Mathematics for Physics

Status: Living page — rigorous core, actively curated at the analysis/QFT seam Last updated: 2026-06-08

This page treats mathematics not as a neutral notation but as the substrate of physical theory, examined layer by layer for where it is a theorem versus where it is a heuristic. The organizing question throughout is epistemic: which mathematical structures underlying physics are rigorously established, which are productive abuses of notation, and which mark genuine open frontiers. See EPISTEMICS.md for the marker conventions and THEORY_MAP.md for how this domain sits beneath the others.

Scope

Covered: linear algebra and infinite-dimensional Hilbert spaces; functional analysis (unbounded self-adjoint operators, the spectral theorem, rigged Hilbert spaces); operator algebras (C*- and von Neumann algebras, Tomita–Takesaki modular theory); differential geometry (manifolds, fiber bundles, connections, curvature) as the geometry of gauge theory and gravity; Lie groups, Lie algebras, and representation theory; measure theory and probability; topology and its physical role (anomalies, instantons, topological phases); the ill-defined path-integral measure and the rigor gap of interacting QFT; constructive and algebraic QFT; category theory and higher structures (TQFT, factorization algebras); and the "unreasonable effectiveness" meta-question.

Excluded: detailed physical predictions, numerical methods, and the internal physics of particular models except as illustrations of mathematical structure. Those live in the domains/ pages proper.

Core formalism

1. Linear algebra and Hilbert space

The kinematic arena of quantum mechanics is a separable complex Hilbert space $\mathcal{H}$: a complex vector space with inner product $\langle\cdot,\cdot\rangle$ (antilinear in the first slot by physics convention), complete in the induced norm $\|\psi\|=\sqrt{\langle\psi,\psi\rangle}$. Separability (a countable orthonormal basis exists) is a standing background assumption. The canonical model is $L^2(\mathbb{R}^n, d^n x)$. ESTABLISHED

States are not vectors but rays — unit vectors up to phase, i.e. points of projective space $\mathbb{P}(\mathcal{H})$, equivalently rank-one projections. Wigner's theorem ESTABLISHED states that any bijection of $\mathbb{P}(\mathcal{H})$ preserving transition probabilities $|\langle\psi,\phi\rangle|^2$ is implemented on $\mathcal{H}$ by an operator that is either unitary or antiunitary. Combined with continuity (one-parameter groups) this forces unitarity, and via Stone's theorem a self-adjoint generator.

2. Unbounded operators and the spectral theorem

Observables are self-adjoint operators. The subtlety separating physics from naïve linear algebra is that the fundamental observables — position $\hat x$, momentum $\hat p = -i\hbar\, d/dx$, Hamiltonians — are unbounded and defined only on a dense domain $D(A)\subsetneq\mathcal{H}$. One must distinguish:

• symmetric: $\langle A\psi,\phi\rangle=\langle\psi,A\phi\rangle$ for $\psi,\phi\in D(A)$;
• self-adjoint: symmetric and $D(A)=D(A^\ast)$.

Only self-adjointness — not mere symmetry ("Hermiticity") — guarantees a real spectrum and unitary evolution $e^{-itA}$. The gap is measured by the deficiency indices $n_\pm = \dim\ker(A^\ast \mp i)$: a symmetric operator is essentially self-adjoint iff $n_+=n_-=0$, and admits self-adjoint extensions iff $n_+=n_-$. Different extensions are genuinely different physics (boundary conditions; the particle on a half-line; singular potentials). ESTABLISHED

Spectral theorem (multiplication form). For any self-adjoint $A$ there is a measure space $(M,\mu)$, a unitary $U:\mathcal{H}\to L^2(M,\mu)$, and a real measurable $f$ with $UAU^{-1}$ = multiplication by $f$. Equivalently there is a projection-valued measure $E(\cdot)$ on the Borel sets of $\mathbb{R}$ with

$A=\int_{\sigma(A)}\lambda\,dE(\lambda),\qquad f(A)=\int f(\lambda)\,dE(\lambda).$

