, Compatibility and Disturbance

A simple way to appreciate the operational content of the microcausality condition in standard QFT rests on the Lüders theorem [21, 22]: A nonselective Lüders measurement of a discrete sharp observable A=∑iλiPiA=\sum_i\lambda_iP_i, resulting in a state update ρ↦∑iPiρPi\rho\mapsto\sum_iP_i\rho P_i, leaves the statistics of a second sharp observable BB undisturbed in every state ρ\rho if and only if [A,B]=0[A,B]=0. For observables with continuous spectrum, this holds for every discretisation [23]. For general POVMs commutativity is replaced by compatibility, which implies commutativity only in special cases.

Definition IV.1. Consider POVMs AA and BB on measurable spaces (ΩA,FA)(\Omega_A,\mathcal{F}_A) and (ΩB,FB)(\Omega_B,\mathcal{F}_B) respectively (which for the purposes of this paper, we may assume are given as (R4,Bor⁡(R4))(\mathbb{R}^{4},\operatorname{Bor}(\mathbb{R}^{4}))). AA and BB are called compatible if there is a joint POVM MM on ΩA×ΩB\Omega_A\times\Omega_B such that for all X∈FAX\in\mathcal{F}_A and all Y∈FBY\in\mathcal{F}_B, we have A(X)=M(X×ΩB)A(X)=M(X\times\Omega_B) and B(Y)=M(ΩA×Y)B(Y)=M(\Omega_A\times Y).

This operational definition of compatibility requires the existence of joint probability distributions for AA and BB in all states, with the distributions for AA and BB obtained as marginals. If at least one of AA and BB is projection valued, the compatibility of AA and BB is equivalent to their commutativity (in the sense that [A(X),B(Y)]=0[A(X),B(Y)]=0 for all XX and YY); however, there are many examples of POVMs that are compatible but fail to commute [24, 25].

For POVMs the relationship between compatibility, (non-)disturbance and commutativity is nuanced. If some measurement of AA leaves BB undisturbed, then AA and BB are compatible (the sequential measurement, AA first and then BB, supplies a joint POVM with marginals AA and BB), but the converse fails, and there are compatible pairs for which every measurement of one POVM disturbs the other [26]. Compatibility here is therefore not a statement of relativistic causality; the latter depends on the specifics of the available measurements even for compatible observables. The dependence of causal behaviour on how a measurement is implemented, rather than on the measured observable alone, is already present in ordinary QFT, where microcausality holds exactly. Let A=∑iλiPi∈A(O1)A=\sum_i\lambda_iP_i\in\mathcal{A}(\mathcal{O}_1) be sharp and local and consider the instrument that measures AA and, upon obtaining outcome λi\lambda_i, re-prepares the system in state σi\sigma_i. Here the instrument is then Ii(ρ)=tr⁡[Piρ]σi\mathcal{I}_i(\rho)=\operatorname{tr}[P_i\rho]\sigma_i, and in the Heisenberg picture its nonselective channel is

Λ(B)=∑itr⁡[σiB]Pi.(48)\Lambda(B)=\sum_i\operatorname{tr}[\sigma_iB]P_i. \tag*{(48)}

This instrument measures AA with the same outcome statistics as the Lüders instrument, yet Λ(B)≠B\Lambda(B)\ne B for essentially every BB, independently of the localisation of BB — this measurement signals, even though AA commutes with every causally disjoint observable. Indeed, we see that Λ(B)∈A(O1)\Lambda(B) \in\mathcal{A}(O_1), so B=Λ(B)B = \Lambda(B) implies B∈A(O1)∩A(O2)B \in\mathcal{A}(O_1) \cap\mathcal{A}(O_2), which consists only of multiples of the unit under the extended locality property that holds for local nets of von Neumann algebras under standard conditions [27].

Causality is therefore not a property of the measured observable, but of the measurement operation, and must be formulated in terms of the localisation properties of the operations implementing the measurement. One way to proceed is to assume that the measurement is conducted by a probe QFT locally coupled to the system QFT with the coupled theory also described as a QFT [28], but for the moment we take a more general viewpoint. In Definition IV.3, we give a notion of locality for operations and show that, in the relativised theory, it yields an operational form of microcausality whose accuracy is set by the frame. First, however, we note that relativisation preserves compatibility.

Proposition IV.2. Let AA and BB be compatible POVMs and let T:B(H)→B(K)T : \mathcal{B}(\mathcal{H}) \to\mathcal{B}(\mathcal{K}) be a normal unital completely positive map. Then T∘AT \circ A and T∘BT \circ B are compatible.

The proof follows immediately from noting that if MM is a joint POVM for AA and BB, then T∘MT \circ M is a joint POVM for T∘AT \circ A and T∘BT \circ B [29]. The same conclusion holds for subnormalised POVMs and subunital normal completely positive maps, after making the appropriate changes.

