This paper interprets Haag’s theorem as more than a technical obstruction to the interaction picture in relativistic quantum field theory. It argues that the theorem reveals a deeper structural limitation of global representation-based ontology: an interacting QFT cannot, under the assumptions of the theorem, be understood as a simple unitary deformation of a free theory within one fixed Fock representation. The paper connects this lesson to algebraic quantum field theory, superselection sectors, locally covariant QFT, Wilsonian renormalization, and effective field theory. It proposes that these developments all point toward a common principle: physical objectivity is not grounded in a privileged representation, but in structural content preserved under admissible transformations between perspectives, representations, sectors, scales, or regions. The paper does not claim to solve Haag’s theorem or construct interacting QFTs. Its aim is interpretive: to show how Haag’s theorem is naturally intelligible within the observer-equivariant framework, where objectivity is understood as invariant or equivariant structure rather than as a view from one absolute representation.
This paper interprets Haag’s theorem as more than a technical obstruction to the interaction picture in relativistic quantum field theory. It argues that the theorem reveals a deeper structural limitation of global representation-based ontology: an interacting QFT cannot, under the assumptions of the theorem, be understood as a simple unitary deformation of a free theory within one fixed Fock representation. The paper connects this lesson to algebraic quantum field theory, superselection sectors, locally covariant QFT, Wilsonian renormalization, and effective field theory. It proposes that these developments all point toward a common principle: physical objectivity is not grounded in a privileged representation, but in structural content preserved under admissible transformations between perspectives, representations, sectors, scales, or regions. The paper does not claim to solve Haag’s theorem or construct interacting QFTs. Its aim is interpretive: to show how Haag’s theorem is naturally intelligible within the observer-equivariant framework, where objectivity is understood as invariant or equivariant structure rather than as a view from one absolute representation.
Quantum field theory (QFT) is the physical framework that integrates quantum mechanics and the special theory of relativity; it is the basis of many of our best physical theories. QFT's for interacting systems have yielded extraordinarily accurate predictions. Yet, in spite of unquestionable empirical success, the treatment of interactions in QFT raises serious issues for the foundations and interpretation of the theory. This dissertation takes Haag's theorem as a starting point for investigatingthese issues. It begins with a detailed exposition and analysis of different versions ofHaag's theorem. The theorem is cast as a reductio ad absurdum of canonical QFT prior to renormalization. It is possible to adopt different strategies in response to this reductio: (1) renormalizing the canonical framework; (2) introducing a volume i.e., long-distance) cutoff into the canonical framework; or (3) abandoning another assumption common to the canonical framework and Haag's theorem, which is the approach adopted by axiomatic and constructive field theorists. Haag's theorem doesnot entail that it is impossible to formulate a mathematically well-defined Hilbert space model for an interacting system on infinite, continuous space. Furthermore, Haag's theorem does not undermine the predictions of renormalized canonical QFT; canonical QFT with cutoffs and existing mathematically rigorous models for interactions are empirically equivalent to renormalized canonical QFT. The final two chapters explore the consequences of Haag's theorem for the interpretation of QFT with interactions. I argue that no mathematically rigorous model of QFT on infinite, continuous space admits an interpretation in terms of quanta (i.e., quantumparticles). Furthermore, I contend that extant mathematically rigorous models for physically unrealistic interactions serve as a better guide to the ontology of QFT than either of the other two formulations of QFT. Consequently, according to QFT, quanta do not belong in our ontology of fundamental entities.
This paper argues that observer equivariance is best understood not merely as an interpretive motif attached to particular physical theories, but as a higher-order constraint on the admissible form of physical theory. If physical law is to count as genuinely public, it must be transportable across admissible changes of perspective without loss of objective content. On this view, observer equivariance constrains not only the covariance of equations, but also the admissibility of representational frameworks, the structure of weighting schemes, the role of compatibility conditions, and the interpretation of residual freedom in theory construction. The argument is synthetic, drawing together earlier work on shared physical law, translational and spectral form, path-integral measure selection, Born-type weighting, and Machian boundary constraints. Three levels of claim are kept distinct throughout: conceptual claims about the public form of physics, partial formal results (such as orbitwise rigidity of weighting schemes and conditional uniqueness of quadratic weighting under additional structural assumptions), and conjectural extensions (most notably to renormalization-related residual freedom). The main philosophical consequence is a strengthened, prospective motivation for epistemic structural realism: the structural content of physics is not merely what happens to survive theory change, but what must already be stably coordinable across admissible perspectives if public physical knowledge is to be possible at all. This route is contrasted with Worrall's retrospective appeal to theory change, with Ladyman and French's ontic structural realism, and with Rovelli's relational quantum mechanics. Keywords: observer equivariance, epistemic structural realism, philosophy of physics, symmetry and invariance, intersubjectivity, path-integral measure, Born rule, Machian inertia, renormalization.
This dissertation gives a general account of the properties of unitarily inequivalent representations (UIRs) in both canonical quantum field theory and algebraic quantum field theory. A simple model is constructed and then used to show how to build a broad spectrum of UIRs including a version of Haag’s theorem. Haag and Kastler,P, two of the founding fathers of algebraic quantum field theory, argue that the problems posed by UIRs are solved by adopting a notion of equivalence that is weaker than unitary equivalence, which they refer to as physical equivalence. In the dissertation, it is shown that their notion does not provide a suitable classificatory schema. Some of the most important physical representations fail to satisfy the mathematical conditions of their notion. However, Haag and Kastler's notion has an unexpected connection with classical observables. A theorem is proven in which two representations make the same predictions with respect to all classical observables if and only if they satisfy their notion of physical equivalence. Following Haag and Kastler's lead, it was claimed by most proponents of algebraic quantum field theory that all physical content resides in a specific class of observables. It is shown in the dissertation that such claims are exaggerated and misleading. UIRs are used to elucidate the nature of quantum field theory by showing that UIRs have different expectation values for some classical observables of the system, such as temperature and chemical potential, which are not in Haag and Kastler’s specific class. It is shown how UIRs may be used to construct classical observables. To capture the physical content of quantum field theory it is shown that a much larger algebra than that of Haag and Kastler is necessary. Finally, the arguments that UIRs are incommensurable theories are shown to be flawed. The lesson of UIRs is that the mathematical structures in both canonical quantum field theory and Haag and Kastler’s version of algebraic quantum field theory are not sufficient to capture all of the physical content that UIRs represent. A suitable algebraic structure for quantum field theory is provided in the dissertation.
Haag’s theorem has been interpreted as establishing that quantum field theory cannot consistently represent interacting fields. Earman and Fraser have clarified how it is possible to give mathematically consistent calculations in scattering theory despite the theorem. However, their analysis does not fully address the worry raised by the result. In particular, I argue that their approach fails to be a complete explanation of why Haag’s theorem does not undermine claims about the empirical adequacy of particular quantum field theories. I then show that such empirical adequacy claims are protected from Haag’s result by the techniques that are required to obtain theoretical predictions for realistic experimental observables. I conclude by showing how Haag’s theorem is illustrative of a general tension between the foundational significance of results that can be obtained in perturbation theory and non-perturbative characterizations of the content of quantum field theory. 1 Introduction2 Haag’s Theorem and the Interaction Picture3 Earman and Fraser on the Success of Scattering Theory4 Haag’s Theorem and Empirical Adequacy5 Conclusion
Observer-Equivariant Ontology (v1) develops a compact ontological and methodological stance tailored to modern physics: the fundamental objects are not “systems in themselves” but observer perspectives related by symmetry. The paper formulates this as an equivalence requirement across admissible perspectives and uses it to motivate why the epistemic core of physics is structurally invariant content (relations, constraints, symmetries), rather than observer-dependent representations. The text is written as a conceptual companion to the more technical parts of the series. It clarifies: (i) what is being quantified over when physics speaks about “observers” or “frames”, (ii) how invariance and gauge redundancy enter as structural constraints on admissible descriptions, and (iii) how a perspective-based ontology can be stated without committing to instrumentalism. Where needed, the paper points to the technical manuscripts for constructions such as symmetry-constrained measures and history groupoids. This upload is version v1 of the paper. Version DOI (this record): 10.5281/zenodo.17929468 Series DOI (all versions): 10.5281/zenodo.17929467
Observer Equivariance (OE) is proposed as an interpretation of quantum theory based on a single structural principle: objectivity is invariance under changes of observer perspective. The framework is formulated using a projection p:\mathcal O \to \mathcal S, where \mathcal O is a groupoid of observer perspectives (descriptions) and \mathcal Srepresents structural content intended to be objective. Within this setting, the paper links three familiar pillars of quantum theory as consequences of one invariance requirement: Symmetry. Physical symmetries are characterized as automorphisms of \mathcal S that are liftable to \mathcal O in a way compatible with the relevant gauge freedom. This recovers Wigner/Uhlhorn-type constraints from a general “liftability” viewpoint. Born weights. Event weights are constrained by invariance under a minimal internal “mixing” operation (unitary change of basis within a fixed structural event), together with coarse graining and phase invariance. A proof sketch shows how this yields a functional equation that forces quadratic (Born-type) weights. Uncertainty. Translational perspective changes motivate a Fourier/Pontryagin-duality analysis in which Plancherel/Parseval norm preservation selects an L^2 structure and yields standard Fourier uncertainty relations as representation-theoretic constraints. The article positions OE relative to QBism and Relational Quantum Mechanics: it shares the view that “facts” and state assignments are perspective-indexed, while differing in that symmetry constraints and the probability calculus are treated as structural necessities derived from invariance, rather than as purely epistemic or decision-theoretic postulates. DOI (this version): 10.5281/zenodo.18221625 Concept DOI (all versions): 10.5281/zenodo.18221624 AI use: AI systems (ChatGPT, Gemini, Grok, and Claude) were used for drafting assistance, language editing, and mathematical polishing. The author reviewed the manuscript and takes full responsibility for all claims and any errors.