This is what makes "probability that a measurement of $A$ lies in $\Delta$" equal $\|E(\Delta)\psi\|^2$ — the Born rule on the spectral side. ESTABLISHED

Stone's theorem. Strongly continuous one-parameter unitary groups $U(t)$ are in bijection with self-adjoint generators: $U(t)=e^{-itA}$, with $A\psi=i\lim_{t\to0}(U(t)\psi-\psi)/t$ on the domain where the limit exists. This underwrites time evolution (generator: the Hamiltonian) and translations (generator: momentum). ESTABLISHED

Rigged Hilbert spaces. The Dirac eigenkets $|x\rangle,|p\rangle$ are not elements of $\mathcal{H}$ (continuous spectrum has no eigenvectors). Their rigorous home is a Gelfand triple $\Phi\subset\mathcal{H}\subset\Phi'$, with $\Phi$ a nuclear space (e.g. Schwartz functions) and $\Phi'$ its dual (tempered distributions); "eigenkets" are distributional eigenvectors in $\Phi'$ via the nuclear spectral theorem. It is ESTABLISHED that this makes Dirac's bra-ket bookkeeping rigorous; the physicist's manipulations are heuristic shorthand for it. See domains/quantum-mechanics.md.

3. Operator algebras

The algebraic approach takes the observables themselves as primary, rather than fixing $\mathcal{H}$.

A C*-algebra is a complex Banach algebra $\mathcal{A}$ with involution $\ast$ satisfying the C*-identity $\|a^\ast a\|=\|a\|^2$. The Gelfand–Naimark theorem ESTABLISHED shows every abstract C*-algebra is isometrically $\ast$-isomorphic to a norm-closed $\ast$-subalgebra of $B(\mathcal{H})$; the commutative case is $C_0(X)$ for locally compact Hausdorff $X$ — the seed of noncommutative geometry ("spaces" $\leftrightarrow$ commutative C*-algebras, so general C*-algebras are "noncommutative spaces"). States are positive normalized linear functionals $\omega$, and the GNS construction reconstructs from $(\mathcal{A},\omega)$ a representation $\pi_\omega$ on a Hilbert space with cyclic vector $\Omega$ such that $\omega(a)=\langle\Omega,\pi_\omega(a)\Omega\rangle$. ESTABLISHED

A von Neumann algebra is a $\ast$-subalgebra $M\subset B(\mathcal{H})$ equal to its own double commutant, $M=M''$ (equivalently weakly closed, by the bicommutant theorem). Murray–von Neumann classified factors (trivial center) into types $\mathrm{I}_n,\mathrm{I}_\infty$ (ordinary QM, $B(\mathcal{H})$), $\mathrm{II}_1,\mathrm{II}_\infty$ (continuous trace), and $\mathrm{III}$ (no trace). ESTABLISHED A structurally crucial fact for QFT: the local algebras of relativistic quantum field theory are generically type III$_1$ factors. This is why there is no global number operator on a single local algebra, no normalizable vacuum localized there, and why the Reeh–Schlieder theorem holds (the vacuum is cyclic and separating for any local algebra). ESTABLISHED See domains/quantum-field-theory.md.

Tomita–Takesaki modular theory. Given a von Neumann algebra $M$ with a cyclic and separating vector $\Omega$, the closure of $S_0: a\Omega\mapsto a^\ast\Omega$ has polar decomposition $S=J\Delta^{1/2}$, yielding the modular conjugation $J$ and the positive modular operator $\Delta$. The deep theorems: $JMJ=M'$ (the commutant), and the modular automorphism group $\sigma_t(a)=\Delta^{it}a\Delta^{-it}$ maps $M$ to itself. The state $\omega=\langle\Omega,\cdot\,\Omega\rangle$ satisfies the KMS condition at inverse temperature $\beta=1$ for $\sigma_t$ — a structural identity between equilibrium statistical mechanics (KMS states) and the intrinsic dynamics a state induces on its algebra. ESTABLISHED This underlies the Bisognano–Wichmann theorem (the Rindler/Unruh modular flow is the Lorentz boost) and modern work on entropy and energy conditions. See domains/statistical-mechanics.md.