Applying the completely positive map Tψ=ηψ∘YT_\psi= \eta_\psi\circ\mathcal{Y} to a joint PVM makes the contrast between commutativity and compatibility explicit. For causally disjoint ff and gg the Weyl operators W(f)W(f) and W(g)W(g) commute, and there is a joint PVM PP whose marginals are the spectral measures PfP_f and PgP_g of W(f)W(f) and W(g)W(g). By Proposition IV.2, Tψ∘PT_\psi\circ P is a joint POVM for the compatible pair Tψ∘PfT_\psi\circ P_f and Tψ∘PgT_\psi\circ P_g, whose first moments are the non-commuting operators BfψB_f^\psi and BgψB_g^\psi, as seen by the calculation

∫Cλ d(Tψ∘Pf)(λ)=Tψ(∫Cλ dPf(λ))=Tψ(W(f))=Bfψ.(49)\int_{\mathbb{C}} \lambda\,\mathrm{d}(T_\psi\circ P_f)(\lambda) = T_\psi\left(\int_{\mathbb{C}} \lambda\,\mathrm{d}P_f(\lambda)\right) = T_\psi(W(f)) = B_f^\psi. \tag*{(49)}

Thus relativisation (and restriction) may destroy commutativity, but nonetheless preserves compatibility.

We now turn to disturbance. Signalling concerns the nonselective channel defined by an instrument, and so we work directly at the level of channels. In addition to the localisation of observables, the localisation of operators is also part of the philosophy of local quantum physics; operations implemented by elements of a local algebra appear already in Haag and Kastler [20]. That the locality of an observable does not by itself ensure the causal behaviour of its measurement was stressed by Sorkin [30] (see also [31]), and the identification of operations that are both causal and physically realisable has since been pursued from several directions: measurement schemes in which the system is coupled locally to a probe [28, 32]; the direct analysis of state-update rules for scalar fields [33]; the construction of regularised, non-signalling measurements of field operators and their spacetime-extended generalisations [34, 35, 36]; and the general study of causal and realisable local operations [37].

Definition IV.3. By an operation we mean a normal unital completely positive map Λ:B(H)→B(H)\Lambda: \mathcal{B}(\mathcal{H}) \to\mathcal{B}(\mathcal{H}) (a channel, in the Heisenberg picture). Every operation admits a Kraus decomposition

Λ(B)=∑i≥0Ki∗BKi,∑i≥0Ki∗Ki=id,(50)\Lambda(B) = \sum_{i \ge0} K_i^* B K_i, \qquad\sum_{i \ge0} K_i^* K_i = \mathrm{id}, \tag*{(50)}

with both sums converging σ\sigma-weakly [38]. An operation Λ\Lambda is local in a bounded region O⊂MO \subset M if it admits such a decomposition with Ki∈A(O)K_i \in\mathcal{A}(O) for every ii, and almost local if it admits one in which every KiK_i is almost local. The Kraus decomposition of a given operation is not unique; locality and almost locality require one decomposition of the stated form; Remark IV.5 below pertains to every decomposition.

We caution immediately that ‘locality’ of an operation does not imply that it can be realised by local physical interactions within the given region. More comments on this appear below. (Note also that our terminology here differs from that of [37], in which an instrument is said to be local to a compact set KK if its measurement channel acts trivially on the algebra associated with the causal complement of KK; see however Sec. IV of [37] for situations in which their definition implies that the instrument is local, in our sense, to any region containing KK.) The identity channel (a single Kraus operator K0=idK_0 = \mathrm{id}) is local in every region. The Lüders channel of a discrete POVM in A(O1)\mathcal{A}(O_1) — effects A1,…,An∈A(O1)A_1,\ldots,A_n \in\mathcal{A}(O_1) with ∑i=1nAi≤id\sum_{i=1}^{n} A_i \le\mathrm{id}, completed by A0:=id−∑i=1nAiA_0 := \mathrm{id} - \sum_{i=1}^{n} A_i — is local in O1O_1: its Kraus operators are Ai1/2∈A(O1)A_i^{1/2} \in\mathcal{A}(O_1) for 0≤i≤n0 \le i \le n, since the continuous functional calculus preserves A(O1)\mathcal{A}(O_1). Thus the Lüders channel of every discrete local POVM is local in the sense of Definition IV.3. The measure-and-prepare channel of (IV.1) is, by contrast, local in no bounded region, as part (i) of the following proposition shows.

Proposition IV.4. Let Λ\Lambda be an operation with Kraus operators KiK_i as in Definition IV.3.

(i) If Λ\Lambda is local in O1O_1 and O2O_2 is causally disjoint from O1O_1, then Λ(B)=B\Lambda(B) = B for every B∈A(O2)B \in\mathcal{A}(O_2).

