Article

The Two-Spin Enigma: From the Helium Atom to

Quantum Ontology
Philippe Grangier 1, * , Alexia Auffèves 2 , Nayla Farouki 3 , Mathias Van Den Bossche 4 and Olivier Ezratty 5

1 Laboratoire Charles Fabry, Institut d’Optique Graduate School, Centre National de la Recherche Scientifique,
Université Paris Saclay, 91127 Palaiseau, France
2 MajuLab International Joint Research Laboratory, Centre for Quantum Technologies, National University of
Singapore, Singapore 117543, Singapore; alexia.auffeves@cnrs.fr
3 Independent Researcher, 38054 Grenoble Cedex, France; naylafarouki@yahoo.fr
4 Thales Alenia Space, 26, Avenue J.-F. Champollion, 31037 Toulouse, France;
mathias.van-den-bossche@thalesaleniaspace.com
5 EPITA Research Laboratory, 14-16 Rue Voltaire, 94270 Le Kremlin-Bicêtre, France; olivier@oezratty.net
* Correspondence: philippe.grangier@cnrs.fr

Abstract: The purpose of this article is to provide a novel approach and justification of the idea
that classical physics and quantum physics can neither function nor even be conceived without the
other—in line with ideas attributed to, e.g., Niels Bohr or Lev Landau. Though this point of view
may contradict current common wisdom, we will show that it perfectly fits with empirical evidence,
and can be maintained without giving up physical realism. In order to place our arguments in a
convenient historical perspective, we will proceed as if we were following the path of a scientific
investigation about the demise, or vanishing, of some valuable properties of the two electrons in the
helium atom. We will start from experimentally based evidence in order to analyze and explain the
physical facts, moving cautiously from a classical to a quantum description, without mixing them
up. The overall picture will be that the physical properties of microscopic systems are quantized, as
initially shown by Planck and Einstein, and that they are also contextual, i.e., they can be given a
physical sense only by embedding a microscopic system within a macroscopic measurement context.

Keywords: quantum physics; helium; electron spin; contextuality; operator algebra

Citation: Grangier, P.; Auffeves, A.; 1. Prologue

Farouki, N.; Van Den Bossche, M.; “Quantum mechanics occupies a very unusual place among physical theories: it contains
Ezratty, O. The Two-Spin Enigma: classical mechanics as a limiting case, yet at the same time it requires this limiting case
From the Helium Atom to Quantum for its own formulation." L.D. Landau and E.M. Lifshitz, Quantum Mechanics [1].
Ontology. Entropy 2024, 26, 1004.
https://doi.org/10.3390/e26121004
“When you have eliminated all the impossible, then whatever remains, however improba-
ble, must be the truth." A. Conan Doyle, The Case-Book of Sherlock Holmes [2].
Received: 12 October 2024
Accepted: 7 November 2024
In 1923, Max Born gave a series of lectures at the University of Göttingen, from which
Published: 22 November 2024 he drew a monograph entitled The Mechanics of the Atom [3]. The second edition of this book
was published in 1927 in English, just after the founding articles on quantum mechanics,
and he wrote the following in the Preface:
“Since the original appearance of this book in German, the mechanics of the atom has
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
developed with a vehemence that could scarcely be foreseen. (. . . However,) it seems
This article is an open access article
to me that the time is not yet arrived when the new mechanics can be built up on its
distributed under the terms and
own foundations, without any connection with classical theory. It would be giving a
conditions of the Creative Commons wrong view of the historical development, and doing injustice to the genius of Niels Bohr,
Attribution (CC BY) license (https:// to represent matters as if the latest ideas were inherent in the nature of the problem, and to
creativecommons.org/licenses/by/ ignore the struggle for clear conceptions which has been going on for twenty-five years."
4.0/).

Entropy 2024, 26, 1004. https://doi.org/10.3390/e26121004 https://www.mdpi.com/journal/entropy

Entropy 2024, 26, 1004 2 of 13

As we now approach the hundredth anniversary of these historic discoveries, can we

say, as Max Born intended, that “the time is arrived when the new mechanics can be built up on
its own foundations, without any connection with classical theory"? The argument we want to
develop here, and which—as Max Born also wanted—somehow does justice to the genius
of Niels Bohr, is that this goal is in fact impossible to achieve. Indeed, based on a century of
theoretical and experimental progress, our conclusion is that classical physics and quantum
physics can neither function nor even be conceived without the other. Following the quotes
above by Landau, Lifshitz, and Conan Doyle, this is a quite improbable situation indeed;
nevertheless, it might well be the truth—we will see below why.
The general level of this article corresponds to that of an undergraduate physics
course. First-year quantum mechanics students should easily follow the presentation,
which deviates from textbooks more in the physical argumentation than in the technical
content. Specific non-textbook terminology is introduced with bold font.

