---
source_pdf: Density_Field_Dynamics_Resolves_the_Penrose_Superposition_Paradox.pdf
title: "Density Field Dynamics Resolves the Penrose Superposition Paradox"
site: https://densityfielddynamics.com/
author: Gary Alcock
framework: Density Field Dynamics (DFD)
format_note: "Markdown extracted from the source PDF for clean AI ingestion. No editorial changes; mathematical typography in the PDF is authoritative."
---

Density Field Dynamics Resolves the Penrose Superposition Paradox
Gary Alcock
Independent Researcher, Los Angeles, CA, USA
(Dated: September 19, 2025)
Penrose has argued that a quantum superposition of mass distributions leads to a structural inconsistency: in general relativity, each branch would source a distinct spacetime geometry, whereas
quantum mechanics allows only a single state until collapse. We show that Density Field Dynamics (DFD), a scalar-field completion of Einstein’s 1911–12 variable-c program, avoids this paradox
entirely. In DFD there is no manifold branching: superposed mass distributions source a single
2
classical (c-number) refractive field ψ, which governs both light (n = eψ ) and matter (a = c2 ∇ψ).
In the weak-field linear regime (µ → 1), ψ is the convex sum of the branch fields; in the full quasilinear regime, monotonicity of the crossover function µ ensures existence and uniqueness of a single
solution. Thus DFD is structurally compatible with quantum superposition, unlike GR, and the
decisive discriminator remains laboratory testability: the co-located cavity–atom redshift comparison at two altitudes, where GR predicts zero slope and DFD predicts a geometry-locked slope of
O(∆Φ/c2 ) ∼ 10−14 per 100 m.

I.

INTRODUCTION

Penrose has long emphasized a tension between general relativity (GR) and quantum mechanics (QM) [1, 2]. If
a macroscopic object is placed in spatial superposition, GR demands that each branch source its own spacetime
curvature, while QM maintains only a single quantum state until measurement. This “two spacetimes vs. one Hilbert
space” contradiction underpins Penrose’s proposal that gravity induces wavefunction collapse.
Density Field Dynamics (DFD) [13–15] replaces curved spacetime with a single classical (c-number) scalar refractive
2
field ψ(x). Photons propagate with index n = eψ , matter accelerates as a = c2 ∇ψ, and ψ obeys the quasilinear elliptic
equation
" 
#



|∇ψ|
8πG
∇· µ
(1)
∇ψ = − 2 ρ − ρ̄ ,
a⋆
c
with µ → 1 in the weak-field regime (and a⋆ the characteristic deep-field acceleration scale). Normalization reproduces
GR’s weak-field optical tests [3], while the µ-family enforces scale symmetry, ellipticity, and convex energy density [9].
II.

SUPERPOSITION SOURCES IN DFD

Let the quantum state of a mass distribution be |Ψ⟩, with density operator ρ̂(x). The effective source entering (1)
is the expectation value
ρeff (x) = ⟨Ψ|ρ̂(x)|Ψ⟩.

(2)

For a superposition of two localized packets |L⟩, |R⟩ with |Ψ⟩ = a|L⟩ + b|R⟩,
ρeff ≃ |a|2 ρL + |b|2 ρR + 2 Re(a∗ b ρLR ) ,

(3)

where the interference term ρLR is exponentially suppressed for well-separated packets. In the linear (Poisson) regime
(µ → 1), the field solution is
ψ ≃ |a|2 ψL + |b|2 ψR ,

(4)

a convex sum of branch fields. In the full nonlinear regime, monotone µ ensures uniform ellipticity; existence and
uniqueness follow by variational methods (Sec. VII). Thus there is always a single ψ field—a weighted combination
of branch contributions—ensuring no manifold branching.
A.

Semiclassical sourcing (but not semiclassical GR)

DFD sources a classical scalar field ψ by ρeff = ⟨ρ̂⟩, yet the geometry is never promoted to an operator; there is
no ĝ. This is not the semiclassical Einstein equation Gµν = 8πG⟨T̂µν ⟩. Instead, the optical metric for light is n = eψ

2

µ(x)

1

x
µ(x) = 1+x
µ(x) = tanh(x)

0.5

0
10−3

10−2

10−1

100

101

102

x = |∇ψ|/a⋆
FIG. 1. Representative crossovers: linear for x ≫ 1, scaling µ ∼ x for x ≪ 1. Both preserve a unique classical ψ.
1
ψ(x) (arb.)

0.8

ψL
ψR
Convex sum ψ

0.6
0.4
0.2
0
−2

−1.5

−1

−0.5

0
Position x

0.5

1

1.5

2

FIG. 2. Illustrative ψ-profiles for two separated packets in 1D. Only the convex-sum field exists in DFD.
2

(Euclidean background with refractive structure), and matter follows a = c2 ∇ψ. Hence no operator-valued geometry
arises, and Penrose’s paradox does not materialize.

B.

Linear vs. Nonlinear Regimes

DFD’s µ(x) crossover unifies two limits with a single PDE and a single ψ: (i) the high-gradient (solar-system) regime
µ → 1, where the equation reduces to a linear Poisson problem; and (ii) the deep-field (galactic) regime µ(x) ∼ x,
which yields scale-free behavior |∇ψ| ∝ 1/r and flat rotation curves. In both regimes ψ is a classical field determined
by ρeff ; no operator-valued geometry arises.
C.

Worked example: superposed grain of sand

Consider m ∼ 10−7 kg in a spatial superposition with branch centers separated by d ∼ 1 µm. In GR, two geometries
are implicated. In DFD, ρeff ≈ 12 (ρL + ρR ) and the weak-field solution is ψ = 12 (ψL + ψR ) by (4). The acceleration
field a = (c2 /2)∇ψ and optical index n = eψ are single-valued; no paradox arises.

III.

QUANTUM EVOLUTION AND CONTINUITY

Matter wavefunctions evolve with
iℏ∂t Ψ = −



ℏ2
∇· e−ψ ∇Ψ + mΦ Ψ,
2m

2

Φ = − c2 ψ.

(5)

Define the current
J=



ℏ −ψ ∗
e
Ψ ∇Ψ − Ψ∇Ψ∗ .
2mi

(6)

3
TABLE I. Illustrative error budget for ∆R/R at the 10−14 per 100 m level.
Systematic

Target (frac.)

Cavity dispersion (dual-λ)
Cavity elastic sag / flips
Atom transition sensitivity
Comb transfer noise
Thermal gradients / birefringence

Control handle

−15

≲ 3 × 10
Dual-wavelength bound
≲ 3 × 10−15
180◦ orientation flips + model
≲ 3 × 10−15
Co-trapped species calibration
≲ 1 × 10−16
Stabilized links + counters
≲ 3 × 10−15 Active stabilization; polarization checks

Multiplying (5) by Ψ∗ and subtracting the conjugate equation yields
∂t |Ψ|2 + ∇· J = 0,

(7)

so probability is conserved. Equivalently, the kinetic operator can be written
−

ℏ2
1 2
∇·(e−ψ ∇) =
p̂ ,
2m
2m ψ

p̂ψ ≡ −iℏ e−ψ/2 ∇ e−ψ/2 ,

(8)

which is self-adjoint on the natural domain (e.g. square-integrable functions with appropriate boundary conditions)
under the flat measure. For bounded domains, impose Dirichlet or Neumann conditions on Ψ; for R3 require Ψ, ∇Ψ ∈
L2 with e−ψ bounded and positive.
IV.

LABORATORY DISCRIMINATOR

While the theoretical resolution is complete, experimental verification remains the decisive test of which theory
nature follows. In a nondispersive band, an evacuated cavity with frequency fcav ∝ c1 /L and a co-located atomic
clock fat respond differently to ψ. Across an altitude change ∆h,
∆Φ
∆R
=ξ 2 ,
R
c

fcav
.
fat

(9)

∆R
≈ 1.1 × 10−14 per 100 m.
R

(10)

R≡

In GR, ξ = 0; in DFD, ξ ≃ 1 [18]. For Earth’s surface,

This level is achievable with state-of-the-art optical metrology [7, 8]. Matter-wave interferometry provides a second
discriminator: DFD predicts a T 3 phase scaling in long-baseline atom interferometers, yielding a ∼ 2 × 10−11 rad
signal at T = 1 s, within reach of current facilities [17].
A.

GLS 4 → 3 slope extraction (sector-resolved)
(M )

(S)

Using two cavity materials (e.g. ULE, Si) and two atomic species (e.g. Sr, Yb), form four ratios R(M,S) = fcav /fat
(M )
(S)
at two altitudes. The observable slopes are ∆R/R = ξ (M,S) ∆Φ/c2 with ξ (M,S) = αw − αL − αat . A generalized
least squares (GLS) fit over the four slopes identifies the three combinations (δtot , δL , δat ) with internal consistency
and covariance control:
ULE
Sr
δtot ≡ αw − αL
− αat
,

Si
ULE
δL ≡ αL
− αL
,

Yb
Sr
δat ≡ αat
− αat
.

(11)

GR predicts all three δ’s vanish; DFD predicts a nonzero δtot in a nondispersive band.
V.
A.

DISCUSSION

Collapse models (GRW/CSL) vs. DFD

GRW and CSL add stochastic, non-unitary collapse terms to resolve the GR/QM tension [6]. DFD requires no
such postulates: the background is a classical c-number field ψ, so there is never more than one geometry to begin
with. Penrose’s structural paradox is absent without modifying the Schrödinger equation stochastically.

4

∆h ≈ 100 m

Cavity

Atom

Frequency comb

Cavity

Atom

Ground
FIG. 3. Sector-resolved cavity–atom LPI test: GR predicts zero slope, DFD predicts ∆R/R ∼ 10−14 per 100 m.

TABLE II. How different approaches treat superposed mass distributions.
Approach

Geometry in superposition

Resolution mechanism

GR + QM
GRW/CSL
Decoherence
DFD

Two spacetimes
One spacetime
Two spacetimes
One ψ field

Structural paradox (Penrose)
Stochastic collapse postulate
Environment hides interference
Convex-sum sourcing; unique PDE

B.

Decoherence vs. DFD

Environmental decoherence suppresses interference but does not remove the two-geometry issue in GR. DFD never
produces branch geometries: superposed sources create one ψ fixed by ρeff . Thus decoherence is relevant to experimental visibility, not to resolving a structural inconsistency.

C.

Cosmological implications

The same optical-metric mechanism impacts cosmography: line-of-sight inhomogeneities bias optical distances,
inducing a directional H0 anisotropy tied to ψ-weighted density gradients [13]. This links laboratory falsification to
large-scale observables.

D.
VI.

Comparison with other approaches

PENROSE PARADOX VS. DFD (SCHEMATIC)

GR
L

DFD
R

“two spacetimes”

Inconsistent superposition

L

R
convex sum

Single ψ: |a|2 ψL + |b|2 ψR

FIG. 4. Schematic contrast. In GR, superposed matter implies two geometries (paradox). In DFD, the scalar ψ is unique,
formed from the weighted density distribution.

5
VII.

WELL-POSEDNESS (EXISTENCE & UNIQUENESS)

Theorem (well-posedness). Let µ : R+ → R+ be continuous, monotone increasing, and satisfy 0 < µmin ≤ µ(·) ≤
µmax < ∞ on compact subdomains of interest. Given ρ ∈ L2loc and suitable boundary conditions, Eq. (1) admits a
1,2
unique weak solution ψ ∈ Wloc
.
R
′
Sketch. Define the energy functional E[ψ] = d3 x F(|∇ψ|) − 8πG
c2 ψ(ρ − ρ̄) with F (y) = µ(y/a⋆ )y. Monotonicity of
µ implies convexity of F and coercivity on appropriate Sobolev spaces. The direct method of the calculus of variations
yields a minimizer; uniqueness follows from strict convexity [9]. For µ → 1 one recovers the linear Poisson theory; for
µ ∼ x deep-field scaling holds.

VIII.

CONCLUSION

Penrose’s paradox arises only if mass superpositions imply multiple geometries. In DFD, superpositions source one
classical ψ field, ensuring consistency with quantum mechanics. The debate moves from philosophy to experiment:
a co-located cavity–atom comparison at two altitudes, together with matter-wave interferometry, can decide between
GR and DFD with current precision.

Appendix A: Continuity and Hermiticity (derivation details)

Starting from (5), multiply by Ψ∗ and subtract the conjugate equation:
Ψ∗ iℏ∂t Ψ − Ψ(−iℏ∂t Ψ∗ ) = −

i
ℏ2 h ∗
Ψ ∇·(e−ψ ∇Ψ) − Ψ∇·(e−ψ ∇Ψ∗ ) .
2m

(A1)

Using ∇ · (f A) = f ∇ · A + ∇f · A and rearranging gives ∂t |Ψ|2 + ∇ · J = 0 with J as in (6). Writing the kinetic
1 2
operator as 2m
p̂ψ with p̂ψ = −iℏe−ψ/2 ∇e−ψ/2 shows self-adjointness on the natural domain (Dirichlet/Neumann for
bounded regions; L2 decay at infinity).

Appendix B: Existence/Uniqueness (variational details)

Let A(ψ) = ∇· (µ(|∇ψ|/a⋆ )∇ψ). Assuming µ monotone and bounded away from zero, A is uniformly elliptic. The
functional E[ψ] is convex and coercive, admitting a minimizer; Gateaux differentiability yields (1) in weak form; strict
convexity implies uniqueness [9].

[1] R. Penrose, “On gravity’s role in quantum state reduction,” Gen. Rel. Grav. 28, 581 (1996).
[2] R. Penrose, Fashion, Faith, and Fantasy in the New Physics of the Universe (Princeton University Press, 2014).
[3] C. M. Will, Theory and Experiment in Gravitational Physics, 2nd ed. (Cambridge University Press, 2018).
[4] R. M. Wald, General Relativity (University of Chicago Press, 1984).
[5] C. Kiefer, Quantum Gravity (Oxford University Press, 2007).
[6] A. Bassi, K. Lochan, S. Satin, T. P. Singh, and H. Ulbricht, “Models of wave-function collapse, underlying theories, and
experimental tests,” Rev. Mod. Phys. 85, 471 (2013).
[7] A. D. Ludlow, M. M. Boyd, J. Ye, E. Peik, and P. O. Schmidt, “Optical atomic clocks,” Rev. Mod. Phys. 87, 637 (2015).
[8] C. W. Chou, D. B. Hume, T. Rosenband, and D. J. Wineland, “Optical clocks and relativity,” Science 329, 1630 (2010).
[9] L. C. Evans, Partial Differential Equations, 2nd ed. (AMS, 2010).
[10] D. Giulini and A. Großardt, “Gravitationally induced inhibitions of dispersion of wave packets,” Class. Quantum Grav.
28, 195026 (2011).
[11] S. Bose, A. Mazumdar, G. W. Morley, et al., “Spin entanglement witness for quantum gravity,” Phys. Rev. Lett. 119,
240401 (2017).
[12] C. Marletto and V. Vedral, “Gravitationally-induced entanglement between two massive particles is sufficient evidence of
quantum effects in gravity,” Phys. Rev. Lett. 119, 240402 (2017).
[13] G. Alcock, “Density Field Dynamics: Completing Einstein’s 1911–12 Variable-c Program with Energy-Density Sourcing
and Laboratory Falsifiability,” Zenodo: 17118387 (2025).

6
[14] G. Alcock, “Sector-Resolved Test of Local Position Invariance with Co-Located Cavity–Atom Frequency Ratios,” preprint
(under review at Metrologia), Zenodo record forthcoming (2025).
[15] G. Alcock, “Strong Fields and Gravitational Waves in Density Field Dynamics: From Optical First Principles to Quantitative Tests,” Zenodo: 17115941 (2025).
[16] G. Alcock, “Density Field Dynamics and the c-Field: A Three-Dimensional, Time-Emergent Dynamics for Gravity and
Cosmology,” Zenodo: 16900767 (2025).
[17] G. Alcock, “Matter-Wave Interferometry Tests of Density Field Dynamics,” Zenodo: 17150358 (2025).
[18] G. Alcock, “A Sharp, Testable Slope Prediction for a Sector-Resolved Cavity–Atom LPI Test,” preprint (2025).