(ii) Suppose Λ\Lambda is almost local with a finite Kraus decomposition K0,…,KnK_0,\ldots,K_n, let BB be almost local, fix a future-pointing unit timelike vector uu, and for x∈Mx\in M let Λx:=αx∘Λ∘αx−1\Lambda_x:=\alpha_x\circ\Lambda\circ\alpha_{x}^{-1} denote the translated operation. Then, for every N∈NN\in\mathbb{N}, there is CN>0C_N>0 such that, for all L>0L>0 and all x1,x2∈Mx_1,x_2\in M satisfying (III.15),

∥Λx1(αx2(B))−αx2(B)∥≤CNL−N.(51)\left\|\Lambda_{x_1}\left(\alpha_{x_2}(B)\right)-\alpha_{x_2}(B)\right\|\le C_NL^{-N}. \tag*{(51)}

Proof. Since ∑iKi∗Ki=id\sum_i K_i^*K_i=\mathrm{id}, with all sums converging σ\sigma-weakly,

Λ(B′)−B′=∑iKi∗[B′,Ki].(52)\Lambda(B')-B'=\sum_i K_i^*[B',K_i]. \tag*{(52)}

For (i), take B′=B∈A(O2)B'=B\in\mathcal{A}(O_2); each Ki∈A(O1)K_i\in\mathcal{A}(O_1), so every term vanishes by microcausality. For (ii), the decomposition is finite and ∥Ki∥≤1\|K_i\|\le1, so, using that Λx1\Lambda_{x_1} has Kraus operators αx1(Ki)\alpha_{x_1}(K_i),

∥Λx1(αx2(B))−αx2(B)∥≤∑i=0n∥[αx2(B),αx1(Ki)]∥,(53)\left\|\Lambda_{x_1}\left(\alpha_{x_2}(B)\right)-\alpha_{x_2}(B)\right\|\le\sum_{i=0}^{n}\left\|[\alpha_{x_2}(B),\alpha_{x_1}(K_i)]\right\|, \tag*{(53)}

and Lemma III.5 bounds each term by CN′L−NC'_NL^{-N}, uniformly in x1,x2x_1,x_2 as in the statement. □\square

Remark IV.5. Part (i) admits a converse, showing that the notion of locality in Definition IV.3 is forced by causality. Suppose Λ(B)=B\Lambda(B)=B for every B∈A(O2)B\in\mathcal{A}(O_2) and every bounded O2O_2 causally disjoint from O1O_1. For any Kraus decomposition of Λ\Lambda and any such BB,

∑i(KiB−BKi)∗(KiB−BKi)=Λ(B∗B)−Λ(B∗)B−B∗Λ(B)+B∗B=0,(54)\sum_i (K_iB-BK_i)^*(K_iB-BK_i)=\Lambda(B^*B)-\Lambda(B^*)B-B^*\Lambda(B)+B^*B=0, \tag*{(54)}

since B∗B^* and B∗BB^*B also lie in A(O2)\mathcal{A}(O_2); a vanishing sum of positive terms, whence [Ki,B]=0[K_i,B]=0 for every ii [39, 40]. Thus each KiK_i lies in the commutant of the von Neumann algebra A(O1′)\mathcal{A}(O_1') generated by the local algebras A(O2)\mathcal{A}(O_2) of bounded regions O2O_2 causally disjoint from O1O_1. Where Haag duality holds for O1O_1 (for instance, when O1O_1 is a double cone in a theory satisfying Haag duality), A(O1′)′=A(O1)\mathcal{A}(O_1')'=\mathcal{A}(O_1), so every Kraus operator of Λ\Lambda lies in A(O1)\mathcal{A}(O_1): the operations that cannot signal out of O1O_1 are precisely those local in O1O_1. This observation appears as Proposition 49 of [37]; the underlying fact—that a channel acting as the identity on a ∗*-algebra has Kraus operators in its commutant, is due to Lindblad [41], whose proof rests on the equality situation of the Schwarz inequality for completely positive maps [39]. The same identity, applied to the spectral projections of a sharp observable, yields the nontrivial part of the Lüders theorem.

We note that Proposition IV.4 and Remark IV.5 concern two regions: as Sorkin observed [30] (see [37 Sec. II.E] for a formulation in terms of channels), locality of the Kraus operators alone does not preclude superluminal signalling in protocols involving three or more regions, and the characterisation of the operations that are both physically realisable and fully causal is delicate [28, 33, 37]; these matters lie beyond the scope of this paper, but it is worth emphasising that the same issues also arise in classical field theories [42].

To bring measurements of relativised observables within the scope of part (ii), it remains to verify that the Kraus operators of their Lüders channels are again almost local.