4. Differential geometry and gauge theory

A smooth manifold $M$ models spacetime/configuration space. The objects: tangent/cotangent bundles, tensor fields, the exterior algebra $\Omega^k(M)$ with exterior derivative $d$ ($d^2=0$), Stokes' theorem $\int_M d\omega=\int_{\partial M}\omega$, and de Rham cohomology $H^k_{dR}(M)=\ker d/\operatorname{im} d$. For gravity one adds a (pseudo-)Riemannian metric $g$, the unique torsion-free metric (Levi-Civita) connection $\nabla$, curvature $R^\rho{}_{\sigma\mu\nu}$, and Einstein's equations $G_{\mu\nu}+\Lambda g_{\mu\nu}=8\pi G\,T_{\mu\nu}$. See domains/general-relativity.md.

Principal bundles and connections are the geometry of gauge theory. A principal $G$-bundle $P\xrightarrow{\pi}M$ carries a free fiberwise $G$-action. A connection is a Lie-algebra-valued 1-form $A\in\Omega^1(P,\mathfrak g)$ (the gauge potential); its curvature is

$F=dA+\tfrac12[A,A]=dA+A\wedge A,\qquad dF+[A,F]=0\ \ (\text{Bianchi}).$

Matter fields are sections of associated vector bundles; the gauge-covariant derivative is $D=d+\rho(A)$ in representation $\rho$. Gauge transformations act as $A\mapsto gAg^{-1}-(dg)g^{-1}$, $F\mapsto gFg^{-1}$, and the Yang–Mills action is

$S_{YM}=\frac{1}{2g^2}\int_M \operatorname{tr}(F\wedge\star F).$

The dictionary "gauge field = connection on a principal bundle; field strength = curvature; Wilson line = parallel transport / holonomy" is exact. ESTABLISHED Characteristic classes (Chern, Pontryagin, Euler) built from $F$ are topological invariants; e.g. the instanton number / second Chern number $\frac{1}{8\pi^2}\int\operatorname{tr}(F\wedge F)\in\mathbb{Z}$.

Index theory. The Atiyah–Singer index theorem equates the analytic index of an elliptic operator (e.g. a Dirac operator $\slashed D$ coupled to a gauge field) with a topological integral of characteristic classes, controlling fermion zero modes, chiral anomalies, and instanton contributions:

$\operatorname{ind}(\slashed D)=\dim\ker\slashed D-\dim\operatorname{coker}\slashed D=\int_M \hat A(M)\wedge\operatorname{ch}(F).$ ESTABLISHED

5. Lie groups and representation theory

Continuous symmetries are Lie groups $G$; their infinitesimal version are Lie algebras $\mathfrak g$ with bracket $[\cdot,\cdot]$, related by $\exp:\mathfrak g\to G$. Physical states transform in unitary representations. Structural results: Peter–Weyl (compact groups), highest-weight classification of finite-dimensional irreps of semisimple Lie algebras (Cartan–Weyl, root/weight lattices), and induced representations for noncompact groups. ESTABLISHED

The relativistic kinematic backbone is Wigner's classification: relativistic particles are unitary irreducible representations of the Poincaré group $\mathrm{ISO}(1,3)^\uparrow_+ = \mathbb{R}^{1,3}\rtimes\mathrm{SL}(2,\mathbb{C})$, labeled by mass $m\ge 0$ (Casimir $P^2$) and spin/helicity (little group: $\mathrm{SU}(2)$ for $m>0$, $\mathrm{ISO}(2)$/helicity for $m=0$). The need for projective representations and the universal cover $\mathrm{SL}(2,\mathbb{C})\to\mathrm{SO}(1,3)$ is why half-integer spin (spinors) exist. ESTABLISHED See domains/particle-physics.md.

