A causal derivation of the algebraic approach to quantum systems

Nick Ormrod · 2025

Read A causal derivation of the algebraic approach to quantum systems by Nick Ormrod (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-01111.txt.

Introduction

John Ormrod wrote a physics paper in 2025 giving a causal derivation of the algebraic approach to quantum systems. The algebraic approach represents a quantum system by an operator algebra, a structured set of quantities representable on a Hilbert space. That representation is usually simply assumed. Ormrod opens with an apparent quantum system that no operator algebra naturally represents, casting doubt on the assumption.

He proposes an alternative, the causal view: a quantum system is defined by the causal structure of its unitary dynamics alone. He proves that every system so defined corresponds to a unique von Neumann algebra. He extends the result to a 'classical quantum system,' inspired by quantum Darwinism, and shows it too corresponds to a unique, commutative von Neumann algebra. He contrasts the causal view with an incompatible epistemic view, which reads quantum systems as states of knowledge.

Catalog record

CREATORNick Ormrod
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE53 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-01111

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A causal derivation of the algebraic approach to quantum systems Preprint: APS/123-QED Nick Ormrod Abstract It is commonly assumed that every quantum system is represented by some algebra of operators. Doubt is cast on this assumption by what appears, at first glance, to be a reasonable candidate for a quantum system that is not naturally represented by any algebra. To resolve this puzzle, this work draws inspiration from recent frameworks for causal modelling in quantum theory to propose a “causal view” of quantum systems. The causal view defines quantum systems purely in terms of the causal structure of the unitary dynamics. The algebraic representation of quantum systems is derived from the causal view: it is proven that every quantum system corresponds to a unique von Neumann algebra of operators. The causal view is extended with a definition of a “classical quantum system” inspired by quantum Darwinism. It is shown that such a system corresponds to a unique commutative von Neumann operator algebra, completing the derivation of the traditional algebraic approach to quantum systems from the causal view. The causal view is contrasted with the “epistemic view” of quantum systems, which is incompatible with the algebraic approach. I Introduction Traditionally, it has often been assumed that the most general sort of quantum system is represented by an algebra of operators (most famously in algebraic quantum field theory Haag and Kastler 1964 , more recently in quantum information and foundations Viola et al. 2001 ; Zanardi 2001 ; Zanardi et al. 2004 ; Ali Ahmad et al. 2022 ; Vanrietvelde et al. 2025 ), and that the algebra is commutative when the quantum system is “classical”. We call this the algebraic approach to quantum systems . This paper asks how, or even whether, the algebraic approach can be justified. The algebraic approach recovers the familiar notion of a system as a Hilbert space, since any Hilbert space \mathcal{H} can be identified with the algebra of bounded operators \mathcal{B}(\mathcal{H}) . But the algebraic approach is also more general. If a superselection rule forbids superpositions of states with respect to a preferred direct sum decomposition \mathcal{H}=\bigoplus_{i}\mathcal{H}_{i} of the Hilbert space, then the superselected system can be identified with the algebra of block-diagonal bounded operators \bigoplus_{i}\mathcal{B}(\mathcal{H}_{i}) . But is the algebraic approach general enough? To motivate this question, we note that there are certain sets of operators in quantum theory that are not operator algebras in their own right, do not naturally correspond to operator algebras, and yet do initially seem plausible as candidates for quantum systems. For example, consider the set of all real linear combinations of the Pauli observables I , X , and Z on a qubit: \mathscr{R}={\rm rspan}(I,X,Z). (1) To see why one might reasonably suggest that \mathscr{R} defines a subsystem of the qubit, suppose that an observer Polly performs many measurements of X and Z on an ensemble of identically prepared qubits in order to determine the expectation values \langle X\rangle_{\rho} and \langle Z\rangle_{\rho} . From these expectation values she can immediately deduce the expectations of all observables in \mathscr{R} . She cannot in general deduce the expectation values of any observables outside of this set (she can in special cases, but see Appendix A ). Therefore, \mathscr{R} is a full list of observables whose expectation values are known to some possible observer . One might think that this last fact alone qualifies a set of observables to represent a quantum system. We call this the epistemic view of systems (not to be confused with the logically unrelated epistemic view of quantum states defended in e.g. Spekkens 2007 ). On the epistemic view, \mathscr{R} represents a quantum system. Corroborating this conclusion, in the context of generalized probabilistic theories (GPTs) Hardy 2001 ; Barrett 2007 , Schmid et al. 2024 proposes a formal definition of a “GPT subsystem” according to which \mathscr{R} is a GPT subsystem of the qubit. 1 1 1 However, the authors of Schmid et al. 2024 take care to note that the notion of a GPT system is necessarily highly general since arbitrary GPTs lack the algebraic structure that traditionally features in the definition of a quantum system. They do not take a position on whether all GPT subsystems in a formulation of quantum theory as a GPT are quantum subsystems: “the rebit is typically not considered to be a quantum subsystem of a qubit, but it makes sense to consider it a GPT subsystem” (original emphasis). Furthermore, \mathscr{R} defines the most elementary system (a “rebit”) in a restricted version of quantum theory known as “real quantum theory” Stueckelberg 1960 ; Caves et al. 2001 . However, \mathscr{R} does not naturally correspond to any algebra. Formally, there is no (von Neumann or C^{*} -) algebra \mathcal{A} such that \mathscr{R} is the full set of its observables, \mathscr{R}={\rm Herm}(\mathcal{A}) . An easy way to see this is to note that any algebra contains arbitrary complex rescalings and products of its members. Since \mathscr{R} contains X and Z , any algebra containing \mathscr{R} must also contain Y=iXZ . However, \mathscr{R} does not contain Y . But should this prevent us from viewing \mathscr{R} as a system? There is no obvious, direct operational meaning of the product of two observables (which in general is not an observable itself), so it is not obvious why it should be a problem that X,Z\in\mathscr{R} but Y\not\in\mathscr{R} . To justify the algebraic approach to systems, one needs to convincingly explain what is “wrong” with viewing \mathscr{R} as a system, which in part means explaining why closure under products is relevant for systemhood. Rising to this challenge, this paper draws inspiration from the recent “causal turn” in the foundations of quantum physics. Among other things, the inadequacy of standard causal models Pearl 2009 ; Spirtes et al. 2000 for explaining quantum correlations Wood and Spekkens 2015 ; Cavalcanti 2018 ; Pearl and Cavalcanti 2021 and the possibility of “indefinite causal order” Hardy 2013 ; Chiribella et al. 2009 ; Chiribella et al. 2013 ; Oreshkov et al. 2012 have led many to suspect that a radical concept of causation plays an essential role in quantum theory (see e.g. Spekkens 2015 ). This view has motivated the development of intrinsically quantum frameworks for causal modelling Pienaar and Brukner 2015 ; Costa and Shrapnel 2016 ; Allen et al. 2017 ; Pienaar 2019 ; Pienaar 2020 ; Barrett et al. 2020 ; Barrett et al. 2021 ; Ormrod et al. 2023 , and more recently an interpretation of the theory in which causation is more fundamental than states Ormrod and Barrett 2024 . This paper contributes a causal view of quantum systems , formally defining a quantum system in terms of the causal structure of unitary interactions. Heuristically, the causal view tells us that a system can be thought of as a full list of observables that are accessible to some possible observer . Also inspired by quantum Darwinism Zurek 2003 ; Zurek 2009 , the causal view defines a classical quantum system as a full list of observables that are independently accessible to a pair of possible observers. Again, the intuition here is formally pinned down in terms of causal influences through unitary interactions. This paper

Read the complete text