Lemma IV.6. Let EE be an almost local effect. Then E1/2E^{1/2} and (id−E)1/2(\mathrm{id}-E)^{1/2} are almost local.

Proof. Fix z∈Mz\in M and a future-pointing unit timelike vector uu, and let ER∈A(CR(z,u))E_R\in\mathcal{A}(C_R(z,u)) be local approximants, which may be taken self-adjoint, with εR:=∥E−ER∥=O(R−∞)\varepsilon_R:=\|E-E_R\|=O(R^{-\infty}). Let f(t)=min⁡{max⁡{t,0},1}f(t)=\min\{\max\{t,0\},1\}; then FR:=f(ER)∈A(CR(z,u))F_R:=f(E_R)\in\mathcal{A}(C_R(z,u)) are effects with ∥ER−FR∥≤εR\|E_R-F_R\|\le\varepsilon_R by the functional calculus, so ∥E−FR∥≤2εR\|E-F_R\|\le2\varepsilon_R. Since the square root is monotone and vanishes at 00, ∥E1/2−FR1/2∥≤∥E−FR∥1/2\|E^{1/2}-F_R^{1/2}\|\le\|E-F_R\|^{1/2} [43 Thm. X.1.1] (the proof there is dimension-independent, using only operator monotonicity of the square root), which is O(R−∞)O(R^{-\infty}); as FR1/2∈A(CR(z,u))F_R^{1/2}\in\mathcal{A}(C_R(z,u)), E1/2E^{1/2} is almost local. Similarly, ∥(id−E)1/2−(id−FR)1/2∥≤∥E−FR∥1/2\|(\mathrm{id}-E)^{1/2}-(\mathrm{id}-F_R)^{1/2}\|\le\|E-F_R\|^{1/2} and (id−FR)1/2∈A(CR(z,u))(\mathrm{id}-F_R)^{1/2}\in\mathcal{A}(C_R(z,u)), giving the second claim. □\square

Corollary IV.7. *Suppose ρψ∈S(R4)\rho_\psi\in\mathcal{S}(\mathbb{R}^4), let A1,…,An∈A(O1)A_1,\ldots,A_n\in\mathcal{A}(O_1) be effects with ∑i=1nAi≤id\sum_{i=1}^{n}A_i\le\mathrm{id}, and put Ei=ηψ(Ξ(Ai))E_i=\eta_\psi(\Xi(A_i)) and E0:=id−∑i=1nEiE_0:=\mathrm{id}-\sum_{i=1}^{n}E_i. Then the nonselective Lüders channel of the relativised observable {Ei}i=0n\{E_i\}_{i=0}^{n},

Λ(B′)=∑i=0nEi1/2B′Ei1/2,(55)\Lambda(B')=\sum_{i=0}^{n}E_i^{1/2}B'E_i^{1/2}, \tag*{(55)}

is an almost local operation, and its translate Λx1\Lambda_{x_1} is the Lüders channel of the translated observable {αx1(Ei)}i=0n\{\alpha_{x_1}(E_i)\}_{i=0}^{n}. Consequently, by Proposition IV.4(ii), a nonselective Lüders measurement of {Ei}\{E_i\} alters the expectation of any relativised local observable at causal distance LL in the sense of (III.15) by at most O(L−∞)O(L^{-\infty}).

Proof. Since Ψ\mathsf{\Psi} is completely positive and subunital and ηψ\eta_{\psi} is unital and completely positive, each EiE_i (1≤i≤n1 \le i \le n) is an effect, almost local by Proposition III.4, and hence so is E:=∑i=1nEi=ηψ(Ψ(∑iAi))≤id⁡E := \sum_{i=1}^{n} E_i = \eta_{\psi}\left(\mathsf{\Psi}\left(\sum_i A_i\right)\right) \le\operatorname{id}, a sum of approximants for the EiE_i (taken with a common zz and uu) serving as approximants for EE. By Lemma IV.6, the Kraus operators Ei1/2E_i^{1/2} (1≤i≤n1 \le i \le n) and E01/2=(id⁡−E)1/2E_0^{1/2} = (\operatorname{id} - E)^{1/2} are almost local. Hence Λ\Lambda is almost local, and since αx1\alpha_{x_1} commutes with the functional calculus, Λx1\Lambda_{x_1} is as claimed. Proposition IV.4(ii) applies.

Proposition IV.4 and Corollary IV.7 together provide what is left of of microcausality in the relativised theory: no operation with a finite almost local Kraus decomposition, including the Lüders measurements of every finite-outcome relativised local observable, can affect the expectation of a distant relativised observable by more than O(L−∞)O(L^{-\infty}). We stress that (IV.4) should not be read as asserting the possibility of superluminal signalling. By Proposition II.1 and Lemma III.1, relativised observables admit no sharp localisation in bounded regions, so that the spacetime labels appearing in any relative no-signalling statement are themselves defined only to the accuracy set by the frame state.

