Quantum mechanics over real numbers fully reproduces standard quantum theory
Abstract: Standard quantum mechanics employs complex Hilbert spaces, but whether complex numbers are fundamental or merely convenient has long been debated. For decades, real-valued equivalents were considered mathematically possible but cumbersome. However, a landmark 2021 result claimed that any quantum theory based on real numbers is experimentally falsifiable via network Bell experiments. Yet, it remains an open question whether this falsification applies to all real-valued theories. Here we show that this conclusion rests on an incomplete real formulation, and we present a rigorous real-valued framework that perfectly reproduces all predictions of standard quantum mechanics, i.e. standard quantum mechanics. We demonstrate that the standard real tensor product ($\otimes_{\mathbb{R}}$) used in previous no-go theorems is algebraically incompatible with the rich structure of standard quantum mechanics. We present a real framework based on \ka space and prove that it is exactly isomorphic to standard quantum mechanics via an explicit bijection $γ$. The isomorphism extends to composite systems through a symplectic composition rule $\otimes{\ks}$ that replaces the Kronecker product. Consequently, our formulation achieves the maximal $\mathrm{CHSH}_{3}$ violation of $6\sqrt{2}$ using purely real variables, directly contradicting previous falsification claims. These results demonstrate that complex numbers are not fundamentally required by nature; rather, they encode a deeper real geometric structure that governs quantum interference and entanglement, settling this long debate.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Explain it Like I'm 14
Plain‑English Summary of the Paper
1) What is this paper about?
Quantum mechanics usually uses complex numbers (numbers with a real part and an “imaginary” part involving ). For a long time, people have wondered: are complex numbers truly necessary, or just a convenient shortcut?
This paper shows that you can describe all of quantum mechanics using only real numbers—if you organize them in the right way. The authors build a careful “real-number version” of quantum theory that makes exactly the same predictions as the usual complex-number version, even in tricky experiments that test quantum entanglement.
2) What questions are they asking?
In simple terms, the authors ask:
- Can every part of quantum mechanics be done with real numbers instead of complex numbers?
- If so, how do we correctly combine separate quantum systems (like two particles) using real numbers so that we still get the right physics?
- Do certain experiments that were thought to prove complex numbers are necessary actually just rely on using the wrong “glue” to combine real-number systems?
3) How did they approach it?
Think of this like translating a book from one language (complex numbers) into another (real numbers) without changing the story.
- The translator (called “γ”, gamma) turns complex-number objects into real-number objects and back again, perfectly. This ensures single systems (one particle) look and behave the same in both languages.
- For combining systems, the key is the “glue” (the tensor product). In usual quantum mechanics, the complex-number tensor product ties two systems together in a very specific way. If you naïvely replace complex numbers with real ones and use the ordinary real-number glue (the Kronecker product), you get the wrong combined space and wrong predictions.
- The authors introduce the correct real-number “glue,” called the symplectic tensor product (written ⊗K). It’s specially designed to mirror how complex numbers combine, but it uses only real numbers.
- Their real-number space is called a Kähler space. You can think of it as:
- a real vector space (a set of arrows you can add and scale),
- with a way to measure lengths/angles (the metric g),
- a way to record “twisty” area/phase information (the symplectic form ω),
- and a built‑in “turn by 90°” operation J that plays the role of multiplying by .
Together, these three ingredients (g, ω, J) let the real space imitate all the features of complex numbers—interference, phases, and entanglement—without ever writing a complex number.
Analogy:
- Complex numbers are like coordinates (x, y) in a plane; multiplying by is like rotating by 90°. The authors keep everything in real coordinates but include a fixed “rotate 90°” tool J so they never need explicitly.
- Combining systems is like snapping together LEGO pieces. Using the wrong connector (the standard real Kronecker product) breaks the model. Using the right connector (the symplectic product ⊗K) makes the pieces fit exactly as they do in the complex-number set.
4) What did they find, and why is it important?
Main findings:
- They construct an exact, two‑way translation (the map γ) between the usual complex‑number quantum theory and their real‑number Kähler theory.