2. The Mysteries of Helium

The inability of classical physics to explain the structure of atoms, as well as the
radiation they emit, is vividly illustrated in Max Born’s book, cited above. By adding to the
classical equations the heuristic recipe of angular momentum quantification proposed by
Niels Bohr, it is possible to compute the radiation—the spectrum—emitted by hydrogen
and some other (so-called hydrogenoid) atoms. But these calculations spectacularly fail for
helium, the simplest atom after hydrogen, with only two electrons—and Max Born’s book
demonstrates that this is not due to a miscalculation by early-twentieth-century theorists.
Mathematics is not magic, and correct calculations based on a physically false theory cannot
provide results in accordance with experiments.
But other experiments and ideas appeared during the 1920s: the Stern–Gerlach experi-
ment proposed by Otto Stern and carried out with Walther Gerlach in 1922; the hypothesis
by George Uhlenbeck and Samuel Goudsmit in 1925 of an intrinsic angular momentum of
the electron, called spin a little later by Wolfgang Pauli; and finally the exclusion principle,
by Pauli again, applied first to electrons, which are particles with half-integer spin, as de-
tailed in standard textbooks [1,4,5]. Finally, in 1925–26, the situation changed, as Max Born
recounts in the previous section, and the solution appeared, “with a vehemence that could
scarcely be foreseen". The two electrons in the helium atom are actually two-spin 1/2 parti-
cles, which are strongly coupled by their electrostatic interaction and subject to the Pauli
exclusion principle. It was understood that their state must be described mathematically by
a so-called wave function, including their positions and spins, which must change sign (we
say: be antisymmetric) by permutation of the two electrons. There are thus two families of
states: those that are symmetric for the spatial part of the wave function, depending on the
electrons’ positions, and antisymmetric for the spin; and those that are antisymmetric for
the spatial part and symmetric for the spin.
Let us then focus on the lowest energy state of helium, called the ground state, which
is symmetric for the spatial part and antisymmetric for the spin. The states of a spin 1/2
particle are traditionally denoted |+⟩ and |−⟩, corresponding to the measurement results
+h̄/2 and −h̄/2 for the spin component along an arbitrary direction, often chosen along an
Oz axis. The symbol h̄ = h/2π corresponds to a quantum of angular momentum, where
h ≃ 6.63 10−34 J.s is Planck’s constant. For two electrons we can thus define four states,
| + +⟩, | + −⟩, | − +⟩, and | − −⟩, where the first ± relates to one of the electrons and the
second relates to the other. The corresponding values ±h̄/2 could in principle be obtained
from independent measurements on the two electrons. None of these four states changes
sign by permutation of the two electrons, so none of them are suitable for the ground state
of helium.
Then, in order to describe this state, we have to consider that in quantum formalism
the symbol |ψ⟩ for a quantum state, as introduced by Dirac [1,4,5], actually designates a
vector in a mathematical sense: this means that linear combinations such as ( a|+⟩ + b|−⟩)
for one spin or (c| + +⟩ + d| + −⟩ + e| − +⟩ + f | − −⟩) for two spins are other possible
Entropy 2024, 26, 1004 3 of 13

states, with the letters designating

√ complex numbers. One can then see that the state
|s⟩ = (| + −⟩ − | + −⟩)/ 2 provides a solution: by swapping the two electrons, we change
|s⟩ to −|s⟩. This state |s⟩ is called a ‘spin singlet’, we will see why later, and the calculation,
taking |s⟩ as the spin part of the electrons’ state, gives the correct values of the helium
energy levels—a huge success for early quantum mechanics.
Now here is the enigma, which is the question that has painfully plagued physicists
for 100 years: if the electron pair is in the state |s⟩, what is the value of the projection
component of the spin of the first electron along the Oz axis? And it unfortunately turns
out that the formalism we have introduced above is completely incapable of answering
this question. So let us be more concrete, and ask the following: in the |s⟩ state, what value
are we going to find by measuring the projection component of the spin of the first electron
along the Oz axis? Then, there is an answer: we can randomly find either +h̄/2 or −h̄/2,
with probabilities both equal to 1/2.
From here on, physicists tear themselves apart, as shown by many, often conflicting,
interpretations on quantum mechanics which are documented in the Stanford Encyclopedia
of Philosophy [6–14] or in [15]. Some physicists say that if we find +h̄/2, it means that this
projection was already worth +h̄/2 before we measured it, because it cannot be otherwise,
right? But some others answer that no, not at all, before the measurement the value of the
projection was simply not defined—whatever it means. And still others say that, in fact, the
measurement never happens; we have a branching of universes, one where the projection
is +h̄/2, and another where it is −h̄/2. Still others claim that the result does not really
exist, but that it is only a subjective bet made by the agent who made the measurement.
And if we ask ‘but does the state |s⟩ really exist’, some will say that this state may be ontic,
epistemic, or epi-ontic. . . which does not really help.
It is thus clear that a terrible slippage has occurred along the way, and that we must
take a few steps back and revisit the whole issue. In the line of our investigation, the enigma
is then as follows: What caused the demise of the spin components of the individual
electrons from our world? Did this really happen, and if yes, why and when? This question
amounts to no less than clarifying what exists, what is real, and what is objective in a
quantum phenomenon. And this is not mere philosophy: at the moment where quantum
technologies bloom and concepts of quantum physics must be assimilated by engineers,
these points have to be clarified—or more stuff that was thought to be real will keep on
vanishing. As a matter of facts, engineers, like enquirers, do not work with metaphors
that shed light on a deep but mysterious mathematical formalism. They need predictable
and repeatable facts on which to build intuition and simple assessment rules because
their action is at another complexity level, the one of building devices by having many
elementary systems work together.

3. One Spin Is Almost Fine, Two Spins Are Really Weird

3.1. Further Investigations
Given the previous conundrum, it is appropriate to seek more expert advice, so let us
refer to textbook quantum mechanics (TBQM) for more explanations [1,4,5]. In the previous
part, we introduced the states |+⟩ and |−⟩, corresponding to the results +h̄/2 and −h̄/2 of
a measurement of the projection of the electron’s spin in an Oz direction. Measuring the
spin of an electron in an atom is not easy, but it can be conducted for atoms with a single
‘active’ electron. This is the principle of the Stern and Gerlach experiment we have already
talked about, initially carried out with silver atoms. The Oz direction of the measurement
corresponds to the direction of a magnetic field gradient that the atom experiences in the
device, and must be chosen beforehand; a measurement along Oz does not tell what the
result would be in another direction, Ox or Oy for example (see Figure 1). Nevertheless,
repeating the same measurement along Oz will—not surprisingly—give the same result.
So, from now on we will add a subscript indicating the measurement direction, such as
|± x ⟩ for Ox, and |±z ⟩ for Oz.
Entropy 2024, 26, 1004 4 of 13

Figure 1. Stern–Gerlach magnets measuring the spin components along Oz (left), along Ox (center),
and along Oz/Ox/Oz (right). In this last case, the results, −h̄/2, are blocked by the dark square
screens. © Manuel Joffre, Ecole Polytechnique.

We can then try to chain several measurements: first along Oz, which gives, let us say,
+h̄/2, then along Ox, then along Oz again. Theory and experiment say that the second
measurement, along Ox, gives a random result ±h̄/2, with probabilities of 1/2. Why not, so
let us assume that we find +h̄/2, and let us go back to the projection component along Oz
that we knew before; but if we make the measurement again along Oz, we find a random
result ±h̄/2, with probabilities of 1/2: everything happens as if measuring along Ox had
erased the previously known result along Oz ! So, already with a single-spin 1/2 particle,
a strange vanishing of some properties can be observed—maybe a hint of what is going on
with the more tricky two-spin situation.
Another very strange fact is this: suppose Alice randomly selects a particle in any one
of the four states, |± x ⟩ or |±z ⟩, and sends it to Bob without any further indication. Bob
is then unable to determine the state of the particle without error! Indeed, if he chooses
(by chance) the same measurement axis as the one selected by Alice, he will find the
right result, the one she has sent, but if he chooses another axis, he will find a random
result, uncorrelated with what Alice sent. This impossibility, which also applies to any
eavesdropper, is at the basis of the technique called quantum cryptography, or, more
precisely, prepare-and-measure quantum key distribution. But then, how it is possible for
Bob to obtain any useful information? This can be performed by Alice and Bob publicly
exchanging their choice of axis after Bob has received the particle, and keeping only the
results where they made the same choice; the revealed information is then useless for a
possible eavesdropper, since the particle is no longer there [16].
The appearance of these random results raises several questions:
- Is it possible to find, in one way or another, situations where a measurement provides
a deterministic result, i.e., one that can be predicted with certainty? This seems
mandatory in order to attribute a physical reality to the objects we consider, and in the
framework of our inquiry, we need actual and objective facts, not ghostly ones.
- If Bob cannot fully identify the state of the particle (the one known and sent by Alice)
by making measurements of that particle, is it legitimate to say that |ψ⟩ is the ‘complete
state of the particle’?
Usual TBQM attributes a ‘complete quantum state’ to a particle, so its answer to
the second question above is affirmative, but it also says that despite this completeness,
Bob’s measurement result is generally not deterministic. Then, moving on to two particles,
the contradictions accumulate, culminating in the strange situation mentioned at the end of
the previous section. Therefore, unfortunately, one has to conclude that the expert advice
from TBQM did not really help, but rather increased the confusion. So we will now look
for different answers to these questions by saying that the quantum state may not be so
complete after all—and show that we can thereby make one step towards elucidating this
two-spin enigma, in helium and everywhere else.
Entropy 2024, 26, 1004 5 of 13

3.2. Some Clues and First Answers to the Questions

In the rest of this article, we will reason about idealized experiments, in the sense
that they are possible without contradicting anything we know about quantum mechanics
(QM), but can be very difficult to perform in practice. Such reasoning has been used since
the early days of QM, under the name of thought experiments, or Gedankenexperiments.
In fact, it turns out that nowadays the progress in the mastery of quantum systems is such
that a very large number of these thought experiments have been carried out, with such a
good degree of approximation that they can be considered to experimentally validate the
reasoning initially carried out in principle.
This is particularly the case for what are now called ‘Quantum Non-Demolition mea-
surements’, or QND measurements, in which the quantum state of a particle is identified
while leaving that particle in the observed state afterwards [17]. These measurements are
sometimes referred to as von Neumann’s ideal measurements, because John von Neu-
mann used them extensively from a theoretical point of view [18]; but it is now possible
to consider them well-established experimental facts [19–21]. They have the considerable
advantage of giving a clearly positive answer to our first question, i.e., whether it is possible
to obtain, in one way or another, deterministic results. The answer is yes, because it is
enough to repeat a QND measurement on the same particle or system to find the same
result again with certainty. Let us insist on the essential point that it is not only a question
of chaining together different measurements on the same system, which leads to random
results as we have written before: it is a question of repeating the same measurement on
the same system. This certainty may not be as general as we might have liked, because it is
destroyed by making a different measurement, as said before, but it is consistent with what
we learn in QM textbooks and observe in the lab, so it will allow us to move forward.
The answer to the second question, ‘if Bob cannot find the state of the particle by
measuring that particle, is it legitimate to say that it is indeed the state of the particle?’ is
going to take us further off the beaten track. We have already seen that if Alice gives Bob the
particle without indicating its state, Bob cannot return to that state with certainty. On the
other hand, if Alice gives Bob the particle without indicating its state, but also indicating the
direction of measurement she used, Ox for example, then Bob can identify with certainty
which state, |+ x ⟩ or |− x ⟩, was prepared by Alice. We thus find the same certainty as
before, provided again that the correct measurement is repeated, i.e., that Alice’s and Bob’s
axes agree. Introducing a bit of terminology, we will call the set of classical parameters
defining Bob’s action a ‘measurement context’, or, more briefly, a context denoted by C and
materialized by a macroscopic classical device. We can thus say that the physical object that
owns a certain and reproducible measurement result is not the system (the particle) alone,
but the system within a context. In this framework, the possible results of a specified
measurement, attributed to a system within a context, are called modalities [22].
If Bob has the quantum particle and knows the context used by Alice to prepare
the particle, then he can identify an associated modality. For a single spin, the context is
defined by the direction of the projection, Oz for example, and there are only two mutually
exclusive results, ±h̄/2. A possible modality is then +h̄/2 along Oz, corresponding to the
vector |+z ⟩ introduced earlier. For a single spin, it seems that there is no difference between
a state and a modality, but we will see later that the distinction between the usual (textbook)
state vector and the modality becomes essential for larger systems, e.g. for two spins.
To summarize this section, we have found some determinism in the behavior of
the physical objects used in QM, if we accept that they are in fact (quantum) systems
within (classical) contexts. Then, the very existence of the victim of the demise—the
spin component of the electron—may become uncertain indeed, because it is no longer a
property belonging to the electron alone. Clearly, the modality may change if we change
the context—and a new modality will show up in a new context, and will be then certain
and repeatable. But before exploiting this major result in our inquiry, let us return once
again to helium, and more precisely to the other family of helium states, those that are
Entropy 2024, 26, 1004 6 of 13

antisymmetric for the spatial part and symmetric for the spin—that is, their spin state does
not change its sign by permutation of the two electrons.

3.3. The Solution to the Enigma—Or Not?

The four possible states, | +z +z ⟩, | +z −z ⟩, | −z +z ⟩, and | −z −z ⟩, for the two spins
correspond to separate measurements made on the two electrons, and do not all fit with
the symmetry or antisymmetry condition imposed by the Pauli exclusion √ principle. We
have already seen that the vector |s⟩ = (| +z −z ⟩ − | −z +z ⟩)/ 2, the singlet state, is
antisymmetric. But what are the symmetric states? One can see √ that | +z +z ⟩ and | −z −z ⟩
are suitable, as well as the state |r ⟩ = (| +z −z ⟩ + | −z +z ⟩)/ 2: we therefore have three
symmetric states, globally called triplet states.
We have already said that the symbols |ψ⟩ are vectors; now, it must be specified
that the four vectors | ±z ±z ⟩ are in fact orthogonal, and correspond to measurement
results, modalities as defined above, which are mutually exclusive: only one of these four
results can be obtained when measuring the components along Oz of the spin of each
electron. Let us then consider the four states, | +z +z ⟩ and |s⟩, |r ⟩, | −z −z ⟩. An elementary
calculation shows that they are also orthogonal, and thus also describe mutually exclusive
measurement results, but what is the associated measurement? A standard algebraic
calculation in QM shows that this is the measurement of the quantities ⃗S2 and Sz , where
⃗S is the total spin of the two electrons ⃗S = ⃗S1 + ⃗S2 ; it is associated with the two quantum
numbers S = 0 and S = 1 (see Figure 2). For S = 0, a measurement of the total spin along Oz
(or an arbitrary axis) gives the value 0: this is in fact the state |s⟩. For S = 1, a measurement
of the total spin along Oz gives the values +h̄ for the state | +z +z ⟩, 0 for the state |r ⟩,
and −h̄ for the state | −z −z ⟩. So, we again have four mutually exclusive results, which are
noted |S = 0, m = 0⟩, |S = 1, m = 1⟩, |S = 1, m = 0⟩, and |S = 1, m = −1⟩, where m is the
result mh̄ of the measurement. We note also that we can write |S = 1, m = 1⟩ = | +z +z ⟩
and |S = 1, m = −1⟩ = | −z −z ⟩: these are therefore the same ‘state vectors’, but not the
same modalities, since these results appear in completely different measurement contexts,
corresponding either to measurements on the separate spins or to a global measurement
of the total spin. This distinction is quite important [23], and we will return to it later; see
Figure 2 for a summary.

M=2×2 | + +⟩
4 orthonormal spin
first and second spin | + −⟩ basis states
first up and down
context first particle second particle
combinations | − +⟩
spin projection spin projection
= ½ or -½ = ½ or -½ | − −⟩ these contexts share 2
along 0z along 0z identical states
1,1 = | + +⟩ (in blue), which are
S=1 named “extravalent
M=3+1 ! modalities”
triplet 1,0 = "
(| + −⟩ + | − +⟩) = |+⟩
(symmetric
second states)
1, −1 = | − −⟩ another 4 orthonormal
context
triplet singlet S=0 spin basis states
!
states state singlet 0,0 = "
(| + −⟩ − | − +⟩) = |,⟩
(antisymmetric
state) this state is entangled
when viewed using the
total spin first context and it is a
M = system dimension total electron
projection (mS)
spin (S) modality in the second
on the z axis
2 context
S Sz

Figure 2. Various quantum states relevant for two electrons. All |±⟩ mean |±z ⟩. The total spin ⃗S2 is
1
the sum of the squares of the three spin projections (S2x + Sy2 + Sz2 ) and it takes the values S(S + 1)h̄2 .
In this figure, M is the total number of possible, mutually exclusive measurement outcomes—or
modalities—in each context.

Though the inquiry is clearly making progress, we will conclude this part with a new
conundrum by again considering the states/modalities |s⟩ and |r ⟩, perfectly defined for a
measurement of the total spin, and associated with well-defined energy levels of the helium
Entropy 2024, 26, 1004 7 of 13

atom. As we have already seen, the question ‘what is the value of the component of the
spin along Oz for only one of the two electrons’ has no answer within the framework of the
formalism of usual QM we have introduced. We can reconcile ourselves with this statement
by saying that of course there is no answer, but this is not really shocking since we are
talking about two electrons within the same atom, very strongly coupled by electrostatic
interaction, so it is not very surprising that they adopt a global behavior. In addition, there
is no clear way to measure each electron’s spin within the helium atom.
Yet the ever-doubting physicist will ask the following question: ok, but can’t we have
the two electrons, or any two spin 1/2 particles, very far apart, and still in the state |s⟩ or in
the state |r ⟩? This can be conducted in suitable experiments, as detailed in the Appendix A.
What, then, is the meaning of the statement that the component of its spin along Oz has
no value, or a completely random value? And what happens if we compare the results
of separate measurements on the two spins? Then, we have to dive into an even deeper
mystery, that of quantum entanglement, Verschränkung in German, with two famous
papers published in 1935: ‘Can Quantum-Mechanical Description of Physical Reality Be
Considered Complete?’, by Einstein, Podolsky, and Rosen (EPR) [24], and Schrödinger’s
almost desperate article, though resigned to the quantum weirdness, ‘Die gegenwärtige
Situation in der Quantenmechanik (The present situation in quantum mechanics)’ [25].
The cause of desperation of these great physicists was the following: the spin components
of the two remote electrons, which were supposed to have “no value”, turn out to be
strongly correlated! But how can it be, if their values are undetermined? Did QM fall from
Charybdis to Scylla? It is this new extended enigma that will now have to be solved.