Thus far we have considered how the local and causal properties of an observable fare under relativisation. We now turn to the question of the physical requirements of the frame for the relative and absolute descriptions to agree closely. Relative to a frame state ψ\psi, the actually realisable counterpart of a system effect AA is the restricted relative observable Tψ(A)=ηψ(Ψ(A))T_{\psi}(A) = \eta_{\psi}(\mathsf{\Psi}(A)), and the quality of approximation is given by D=∥Tψ(A)−A∥D = \lVert T_{\psi}(A) - A\rVert. In the quantum mechanical setting, bounds on DD were obtained in [44], where good approximation was shown to require a broad distribution of the symmetry generator in the frame state. For translations in Minkowski space the relevant property of the frame is its ability to resolve a displacement yy, quantified by the overlap cψ(y)=(ψ,U(y)ψ)c_{\psi}(y) = (\psi, U(y)\psi) of the frame state with its translate: ∣cψ(y)∣\lvert c_{\psi}(y)\rvert close to 1 means that the frame cannot resolve yy. The following proposition shows that if AA is sensitive to the displacement yy, then accurate recovery of AA forces ∣cψ(y)∣\lvert c_{\psi}(y)\rvert to be small. The bound also involves the unsharpness V(A)=∥A−A2∥V(A) = \lVert A - A^2\rVert of AA, which vanishes exactly when AA is a projection.

Proposition IV.8. Let AA be a system effect, suppose the action α\alpha is implemented by a strongly continuous unitary representation USU_S of Rd\mathbb{R}^d, and put

Tψ(A)=ηψ(Ψ(A)),D=∥Tψ(A)−A∥,V(A)=∥A−A2∥.(56)T_{\psi}(A) = \eta_{\psi}(\mathsf{\Psi}(A)), \qquad D = \lVert T_{\psi}(A) - A\rVert, \qquad V(A) = \lVert A - A^2\rVert. \tag*{(56)}

For y∈My \in M, let cψ(y)=(ψ,U(y)ψ)c_{\psi}(y) = (\psi, U(y)\psi). If cψ(y)≠0c_{\psi}(y) \ne0, then

∥αy(A)−A∥≤2D+21−∣cψ(y)∣2∣cψ(y)∣2D+V(A).(57)\lVert\alpha_y(A) - A\rVert\le2D + \frac{2\sqrt{1-\lvert c_{\psi}(y)\rvert^2}}{\lvert c_{\psi}(y)\rvert}\sqrt{2D + V(A)}. \tag*{(57)}

Proof. Set X=Ψ(A)X = \mathsf{\Psi}(A) and η=ηψ\eta= \eta_{\psi}. Since 0≤A≤id⁡0 \le A \le\operatorname{id} and Ψ\mathsf{\Psi} is completely positive and subunital, XX is an effect. The map η\eta is unital and completely positive, so the estimate given in [44 eq. (9)] gives

∥η(X2)−η(X)2∥≤2D+V(A).(58)\lVert\eta(X^2) - \eta(X)^2\rVert\le2D + V(A). \tag*{(58)}

Moreover XX is invariant, so [X,US(y)⊗U(y)]=0[X,U_S(y)\otimes U(y)] = 0, while

η(US(y)⊗U(y))=cψ(y)US(y).(59)\eta(U_S(y)\otimes U(y)) = c_{\psi}(y)U_S(y). \tag*{(59)}

The Cauchy–Schwarz inequality for completely positive maps [45 Lem. 1] gives

∥[Tψ(A),cψ(y)US(y)]∥≤21−∣cψ(y)∣22D+V(A).(60)\lVert[T_{\psi}(A),c_{\psi}(y)U_S(y)]\rVert\le2\sqrt{1-\lvert c_{\psi}(y)\rvert^2}\sqrt{2D+V(A)}. \tag*{(60)}

Dividing by ∣cψ(y)∣\lvert c_{\psi}(y)\rvert and using ∥[A,US(y)]∥≤∥[Tψ(A),US(y)]∥+2D\lVert[A,U_S(y)]\rVert\le\lVert[T_{\psi}(A),U_S(y)]\rVert+ 2D gives (IV.10), up to the sign convention in the implementation of translations. □

Remark IV.9. If AA varies appreciably under a translation yy, then accurate recovery of AA by relativisation followed by restriction requires DD to be small and hence the frame state to have sufficiently small ∣cψ(y)∣=∣(ψ,U(y)ψ)∣\lvert c_{\psi}(y)\rvert= \lvert(\psi,U(y)\psi)\rvert. In other words, and bearing in mind that cψ(0)=1c_{\psi}(0)=1, the frame must itself resolve the displacement yy. Since cψc_{\psi} is related to the QRF energy-momentum distribution, resolving shorter displacement scales requires broader energy-momentum support.