- They prove that when you use their symplectic product ⊗K, combined systems behave exactly the same as in standard quantum mechanics.
- They show their real‑number theory achieves the same maximum score in a network Bell test (called CHSH3) as the complex‑number theory: . In 2021, other researchers claimed real‑number quantum theories couldn’t reach that score—but that claim used the wrong real-number “glue.” With the correct glue, the real theory matches the complex one.
Why it matters:
- This settles a long‑running debate: complex numbers aren’t strictly “required by nature” as separate ingredients. Instead, they are a neat way to describe a deeper real geometric structure (lengths, twists, and 90° turns) that fully captures quantum behavior.
- It means experiments can’t distinguish the two descriptions when the real-number version is done correctly. The physics—the predictions—are the same.
5) What’s the bigger picture?
- Complex numbers remain extremely useful and natural for calculations. But this paper shows they are not fundamental “things” you must assume; they can be replaced by real-number geometry that encodes the same information.
- This perspective can clarify how we think about quantum phases, interference, and entanglement: they come from the real geometric structure (g, ω, J) rather than from complex numbers themselves.
- It also corrects a misunderstanding in earlier work and prevents future confusion: when building real-number versions of quantum theory, you must use the right way to combine systems (the symplectic tensor product), or you will get wrong predictions.
In short: the authors build a carefully designed real-number framework that perfectly mirrors standard quantum mechanics, even in advanced entanglement tests. This shows complex numbers are a language choice, not a physical necessity, because the “imaginary” behavior is already encoded in the geometry of real spaces.
Knowledge Gaps
Knowledge gaps, limitations, and open questions
The paper makes a strong isomorphism claim for finite-dimensional systems and demonstrates a specific Bell-inequality case. The following concrete gaps and open problems remain for a complete, general, and operationally robust theory.
- Infinite-dimensional extension:
- Generalize the γ/γ⁻¹ isomorphism and the symplectic tensor product ⊗K to separable infinite-dimensional Hilbert spaces.
- Treat unbounded operators, domains, closures, and spectral calculus; derive Stone’s theorem and dynamics for time-dependent Hamiltonians in Kähler space.
- Handle continuous-variable systems (Gaussian states/channels, CCRs) rigorously.
- C*- and von Neumann algebra structure:
- Prove that γ is a ∗-isomorphism: γ(A†) = γ(A)† and γ preserves positivity and norms at the C*-algebra level.
- Establish a one-to-one correspondence between CPTP maps on H and “Kähler-CP” maps on K; give Kraus forms in K and verify complete positivity.
- Partial trace and marginals:
- Fully formalize the “tensor contraction” T as the counterpart of partial trace in K, including basis independence, positivity/traceness preservation, and complete positivity for reduced states.
- Prove functorial compatibility T ∘ (* ⊗K ·) = (* ⊗K T(·)) wherever well-defined.
- Monoidal coherence and multipartite structure:
- Prove associativity, symmetry (swap), and coherence of ⊗K for arbitrary multipartite systems (explicit associators/braidings and pentagon/hexagon identities).
- Demonstrate that local tomography holds in K (state determination from local measurement statistics) and that ⊗K satisfies the expected operational properties.
- Dependence on the choice of complex structure J:
- Clarify uniqueness up to O(2N) ∩ Sp(2N,ℝ) transformations and how different local choices of J_A, J_B impact ⊗K on AB.
- Provide operational criteria/gauge conditions to align subsystem complex structures and show misalignment is physically unobservable (gauge) rather than predictive.
- Antiunitary symmetries and Wigner’s theorem:
- Construct explicit K-space representations of antiunitaries (time reversal, charge conjugation) and prove Wigner’s theorem within the K framework.
- Analyze locality/globality of the J ↦ −J involution in composites and its interplay with ⊗K.
- Fermions, grading, and superselection:
- Extend the formalism to CAR algebras, parity superselection, and graded tensor products; reconcile ⊗K with the fermionic composition law and Majorana representations.