4. Predictive Incompleteness of |ψ⟩ with No Context

4.1. The Bell Perspective
The problems raised by Einstein and his colleagues in 1935 addressed the almost philo-
sophical foundations of QM but had no direct experimental implications. The situation
changed in 1964 when John Bell proposed inequalities, deduced from hypotheses very close
to those discussed by EPR, and which could be tested experimentally on entangled pairs of
particles [26]. This stimulated a long series of experiments, starting in 1972, with highlights
in 1982 and 2015 [27], that fortunately concluded with the Nobel Prize awarded in 2022 to
Alain Aspect, John Clauser, and Anton Zeilinger ‘for experiments with entangled photons,
establishing the violation of Bell’s inequalities and pioneering quantum information sci-
ence’. We are not going to tell this story here, since it has been the subject of many articles
and books, but rather consider the following question: what is ultimately the content of
Bell’s hypotheses, leading to these inequalities, and now dismissed by observations? Many
possible answers have been proposed, but we will now develop one that is in line with the
content of the previous sections.
A possible understanding of Bell’s perspective says that regardless of the limitations of
actual measurements, the underlying reality should have a definite configuration, described
by some local hidden variables (LHVs), that fully specifies everything that can/will happen.
The existence of a physical mechanism producing an effect from LHVs as a cause is very
generally true in classical physics, and it is also built on the use of classical probability
theory. It means that even if the values of the LHV are not known, they are supposed to
exist as the ‘hidden cause’. Such an explanatory mechanism is generally called the com-
pleteness of inference, or predictive completeness, given the hidden parameters [28]. Let
us emphasize that this predictive completeness does not mean fully knowable determinism
but ontological determinism, i.e., the existence of a fundamental underlying predictable
mechanism, as expressed by Laplace’s demon or by Einstein’s famous quote “God does
not play dice”. What is at stake is thus the possibility of specifying everything that will
happen, in a classically deterministic or at least in a probabilistic theory, which is also
assumed by Bell to be local. Equivalently, one may say that randomness has an “ignorance
interpretation”: like in tossing a coin, the result is random, but in principle knowing all
Entropy 2024, 26, 1004 8 of 13

the initial parameters and the laws of dynamics should allow one to predict the result
with certainty.
Now, without further ado or mathematics, it is clear that the usual quantum state
vector, as introduced in the previous section for describing the spin systems and generically
denoted by |ψ⟩, does not fit in this framework. Specifying everything that may happen
can only be performed once the measurement context has been specified. More precisely,
this measurement context is required, in addition to |ψ⟩, in order to define a standard
probability distribution over a set of mutually exclusive events. In this sense, we can say
that for the spin systems the quantum state alone is predictively incomplete. Specifying
all these possible events (i.e., the possible measurement results) is equivalent to giving the
context, and |ψ⟩ allows one to calculate their respective probabilities of occurring. Since
this is obtained by specifying both |ψ⟩ and the context, we have to acknowledge that what
QM allows are contextual inferences [28].
Though this just tells how quantum mechanics works, it may appear shocking to both
a classical physicist and a (naive) quantum physicist. A classical physicist has a hidden
cause of randomness and correlations (Bell’s LHV) deeply rooted in their mind, so saying
that there are no such variables in quantum physics is hard to accept. On the other hand,
a quantum physicist has deeply rooted in their mind that |ψ⟩ is complete, in the sense that
there is no underlying hidden cause to be sought—as it was taught by Niels Bohr, and as
it is still true. But saying that the quantum state vector |ψ⟩ is incomplete as long as the
context has not been specified, though it is operationally obvious in quantum mechanics,
and does not contradict Bohr’s message, may be quite hard to accept. Why? Because
we would have then to give up the idea that |ψ⟩ fully specifies a physical object, which
is the quantum system under study, and to admit that the actual physical object has to
be the quantum system within a classical context. This appears so shocking that the
quantum physicist may be tempted to consider instead ‘non-local influences’, or other
weird explanations, in order to violate Bell’s inequalities; but this leads to the contradictory
points of views presented before, and it is actually not needed if predictive incompleteness
is acknowledged [28].

4.2. Mutual Consistency of the Classical and Quantum Descriptions of the Physical World
The idea that context is needed to obtain a complete description of what can be ob-
served on a quantum system has often been confused with statements about the subjectivity
of the quantum state, or about the role of the consciousness of the observer [15]. There is
no need for that either, and a physical object remains perfectly objective, but it is still a
system within a context. This requires one, however, to recognize that there is no such thing
as absolute objectivity of predictions about the system alone; this departs from the usual,
classical, apprehension of the universe, taken as a whole, with all its logical and ontological
necessity as Newton and Einstein conceived it. The quantum requirement for a contextual
description of physical objects is related to many issues in quantum physics, and it has
been spelled out initially in the article ‘Contextual objectivity: a realistic interpretation of
quantum mechanics’ (2001) [29], which may provide us with a more suitable, non-classical
view about the world we live in [30].
There is clearly a price to pay for the idea that there are well-defined systems and
contexts: if this is the case, there must be a physical separation between them, usually
known as the “Heisenberg cut”, where the applicable laws change from quantum to
classical. This creates a dualist view of the physical world that does not seem satisfactory,
and there have been many attempts to find ways by which the classical world should
‘emerge’ from the quantum one—if any. This was also the content of Max Born’s initial
request, to build up ‘the new mechanics on its own foundations, without any connection
with classical theory’. But it can also be said, as Bohr and Landau did, that there is no
way to formulate quantum mechanics without referring to classical concepts. For instance,
Landau and Lifshitz [1] write “It is clear that, for a system composed only of quantum objects, it
would be entirely impossible to construct any logically independent mechanics (. . . ) Thus quantum
Entropy 2024, 26, 1004 9 of 13

mechanics occupies a very unusual place among physical theories: it contains classical mechanics as
a limiting case, yet at the same time it requires this limiting case for its own formulation”.
This leads to the idea that classical and quantum descriptions are both required
from the beginning and are simultaneously needed for mutual consistency [31]. Actually,
classical descriptions without quantum descriptions are unable to explain the structure of
atoms, as seen above, but quantum descriptions without classical descriptions are unable
to say clearly what a quantum state describes, and when extrapolated to the macroscopic
world they become lost in bizarre predictions that clash with empirical evidence.
An apparent difficulty for unifying classical and quantum physics is that their mathe-
matical frameworks seem incompatible – but this is not correct, and a unified formalism,
using operator algebras, is actually possible and was already suggested by John von Neu-
mann in 1939. Though this extension of the standard quantum formalism is mathematically
nontrivial, it definitively deserves a closer look by quantum physicists—after being placed
into the appropriate ontological framework introduced above. This has been discussed in
several recent articles [32].

4.3. Recovering Real Properties: Modalities vs. State Vectors

An important step in our discussion has been to introduce a specific word representing
a physical property or set of physical properties, attributed to a system within a context—we
call it a modality. Clearly, a modality is different from a usual quantum state vector |ψ⟩,
which does not specify the context. It is important to note that for a single-spin 1/2 particle,
just giving a quantum state vector like |+z ⟩ actually specifies the context Oz, and there is
only one other mutually exclusive modality, that is, |−z ⟩. But for larger values of the spin,
or for more particles, there are more than two mutually exclusive modalities in a given
context. In that case, a given quantum state vector may appear in infinitely many contexts,
as was the case (for two different contexts) with | +z +z ⟩ = |S = 1, m = 1⟩ in the example
given above.
One should also note √ that the quantum state vectors √ |±z ⟩ and |± x ⟩ are related by
|+ x ⟩ = (|+z ⟩ + |−z ⟩)/ 2 and |− x ⟩ = (|+z ⟩ − |−z ⟩)/ 2: the state vectors along Ox
are linear superpositions of the state vectors along Oz. It is often said that ‘quantum
superposition are like being in the two states |+z ⟩ and |−z ⟩ at the same time’, which clearly
makes no sense: the modalities associated with |± x ⟩ are certainties in the Ox context,
mathematically expressed as linear combinations of state vectors |±z ⟩ that are certainties in
the Oz context. It is also useful to keep in mind that the state |s⟩ appears entangled in the
context where the two spins are described and measured separately, but not in the context
based on the total spin with S = 1 or S = 0. Entanglement is thus related to the choice
of a description and measurement context (technically, a tensor product structure) for the
system of interest. Obviously, the description using |±, ±⟩ appears more ‘natural’ when
the two spins are spatially separated.
Putting everything together, the theoretical tools used so far consist of associating
orthogonal vectors with mutually exclusive modalities (in the same context), and the
same vector with mutually certain modalities (in different contexts). Introducing some
vocabulary, the non-obvious physical phenomena that certainty can be transferred between
contexts is called extracontextuality [23]. Then, the property of ‘being mutually certain’
defines an equivalence relation between modalities, which is called extravalence—think
again of the relation | +z +z ⟩ = |S = 1, m = 1⟩.
We come thus to the conclusion that the quantum state vector |ψ⟩ is a mathematical
object associated with an extravalence class of modalities. This establishes a clear sepa-
ration between mathematical objects like |ψ⟩ and physical objects that are systems within
contexts and carry well-defined and measurable properties that correspond to modalities.
These sentences require some thinking to be fully appreciated, but they do refer to the
ontology of physical objects (systems and contexts) and their properties (modalities) within
the non-classical but realistic framework provided by contextual objectivity.
Entropy 2024, 26, 1004 10 of 13

To complete the picture, we need a fundamental postulate, called contextual quanti-

zation, which says that the maximum number M of mutually exclusive modalities is a
property of the quantum system, and it is the same in any relevant context.
We have already seen an example of this behavior with the two electron spins, where
M = 4 for several possible contexts—this is actually the case for any such context. Given
this postulate, it can be shown that the relation between modalities in different contexts can
only be probabilistic [33]. The fixed value of M also allows one to associate the system with
an M-dimensional vector space—actually a Hilbert space [34]. In this picture, there is no
trouble in using the standard projection postulate of textbook QM, which appears simply
as an update of the probability distribution associated with |ψ⟩, for a given context.
The crucial point is that this probability distribution is non-classical and contextual,
and it becomes a computable physical meaning only once the actual context has been
specified [33]. For the two-spin enigma, the violation of Bell’s inequalities is attributed to the
predictive incompleteness of |ψ⟩, and it does not require any non-local influence, but rather
a contextual inference, as explained in detail in [28]. So, there are no ‘hidden variables’ to
be looked for, and the context, completing |ψ⟩, specifies the “very conditions which define
the possible types of predictions regarding the future behavior of the system”, as stated
by Bohr [35]. Mixing up probabilities calculated in different contexts is counterfactual
and leads to contradictions. There is no doubt that these explanations are quite remote from
classical physics, but again they make full sense in the framework of contextual objectivity.

5. Epilogue
Since the purpose of this paper is to illustrate and discuss quantum ideas using two-
spin 1/2 particles, we did not reconstruct the full quantum formalism yet, and explanations
of basic tools like unitary transforms and Born’s rule are still missing. But there is not much
choice left, because the framework defined above fits with the hypotheses of powerful math-
ematical theorems, establishing the need for unitary transformations (Gerhard Uhlhorn’s
theorem [36]) and for Born’s rule (Andrew Gleason’s theorem [37]). These arguments have
been published elsewhere [34] and will not be reproduced here. Also, systems and contexts
should be part of a unified mathematical formalism, going beyond standard textbook
QM. Operator algebra, introduced by Murray and von Neumann [38,39] and considerably
developed since then, can provide such a formalism and give a mathematical status to the
Heisenberg cut, as has been shown in [32].
So finally, what should we say to the layman to bring our investigation to a successful
conclusion and to replace misleading statements such as “a quantum superposition is like
being in two states at the same time”, or “quantum entanglement is like an instantaneous
action at a distance”? It could be the following:
The physical properties of microscopic systems are quantized, as initially shown by Planck
and Einstein, and they are also contextual; this means that they can be given a physical
sense only by embedding a microscopic system within a macroscopic context that specifies
how the system is observed, as proposed by Bohr. Such a behavior is quite remote from
the non-quantized and non-contextual perspective of classical physics, and it requires
developing a specific contextual probabilistic theory, which is basically what quantum
mechanics has been doing.
To conclude, let us emphasize that QM follows from the conjunction of quantization
and contextuality; these two properties taken separately may have some classical analogue,
and therefore they do not have the same constraining power when being combined.

Author Contributions: Writing—original draft, P.G.; Writing—review & editing, A.A., N.F., M.V.D.B.
and O.E. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Entropy 2024, 26, 1004 11 of 13

Data Availability Statement: This is theoretical research, so no new data were created.
Acknowledgments: P.G. thanks Franck Laloë, Roger Balian, and Karl Svozil for many interesting and
useful discussions, and Manuel Joffre for permission to use Figure 1 from his Quantum Mechanics
course at Ecole Polytechnique. MvdB thanks the Thales engineers for their candid but actually
extremely relevant questions. The authors thank Michel Kurek and Camilia Ben Messaouda for
proofreading and comments on the paper.
Conflicts of Interest: Author Mathias Van Den Bossche is employed by the company Thales Ale-
nia Space. The remaining authors declare that the research was conducted in the absence of any
commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A. Another View on Two-Spin 1/2 Particles

In the above text, we use the example of the helium atom to introduce the four states
| +z +z ⟩, |r ⟩, |s⟩, and | −z −z ⟩, which can also be written as |S = 1, m = 1⟩, |S = 1, m = 0⟩,
|S = 0, m = 0⟩, and |S = 1, m = −1⟩. In the helium atom, these states are imposed by the
Pauli exclusion principle and can be identified because they correspond to specific energy
levels. On the other hand, the context corresponding to the separated states of the two
electrons, | +z +z ⟩, | +z −z ⟩, | −z +z ⟩, and | −z −z ⟩, is not accessible within the helium
atom, since this would require one to have individual access to the electrons by taking them
far away from the atom and thus changing their energy state.
So, it would be nice to exhibit another system, where all these states can be manipu-
lated and measured. This has been proposed by photo-dissociating a Hg2 molecule in a
singlet state and carrying out spin measurements on the two remote fragments [40]. But this
turned out to be a quite difficult experiment, and another more feasible approach is to
consider that a spin 1/2 particle is actually a quantum bit, a qubit in a quantum computer,
and identify |+z ⟩ with the logical state |1⟩ and |−z ⟩ with the logical state |0⟩.
The question is then to move from the four states (‘uncoupled basis’) |1, 1⟩, |1, 0⟩,
|0, 1⟩, and |0, 0⟩, usually√ called the computational
√ basis, to the four states (‘coupled basis’)
|1, 1⟩, (|1, 0⟩ + |0, 1⟩)/ 2, (−|1, 0⟩ + |0, 1⟩)/ 2, and |0, 0⟩. This can be carried out by the
following 4 × 4 unitary matrix:

1 0√ 0√ 0
 
 0 1/ 2
√ −1/√ 2 0 
G=
 0 1/ 2

1/ 2 0 
0 0 0 1

which can be implemented by a sequence of quantum gates, as shown in Figure A1. This
shows explicitly that, as written in the text, two qubits can be set or measured in the state
|s⟩ or in the state |r ⟩ outside the helium atom.

𝐻 𝐻

Figure A1. Sequence of one- and two-qubit quantum gates and their matrix unitaries acting on
two qubits, to move between the coupled and uncoupled basis. The first and last gates are Hadamard
gates, the second and forth Controlled-NOT gates, and the middle ones are specific non-Clifford
one-qubit gates.
Entropy 2024, 26, 1004 12 of 13

An arbitrary quantum state can thus be measured in either the computational √ basis
{|1, 1⟩, √
|1, 0⟩, |0, 1⟩, |0, 0⟩} or in the coupled basis {|1, 1⟩, (|1, 0⟩ + |0, 1⟩)/ 2, (−|1, 0⟩ +
|0, 1⟩)/ 2, |0, 0⟩}. More precisely, the nontrivial measurement in the coupled basis can
be realized by applying G † to the two qubits, carrying out a QND measurement in the
uncoupled (computational) basis, which is in principle easy, and then applying G to recover
the measured state in the coupled basis. Therefore, such a given sequence of gates, building
up a unitary transformation, implements a controlled change of context, since not only one
but a whole set of mutually orthogonal states are mapped onto another set of mutually
orthogonal states. This has the advantage that modalities can be defined up to a unitary
transform, in the case where the appropriate context cannot be implemented. Though not so
easy to implement in practice, this two-way procedure (G † /QND/G) is useful in quantum
computing and shows that thinking about physical modalities, associated with a context,
may be more enlightening than just thinking about mathematical state vectors, which
represent extravalence classes of modalities occurring in different contexts.

References
1. Landau, L.D.; Lifshitz, E.M. Quantum Mechanics. 1965. Available online: https://archive.org/details/ost-physics-landaulifshitz-
quantummechanics (accessed on 12 May 2023).
2. Conan Doyle, A. The Case-Book of Sherlock Holmes; John Murray: London, UK, 1927.
3. Born, M. The Mechanics of the Atom; G. Bell and Sons: London, UK, 1927.
4. Feynman, R. The Feynman Lectures on Physics, Volume III: Quantum Mechanics. 1963. Available online: https://www.
feynmanlectures.caltech.edu/ (accessed on 12 May 2023).
5. Cohen-Tannoudji, C.; Diu, B.; Laloë, F. Quantum Mechanics; 3 Volumes; John Wiley & Sons: Hoboken, NJ, USA, 2019.
6. Faye, J. Copenhagen Interpretation of Quantum Mechanics. Available online: https://plato.stanford.edu/entries/qm-copenhagen/
(accessed on 12 May 2023).
7. Bacciagaluppi, G. The Role of Decoherence in Quantum Mechanics. 2020. Available online: https://plato.stanford.edu/entries/
qm-decoherence/ (accessed on 12 May 2023).
8. Goldstein, S. Bohmian Mechanics. 2021. Available online: https://plato.stanford.edu/entries/qm-bohm/ (accessed on 12 May
2023).
9. Ghirardi, G.; Bassi, A. Collapse Theories. 2020. Available online: https://plato.stanford.edu/entries/qm-collapse/ (accessed on
12 May 2023).
10. Barrett, J. Everettian Quantum Mechanics. 2023. Available online: https://plato.stanford.edu/entries/qm-everett/ (accessed on
12 May 2023).
11. Vaidman, L. Many-Worlds Interpretation of Quantum Mechanics. 2021. Available online: https://plato.stanford.edu/entries/qm-
manyworlds/ (accessed on 12 May 2023).
12. Lombardi, O.; Dieks, D. Modal Interpretations of Quantum Mechanics. 2021. Available online: https://plato.stanford.edu/entries/
qm-modal/ (accessed on 12 May 2023).
13. Laudisa, F.; Rovelli, C. Relational Quantum Mechanics. 2019. Available online: https://plato.stanford.edu/entries/qm-relational/
(accessed on 12 May 2023).
14. Healey, R. Quantum-Bayesian & Pragmatist Views of Quantum Theory. 2022. Available online: https://plato.stanford.edu/
entries/quantum-bayesian/ (accessed on 12 May 2023).
15. Laloë, F. Do We Really Understand Quantum Mechanics? Cambridge University Press: Cambridge, UK, 2012.
16. Bennett, C.H.; Brassard, G. Quantum cryptography: Public key distribution and coin tossing. In Proceedings of the IEEE
International Conference on Computers, Systems and Signal Processing, Bangalore, India, 9–12 December1984; Volume 175, p. 8.
17. Braginsky, V.B.; Khalili, F.Y. Quantum nondemolition measurements: The route from toys to tools. Rev. Mod. Phys. 1996, 68, 1.
[CrossRef]
18. von Neumann, J. Mathematische Grundlagen der Quantenmechanik; Springer: Berlin/Heidelberg, Germany, 1932; English translation
Mathematical Foundations of Quantum Mechanics; Princeton University Press: Princeton, NJ, USA, 1955.
19. Grangier, P.; Levenson, J.A.; Poizat, J.P. Quantum Non-Demolition Measurements in Optics. Nature 1998, 396, 537. [CrossRef]
20. Haroche, S.; Raimond, J.M. Exploring the Quantum; Oxford University Press: Oxford, UK, 2006.
21. Hume, D.B.; Rosenband, T.; Wineland, D.J. High-Fidelity Adaptive Qubit Detection through Repetitive Quantum Nondemolition
Measurements. Phys. Rev. Lett. 2007, 99, 120502. [CrossRef] [PubMed]
22. Auffèves, A.; Grangier, P. Contexts, Systems and Modalities: A new ontology for quantum mechanics. Found. Phys. 2016, 46,
121–137. [CrossRef]
23. Auffèves, A.; Grangier, P. Extracontextuality and extravalence in quantum mechanics. Phil. Trans. R. Soc. A 2018, 376, 20170311.
[CrossRef]
24. Einstein, A.; Podolsky, B.; Rosen, N. Can Quantum-Mechanical Description of Physical Reality Be Considered Complete? Phys.
Rev. 1935, 47, 777. [CrossRef]
Entropy 2024, 26, 1004 13 of 13

25. Schrödinger, E.W. Die gegenwärtige Situation in der Quantenmechanik (The present situation in quantum mechanics). Naturwis-
senschaften 1935, 23, 807. [CrossRef]
26. Bell, J.S. On the Einstein Podolsky Rosen Paradox. Physics 1964, 1, 195–200. [CrossRef]
27. Aspect, A. Closing the Door on Einstein and Bohr’s Quantum Debate. Physics 2015, 8, 123. Available online: https://physics.aps.
org/articles/v8/123 (accessed on 15 May 2015). [CrossRef]
28. Grangier, P. Contextual inferences, nonlocality, and the incompleteness of quantum mechanics. Entropy 2021, 23, 1660. [CrossRef]
29. Grangier, P. Contextual objectivity: A realistic interpretation of quantum mechanics. Eur. J. Phys. 2002, 23, 331. [CrossRef]
30. Farouki, N.; Grangier, P. The Einstein-Bohr debate: Finding a common ground of understanding? Found. Sci. 2021, 26, 97–101.
[CrossRef]
31. Grangier, P. Completing the quantum formalism in a contextually objective framework. Found. Phys. 2021, 51, 76. [CrossRef]
32. Van Den Bossche, M.; Grangier, P. Postulating the Unicity of the Macroscopic Physical World. Entropy 2023, 25, 1600. [CrossRef]
33. Auffèves, A. and P. Grangier, P. What is quantum in quantum randomness? Phil. Trans. R. Soc. A 2018, 376, 20170322.
34. Auffèves, A.; Grangier, P. Revisiting Born’s rule through Uhlhorn’s and Gleason’s theorems. Entropy 2022, 24, 199. [CrossRef]
35. Bohr, N. Can Quantum-Mechanical Description of Physical Reality be Considered Complete? Phys. Rev. 1935, 48, 696. [CrossRef]
36. Uhlhorn, U. Representation of symmetry transformations in quantum mechanics. Arkiv für Fysik 1962, 23, 307–340.
37. Gleason, A.M. Measures on the Closed Subspaces of a Hilbert Space. J. Math. Mech. 1957, 6, 885–893. [CrossRef]
38. Murray, F.J.; von Neumann, J. On Rings of Operators IV. Ann. Math. 1943, 44, 716. [CrossRef]
39. von Neumann, J. On infinite direct products. Compos. Math. 1939, 6, 1–77. Available online: http://www.numdam.org/item?id=
CM_1939__6__1_0 (accessed on 8 September 2019).
40. Fry, E.S.; Walther, T. Fundamental tests of quantum mechanics. Adv. At. Mol. Opt. Phys. 2000, 42, 1.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.