To quantify this last point slightly more, we argue as follows. Recalling that cψ(y)=(ψ,U(y)ψ)c_{\psi}(y) = (\psi,U(y)\psi), we note that

∣cψ(y)∣≥Re⁡cψ(y)=∫cos⁡(p⋅y) dμψ(p)≥(1−Mψ,θ)cos⁡(θ)−Mψ,θ(61)\lvert c_{\psi}(y)\rvert\ge\operatorname{Re} c_{\psi}(y) = \int\cos(p \cdot y)\,\mathrm{d}\mu_{\psi}(p) \ge(1-M_{\psi,\theta})\cos(\theta)-M_{\psi,\theta} \tag*{(61)}

holds for any 0<θ<π/20 < \theta< \pi/2, where Mψ,θ=μψ({p∈V‾+:∣p⋅y∣>θ})M_{\psi,\theta} = \mu_{\psi}(\{p \in\overline{V}^{+} : |p \cdot y| > \theta\}), which follows from the rather crude bound that cos⁡(p⋅y)≥cos⁡θ\cos(p \cdot y) \ge\cos\theta for ∣p⋅y∣≤θ|p \cdot y| \le\theta and cos⁡(p⋅y)≥−1\cos(p \cdot y) \ge-1 otherwise. Supposing that ∣cψ(y)∣<ϵ|c_{\psi}(y)| < \epsilon and rearranging,

Mψ,θ≥cos⁡(θ)−ϵ1+cos⁡(θ).(62)M_{\psi,\theta} \ge\frac{\cos(\theta)-\epsilon}{1+\cos(\theta)}. \tag*{(62)}

If θ\theta is chosen small but nonzero, and ϵ≪1\epsilon\ll1, then Mψ,θ≳12M_{\psi,\theta} \gtrsim\frac{1}{2}. Thus, a significant proportion of the energy-momentum distribution of ψ\psi lies in the set {p∈V‾+:∣p⋅y∣>θ}\{p \in\overline{V}^{+} : |p \cdot y| > \theta\}. If, for example, yy is a timelike vector, then all momenta p∈V‾+p \in\overline{V}^{+} with ∣p⋅y∣>θ|p \cdot y| > \theta have energy greater than θ/y⋅y\theta/\sqrt{y \cdot y} in the rest-frame of yy. Alternatively, if yy is spacelike and ∣p⋅y∣>θ|p \cdot y| > \theta then pp has energy greater than θ/−y⋅y\theta/\sqrt{-y \cdot y} in any frame for which yy is purely spatial, i.e., orthogonal to the observer’s 4-velocity. To see this, let uu be the 4-velocity, then p∈V‾+p \in\overline{V}^{+} implies (p⋅u)2≥(p⋅y)2/(−y⋅y)≥θ2/(−y⋅y)(p \cdot u)^2 \ge(p \cdot y)^2/(-y \cdot y) \ge\theta^2/(-y \cdot y). In either case, resolving short spacetime displacements requires high energies in an associated frame of reference.

Concluding Remarks

The spectrum condition obstructs the existence of ideal reference frames for spacetime translations in QFT: translation-covariant POVMs are never projection-valued, and they fail the norm-1 property in many cases. Relative to such a frame, local observables become almost local, exact microcausality is replaced by rapid commutator decay controlled by the localisation density of the frame state. From an operational standpoint, in which spacetime structure is abstracted from relations among observables, the results above indicate that the causal structure relative to any realistic frame is intrinsically smeared (Remark III.7), and the sharp light cone appears as an idealised limit of frame localisation - a limit which, by Corollary II.2, no physical frame satisfies. We leave the systematic development of this perspective, and the associated question of the emergence of classical causal structure, to future work.

Several further questions remain open. The frames with unbounded hh, the only remaining candidates for the norm-1 property (Remark II.10), need a physical characterisation. Relativised two-point functions of Hadamard states may provide a natural class of examples. Whether the operational compatibility preserved under relativisation excludes obstructions to measurements of the kind identified by Sorkin [30] is open, and connects with the measurement framework of [28]. The relation between causal and realisable operations is delicate already in the absolute setting [37] and we leave the relational setting for the future. Finally, we note that the necessary unsharpness of time observables may play a role in the ubiquity of the phenomenon of type change [46, 47, 48, 49].

Note added. During the completion of this work, we became aware of the work of Agullo and Mele [50], who investigate QRFs also in a QFT setting with Gaussian-modulated POVMs as frame observables, which do not satisfy the norm-1/localisability property. We look forward to comparing our findings to theirs, particularly in connection with Corollaries II.8 and II.9. We also became aware of Freidel and Kirklin [51], in which the quantum reference frame for reparametrisations of a gravitational null ray is necessarily non-ideal; positivity of the null energy excludes a sharp frame observable. Though the context is different, this gives further indication that unsharp QRFs are likely to be typical in relativistic quantum settings quite generally.