6. Measure, probability, and the path integral

Classical probability is Kolmogorov's: a measure space $(\Omega,\mathcal F,P)$ with $P(\Omega)=1$, random variables as measurable functions, expectation as Lebesgue integral. Quantum theory generalizes this to a noncommutative probability over a von Neumann algebra with a state, and to POVM-valued observables. ESTABLISHED See domains/information-theory.md.

The path integral $\langle q_f|e^{-iHT/\hbar}|q_i\rangle = \int\mathcal{D}q\, e^{iS[q]/\hbar}$ is rigorous in two regimes and heuristic in general:

• QM (finitely many degrees of freedom), Euclidean ($t\to-i\tau$): the Feynman–Kac formula expresses $e^{-\tau H}$ via the genuine Wiener measure on continuous paths — fully rigorous. ESTABLISHED
• The Lorentzian functional measure $\mathcal{D}q\,e^{iS}$ is not a measure: there is no translation-invariant $\sigma$-additive measure on infinite-dimensional spaces, and $e^{iS}$ oscillates without absolute convergence. ESTABLISHED that no such measure exists; the object is heuristic.
• In QFT the would-be measure $\mathcal{D}\phi$ lives on distributions, requires renormalization, and exists rigorously only for super-renormalizable models in low dimensions (constructive QFT: $P(\phi)_2$, $\phi^4_3$; Glimm–Jaffe, Nelson, Osterwalder–Schrader). The Osterwalder–Schrader axioms give precise conditions (reflection positivity above all) under which a Euclidean measure reconstructs a relativistic QFT. ESTABLISHED

7. Topology in physics

Topology enters through global features invisible to local field equations: homotopy groups classify defects and solitons ($\pi_1$ vortices, $\pi_2$ monopoles, $\pi_3$ Skyrmions/instantons via $\pi_3(G)=\mathbb{Z}$ for simple $G$); cohomology and characteristic classes classify anomalies and flux quantization (Dirac quantization $\leftrightarrow c_1\in H^2(M,\mathbb{Z})$); $K$-theory and twisted/equivariant variants classify topological phases of matter and D-brane charges. The integer quantum Hall conductance is a first Chern number (TKNN invariant). Anomalies are computed by index theory and descend via the Stora–Zumino descent equations. ESTABLISHED

8. Category theory and higher structures

Functorial QFT (Atiyah–Segal axioms) recasts a TQFT as a symmetric monoidal functor $Z:\mathrm{Bord}_n\to\mathcal{C}^{\otimes}$: manifolds $\mapsto$ vector spaces, bordisms $\mapsto$ linear maps, gluing $\mapsto$ composition. The cobordism hypothesis (Baez–Dolan; proof program by Lurie) characterizes fully extended TQFTs by their value on a point — a fully dualizable object in a symmetric monoidal $(\infty,n)$-category. The framework is rigorous and Lurie's proof is widely accepted INFERENCE, but the full written details remain less universally vetted than a textbook theorem. Factorization algebras (Costello–Gwilliam) rigorously encode the operator-product structure of perturbative QFT; this is ESTABLISHED as mathematics, though its full equivalence to physicists' nonperturbative QFT is only partial.

Foundational assumptions

See ASSUMPTIONS_LEDGER.md for the cross-domain rollup.

Domain of validity

Each layer is a theorem in a sharp regime and a heuristic beyond it.

