Quantum Mechanics as
Classical Physics
Charles Sebens
University of Michigan
July 31, 2013
Dirk-André Deckert
Michael Hall
Howard Wiseman
UC Davis
Griffith University
Griffith University
2
A Strange Interpretation of QM
 Deterministic, no collapse
 No wave function, no Schrödinger equation
 Many worlds
 Worlds do not branch
 The number of worlds is finite
 Particles follow Bohmian trajectories
 Equation of motion of the form 𝐹 = 𝑚𝑎
This is a development of the hydrodynamic interpretation, originally proposed by
Madelung (1927) and developed by Takabayasi (e.g., 1952) among others.
3
Outline
I.
A Series of Solutions to the Measurement Problem
I.
The Many-worlds Interpretation
II.
Bohmian Mechanics
III.
Prodigal QM
II. Newtonian Quantum Mechanics
III. A Strength: Probability
IV. A Weakness: Non-quantum States
4
Refresher: The Double-slit Experiment
5
Solution 1: Everettian QM
What there is (Ontology):
The Wave Function
What it does (Laws):
Schrödinger Equation
6
Solution 2: Bohmian QM
Ontology:
The Wave Function
Particles
Laws:
Schrödinger Equation
Guidance Equation
7
Solution 3: Prodigal QM
Ontology:
The Wave Function
Particles (in many worlds)
Laws:
Schrödinger Equation
Guidance Equation
This is a significantly altered variant of the proposal in Dorr (2009).
8
Two Important Characters: 𝜌 & 𝑣𝑘

is the density of worlds in configuration
space, scaled so that
.
 By hypothesis,

.
gives the velocity of the 𝑘-th particle in the
world where particles are arranged
.
 The evolution of 𝜌 is determined by the velocities of the
particles in the various worlds by a continuity equation:
9
The Evolution of 𝑣𝑘
From the following facts one can derive the double boxed equation below.
 Worlds are distributed in accordance with psi-squared:
 The Schrödinger equation:
 The guidance equation:
The quantum
potential Q.
This equation gives a way of calculating the evolution of 𝑣𝑘 , and
hence 𝜌, which never references the wave function Ψ.
10
Solution 4: Newtonian QM
Ontology:
Particles (in many worlds)
Law:
Newtonian Force Law
11
Review of the Alternatives
12
The Wave Function in Newtonian QM
The wave function is not part of the fundamental ontology in
Newtonian QM. Still, one can introduce an object Ψ which
conforms to the following conditions:
 Worlds are distributed in accordance with psi-squared
 The guidance equation is obeyed
13
The Quantitative Probability
Problem for Everettian QM
Alice makes 10 measurements of the z-spin of the following state:
In a highly idealized scenario, 210 = 1,024 branches…
 In 252 branches (25%) she will see half up, half down.
 In 10 branches (1%) she will see nine up, one down (which should
be most likely, happening about 40% of the time).
In general, most branches don’t exhibit Born Rule statistics.
14
No Similar Problem for Newtonian QM
 Since the wave function is not fundamental but introduced as
a means of summarizing information about 𝜌 & 𝑣𝑘 , it is
impossible for the density of worlds in a region of
configuration space to deviate from psi-squared. Most worlds
are in high-amplitude regions.
 In Newtonian QM the uncertainty present is self-locating
uncertainty. One needs to determine which world they are in
since many are consistent with any particular set of
experiences.
 Since, in general, many worlds will be consistent with one’s
experiences, each should be assigned equal credence (given
the state of the universe; see Elga 2004).
15
Versus Bohmian Mechanics
 In Newtonian QM, each world follows a Bohmian trajectory
through configuration space. So, if Bohmian mechanics can
reproduce the predictions of textbook QM, Newtonian QM
should be able to also.
 Some Bohmian trajectories don’t predict quantum statistics.
 The correct long-run quantum statistics will be observed if the
universe satisfies the quantum equilibrium hypothesis.
Quantum Equilibrium Hypothesis (Teufel 2013, rough version):
“The initial wave function Ψ(0) and configuration 𝑄(0) of the "universe" are
such that the empirical distributions of subsystem configurations 𝑋𝑖 (𝑡𝑖 ) of
(different) subsystems at (different) times 𝑡𝑖 with the same conditional wave
function 𝜑𝑐𝑜𝑛𝑑 are close to the |𝜑𝑐𝑜𝑛𝑑 |2 -distribution.”
