# References — Chapter ᚍ(K): The listening room *Teclam Áirme, the Mathematical Compendium — a copy for AI readers of https://www.teclamairme.com/#ref-ch-15, as of 2026-09-30. Links are relative to https://www.teclamairme.com/fortheaireaders/.* [Guide for AI readers](index.md) · [The chapter](ch15.md) *Each entry: one plain sentence saying exactly what it is cited for, its status, then every source with a note saying what that source actually carries. "Own computation" / "Own reading" entries are the author's own work, not outside sources.* ### 15.1 {#r-15.1} A Hilbert space is a complete inner-product space; l² and L² are standard examples, and the inner product defines norm and orthogonality. *Status: confirmed* - [Encyclopedia of Mathematics, Hilbert space](https://encyclopediaofmath.org/wiki/Hilbert_space) — defines Hilbert space, gives l² and L² as examples, and defines orthogonality by vanishing inner product. Its numbered axioms require the space to be infinite-dimensional, but the article adds that this condition is often omitted, so the source carries BOTH conventions. Recorded because this entry is cited in support of the wide definition while its headline axioms give the narrow one; see 15.24. ### 15.2 {#r-15.2} Hilbert’s integral-equation work around 1909 established the basis for the infinite-dimensional space bearing his name; von Neumann made abstract Hilbert space the rigorous common setting for quantum mechanics and published his foundational book in 1932. *Status: confirmed* - [MacTutor, David Hilbert](https://mathshistory.st-andrews.ac.uk/Biographies/Hilbert/) — Dates Hilbert’s integral-equation work to about 1909 and connects it directly to infinite-dimensional Hilbert space and quantum mechanics. - [Encyclopedia of Mathematics, Hilbert space](https://encyclopediaofmath.org/wiki/Hilbert_space) — Credits Hilbert with l² and L² in his integral-equation work and von Neumann, Riesz and Stone with the abstract definition. - [MacTutor, The quantum age begins](https://mathshistory.st-andrews.ac.uk/HistTopics/The_Quantum_age_begins/) — Records von Neumann’s 1932 rigorous operator-algebra foundation for quantum theory. Editorial note: The inherited 1906–1912/1927/1929 compressed chronology was replaced by the dates these sources actually support. ### 15.3 {#r-15.3} Bra-ket notation represents inner products as ⟨a|b⟩; orthogonal vectors have zero inner product, and the computational basis states of a qubit are orthonormal. A bra times a ket is a scalar, and the standard basis vectors form an orthonormal basis. *Status: confirmed* - [Dirac, A new notation for quantum mechanics](https://doi.org/10.1017/S0305004100021162) — Primary 1939 paper introducing Dirac’s notation. - [Encyclopedia of Mathematics, Hilbert space](https://encyclopediaofmath.org/wiki/Hilbert_space) — Defines orthogonality as a zero inner product. - [IBM Quantum Learning, Quantum information](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/single-systems/quantum-information) — Gives the standard qubit bases, inner products and measurement probabilities. - [IBM Quantum Learning, Basics of quantum information: Inner products and projections](https://quantum.cloud.ibm.com/learning/courses/basics-of-quantum-information/quantum-circuits/inner-products-and-projections) — “We also have that the product of a bra vector and a ket vector, viewed as matrices either having a single row or a single column, results in a scalar.”; “for any classical state set Σ, the set of all standard basis vectors {|a⟩ : a∈Σ} is an orthonormal basis.” ### 15.4 {#r-15.4} Orthogonal pure quantum states can be perfectly distinguished, whereas non-orthogonal states cannot; the same framework yields the no-cloning limitation. Any two orthogonal states can be discriminated perfectly, non-orthogonal ones cannot; a cloning unitary can work only for states that are identical up to phase or orthogonal. *Status: confirmed* - [IBM Quantum Learning, Quantum circuits](https://learning.quantum.ibm.com/course/basics-of-quantum-information/quantum-circuits) — the lesson introduction: it introduces inner products, orthogonality and projective measurements, and says “we’ll derive fundamental limitations on quantum information, including the no-cloning theorem and the impossibility to perfectly discriminate non-orthogonal quantum states” (the derivations are quoted on the Limitations line below). - [IBM Quantum Learning, Basics of quantum information: Limitations on quantum information](https://quantum.cloud.ibm.com/learning/courses/basics-of-quantum-information/quantum-circuits/limitations-on-quantum-information) — “if we have two quantum states |ψ⟩ and |φ⟩ that are not orthogonal … then it’s impossible to discriminate them perfectly”; “It is possible, by the way, to perfectly discriminate any two states that are orthogonal”; “The no-cloning theorem shows it is impossible to create a perfect copy of an unknown quantum state.” - [No-cloning theorem, Wikipedia](https://en.wikipedia.org/wiki/No-cloning_theorem) — read as raw wikitext: “Hence by the Cauchy–Schwarz inequality either |φ⟩ = e^{iβ}|ψ⟩ or |φ⟩ is orthogonal to |ψ⟩. However, this cannot be the case for two arbitrary states. Therefore, a single universal U cannot clone a general quantum state.” ### 15.5 {#r-15.5} The Born rule assigns probabilities from squared amplitudes; Born proposed the statistical interpretation in 1926 and received half of the 1954 Nobel Prize for it. *Status: confirmed* - [Nobel Prize, Physics 1954 summary](https://www.nobelprize.org/prizes/physics/1954/summary/) — Official award citation credits Born for the statistical interpretation of the wavefunction. - [Max Born, Nobel Lecture](https://www.nobelprize.org/uploads/2018/06/born-lecture.pdf) — Born’s primary lecture states that squared amplitudes give state probabilities and recounts his 1926 statistical interpretation. ### 15.6 {#r-15.6} The Hadamard gate sends |0⟩ to |+⟩=(|0⟩+|1⟩)/√2, giving overlap 1/√2 with |0⟩ and measurement probability 1/2. *Status: confirmed* - [IBM Quantum Learning, Quantum information](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/single-systems/quantum-information) — Gives H|0⟩=|+⟩ and the equal computational-basis probabilities. ### 15.7 {#r-15.7} The Pauli X matrix flips |0⟩ and |1⟩, serves as the quantum NOT gate, and squares to the 2×2 identity; Pauli introduced the matrices in 1927. *Status: confirmed* - [Encyclopedia of Mathematics, Pauli matrices](https://encyclopediaofmath.org/wiki/Pauli_matrices) — Gives the matrices, their 1927 attribution, and their algebra, including σᵢ²=I. - [IBM Quantum Composer reference](https://quantum.cloud.ibm.com/docs/en/guides/composer) — Identifies Pauli X as the standard NOT gate. ### 15.8 {#r-15.8} Grassmann variables anticommute and square to zero and are used for fermionic path integrals; Pauli formulated the exclusion principle in 1925, and it applies to fermions. Grassmann variables anticommute and square to zero, and their anticommutation is that of the Pauli exclusion principle. *Status: confirmed* - [MIT OpenCourseWare, Path Integrals of Fermions](https://ocw.mit.edu/courses/8-323-relativistic-quantum-field-theory-i-spring-2023/resources/ocw_8323_lecture19_2023apr19_mp4/) — Graduate lecture covers Grassmann anticommutation and fermionic path integrals. - [Nobel Prize, Wolfgang Pauli facts](https://www.nobelprize.org/prizes/physics/1945/pauli/facts/) — Dates Pauli’s exclusion principle to 1925 and states its same-state prohibition. - [Nobel Prize, Physics 