1. Hilbert-space QM with the spectral/Stone machinery is fully rigorous for finitely many degrees of freedom with well-posed self-adjoint Hamiltonians; the subtleties (domains, self-adjoint extensions) are controllable. ESTABLISHED
2. Stone–von Neumann uniqueness holds only for finitely many canonical pairs. With infinitely many degrees of freedom (QFT, the thermodynamic limit) there are uncountably many unitarily inequivalent representations — not a pathology but the correct setting for distinct phases, superselection, and spontaneous symmetry breaking. This is precisely where the "one Hilbert space, one Fock vacuum" intuition breaks. ESTABLISHED
3. Differential geometry / gauge theory as connections is exact classically and as a quantum kinematic framework; its quantization (the Yang–Mills measure) is rigorous only perturbatively in 4D.
4. Constructive QFT rigorously builds interacting models in $d=2$ ($P(\phi)_2$, Yukawa$_2$, Thirring) and $d=3$ ($\phi^4_3$), exploiting super-renormalizability; $d=4$ interacting theories are not constructed. ESTABLISHED
5. Algebraic QFT (Haag–Kastler nets) is a rigorous structural theory wherever local nets exist, including QFT on a fixed curved background; it does not by itself construct interacting 4D dynamics.
6. Euclidean methods / OS reconstruction are valid when reflection positivity holds and the continuation back to Lorentzian signature is justified — generically true for static/stationary spacetimes, problematic for generic Lorentzian/cosmological geometries.
7. The whole edifice presupposes a fixed (possibly curved) background spacetime. It has no established validity once the metric itself is quantized: quantum gravity lies outside the domain of all these frameworks.

Where it breaks down

• Interacting QFT in 4D (Yang–Mills, the Standard Model). No rigorous nonperturbative construction exists. Perturbation series are asymptotic (Dyson) and generically divergent; QED has a Landau pole and $\phi^4_4$ is provably trivial (Aizenman 1981; Aizenman–Duminil-Copin 2021), i.e. its continuum limit is free — strongly suggesting survival only as a cutoff EFT. A true incompleteness of the rigorous program (the Clay mass-gap problem), not a logical inconsistency. OPEN
• The Lorentzian functional measure. Literal nonexistence; meaning only via Euclidean continuation, oscillatory-integral surrogates, or perturbation theory. Domain-of-validity / abuse-of-notation issue. ESTABLISHED
• Quantization ambiguities. The classical $\to$ quantum map is not a homomorphism: the Groenewold–van Hove no-go theorem proves there is no quantization map respecting $\{\cdot,\cdot\}\mapsto\frac{1}{i\hbar}[\cdot,\cdot]$ on the whole Poisson algebra. Operator ordering, choice of polarization (geometric quantization), and self-adjoint extensions are genuine ambiguities fixed only by extra physical input. A fundamental obstruction, not a gap to be closed. ESTABLISHED
• Wick rotation in generic curved/Lorentzian spacetimes. The Euclidean$\leftrightarrow$Lorentzian dictionary requires a real Euclidean section and reflection positivity, generically absent in time-dependent or cosmological spacetimes (e.g. de Sitter). Without it, neither constructive techniques nor OS reconstruction apply. OPEN in those settings.
• Gauge-fixing and the global configuration space. The Gribov ambiguity: in non-abelian gauge theory no global continuous gauge-fixing section exists (Singer's theorem — the gauge orbit space is topologically obstructed), so Faddeev–Popov gauge-fixing is only locally valid and the BRST/path-integral construction is incomplete nonperturbatively. A genuine topological obstruction. ESTABLISHED
• Measurement vs. unitary evolution. Unitary Schrödinger evolution and projective Born-rule update are two different dynamical laws; reconciling them (the measurement problem) is unresolved. Interpretational, not a mathematical inconsistency — each interpretation is internally consistent but they disagree on ontology. CONTESTED
• QFT + dynamical gravity. Perturbatively quantized GR is non-renormalizable ('t Hooft–Veltman one-loop matter result; Goroff–Sagnotti two-loop pure-gravity divergence). The smooth-manifold + quantized-metric framework breaks at the Planck scale — the outer boundary of this entire domain. OPEN

See GAPS_AND_CONTRADICTIONS.md for how these are classified (inconsistency vs. domain mismatch vs. unsolved-but-consistent).

Open problems (internal)

• Existence of 4D quantum Yang–Mills with a mass gap (Clay Millennium Problem). Construct a nontrivial 4D YM theory satisfying the Wightman/OS axioms with $\sigma(H)\subset\{0\}\cup[\Delta,\infty)$, $\Delta>0$. Expected true on lattice/numerical and confinement grounds; no rigorous proof. OPEN
• Nonperturbative construction of the Standard Model (or any realistic 4D non-asymptotically-free theory). $\phi^4_4$ triviality ESTABLISHED and the QED Landau pole INFERENCE suggest some sectors are only EFTs; lattice chiral fermions and the sign problem add obstructions. OPEN/CONTESTED
• Rigorous nonperturbative definition of the (Lorentzian) 4D path integral / functional measure, beyond known low-dimensional and Euclidean reflection-positive cases. OPEN
• Derivation vs. postulation of the Born rule and resolution of the measurement problem. Gleason and envariance constrain it; Everettian derivations are disputed. OPEN/CONTESTED
• Why complex Hilbert space? Necessity of $\mathbb{C}$ over $\mathbb{R}$ or $\mathbb{H}$, and of the Hilbert-space framework itself. Operational reconstructions (Hardy 2001; Chiribella–D'Ariano–Perinotti) derive QM from axioms, but the complex-field choice and its experimental tests rest on auxiliary independence assumptions. OPEN
• Complete proof/dissemination of the cobordism hypothesis and a nonperturbative higher-categorical formulation of realistic 4D QFT. OPEN to partially settled
• Reconciling QFT with dynamical/quantum gravity at and below the Planck scale; the smooth-manifold substrate may not survive. OPEN
• The "unreasonable effectiveness of mathematics" (Wigner). Not a conjecture with an ordinary truth value but a genuine, unresolved meta-question. Partial deflationary accounts (selection of effective math, symmetry/representation-theoretic constraint, cognitive/anthropic) exist; none is decisive. Flag as a question, never a settled doctrine. OPEN/philosophical

See OPEN_PROBLEMS.md and HYPOTHESES.md.

Connections to other frameworks

• Quantum mechanics — Mathematics supplies the entire kinematic substrate (Hilbert space, self-adjoint observables, spectral theorem, Stone, Gleason, Wigner). The clash is at measurement: the linear/unitary structure is silent on collapse, so the foundational debate is downstream of, not contained in, the mathematics.
• Quantum field theory — Where the mathematics is most powerful and most incomplete: operator algebras (type III), representation theory (Wigner classification), and geometry (gauge connections) are exact, but the dynamics/measure are rigorous only in low dimensions. The Yang–Mills mass gap sits exactly on this seam.
• General relativity and quantum gravity — Differential geometry is shared bedrock. The clash: GR makes the metric dynamical, but every rigorous QFT framework presupposes a fixed background; the smooth-manifold assumption is the suspected casualty.
• Statistical mechanics / thermodynamics — Deep unification via KMS states and Tomita–Takesaki: equilibrium states are modular states of the observable algebra, so "temperature" and "time" share an algebraic origin. Reflection positivity links Euclidean QFT to classical lattice systems; phase transitions correspond to inequivalent representations.
• Classical mechanics — The symplectic geometry of phase space, Poisson brackets, and the Groenewold–van Hove obstruction define the classical $\to$ quantum boundary discussed above.
• Information theory and probability — Quantum theory is a noncommutative generalization of Kolmogorov probability (states on a von Neumann algebra, POVMs); information-theoretic reconstructions try to derive the Hilbert-space formalism, and modular relative entropy feeds back into QFT energy conditions and holography.
• Cosmology — Inherits the Wick-rotation breakdown: defining path integrals and QFT in de Sitter/time-dependent geometries is open.
• Pure mathematics (see the in-text references and domains/mathematics.md cross-links) — Bidirectional and historically transformative: physics (QFT/string theory) conjectured mirror symmetry, Seiberg–Witten invariants, the Jones polynomial via Chern–Simons, and monstrous moonshine, later proved by mathematicians; mathematics supplies index theory, operator algebras, and representation theory. The recurring clash is rigor: physical derivations are heuristic until mathematized, and some (path integrals) resist full rigor.

See UNIFICATION_LANDSCAPE.md and UNIFYING_PRINCIPLES.md for the cross-cutting roles of symmetry, locality, and the algebra/state duality; CONSTANTS_AND_SCALES.md for where these structures meet physical scales.

Key references

Full entries in BIBLIOGRAPHY.md. Standard, real sources only:

• M. Reed and B. Simon, Methods of Modern Mathematical Physics, Vols. I–IV (Academic Press, 1972–1979) — the canonical rigorous reference for functional analysis: self-adjointness and extensions, the spectral theorem, Stone's theorem, scattering.
• R. Haag, Local Quantum Physics: Fields, Particles, Algebras, 2nd ed. (Springer, 1996) — algebraic QFT, type-III local algebras, modular theory, DHR superselection, Reeh–Schlieder, Bisognano–Wichmann.
• J. Glimm and A. Jaffe, Quantum Physics: A Functional Integral Point of View, 2nd ed. (Springer, 1987) — constructive QFT, the rigorous Euclidean path integral, OS axioms, $P(\phi)_2$ and $\phi^4_3$.
• M. Takesaki, Theory of Operator Algebras, Vols. I–III (Springer, 1979/2003) — definitive treatise on C*- and von Neumann algebras, the Murray–von Neumann/Connes type classification, and Tomita–Takesaki modular theory with the KMS condition.
• M. Nakahara, Geometry, Topology and Physics, 2nd ed. (IOP, 2003) — manifolds, fiber bundles, connections, curvature, characteristic classes, Atiyah–Singer, anomalies, instantons.
• A. O. Barut and R. Rączka, Theory of Group Representations and Applications (PWN, 1977) — Lie groups, Lie algebras, unitary representations, induced representations, Wigner's classification.
• A. Jaffe and E. Witten, "Quantum Yang–Mills Theory" (Clay Mathematics Institute Millennium Problem description, 2000) — the official statement of the existence and mass-gap problem.
• M. Aizenman and H. Duminil-Copin, "Marginal triviality of the scaling limits of critical 4D Ising and $\phi^4_4$ models," Annals of Mathematics 194 (2021), 163–235 (with M. Aizenman, Phys. Rev. Lett., 1981) — rigorous triviality of $\phi^4$ in four dimensions.
• J. Lurie, "On the Classification of Topological Field Theories," Current Developments in Mathematics 2008 (2009), 129–280 — the cobordism-hypothesis program.
• K. Costello and O. Gwilliam, Factorization Algebras in Quantum Field Theory, Vols. 1–2 (Cambridge Univ. Press, 2017/2021) — rigorous perturbative QFT via factorization algebras and the BV formalism.
• E. P. Wigner, "The Unreasonable Effectiveness of Mathematics in the Natural Sciences," Comm. Pure Appl. Math. 13 (1960), 1–14 — the original framing of the meta-question.
• Landmark structural theorems: V. Bargmann, "On unitary ray representations of continuous groups," Ann. Math. 59 (1954); A. M. Gleason, "Measures on the closed subspaces of a Hilbert space," J. Math. Mech. 6 (1957); and the Gelfand–Naimark / Segal foundational papers.

See also

• README.md · AGENTS.md · EPISTEMICS.md · THEORY_MAP.md
• OPEN_PROBLEMS.md · GAPS_AND_CONTRADICTIONS.md · HYPOTHESES.md · ASSUMPTIONS_LEDGER.md
• UNIFICATION_LANDSCAPE.md · UNIFYING_PRINCIPLES.md · CONSTANTS_AND_SCALES.md · FINDINGS.md
• GLOSSARY.md · ROADMAP.md · CHANGELOG.md
• Domains: quantum-mechanics.md · quantum-field-theory.md · general-relativity.md · classical-mechanics.md · thermodynamics.md · statistical-mechanics.md · cosmology.md · particle-physics.md · information-theory.md

References

See BIBLIOGRAPHY.md for full citation entries and additional standard sources.