16
Avoiding Anomalous Statistics
 In Newtonian QM, it is also possible that one’s own world
does not exhibit quantum statistics.
 However, it is necessarily true that the majority of worlds in
any universe satisfy the quantum equilibrium hypothesis
since
(by definition of Ψ).
 Thus, one should always expect to be in a world that is in
quantum equilibrium. So, one should expect to see Born Rule
statistics in long-run frequencies of measurements
(see Durr et al. 1992).
17
How Many Worlds?
 We’ve been treating the number of worlds as continuously
infinite, described by a density, 𝜌, and a collection of velocity
fields,
.
 Meanwhile, I’ve claimed that the number of worlds is actually
finite.
 Why finitely many worlds? 𝜌 is easier to interpret.
 How many? Enough that they can be well-modeled by a
continuum. This is similar to the hydrodynamic limit in fluid
mechanics.
 If there isn’t really a continuum, is the force law given earlier
really a fundamental law? No.
18
The Quantization Condition
 Not all collections of 𝜌 & 𝑣𝑘 can be described by a wave function.
 The velocity of a particle is proportional to the gradient of the wave
function’s complex phase.
 For there to exist a wave function which describes* 𝜌 & 𝑣𝑘 , the
velocity field must satisfy the following requirement (see Wallstrom
1994; Takabayasi 1952).
Quantization Condition: Integrating particle momenta along any
closed loop in configuration space gives a multiple of Planck’s constant.
*That is, a wave function for which
by the particle velocities.
and the guidance equation is obeyed
19
Limitations of Newtonian QM
 Yet to be extended to relativistic quantum physics
 Yet to be extended to multiple particles with spin
 We don’t yet have the fundamental law(s)
 The state space is too large in two ways:
 States that violate the Quantization Condition
 States with too few worlds to use the hydrodynamic limit
20
Neat Features of Newtonian QM
 Wave function is a mere summary of the properties of particles
 No superpositions
 No entanglement
 No collapse
 No mention of “measurement” in the laws
 All dynamics arise from Newtonian forces
 The theory is deterministic
 Worlds are fundamental, not emergent (so avoids the need to
explain how people and planets arise as structures in the WF)
 Worlds do not branch (so avoids concerns about personal
identity)
 No qualitative probability problem
 No quantitative probability problem
 Immune to Everett-in-denial objection,* not in denial
21
* See Deutsch (1996), Brown & Wallace (2005).
Neat Features of Newtonian QM
 Wave function is a mere summary of the properties of particles
 No superpositions
 No entanglement
 No collapse
 No mention of “measurement” in the laws
 All dynamics arise from Newtonian forces
 The theory is deterministic
 Worlds are fundamental, not emergent (so avoids the need to
explain how people and planets arise as structures in the WF)
 Worlds do not branch (so avoids concerns about personal
identity)
 No qualitative probability problem
 No quantitative probability problem
 Immune to Everett-in-denial objection, not in denial
22
Neat Features of Newtonian QM
 Wave function is a mere summary of the properties of particles
 No superpositions
 No entanglement
 No collapse
 No mention of “measurement” in the laws
 All dynamics arise from Newtonian forces
 The theory is deterministic
 Worlds are fundamental, not emergent (so avoids the need to
explain how people and planets arise as structures in the WF)
 Worlds do not branch (so avoids concerns about personal
identity)
 No qualitative probability problem
 No quantitative probability problem
 Immune to Everett-in-denial objection, not in denial
23
References
24
End
25
Ontological Options
Option 1: World-particles in Configuration Space
Option 2: World-particles in Configuration Space and 3D Worlds
Option 3: Distinct 3D Worlds
Option 4: Overlapping 3D Worlds
Shown below for two particles in one dimensional space…
26
An Unnatural Constraint
To get a feel for the Quantization Condition, let’s look at the
following case:
Consider the 𝑛 = 2, 𝑙 = 1, 𝑚 = 1 energy eigenstate of an
electron in the Hydrogen atom (treating the force from the
nucleus as an external potential).
If the electrons were circling the 𝑧-axis a little faster or a little
slower, they would not be representable by a wave function.
27
The Orbital
28