1901–2000](https://www.nobelprize.org/nobel_prizes/themes/physics/karlsson/) — Connects the exclusion principle with half-integer-spin fermions; “This principle was later found to be associated with the symmetry of wave functions for particles of half-integer spins in general, distinguishing what is now called fermions from the bosonic particles whose spins are integer multiples of” (the unit is an image on the page, not in its text). - [Grassmann number, Wikipedia](https://en.wikipedia.org/wiki/Grassmann_number) — read as raw wikitext: “the generators and thus the algebra are anti-commuting,” θᵢθⱼ = −θⱼθᵢ; “The squares of the elements vanish”; “the anti-commuting behavior can be strongly identified with the quantum-mechanical behavior of fermions: the anti-commutation is that of the Pauli exclusion principle.” ### 15.9 {#r-15.9} The canonical relation between a position and its own momentum is [x̂, p̂] = iℏI; Born and Jordan supplied the canonical matrix-mechanics relation in 1925. For the three-dimensional form with the Kronecker delta, and therefore for x-position commuting with y-momentum, see 15.23: neither source below prints it. The sign follows [x̂, p̂ₓ] = x̂p̂ₓ − p̂ₓx̂ = iℏI. *Status: confirmed* - [Encyclopedia of Mathematics, Uncertainty principle](https://encyclopediaofmath.org/wiki/Uncertainty_principle) — prints the same-component canonical commutators with the identity operator, [p̂ₓ, x̂] = [p̂_y, ŷ] = [p̂_z, ẑ] = iℏE, with no cross terms and no Kronecker delta. The annotation here used to read “Gives the Cartesian canonical commutators,” which 15.23 contradicts in the same reference list; corrected 17 September 2026. Its sign is the reverse of this book’s: [x̂, p̂] = iℏI gives [p̂ₓ, x̂] = −iℏI. The book does not follow this page’s sign; the Canonical commutation relation line carries it. - [Born–Jordan Quantization and the Equivalence of Matrix and Wave Mechanics](https://arxiv.org/abs/1405.2519) — the abstract dates the Born and Jordan paper to 1925: “The aim of the famous Born and Jordan 1925 paper was to put Heisenberg’s matrix mechanics on a firm mathematical basis.” It does not print the commutation relation itself. - [Canonical commutation relation, Wikipedia](https://en.wikipedia.org/wiki/Canonical_commutation_relation) — read as raw wikitext: “[x̂, p̂ₓ] = iℏI … where [x, pₓ] = x pₓ − pₓ x is the commutator of x and pₓ.” ### 15.10 {#r-15.10} For |+⟩, σz yields +1 and −1 with equal probability and expectation zero; equivalently σz|+⟩=|−⟩ and ⟨+|−⟩=0. {|+⟩, |−⟩} is an orthonormal basis, so ⟨+|−⟩ = 0. *Status: confirmed* - [Encyclopedia of Mathematics, Pauli matrices](https://encyclopediaofmath.org/wiki/Pauli_matrices) — Gives σz=diag(1, −1) and its eigenvalues. - [IBM Quantum Learning, Quantum information](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/single-systems/quantum-information) — defines the plus state and the minus state, and of a standard basis measurement says “Measuring either state produces the same output distribution as the other,” 0 and 1 each with probability 1/2 (their orthogonality is on the Inner products line below). - [IBM Quantum Learning, Basics of quantum information: Inner products and projections](https://quantum.cloud.ibm.com/learning/courses/basics-of-quantum-information/quantum-circuits/inner-products-and-projections) — “{|+⟩,|−⟩} is an orthonormal basis for the 2-dimensional space corresponding to a single qubit.” ### 15.11 {#r-15.11} Self-adjoint observables have real spectrum and orthogonal eigenspaces for distinct eigenvalues; Schrödinger published wave mechanics in 1926 and showed its equivalence to matrix mechanics. The first 1926 paper, Quantisierung als Eigenwertproblem, treated time-independent systems; a third, in May, showed the equivalence with matrix mechanics. *Status: confirmed* - [Encyclopedia of Mathematics, Linear operator](https://encyclopediaofmath.org/wiki/Linear_operator) — States that self-adjoint spectra are real, distinct eigenspaces are orthogonal, and self-adjoint operators represent quantum observables. - [MIT OpenCourseWare, Mathematical formalism of quantum mechanics](https://ocw.mit.edu/courses/18-700-linear-algebra-fall-2013/cc993d0c7953d1aa7d57997b1e43b7da_MIT18_700F13_qntm_mechnc.pdf) — Explains self-adjoint observables, spectral values and the finite-dimensional spectral decomposition. - [MacTutor, Erwin Schrödinger](https://mathshistory.st-andrews.ac.uk/Biographies/Schrodinger/) — Documents Schrödinger’s 1926 papers and his equivalence argument for wave and matrix mechanics; “Schrödinger published his revolutionary work relating to wave mechanics and the general theory of relativity in a series of six papers in 1926”; “In its most general form, the time-dependent Schrödinger equation describes the time evolution of closed quantum physical systems.” - [Erwin Schrödinger, Wikipedia](https://en.wikipedia.org/wiki/Erwin_Schr%C3%B6dinger) — read as raw wikitext: “In January 1926, Schrödinger published in Annalen der Physik the paper “Quantisierung als Eigenwertproblem” (Quantization as an Eigenvalue Problem)… In this paper, he gave a “derivation” of the wave equation for time-independent systems”; “A third paper, published in May, showed the equivalence of his approach to that of Werner Heisenberg’s matrix mechanics.”; “A fourth paper in this series showed how to treat problems in which the system changes with time, as in scattering problems.” ### 15.12 {#r-15.12} Kennard proved ΔxΔp≥ℏ/2 in 1927, Robertson generalized it in 1929, and Gaussian wavepackets attain equality, so min ΔxΔp/ℏ=1/2. *Status: confirmed* - [Encyclopedia of Mathematics, Uncertainty principle](https://encyclopediaofmath.org/wiki/Uncertainty_principle) — Gives the inequality and the Kennard 1927 and Robertson 1929 chronology; its comments: “W. Heisenberg [a1] presented this uncertainty principle in 1927. In the same year E.H. Kennard [a2] found relation (2), while the general relation (1) was proved by H.P. Robertson [a3] in 1929..” - [University of Texas at Austin, Quantum Mechanics](https://farside.ph.utexas.edu/teaching/qmech/qmech.pdf) — States that equality is achieved by Gaussian wavepackets. ### 15.13 {#r-15.13} For the CHSH expression, local hidden-variable correlations are bounded by 2 and quantum correlations by 2√2; quantum experiments can approach the latter value. *Status: confirmed* - [Poh, Joshi, Cerè, Cabello and Kurtsiefer, Approaching Tsirelson’s Bound in a Photon Pair Experiment](https://harvest.aps.org/v2/journals/articles/10.1103/PhysRevLett.115.180408/fulltext) — Primary Physical Review Letters article states both bounds and reports an experiment approaching 2√2. ### 15.14 {#r-15.14} Clauser, Aspect and Zeilinger received the 2022 Nobel Prize for entangled-photon and Bell-inequality experiments; loophole-free Bell tests were reported in 2015. The prize followed the Bell-test experiments; the first loophole-free tests were published in 2015. *Status: confirmed* - [Nobel Prize, Physics 2022 summary](https://www.nobelprize.org/prizes/physics/2022/summary/) — Official award citation. - [NIST, A significant-loophole-free test of Bell’s theorem with entangled photons](https://www.nist.gov/publications/significant-loophole-free-test-bells-theorem-entangled-photons) — Records the 2015 test and observed violation. - [Bell test, Wikipedia](https://en.wikipedia.org/wiki/Bell_test) — read as raw wikitext: “These efforts to experimentally validate violations of the Bell inequalities resulted in John Clauser, Alain Aspect, and Anton Zeilinger being awarded the 2022 Nobel Prize in Physics”; “In 2015 the first three significant-loophole-free Bell-tests were published within three months by independent groups in Delft, Vienna and Boulder.” ### 15.15 {#r-15.15} Fourier presented his heat memoir in 1807 and published the book in 1822; over a full period, sines and cosines are orthogonal, including ∫₀²π sin x cos x dx=0. *Status: confirmed* - [MacTutor, Joseph Fourier](https://mathshistory.st-andrews.ac.uk/Biographies/Fourier/) — Dates the 1807 presentation, the objections by Lagrange and Laplace, and the 1822 publication. - [MIT OpenCourseWare, Orthogonality Relations](https://ocw.mit.edu/courses/18-03sc-differential-equations-fall-2011/618691a1edaa81b0a75303e09b4cb4b5_MIT18_03SCF11_s21_7text.pdf) — Derives the sine-cosine orthogonality integrals over a full period. Editorial note: MIT’s derivation supports the displayed integral and its vanishing. ### 15.16 {#r-15.16} Under a consistent normalization, Plancherel’s theorem makes the Fourier transform unitary on L² and preserves its norm; Plancherel established it in 1910. With f̂(ξ) = ∫f(x)e^(−2πixξ)dx the extension to L² is unitary. *Status: confirmed* - [Encyclopedia of Mathematics, Plancherel theorem](https://encyclopediaofmath.org/index.php?title=Plancherel_theorem) — States the unitary L² extension, norm identity and 1910 date. - [Plancherel theorem, Wikipedia](https://en.wikipedia.org/wiki/Plancherel_theorem) — read as raw wikitext: proven “by Michel Plancherel in 1910”; the transform of an L¹ function “is defined as the Lebesgue integral” f̂(ξ) = ∫ f(x)e^(−2πixξ) dx; its L² extension is an isometry, and “This isometry is actually a unitary map.” ### 15.17 {#r-15.17} The limiting Fourier overshoot at a jump is Si(π)/π−1/2≈0.089489872, and its spike narrows without losing relative height; the chapter leaves the number’s arithmetic classification open. *Status: judgement* - [NIST Digital Library of Mathematical Functions, §6.16(i)](https://dlmf.nist.gov/6.16.i) — Gives the sine-integral formula, limiting overshoot and Gibbs’s 1899 report. - [Wolfram MathWorld, Wilbraham–Gibbs Constant](https://mathworld.wolfram.com/Wilbraham-GibbsConstant.html) — Gives the constant, decimal value and Wilbraham 1848/Gibbs 1899 history; page last updated 2 September 2026. Editorial note: Neither standard reference assigns a rational, irrational, algebraic or transcendental class, and targeted searches found no theorem settling it. The page therefore states only that this review leaves the classification open, rather than claiming an exhaustive proof of the literature’s absence. ### 15.18 {#r-15.18} L² is the complete Hilbert space of square-integrable functions modulo equality almost everywhere; Riesz and Fischer independently proved the relevant L²/ℓ² Fourier-coefficient theorem in 1907. Riesz and Fischer proved the theorem independently in 1907; an element of L² is a class of functions equal almost everywhere. *Status: confirmed* - [Encyclopedia of Mathematics, Hilbert space](https://encyclopediaofmath.org/wiki/Hilbert_space) — Defines the L² example and completeness framework. - [Encyclopedia of Mathematics, Orthogonal series](https://encyclopediaofmath.org/wiki/Orthogonal_series) — Dates the Riesz and Fischer theorem to 1907 and states the L²/ℓ² correspondence. - [Square-integrable function, Wikipedia](https://en.wikipedia.org/wiki/Square-integrable_function) — read as raw wikitext: “The square integrable functions (in the sense mentioned in which a “function” actually means an equivalence class of functions that are equal almost everywhere) form an inner product space.” - [Riesz–Fischer theorem, Wikipedia](https://en.wikipedia.org/wiki/Riesz%E2%80%93Fischer_theorem) — read as raw wikitext: “The theorem was proven independently in 1907 by Frigyes Riesz and Ernst Sigismund Fischer.” ### 15.19 {#r-15.19} Position- and momentum-space wavefunctions are Fourier transforms of one another, and Gaussian wavepackets transform to Gaussians and attain the position-momentum uncertainty minimum. Canonical conjugate quantities are those related by definition such that one is the Fourier transform of the other. *Status: confirmed* - [Durham University, Momentum-space wave function](https://www.maths.dur.ac.uk/users/kasper.peeters/mathphys/momentum_space.html) — Gives the position/momentum Fourier-transform pair and operator representations. - [University of Texas at Austin, Quantum Mechanics](https://farside.ph.utexas.edu/teaching/qmech/qmech.pdf) — Gives Gaussian position/momentum wavepackets and the minimum uncertainty product. - [Canonical commutation relation, Wikipedia](https://en.wikipedia.org/wiki/Canonical_commutation_relation) — read as raw wikitext: “the canonical commutation relation is the fundamental relation between canonical conjugate quantities (quantities which are related by definition such that one is the Fourier transform of another).” ### 15.20 {#r-15.20} The chapter’s displayed inner products, Pauli expectations, trace and Grassmann anticommutator follow directly from the cited definitions and matrices; for the cross-coordinate commutator see 15.23. Grassmann variables anticommute, θ₁θ₂ = −θ₂θ₁, and square to zero. *Status: confirmed* - [IBM Quantum Learning, Quantum information](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/single-systems/quantum-information) — Gives |0⟩, |1⟩, |+⟩, |−⟩ and their measurement and inner-product relations. - [Encyclopedia of Mathematics, Pauli matrices](https://encyclopediaofmath.org/wiki/Pauli_matrices) — Gives the explicit matrices and σᵢ²=I. - [MIT OpenCourseWare, Path Integrals of Fermions](https://ocw.mit.edu/courses/8-323-relativistic-quantum-field-theory-i-spring-2023/resources/ocw_8323_lecture19_2023apr19_mp4/) — A video lecture on fermionic path integrals; the page itself prints no text of the anticommutation rule, which the Grassmann number line carries. - [Encyclopedia of Mathematics, Uncertainty principle](https://encyclopediaofmath.org/wiki/Uncertainty_principle) — Prints the same-component canonical commutators, in the order [p̂ₓ, x̂] and with the opposite sign to this book’s (see 15.9); for the cross-coordinate case see 15.23. - [Grassmann number, Wikipedia](https://en.wikipedia.org/wiki/Grassmann_number) — read as raw wikitext: “the generators and thus the algebra are anti-commuting,” θᵢθⱼ = −θⱼθᵢ; “The squares of the elements vanish”; “the anti-commuting behavior can be strongly identified with the quantum-mechanical behavior of fermions: the anti-commutation is that of the Pauli exclusion principle.” Editorial note: The cited external definitions and matrices support each displayed value. ### 15.21 {#r-15.21} The canonical commutator is operator-valued and has no finite-dimensional matrix realization, since a finite commutator has trace zero while a nonzero scalar multiple of the identity does not. *Status: confirmed* - [Canonical commutation relation, Wikipedia](https://en.wikipedia.org/wiki/Canonical_commutation_relation) — (§ Weyl relations): gives the trace obstruction — the relation Tr(AB) = Tr(BA) “gives a nonzero number on the right and zero on the left” — so [x̂, p̂] = iℏ has no finite-dimensional matrix realization. - [Fritz, On infinite-dimensional state spaces](https://arxiv.org/abs/1202.3817) — States that [x, p]=i can be realized exactly only on an infinite-dimensional Hilbert space. ### 15.22 {#r-15.22} The chapter epigraph is Feynman’s statement that nobody understands quantum mechanics, from the sixth Messenger Lecture, Probability and Uncertainty, delivered at Cornell in 1964 and collected as the sixth chapter of The Character of Physical Law, first published by the BBC in 1965. *Status: confirmed* - [Wikiquote, Richard Feynman](https://en.wikiquote.org/wiki/Richard_Feynman) — gives “I think I can safely say that nobody understands quantum mechanics” from The Character of Physical Law, in that book’s sixth lecture, “Probability and Uncertainty — the Quantum Mechanical View of Nature,” the Messenger Lectures at Cornell, November 1964. - [The Character of Physical Law, Wikipedia](https://en.wikipedia.org/wiki/The_Character_of_Physical_Law) — read as raw HTML: the infobox gives “Published 1965 (BBC)”; the lectures were delivered “at Cornell University, as part of the Messenger Lectures series” and “The BBC recorded the lectures and published a book under the same title the following year”; and the list of the seven topics has as its sixth “Probability and uncertainty – the quantum mechanical view of nature,” which is the sixth chapter the epigraph cites. ### 15.23 {#r-15.23} The three-dimensional canonical commutation relations carry a Kronecker delta, [x̂ᵢ, p̂ⱼ] = iℏδᵢⱼI, so a position coordinate commutes with a different component of the momentum and the two can be measured simultaneously. *Status: confirmed* - [University of Texas at Austin, Quantum Mechanics](https://farside.ph.utexas.edu/teaching/qmech/qmech.pdf) — Chapter 7, Three-Dimensional Quantum Mechanics, p. 95: Eqs. (7.15)–(7.17) display [xᵢ, xⱼ] = 0, [pᵢ, pⱼ] = 0 and [xᵢ, pⱼ] = iℏδᵢⱼ, and the text immediately beneath them says it is “perfectly possible to simultaneously measure two different positions coordinates, or two different components of the momentum,” the only restriction being a coordinate against its own conjugate momentum. Eq. (9.12), p. 116, repeats it as “the fundamental commutation relation for the xᵢ and the pᵢ.” - [Canonical commutation relation, Wikipedia](https://en.wikipedia.org/wiki/Canonical_commutation_relation) — The lead section gives the generalized relation [x̂ᵢ, p̂ⱼ] = iℏδᵢⱼ, whose Kronecker delta vanishes whenever the indices differ. Editorial note: Added by the V14 prose pass to close the chapter’s open WRONG-REF row. 15.9’s two pages do not carry the Kronecker-delta form: the Encyclopedia of Mathematics Uncertainty principle entry prints only the same-component commutators [p̂ₓ, x̂] = [p̂_y, ŷ] = [p̂_z, ẑ] = iℏE, with no cross terms and no delta, and de Gosson (arXiv:1405.2519) gives only the one-dimensional pq − qp = −iℏ1. The claim itself is standard; 15.23 is the source that states it. The University of Texas text was already cited in this chapter at 15.12 and 15.19, and the Wikipedia article at 15.21. ### 15.24 {#r-15.24} Von Neumann required infinite-dimensionality as a named axiom of abstract Hilbert space, deliberately, because that choice makes the space unique up to isomorphism and so establishes the equivalence of matrix and wave mechanics. Modern usage generally drops the requirement, so a finite-dimensional complete inner-product space such as C² counts as a Hilbert space. *Status: confirmed* - [Anthony Duncan, annotated translation of von Neumann, Mathematische Begründung der Quantenmechanik (1927)](https://arxiv.org/pdf/2406.02149) — gives the 1927 axioms A to E verbatim. Axiom C reads “H has infinitely many dimensions: i.e., there are an arbitrarily large number of linearly independent elements of H,” and von Neumann states that the five properties fully determine the space. This is the direct evidence that the infinite-dimensionality was required by name and on purpose. - [Robert B. Griffiths, Hilbert Space Quantum Mechanics, Carnegie Mellon qitd113, section 1.1](https://quantum.phys.cmu.edu/QCQI/qitd113.pdf) — names the split explicitly: the term is “often reserved for an infinite-dimensional inner product space” but “is often used nowadays ... in a way that includes finite-dimensional spaces,” and it calls a two-dimensional Hilbert space a qubit. - [Federico Zalamea, Chasing Individuation: Mathematical Description of Physical Systems, PhD thesis, Université Paris Diderot, 2016, p. 117](https://arxiv.org/pdf/1612.02959) — quotes von Neumann’s 1932 book, in the Beyer translation at p. 45, giving the dimension condition in two forms — C.(n), exactly n linearly independent vectors, and C.(infinity), arbitrarily many — followed by “We obtain a different space R, depending on which we decide upon.” READ THE CHAIN OF CUSTODY BEFORE RELYING ON THIS: it is Zalamea quoting Beyer quoting von Neumann, and Zalamea is a doctoral thesis rather than a peer-reviewed article. The Beyer text itself is lending-restricted and could not be obtained here, and Zalamea’s PDF would not extract for an independent check — two attempts recovered 48 KB of a 2.3 MB file, and the ar5iv HTML redirects to the abstract — so this source is carried on the reviewing agent’s reading and not on the coordinator’s. The title and document type above WERE checked against the arXiv record on 11 September 2026, because the first draft of this entry carried a wrong title. Nothing in the book’s prose or cards rests on it — the card says only what the first source above supports. Editorial note: Added 11 September 2026, closing the deep-research item DR2. The book had been speaking two conventions: chapter T10’s prose said a Hilbert space has infinitely many dimensions, while chapter T15 computed its whole quantum brace in C² and the hilbert card gave the general definition. A survey of eleven sources found ten allowing finite dimensions, four naming Cⁿ or Rⁿ as a Hilbert space outright (Rudin, Conway, Kreyszig, MathWorld), and Nielsen and Chuang section 2.1.4 using the terms interchangeably in finite dimension. The two apparent holdouts do not define against it: Sakurai avoids the phrase rather than restricting it, and Watrous says he prefers “complex Euclidean space” precisely because “Hilbert space” is the more general notion. The book now speaks the wide convention everywhere. Do not claim that von Neumann called a finite-dimensional space a Hilbert space; he did not. ### 15.25 {#r-15.25} Cites the CHSH inequality’s bound of 2 for local hidden-variable correlations and the derivation of the quantum bound 2√2. Clauser, Horne, Shimony and Holt published the inequality in 1969, and experiments have come close to the quantum limit 2√2. A local hidden-variable model computes each side’s outcome from that side’s own setting and a hidden state λ of the common source, never from the other side’s setting. *Status: confirmed* - [CHSH inequality, Wikipedia](https://en.wikipedia.org/wiki/CHSH_inequality) — gives the 1969 Clauser-Horne-Shimony-Holt inequality, its local-hidden-variable bound of 2, and Tsirelson’s quantum bound of 2√2; it is “a constraint—on the statistical occurrence of “coincidences” in a Bell test” and “In practice, the inequality is routinely violated by modern experiments in quantum mechanics.” Its derivation “start[s] with the standard assumption of independence of the two sides,” the hidden variable λ being “drawn from a fixed distribution of possible states of the source,” and writes E(a, b) = ∫ A(a, λ)B(b, λ)ρ(λ) dλ, “where A and B are the outcomes,” so each outcome depends only on its own side’s setting and λ; “The maximum violation predicted by quantum mechanics is 2√2 (Tsirelson’s bound).” - [Approaching Tsirelson’s bound in a photon pair experiment, arXiv](https://arxiv.org/abs/1506.01865) — Poh, Joshi, Ceré, Cabello and Kurtsiefer (2015): “a value 2.82759+-0.00051 is observed” and “the smallest distance with respect to Tsirelson’s bound ever reported, namely, 0.00084+-0.00051”