- Quantum field theory:
- Develop the Kähler-space description for Fock spaces (bosonic and fermionic), creation/annihilation operators, and QFT symmetries; ensure the isomorphism covers QFT dynamics and observables.
- Measurements and state update:
- Specify how general POVMs and instruments (including post-measurement states) are represented in K using g and ω; prove equivalence to the Born rule and Lüders update for all instruments.
- Mixed states and entanglement structure:
- Give necessary and sufficient conditions for K-positivity of density operators and characterize separable vs entangled states directly in K.
- Reconstruct standard entanglement measures (e.g., negativity, concurrence) and show they compute identically in K.
- Network nonlocality beyond CHSH₃:
- Provide a general theorem (not just the b=00 case) showing that all quantum-optimal violations in network Bell scenarios are attainable with the Kähler real formulation, including all Bob outcomes and other network topologies.
- Robustness to noise and imperfections:
- Analyze noisy states/measurements in K and reproduce quantum/classical bounds as functions of visibility; confirm that experimental imperfections preserve indistinguishability between H and K.
- Category-theoretic formulation:
- Connect the construction to dagger-compact (or related) categorical frameworks; give a basis-independent characterization of γ and ⊗K and relate ⊗K to balanced tensor products rigorously.
- Basis and implementation issues:
- Address basis dependence of γ (choice of the 2×2 block {1₂, τ}) and show operational invariance under basis changes; specify how to select/track J in actual implementations without introducing hidden nonlocal resources.
- Computational aspects:
- Develop efficient algorithms/libraries for ⊗K and γ/γ⁻¹ for large-scale systems; quantify computational overhead vs standard complex arithmetic and numerical stability properties.
- Symmetries and conservation laws:
- Map continuous symmetries and Noether charges into K; verify that generators (self-adjoint in H) have correct K analogues and conserve the metric/symplectic structures under dynamics.
- Clarify the status of “tensor contraction” on rectangular objects:
- Prove linearity, invariance, and consistency under regrouping of subsystems; show that using T does not create spurious degrees of freedom or violate normalization in composite reductions.
- Completeness and reproducibility of CHSH₃ analysis:
- Supply full derivations for all Bob outcomes and settings, and provide complete, accessible references for all concurrent works cited (some listed with future years), ensuring independent reproducibility.
- Scope of empirical indistinguishability:
- Articulate whether any physically motivated constraints (e.g., locality of representations, resource limitations, or restrictions on J) could in practice differentiate real-Kähler and complex formulations, and propose concrete tests if such constraints exist.
Practical Applications
Immediate Applications
The results enable concrete changes to software, experiments, and pedagogy without altering any physical predictions of quantum mechanics. Below are actionable use cases that can be pursued now.
- Real-valued quantum simulation backends using Kähler-space maps
- Sectors: software, HPC, quantum computing, computational chemistry/materials
- Tools/products/workflows:
- Implement the explicit maps
γandγ^{-1}and the symplectic tensor product⊗^Kas a library (e.g., a “Kähler-QM” backend for Qiskit, Cirq, PennyLane). - Add real-valued state/operator representations (block-real matrices) with guaranteed equivalence to complex QM.
- Provide kernels to apply gates, compose subsystems, and compute expectations via the metric g without complex arithmetic.
- Assumptions/dependencies:
- Must implement
⊗^K(not the standard real Kronecker product) to preserve equivalence to complex QM. - Numerical performance depends on trade-offs: dimension doubling vs faster real BLAS/GPU kernels.
- Faster classical simulation using real-only linear algebra
- Sectors: HPC, software, quantum algorithm R&D, computational chemistry
- Tools/products/workflows:
- Exploit optimized real GEMM kernels in cuBLAS/oneMKL for large-scale circuit and Hamiltonian simulations (VQE, time-evolution).
- Integrate “real-only” automatic differentiation for variational algorithms, reducing complex AD overhead.
- Assumptions/dependencies:
- Net speed-up is hardware- and problem-dependent; dimension doubling may offset real-kernel gains.
- Requires maintaining the Kähler complex structure J exactly to avoid drift from the physical subspace.
- Corrected analysis pipelines for multipartite/network Bell experiments
- Sectors: experimental physics, quantum foundations, standards/policy
- Tools/products/workflows:
- Update data-analysis software for network Bell tests to use
⊗^Kwhen real representations are employed, preventing spurious state-space enlargement. - Reanalyze CHSH₃-like network experiments with Kähler-compatible composition to avoid misinterpretation of “real-vs-complex” separations.
- Assumptions/dependencies:
- Community acceptance and reviewer/standards guidance to distinguish
⊗^Kfrom the naive real Kronecker product. - Requires modest refactoring of existing experimental toolchains.
- Formal verification and unit tests for quantum software using Kähler identities
- Sectors: quantum software QA, verification, education
- Tools/products/workflows:
- Add real-matrix identity checks (e.g., {σ_xK, σ_yK} = 0, σ_xK σ_yK = J σ_zK) as invariant tests during simulation or compilation.
- Use Kähler-space anti-commutation and composition rules to detect modeling errors in multipartite code.
- Assumptions/dependencies:
- Requires exposing J explicitly in software and preserving it through transformations and numerical operations.
- Unified real-geometric pedagogy and visualization
- Sectors: education, outreach
- Tools/products/workflows:
- Interactive visualizations of quantum states as points/vectors in a real 2N-dimensional space with metric g, symplectic form ω, and complex structure J.
- Curriculum modules that show exactly how complex phases arise from real geometry.
- Assumptions/dependencies:
- Instructors’ training and incorporation into existing QM and QIS courses.
- Real-space formulations for quantum control and estimation in optics and sensing
- Sectors: quantum sensing, photonics, control engineering
- Tools/products/workflows:
- Implement Kähler-space filters and controllers that leverage the symplectic structure (ω) and real parameterizations (useful for Gaussian/continuous-variable approximations).
- Aligns with existing symplectic methods in control while maintaining exact equivalence to complex QM for finite-dimensional subsystems.
- Assumptions/dependencies:
- Integration with existing control toolchains; extension to infinite-dimensional CV systems may require additional care (see Long-Term).
- Re-express device-independent and self-testing proofs in a Kähler-compatible framework
- Sectors: quantum information/cryptography theory
- Tools/products/workflows:
- Rewrite proof scripts and numerical self-tests to avoid hidden assumptions tied to the naive real tensor product.
- Provide templates for DIQKD and certification protocols that are representation-agnostic.
- Assumptions/dependencies:
- Theoretical community consensus and adoption in standard references.
Long-Term Applications
These opportunities build on the paper’s isomorphism and symplectic composition but require further research, scaling, or ecosystem changes.
- Specialized hardware and accelerators for Kähler-QM simulation
- Sectors: semiconductors, HPC, cloud computing
- Tools/products/workflows:
- FPGA/ASIC IP cores that implement
⊗^K, Kähler-aware tensor contractions, and block-real operator algebra. - On-die enforcement of J structure to keep computations within the physical subspace.
- Assumptions/dependencies:
- Demonstrated demand and measurable speed/energy advantages over complex kernels after considering dimension doubling.
- Kähler-native intermediate representations (IR) and compilers
- Sectors: quantum software, compilers
- Tools/products/workflows:
- A compiler IR that tracks (g, ω, J) and composes subsystems with
⊗^K, enabling new optimization passes that exploit O(2N) ∩ Sp(2N,ℝ) symmetries. - Automated detection and removal of unphysical degrees of freedom introduced by naive real tensoring.
- Assumptions/dependencies:
- Coordination with major frameworks (QIR, MLIR dialects) and community buy-in.
- Error correction, decoding, and noise modeling via real symplectic geometry
- Sectors: quantum hardware, error correction R&D
- Tools/products/workflows:
- Explore decoders and noise models that leverage the intersection group U(N) = O(2N) ∩ Sp(2N,ℝ) for Clifford and beyond.
- Real-geometry constraints as priors for noise learning and model reduction.
- Assumptions/dependencies:
- Further theory to demonstrate advantages in decoding performance or robustness.
- Revised standards for multipartite experiment design and certification
- Sectors: standards bodies, national labs, funding agencies
- Tools/products/workflows:
- Update experiment and analysis guidelines to specify representation assumptions (complex vs real with
⊗^K), especially in network nonlocality tests. - Certification protocols that are invariant under the H ↔ K isomorphism.
- Assumptions/dependencies:
- Broad community consensus and inter-lab validation.
- Reassessment of “complex-number advantage” tasks and protocols
- Sectors: quantum communication, complexity theory, resource theories
- Tools/products/workflows:
- Systematic review of tasks asserting separations between real and complex QM; re-derive bounds and optimal strategies in Kähler space.
- Possible new algorithms or tighter bounds when phrased in real symplectic terms.
- Assumptions/dependencies:
- Extensive theoretical work to translate and compare results across frameworks.
- Hybrid classical–quantum modeling and control using a unified real symplectic toolkit
- Sectors: robotics, autonomous systems with quantum sensors, advanced manufacturing
- Tools/products/workflows:
- Co-design of controllers where classical components (already symplectic/Hamiltonian in form) interface with quantum subsystems modeled in Kähler space.
- Assumptions/dependencies:
- Development of robust interfaces between quantum Kähler models and classical control stacks.
- Infinite-dimensional and field-theoretic extensions
- Sectors: quantum optics, high-energy theory, condensed matter
- Tools/products/workflows:
- Extend the isomorphism and
⊗^Kto continuous-variable systems and QFT contexts with rigorous treatment of domains, measures, and topology. - Real-space formulations for Gaussian channels, bosonic codes, and metrology.
- Assumptions/dependencies:
- Mathematical development to ensure completeness and numerical stable implementations.
- Differentiable programming and AI for quantum with real-only variables
- Sectors: AI/ML for quantum, optimization
- Tools/products/workflows:
- Kähler-native training loops (PyTorch/JAX) for VQA/QML that exploit real parameterizations for stability and performance.
- Real-valued gradient flows constrained by (g, ω, J) to preserve physicality during training.
- Assumptions/dependencies:
- Empirical evidence of convergence or stability benefits over complex-parameter approaches.
Cross-cutting assumptions and dependencies
- The equivalence relies on correct implementation of the maps
γ,γ^{-1}, the metric g, symplectic form ω, complex structure J, and especially the symplectic tensor product⊗^K. Using the standard real Kronecker product breaks the equivalence. - Numerical stability must preserve the Kähler structure; mixed-precision or approximate kernels may require structure-preserving projections.
- Performance benefits of real arithmetic are hardware- and problem-specific; dimension doubling can offset raw real-kernel advantages.
- Some claims and pipelines in device-independent testing and self-testing rely on assumptions (e.g., “source independence”). Adoption of Kähler-consistent methods may require revisiting these assumptions and updating community standards.
- Extensions to infinite-dimensional/CV systems and open-system dynamics require additional theoretical development, though the paper’s framework suggests a consistent path forward.
Glossary
- anti-commutation relations: Algebraic relations where two operators anticommute, i.e., their anticommutator equals zero or a scalar multiple of the identity. Example: "These satisfy the a-space anti-commutation relations "
- antiunitary time-reversal operator: A conjugate-linear (antiunitary) symmetry operation implementing time reversal in quantum mechanics. Example: "Wigner's antiunitary time-reversal operator"
- balanced tensor product: A construction in module theory (and categorical tensoring) that quotients by a balancing relation to enforce bilinearity over a base ring/field. Example: "For the connection with the balanced tensor product (see section S8)"
- Bell state: A maximally entangled two-qubit state (e.g., singlet or triplet). Example: "A Bell state maps to"
- CHSH inequality: A bipartite Bell inequality whose quantum maximum is the Tsirelson bound. Example: "can maximally violate the standard CHSH inequality"
- CHSH inequality: A network Bell inequality tailored for a tripartite entanglement-swapping scenario; quantum maximum is . Example: "the inequality - tailored for a tripartite entanglement-swapping network - admits a maximal quantum violation of "
- complex conjugation: The involutive map on complex numbers; in QM it underlies antiunitary symmetries. Example: "carries a fundamental involutive symmetry (complex conjugation)"
- complex Hilbert space: A Hilbert space over the field of complex numbers, the standard setting of quantum mechanics. Example: "Standard quantum mechanics employs complex Hilbert spaces"
- complex structure: A linear map with that equips a real vector space with a compatible complex structure. Example: "they constitute a complex structure on the real doubled space"
- complexification map: A map that reconstructs complex operators/vectors from their real (Kähler) representatives. Example: "Complexification map "
- CPT theorem: The fundamental result that any Lorentz-invariant local quantum field theory with a Hermitian Hamiltonian is invariant under the combined operations of charge conjugation (C), parity (P), and time reversal (T). Example: "the CPT theorem"
- CPTP map: A completely positive trace-preserving linear map, the most general quantum channel on density matrices. Example: "CPTP maps "
- entanglement-swapping network: A networked quantum setup where entanglement is created between parties that never directly interacted via a mediating measurement. Example: "tripartite entanglement-swapping network"
- Kähler space: A real vector space equipped with a compatible triple of metric, symplectic form, and complex structure; equivalent to a complex Hilbert space for QM here. Example: "Definition of a Kähler space"
- Kronecker product: The standard matrix-level tensor product producing block matrices; serves as the tensor product over a given field. Example: "familiar in its matrix guise as the Kronecker product"
- monoidal isomorphism: An isomorphism between two tensor (monoidal) structures that preserves the tensor product and unit up to coherence. Example: "are isomorphic as (monoidal) quantum theories"
- realification map: A map that encodes complex vectors/operators into real block form while preserving algebraic structure. Example: "Realification and complexification maps"
- representation locality: The operational property that local processes admit local representations without hidden nonlocal resources. Example: "a real quantum theory with representation locality is possible"
- ring homomorphism: A structure-preserving map between rings that respects addition and multiplication. Example: "they are ring homomorphisms (appendix)."
- Riemannian metric: A positive-definite inner product on a real vector space, used here as the physical inner product in Kähler space. Example: "the physical inner product is the Riemannian metric "
- self-adjoint/skew-adjoint decomposition: The canonical splitting of a linear operator into its Hermitian and anti-Hermitian parts. Example: "the canonical decomposition of any operator into its self-adjoint and skew-adjoint parts"
- symplectic composition rule: A composition operation on subsystems compatible with the underlying symplectic and complex structures, denoted . Example: "The symplectic composition rule is exactly equivalent to complexifying via , taking the standard complex tensor product $\otimes_{\mathbb{C}$, and realifying via ."
- symplectic form: A non-degenerate, skew-symmetric bilinear form that encodes canonical structure (e.g., phase-space areas). Example: " is a non-degenerate skew-symmetric bilinear form (symplectic form)"
- symplectic geometry: The geometry of spaces endowed with a symplectic form, central to Hamiltonian and Kähler structures. Example: "the symplectic geometry of "
- symplectic tensor product: A tensor product on real (Kähler) spaces that reproduces complex tensoring by respecting the encoded complex structure. Example: "the symplectic tensor product - is distinct from $\otimes_{\mathbb{R}$"
- symmetry group : The unitary group identified as the intersection of orthogonal and symplectic groups in the real (Kähler) picture. Example: "The symmetry group is ."
- tensor contraction: An operation generalizing partial trace that contracts one tensor factor by tracing over it. Example: "where $T[A\otimes C]=(\Tr A)C$ is the tensor contraction (see Appendix)"
- Tsirelson bound: The maximum quantum value for CHSH-type inequalities, equal to for CHSH. Example: "the standard CHSH combination yields $g(\ket{\psi^-}_{K},\hat{C}_{\mathrm{CHSH}^{K}\ket{\psi^-}_{K})=2\sqrt{2}$, the Tsirelson bound."
- -linearity: Linearity over the complex field; crucial for correctly composing subsystems in complex QM. Example: "enforcing -linearity across subsystems."
Collections
Sign up for free to add this paper to one or more collections.