Acknowledgments

We thank Daan Janssen, Kasia Reizner, Jan Glowacki and Samuel Fedida for useful discussions. We also thank Laurent Freidel for pointing out [51]. The work of CJF is supported in part by EPSRC Grant EP/Y000099/1 to the University of York. For the purpose of open access, the authors have applied a creative commons attribution (CC BY) licence to any author accepted manuscript version arising.

References

  1. [1]S. D. Bartlett, T. Rudolph, and R. W. Spekkens, Reference frames, superselection rules, and quantum information, Reviews of Modern Physics 79, 555 (2007).DOI
  2. [2]L. Loveridge, Quantum measurements in the presence of symmetry, Ph.D. thesis, University of York (2012).
  3. [3]L. Loveridge, T. Miyadera, and P. Busch, Symmetry, reference frames, and relational quantities in quantum mechanics, Foundations of Physics 48, 135 (2018).
  4. [4]A.-C. de la Hamette, T. D. Galley, P. A. Höhn, L. Loveridge, and M. P. Müller, Perspective-neutral approach to quantum frame covariance for general symmetry groups (2021), arXiv:2110.13824 [quant-ph].arxiv.org/abs/2110.13824
  5. [5]J. Glowacki, L. Loveridge, and J. Waldron, Quantum reference frames on finite homogeneous spaces, International Journal of Theoretical Physics 63, 137 (2024).
  6. [6]T. Carette, J. Glowacki, and L. Loveridge, Operational quantum reference frame transformations, Quantum 9, 1680 (2025).DOI
  7. [7]F. Giacomini, E. Castro-Ruiz, and Č. Brukner, Quantum mechanics and the covariance of physical laws in quantum reference frames, Nature Communications 10, 494 (2019).
  8. [8]A. Vanrietvelde, P. A. Höhn, F. Giacomini, and E. Castro-Ruiz, A change of perspective: switching quantum reference frames via a perspective-neutral framework, Quantum 4, 225 (2020).DOI
  9. [9]A.-C. de la Hamette and T. D. Galley, Quantum reference frames for general symmetry groups, Quantum 4, 367 (2020).
  10. [10]E. Castro-Ruiz and O. Oreshkov, Relative subsystems and quantum reference frame transformations, Communications Physics 8, 187 (2025).
  11. [11]L. Loveridge and J. Waldron, W*-algebraic quantum reference frames, In preparation (2026).
  12. [12]J. Glowacki and Y. Kuramochi, W*-algebraic integration theory, arXiv:2606.27366 (2026).
  13. [13]M. Toller, Localization of events in space-time, Physical Review A 59, 960 (1999).DOI
  14. [14]S. Mazzuccchi, On the observables describing a quantum reference frame, Journal of Mathematical Physics 42, 2477 (2001).DOI
  15. [15]R. Haag, Local Quantum Physics: Fields, Particles, Algebras, 2nd ed., Theoretical and Mathematical Physics (Springer, Berlin, Heidelberg, 1996).
  16. [16]W. Dybalski and C. Gérard, A criterion for asymptotic completeness in local relativistic QFT, Communications in Mathematical Physics 332, 1167 (2014).
  17. [17]S. Fedida and J. Glowacki, Foundations of relational quantum field theory I: Scalars (2025), arXiv:2507.21601 [quant-ph].arxiv.org/abs/2507.21601
  18. [18]R. F. Streater and A. S. Wightman, PCT, Spin and Statistics, and All That (Princeton University Press, 2000) originally published 1964 by W. A. Benjamin.DOI
  19. [19]R. Gianitrapani, Quantum coordinates of an event in local quantum physics, Journal of Mathematical Physics 39, 5180 (1998).
  20. [20]R. Haag and D. Kastler, An algebraic approach to quantum field theory, Journal of Mathematical Physics 5, 848 (1964).DOI
  21. [21]G. Lüders, Über die Zustandsänderung durch den Meßprozeß, Annalen der Physik 8, 322 (1951).DOI
  22. [22]P. Busch and P. Lahti, Lüders rule, in Compendium of Quantum Physics (Springer, 2009) pp. 356–358.
  23. [23]P. Busch, Unsharp localization and causality in relativistic quantum theory, Journal of Physics A: Mathematical and General 32, 6535 (1999).DOI
  24. [24]P. Busch, Unsharp reality and joint measurements for spin observables, Physical Review D 33, 2253 (1986).pubmed.ncbi.nlm.nih.gov/9956897
  25. [25]P. Busch, P. Lahti, J.-P. Pellonpää, and K. Ylinen, Quantum Measurement (Springer, Cham, 2016).DOI
  26. [26]T. Heinosaari and T. Miyadera, Qualitative noise-disturbance relation for quantum measurements, Phys. Rev. A 88, 042117 (2013).DOI
  27. [27]L. J. Landau, A note on extended locality, Comm. Math. Phys. 13, 246 (1969).DOI
  28. [28]C. J. Fewster and R. Verch, Quantum fields and local measurements, Communications in Mathematical Physics 378, 851 (2020).arxiv.org/abs/1810.06512
  29. [29]T. Heinosaari, J. Kiukas, D. Reitzner, and J. Schultz, Incompatibility breaking quantum channels, Journal of Physics A: Mathematical and Theoretical 48, 435301 (2015).DOI
  30. [30]R. D. Sorkin, Impossible measurements on quantum fields, in Directions in General Relativity: Proceedings of the 1993 International Symposium, Maryland, Vol. 2 (Cambridge University Press, 1993) pp. 293–305.arxiv.org/abs/gr-qc/9302018
  31. [31]L. Borsten, I. Jubb, and G. Kells, Impossible measurements revisited, Phys. Rev. D 104, 025012 (2021).DOI
  32. [32]H. Bostelmann, C. J. Fewster, and M. H. Ruep, Impossible measurements require impossible apparatus, Phys. Rev. D 103, 025017 (2021).DOI
  33. [33]I. Jubb, Causal state updates in real scalar quantum field theory, Physical Review D 105, 025003 (2022).DOI
  34. [34]R. Oeckl, Spectral decomposition of field operators and causal measurement in quantum field theory, J. Math. Phys. 66, 042302 (2025).DOI
  35. [35]R. Oeckl, Causal measurement in quantum field theory: spacetime, Phys. Rev. D 113, 085002 (2026).DOI
  36. [36]R. Oeckl and A. Zampeli, Towards local and compositional measurements in quantum field theory, Quantum 10, 2232 (2026), arXiv:2505.10968.DOI
  37. [37]J. Mandrysch, R. Simmons, and M. Navascués, Causality and realizability of local operations in quantum field theory (2026), arXiv:2607.12976.arxiv.org/abs/2607.12976
  38. [38]K. Kraus, States, Effects, and Operations: Fundamental Notions of Quantum Theory, Lecture Notes in Physics, Vol. 190 (Springer, Berlin, 1983).
  39. [39]M.-D. Choi, A Schwarz inequality for positive linear maps on C*-algebras, Illinois Journal of Mathematics 18, 565 (1974).DOI
  40. [40]A. Arias, A. Gheondea, and S. Gudder, Fixed points of quantum operations, Journal of Mathematical Physics 43, 5872 (2002).DOI
  41. [41]G. Lindblad, A general no-cloning theorem, Letters in Mathematical Physics 47, 189 (1999).
  42. [42]A. Much and R. Verch, Superluminal local operations in quantum field theory: A ping-pong ball test, Universe 9, 447 (2023).DOI
  43. [43]R. Bhatia, Matrix Analysis, 1st ed. (Springer New York, NY, 1997).
  44. [44]T. Miyadera, L. Loveridge, and P. Busch, Approximating relational observables by absolute quantities: a quantum accuracy–size trade-off, J. Phys. A: Math. Theor. 49, 185301 (2016).
  45. [45]B. Janssens, Unifying decoherence and the Heisenberg principle, Letters in Mathematical Physics 107, 1557 (2017).
  46. [46]V. Chandrasekaran, R. Longo, G. Penington, and E. Witten, An algebra of observables for de Sitter space, Journal of High Energy Physics 2023, 82 (2023).
  47. [47]C. J. Fewster, D. W. Janssen, L. D. Loveridge, K. Rejzner, and J. Waldron, Quantum reference frames, measurement schemes and the type of local algebras in quantum field theory, Communications in Mathematical Physics 406, 19 (2025).
  48. [48]J. De Vuyst, S. Eccles, P. A. Höhn, and J. Kirklin, Gravitational entropy is observer-dependent, Journal of High Energy Physics 2025, 146 (2025).DOI
  49. [49]J. De Vuyst, S. Eccles, P. A. Höhn, and J. Kirklin, Crossed products and quantum reference frames: on the observer-dependence of gravitational entropy, Journal of High Energy Physics 2025, 063 (2025).DOI
  50. [50]I. Agullo and F. M. Mele, A proposal for quantum reference frames in quantum field theory and their experimental exploration, In preparation (2026).arxiv.org/abs/2611.21740
  51. [51]L. Freidel and J. Kirklin, Gravitational null rays: Covariant quantization and the dressing time, J. High Energy. Phys 2026, 130 (2026).DOI

Paper details

vlnc:W_vfvnct7w
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 4

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents