Two aspects of Śūnyatā in quantum physics :

relativity of properties and quantum non-separability

Michel Bitbol
Archives Husserl, CNRS/ENS, 45, rue d’Ulm, 75005 Paris, France

In : Siddheshwar Rameshwar Bhatt (ed.), Quantum Reality and Theory of Śūnya,

Springer, 2019

Abstract : The so-called « paradoxes » of quantum physics are easily disposed of as

soon as a momentous philosophical step is taken. This step consists in accepting that
there are no such things as intrinsically existing particles and their intrinsic properties,
but that both particles and properties are relational “observables”. Accordingly,
quantum physics does not offer a “description of the outer world”, but rather a
prescription about how to make probabilistic predictions within a participatory
environment. The latter view (or rather criticism of views) looks quite radical with
respect to standard Western Aristotelian ontology; but it looks entirely natural in the
context of the Indian-Buddhist concept of Pratītyasamutpāda which underpins Śūnyatā.
Special attention will then be devoted to the quantum feature of non-separability or
entanglement. Quantum non-separability displays remarkable similarities with
Pratītyasamutpāda, in view of its lack of affinities with productive causality, and its
convergence with a symmetric relation of co-production. Finally, the meaning of such
twofold parallel between quantum physics and Śūnyatā will be discussed. This parallel
will be related to the similarity of epistemological situation between knowing a world
from which we are not entirely separated, and knowing oneself.

Introduction
That there is a striking analogy between central features of
quantum physics and the Indo-Buddhist concept of Śūnyatā was
pointed out by many authors in the past1. However, most of them
did not go beyond superficial similarities in the vocabulary (e.g.
equating the quantum vacuum with emptiness), or superficial
affinities between the alleged pictures of the world (e.g.
interdependence instead of separation, flux and impermanence
1
F. Capra, The Tao of Physics, Shambala, 1983 ; K. Wilber (ed.), The Holographic Paradigm
and Other Paradoxes: Exploring the Leading Edge of Science, Shambala, 1982 ; B. Alan
Wallace, Choosing Reality : A Buddhist View of Physics and the Mind, Snow Lion, 2003 ; V.
Mansfield, Tibetan Buddhism and Modern Science, Templeton Foundation Press, 2008, etc.
Among these books, the one to which I feel closer by far is Alan Wallace’s. However, I lean
more than he does towards an anti-metaphysical and quasi-instrumentalist construal of
physics. It seems to me that the middle way should not apply to the status of physical theories.
Physical theories are definitely not about Being. They are about our practical possibilities to
anticipate the effects of our actions. They are (a sophisticated) part of the conventional truth,
even though, as we will see in section 4, one can identify structures which are reasonably
isomorphic to the duality of conventional truth and ultimate truth inside them.

1
instead of reifying stability etc.). In this paper, I wish to spell out
some points on which the parallel precisely works beyond mere
family resemblance, and to make a few assumptions about the
origin of this parallel.
As a preliminary, it must be borne in mind that the relevant
parallel should not concern two metaphysical views. It should not
be question of comparing two views of the world (a Buddhist and a
Scientific view), and to find that, surprisingly enough, they are
similar. It should not be question either of two ways (internal and
external, experiential and experimental) of probing into the true
essence of reality, and of expressing this essence by words or
mathematical formalisms which, surprisingly, seem to converge.
Actually, the parallel concerns two of the most radical critiques of
metaphysical views and essentialism2 that have been proposed in
the history of human thought.
Buddhism arose from a deep and basic existential concern, and
did not indulge in metaphysical speculation. Its aim was
therapeutic rather than doctrinal. This skeptical / agnostic tendency
of Buddhism was amplified in Prasangika Madhyamaka, wherein
every single metaphysical view is submitted to a procedure of
reductio ad absurdum which tends to show that this view is in fact
only partial and conventional. Ultimate truth here can by no means
be captured by a metaphysical picture expressed in words or
symbols.
In the same way, quantum physics3 is persistently averse to
metaphysical interpretation. Despite many half-successful attempts
of providing it with a so-called “realist interpretation”, quantum
physics still relies on Bohr’s and Heisenberg’s initial remark
according to which no unified picture of the atomic and subatomic
domain can be derived from it. According to these founding fathers
of quantum mechanics, their theory is no “view” of the micro-
world, but rather a mathematical symbolism intended to predict
probabilistically the outcome of experiments performed at the
micro-scale using macroscopic devices. As we will see in sections
1, 2 and 3 of this paper, most “mysteries” of quantum physics (such
as the Einstein-Podolsky-Rosen paradox4, or the measurement

2
HH the Dalai-Lama, The Universe in a Single Atom, Morgan Road Books, 2005, p. 69
3
A. Zajonc & Greenstein, The Quantum Challenge: Modern Research on the Foundations of
Quantum, Jones & Bartlett Publishers, 1997
4
M. Smerlak & C. Rovelli, “Relational EPR”, Foundations of Physics 37, 427-445, 2007 ; M.
Bitbol, “An analysis of the Einstein-Podolsky-Rosen correlations in terms of events”, Physics
Letters 96A, 66-70, 1983

2
problem5) are defused from the outset if one is ready to accept all
the consequences of this economic construal of quantum
mechanics.
Recent developments in theoretical and experimental research
have strengthened the anti-metaphysical reading of quantum
physics. Quantum information theories, which underly the project
of quantum computers, have been shown to offer an alternative
foundation to quantum physics as a whole6. According to them,
quantum mechanics is nothing more and nothing else than a theory
of the limits of available experimental information. It is by no
means a description of the putative objects of experiments.
Besides, new tests of Bell-like inequalities7 put the very
assumption of “realism” (according to which quantum mechanics is
a possibly incomplete description of the intrinsic properties of real
objects) under increasingly strong pressure.
Then, in sections 4 and 5, special attention will be devoted to a
prominent feature of quantum physics that precisely evokes the
relational background of Śūnyatā. This remarkable feature is non-
separability or entanglement. Quantum non-separability displays
remarkable similarities with Pratītyasamutpāda. Indeed, far from
any kind of asymmetric productive causality, non-separability
strongly evokes a symmetric relation of co-production. But, here
again, instead of ascribing quantum non-separability to the world
as a whole, as if it were an intrinsic feature of the universe, we will
point out that non-separability itself is non-separable. Non-
separability is non-separable from its instrumental conditions of
manifestation (just as, according to Madhyamaka, emptiness itself
is empty).
At the end of the paper, in section 6, the meaning of such
multifarious parallel between quantum physics and Śūnyatā will be
discussed. This parallel will be related to the similarity of

5
U. Mohrhoff, “The world according to quantum mechanics (Or the 18 errors of Henry P.
Stapp)”, Foundations of Physics 32 (2), 217–254 (2002) ; M. Bitbol, “Consciousness,
Situations, and the Measurement Problem of Quantum Mechanics”, NeuroQuantology, 6, 203-
213, 2008
6
A. Zeilinger, « Foundational principle for quantum mechanics » Foundations of Physics, 29,
631-643, 1999 ; C.A. Fuchs, « Quantum mechanics (and only a little more) », in : A.
Khrennikov (ed .), Quantum Theory: Reconsideration of foundations, Växjo University Press,
2002 ; A. Grinbaum, « Elements of information-theoretic derivation of the formalism of
quantum theory », International Journal of Quantum Information, 1, 289–300, 2003.
7
S. Gröblacher,T. Paterek, R. Kaltenbaek, C. Brukner, M. Zukowski, M. Aspelmeyer, & A.
Zeilinger, « An experimental test of non-local realism », Nature, 446, 871-875, 2007

3
epistemological situation between knowing a world from which we
are not entirely separated, and knowing oneself.

1-Some quantum paradoxes: a Śūnyatā resolution

This preliminary section is devoted to showing that the majority

of the so-called “weird paradoxes” of quantum physics dissolve
away and no longer appear “weird”, once they have been
approached relationally rather than absolutely. In other terms, they
dissolve away as soon as one accepts that the microentities are
empty of intrinsic being and determinations.
There is a premise which conditions all these case-studies. It
concerns the evaluation of the epistemological status of quantum
theories. What are these theories, what do they succeed in doing,
and what do we have a right to ask of them? Do they offer an
explanation, or at least a description unequalled in its precision, of
subatomic processes? Have they made it possible to elucidate the
intimate mechanisms of chemical phenomena and nuclear
transformations? Have they penetrated further into the fundamental
nature of things than classical theories, such as the mechanics of
Galileo and Newton, thermodynamics or electromagnetism, have
ever managed to do before? The great misunderstanding which
runs through the debates on quantum physics stems from the fact
that most of the time it is thought that this is the way things must
be: that the three questions listed above must be construed as so
many positive assertions. The belief of Western science in a telos
here claims its due. However, once one accepts that, a whole
cascade of disturbing conclusions burst forth. The world which is
supposedly “revealed” by quantum mechanics has an appearance as
unexpected and ludicrous as Alice’s Wonderland recounted by
Lewis Carroll. And in this case it would seem that all that can be
done is to accept the weirdness.
As it is well known, one of the first surprises is the wave-particle
duality, this strange association between extension and point
localization, between continuity and discontinuity, which is
supposed to describe the “nature” of the new entities that are
sometimes called “quantons”8. Einstein, the first scientist to have
imagined entities of this sort (concerning electromagnetic
radiation), was unable to hide his perplexity on this subject: “Is it

8
Lévy-Leblond J.M. & Balibar F. (1984), Quantique : Rudiments, Paris : Interéditions

4
possible to reconcile the quanta of energy on one hand, and the
Huygens principle on the other? Appearances go against it, but
God seems to have found the trick of it”9. It may be added that
although our spontaneous intuition cannot digest this amalgam
between the discontinuous and the continuous, between quanta and
the “Huygens principle” for wave interference, we ourselves seem
to have found a “trick” which has the reputation of being able to
combine them and thus to “enter into the mind of God”. The
mathematical formalism of Hilbert does indeed make it possible to
establish a connection both with a continuous geometry (by means
of the concept of the spatial amplitude of a probability or a “wave
function”), and with an algebra of discontinuity (by means of the
scheme of quantization). But does this formal derivation really
suffice to reconcile the two contradictory concepts, or does it just
amount to reject them both (setting them back-to-back) and
substitute something quite different in their place?
This paradoxical situation was the first one to alert scientists
about the extreme novelty of the quantum domain. Due to this and
other situations, many physicists consider that there is something
quite extraordinary in the occult nooks and crannies of the world;
something that quantum descriptions give an oblique glimpse of,
but that neither our language nor our imagination can properly
grasp, and that only mathematics makes it possible to circumscribe.
Nevertheless, the very same situations can also be interpreted in a
diametrically opposite fashion. Once they are reconsidered in the
most intellectually economical fashion, all these supposed
“paradoxes” converge towards the possibility that quantum theory
is nothing more than an ingenious but purely formal way of
anticipating experimental information; that it does not offer an
incomplete and cryptic revelation of an invisible and ineffable
reality, but only a method for orienting oneself with respect to that
which shows itself and is said; that instead of penetrating further
into the recondite depths of matter than any previous theories have
managed to do, quantum mechanics is rather a systematic inventory
of its surface. For the so-called wave-particle duality paradox can
be immediately dissolved as soon as one renounces the application
to quantum mechanics of the descriptive, representationalist,
“realist” conception of physical theories.

9
Einstein A. (1989), Oeuvres choisies, 1 Quanta, Seuil

5
 Indeed, assigning a double nature, as wave and as particle, to the
objects called “quantons” is a biased, over-determined and
prejudiced way of expressing a phenomenon which does not a
priori impose either an ontology of waves or an ontology of
particles. The phenomenon in question is the distribution of a large
number of punctual events according to a pattern which is
isomorphic to that which would be produced by the interference of
two waves or the diffraction of a single plane wave passing through
a hole. Bohr himself already criticized treating this sort of
phenomenon in ontological terms when he replaced the assertion of
a wave-particle duality by that of a complementarity of the images
of a wave and a particle. Each image is only relevant, according to
Bohr, with respect to a particular experimental context; and the
contexts which render these two images appropriate are partially
exclusive of each other. But that is not all. It can be shown in a
quite general way that any theory capable of accounting for
phenomena concerning mutually exclusive contexts, predicts
distributions which will have a wave-like appearance10; i.e.
distributions where everything happens as though we are dealing
with waves even though there are no waves at all. In other words,
far from manifesting the absolute wave-like properties of
microscopic entities, the interference behaviour of quantum
phenomena could be the eloquent sign of their epistemic relativity;
far from being an indication about the intrinsic nature and own-
being of the world, the interference behavior is a sign of its
emptiness of own-being; far from bearing witness to the deep
nature of things, the pseudo-wavelike effects could well represent
one of the most salient marks of the superficial, interfacial
character of the phenomena that quantum mechanics makes it
possible to anticipate.
A second surprising feature of quantum physics is the
measurement problem. A famous way of telling the story so as to
bring out its dramatic implications is the paradox of Schrödinger’s
cat11. The most succinct account of this paradox plays on the
contradiction between the state of the cat as described and as
concretely occurring. Quantum mechanics (so it is said) describes

10
Destouches-Février P. (1951), La structure des théories physiques, P.U.F.
11
Schrödinger E. ([1935] 1983), “The present situation in quantum mechanics”, in : Wheeler
J.A. & Zurek W.H. (eds), Quantum Theory and Measurement, Princeton: Princeton University
Press

6
the cat subjected to Schrödinger’s infernal machine12 as being in a
superposed state of being both alive and dead. However in actual
practice it is found that the cat is found to be either alive or dead.
Here, the (supposed) quantum description of the cat does not
accord with what one sees of it. Dozens of solutions have been
proposed to get around this difficulty. One of them consists of
taking the quantum “description” literally, and to suppose that each
of the two terms in the superposition represents a separate
“possible world”: in one of these worlds the cat really is alive (and
the inhabitants of this world see the cat alive), and in the other
world the cat is dead (and the inhabitants find it dead). However
the way out that is currently dominant (called “decoherence”)
amounts to refusing to confront the problem according to the
standard formulation, and to change the formulation in a way that
is so subtle that many scientists are unaware of the sleight of hand:
instead of a problem of compatibility between conjunction and
disjunction, between a plurality of possibilities and the uniqueness
of what actually exists, the problem that is resolved is a problem of
connection between two forms of the calculation of probabilities13.
Yet, this so-called cat’s paradox can also be dissolved at once if
only one accepts once again to follow the lead given by Bohr, and
if one criticizes once again the concept of intrinsic determinations
in favor of one of relational (or dependently arising)
determinations. The apparent contradiction that arises in the “cat’s
paradox” indeed derives from the repeated use of the term “state”,
which actually has two quite different meanings. The superposed
quantum “state” of the cat does not fit with the “state” that is
manifest and observable. This apparent conflict disappears as soon
as we recognize that the quantum “state”, far from indicating what
the cat intrinsically is, only makes it possible to estimate the
chances one has of seeing it in a certain way; that far from
corresponding to a “state” in the full and proper sense of the term,
the quantum “state” vector is nothing other than a symbolic

12
This (imaginary) machine comprises a fragment of radioactive matter having one chance in
two of disintegrating over the time of one hour, and a flask of poison which is released when
the disintegration occurs. If the poison is released, it kills the cat.
13
Lyre H. (1999), “Against Measurement? - On the Concept of Information”, In: P. Blanchard
and A. Jadczyk (eds.), Quantum Future: From Volta and Como to Present and Beyond,
Berlin: Springer; Lyre H. (1999), “Against Measurement? - On the Concept of Information”,
In: P. Blanchard and A. Jadczyk (eds.), Quantum Future: From Volta and Como to Present
and Beyond, Berlin: Springer ; Bitbol M. (2009), “Decoherence and the Constitution of
Objectivity” in : M. Bitbol, P. Kerszberg & J. Petitot (eds.), Constituting Objectivity :
Transcendental Perspectives on Modern Physics, Berlin: Springer

7
instrument making it possible to evaluate the probability of finding
the cat in one or other of its two biological states. Indeed, no-one
has ever required that a probabilistic evaluation should reveal in
advance the actual outcome (in the full and proper sense of the
term) of the event in question; in the same way, no-one should
hope to reveal or to engender the actual observed state of the cat
merely on the basis of the quantum probabilities. The only non-
conventional aspect of quantum theory is the peculiar (non-additive
and interferential) structure of its calculation of probabilities,
which is quite different from the classical calculation, because it is
adapted to the contextuality of microscopic phenomena14. The only
remaining problem thus consists of linking up (at least
approximately) this non-classical structure of probabilities with the
classical additive structure which is valid for the mutually
exclusive events observed in the laboratory. This technical (rather
than fundamental) problem is solved, as we have already indicated,
by the theories of decoherence.
A third disturbing finding was formulated for the first time by
Einstein, Podolsky and Rosen15. Even though the aim of these
authors was to demonstrate the “incompleteness” of quantum
mechanics (its incapacity to describe all the “elements of reality”
attached to physical systems), what posterity has retained from
their reflections is quite different16. The enigma which remains
bears on the explanation of the strange “EPR correlations”
predicted by the “entangled states” of quantum mechanics. How is
one to understand the strict correlation between the values of
observables measured on pairs of particles which were initially in
contact but which are now situated at arbitrarily large distances
from each other? Briefly, the only two explanatory frameworks
which are plausible are (a) common causes and (b) reciprocal
causal influence. But one and the other of these two explanatory
possibilities encounters insurmountable obstacles in quantum
physics. Considering that the origin of the correlations lies in
common causes amounts to asserting that they are inscribed in the
properties of the particles, and that these properties were fixed ever

14
Bitbol M. (1998), “La mécanique quantique comme théorie des probabilités généralisée”, in:
E. Klein & Y. Sacquin (eds.), Prévision et probabilités dans les sciences, Paris : Editions
Frontières
15
Einstein A., Podolsky B. & Rosen N. ([1935] 1983) “Can quantum-mechanical description
of reality be considered complete ?”, in : Wheeler J.A. & Zurek W.H. (eds), Quantum Theory
and Measurement, Princeton: Princeton University Press
16
Espagnat B. d’ (1994), Le réel voilé, Paris : Fayard

8
since the initial moment when the particles were contiguous.
However this option (called local hidden variables) is excluded by
Bell’s theorem17. The other hypothesis, that of reciprocal causal
influences with an arbitrarily large speed (including larger than the
speed of light) has indeed been modeled and tested experimentally
in recent years18; but it has been refuted and must therefore be
rejected in its turn. How is it possible to extricate oneself from this
impasse? Two extreme options remain available. On the basis of
the presupposition of « scientific realism » (according to which
quantum theory describes the properties of things as they really are,
including their inseparability), the only way out is to adopt an
ontological holism. According to this doctrine, space and time are
only emergent deployments of an “implicate order”19 which is pre-
spatial and pre-temporal; and the two distant particles are in truth
distinct manifestations of one and the same universal entity. Their
correlation no longer has to be explained by any sort of
transmission, from the past to the present or from a present here to
a present over there, but simply by a statement of identity. At the
opposite extreme, according to the most radical of the anti-realist
options, there is simply no need to “explain” an instantaneous
correlation at a distance, for the good and simple reason that the
latter has no intrinsic existence. The correlation only ever sees the
day relatively to mechanical and electro-magnetic devices apt to
“provide evidence for it”. Now, that can only come about when the
information concerning one of the correlated properties has had a
sufficient time (at least the time that would be taken by a light
signal) to rejoin the region of space where the information
concerning the other property is available20. No “non-local
influence” need be invoked in this case.
The fourth disturbing finding covers in fact a whole network of
clues that point towards a conception of physical theory that is non-
descriptive and non-representational, but rather purely predictive

17
Bell J.S. (1987), Speakable and Unspeakable in Quantum Mechanics, Cambridge :
Cambridge University Press. Bell’s theorem establishes the incompatibility of quantum
mechanics with certain inequalities (the Bell inequalities) which inevitably result from
theories with local hidden variables.
18
Suarez A. (2000), “Quantum mechanics versus multisimultaneity in experiments with
acousto-optic choice devices”, Physics Letters, A269, 293-302
19
Bohm D. (1984), Wholeness and Implicate Order, London: Ark Paperbacks
20
Bitbol M. (1983), “An analysis of the Einstein-Podolsky-Rosen correlations in terms of
events”, Physics Letters 96A, 66-70; Smerlak M. & Rovelli C. (2007), “Relational EPR”,
Foundations of Physics 37, 427-445

9
and informational21. A large number of experiments (some of
which have actually been carried out, others which are pure
thought-experiments) make it pretty much unthinkable that one
could describe processes which are supposed to have happened
before the actual act of their detection or observation; and this
forces one to trust only the information drawn from such an act by
basing oneself on previous knowledge of the configuration of the
experimental set-up as a whole. I will mention just two of these
experiments: “measurements without interaction”, and “delayed-
choice experiments”22. In the first sort of experiment, information
derived from an absence of interaction between the object and an
intermediate part of the instrument have exactly the same
consequences as those that would result from their actual
interaction23. That is enough to make one think that what counts in
a quantum experiment is not the detail of the hypothetical
processes which may be supposed to occur between the preparation
and the final detection, but rather the informational content that the
whole structure of the apparatus confers on the event of detection.
Indeed, in some spectacular experiments with “delayed choice”, the
object interacts with an elementary measuring agent (for example a
photon), but its so-called “state” depends on decisions that can be
made millions of years later concerning the arrangement of the
device which makes it possible to collect the photon. Unless one
imagines that certain influences can go backwards in time (as
certain physicists have been led to propose (Wheeler 1978)), it
must be recognized that what may be carelessly called “the state of
an object” expresses nothing other than the information made
available by the observational apparatus which gives access to it
after the moment when all the decisions concerning the apparatus
have been taken.

2- A non-essentialist, Śūnyatā-like, strategy to make sense of

quantum theory

21
Brukner C. & Zeilinger A. (2009), “Information Invariance and Quantum Probabilities”,
Foundations of Physics 39, 677–689
22
Scully M.O. & Drühl K. (1982), “Quantum eraser: A proposed photon correlation
experiment concerning observation and ‘delayed choice’ in quantum mechanics”, Physical
Review A 25, 2208–2213, 1982; Elitzur A.C., Dolev S., & Zeilinger A. (2003), “Time-
reversed EPR and the choice of histories in quantum mechanics”, Proceedings of XXII Solvay
Conference in Physics, 452-461, Singapore: World Scientific
23
Elitzur A.C. & Vaidman L. (1993), “Quantum mechanical interaction-free measurements”,
Foundations of Physics, 23, 987-997

10
 If we wish to express the lesson of these reflections in a
deliberately provocative way, we might remark that quantum
mechanics is better understood, in a way that avoids posing
logically insoluble problems, by admitting that it reveals rigorously
nothing about the alleged intimate nature of its objects. After all, if
quantum mechanics is considered as a generalized process of
evaluating probabilities, there is no more reason for it to reveal the
nature of its objects than the classical theory of probabilities has of
revealing the nature of objects to which it is commonly applied
such as dice, of roulette tables, of fluctuations in the financial
market, or the clients of an insurance company. Just like the theory
of probabilities, quantum mechanics is grafted onto the outside
layer of events that it aims at anticipating – without penetrating
into a hypothetical “interior”. Even more than the theory of
probabilities, quantum mechanics rests on the surface of things,
because what it anticipates are not even actual events that will
come about by themselves, but merely potential phenomena which
require a particular experimental setup in order to occur (Bohr
wrote that these phenomena are defined by such an experimental
setup). And that is not yet all. Not only does quantum theory reveal
no intimate nature of things beyond the phenomena, but its success
and its fruitfulness are easily explained by the fact that it
incorporates in its very structure the limits to the exploration of
phenomena. Its success and its fruitfulness come from the fact that
it does not even allow any meaning to the belief that there might be
something deeper to understand behind the superficial screen
which is its own domain of validity. Heisenberg’s indeterminacy
relations can thus be considered as the expression of a limit to any
possible knowledge of the dynamic variables of elementary
particles. But these relations are at the same time a powerful tool of
theoretical exploration which has made it possible to predict,
among other things, the bandwidth of rays of electromagnetic
emission, the life-time of radioactive nuclei, and a number of
striking effects of quantum field theory (such as virtual particles,
the Casimir forces, etc.). Here, the limit to knowledge is not a
matter of a provisional obstacle, but determines the very form of
what is to be known. Relaxing the usual struggle towards
representational sense-making turns out to be a good strategy to
make sense of the efficiency of quantum mechanics.
Indeed, not only is quantum mechanics the superficial prediction
of superficial phenomena, but its redoubling of superficiality is

11
what accounts for its remarkable vocation for universality. If
quantum theory is above all a general procedure for anticipating on
a probabilistic mode the replies to experimental solicitations, or
more precisely for anticipating replies which correspond to the type
and the order of these solicitations, then it ought to be generally
applicable to any domain whatever that is solicited. Now this does
indeed turn out to be the case, which reinforces the initial
“deflationist” interpretations. The recent generalization of quantum
theory, which is applicable to many domains in the human
sciences24 running from decision theory to semantics by way of the
psychology of perception, is a remarkable illustration of this. It
does not matter who or what responds (human beings or things),
the probabilistic structure of the responses is the same. From this
restricted point of view, a set of human beings making choices
which depend on the options which are presented to them, and on
the order of the decisions to be taken, behave exactly like a set of
electrons on which one evaluates several incompatible
observables25. A set of speakers who have to decide on the
meaning of a polysemic word, according to the propositional
contexts, thus behaves exactly like a set of microscopic particles
which violate the Bell inequalities26. There is nothing shocking
about the fact that it should be so, and implies strictly nothing
about any community at the level of their profound being between
electrons and humans; there is only a formal isomorphism in their
situation and their “surface” reactions to being solicited.
To recapitulate, the non-essentialist conception of quantum
mechanics as being doubly “superficial” (both superficially
phenomenal and superficially probabilistic) makes it possible to
dissolve away what are alleged to be the major paradoxes of this
theory; to explain a large part of its effectiveness; and to promote
its universality. As if this were not enough, one can add that this
conception also maintains a remarkable degree of notional and
mathematical simplicity, which contrast strongly with the ever-
increasing sophistication of those ideas which aim at saving a
“realist” interpretation of quantum physics. Why, under these

24
Bruza P., Sofge D., Lawless W., van Rijsbergen C.J. & Klusch M. (eds.) (2009a), Quantum
interaction, Berlin : Springer ; Busemeyer J. & Bruza P., Quantum Models of Cognition and
Decision, Cambridge University Press, 2014
25
Zwirn H. (2009), “Formalisme quantique et préférences indéterminées en théorie de la
décision”, in : M. Bitbol (ed.), Théorie quantique et sciences humaines, Paris : CNRS Editions
26
Bruza P.D., Kitto K., Nelson D. & McEvoy C., (2009b), “Is there something quantum-like
about the human mental lexicon?”, Journal of Mathematical Psychology, 53, 362-377

12
conditions, is this conception not more widely accepted? Why does
it so often find itself opposed by the indignant reactions of certain
physicists who reproach it with “betraying the ideal of science”, of
“breaking the great dream of knowledge”27, of being unacceptable
or even “scandalous”28? Why, even when indignation is absent,
does the exposition of the minimalist conception of quantum theory
give rise to a resigned silence which manifestly expresses a
profound disappointment? There is no doubt that it is because, as
we have felt it coming since our introduction, we are dealing with a
breach of several contracts at the level of a whole civilization. One
of these is a fairly recent contract which, from the sixteenth century
onwards, has instigated a collusion between the desire for a
metaphysical breakthrough upheld by the clerks, and the need for
technological perfectionism of the craftsmen29. Another is a very
ancient contract which has made it an obligation to seek a principle
of understanding appearances in the inmost depths of things30. If
scientific progress does not help our gaze to penetrate to the very
heart of material bodies, and to definitively guarantee technological
effectiveness by laying bare their secret, what is the point of it? If
the progress of knowledge amounts merely to a kaleidoscopic
deployment of the phenomenal skin of things, instead of opening
up a vision of their very flesh and marrow, does it not seem in
vain? It is all very well to recall that all the entities which, in the
history of science, were pompously dignified at the time by the title
of “realities behind appearances” have turned out to be themselves
a matter of : (1) other appearances (or phenomena) revealed by a
new approach, postponing the revelation of what Goethe called the
“Urphänomen”31 to an indefinitely remote utopia; or (2)
mathematical idealities which express some invariants of the
phenomena reconstructed by the intelligence. This simple reminder
is not enough. The “dream of reason” pursues its course; this same
dream that Kant upheld at the beginning of his quest, before
discarding it in his critical philosophy: the dream of managing to
grasp by thought a “representation of things as they are”32. The

27
Stengers I. (1997), Cosmopolitiques 4 : Mécanique quantique, la fin du rêve, Paris : La
Découverte
28
Thom R. (1993), Prédire n’est pas expliquer, Paris : Flammarion
29
Scheler M. (1993), Problèmes de sociologie de la connaissance, Paris : Presses
Universitaires de France
30
Schrödinger E. (1954), Nature and the Greeks, Cambridge : Cambridge University Press
31
Seamon D. & Zajonc A. (eds.) (1998), Goethe’s way of science, SUNY Press
32
Kant I. ([1770] 2004), Inaugural Dissertation of 1770, Whitefish : Kessinger Publishing

13
only possibility to suspend this vain quest would be to change our
cultural representations, by opening it to alternative cultural areas.
The Indo-Buddhist cultural area, with its (anti-)concept of Śūnyatā
is one of the best possible options for this sake.
So, let us now compare the modalities and motivations of both
Buddhist and quantum criticisms of metaphysics33. We will start
our further inquiry in section 3, before the era of Madhyamaka for
Buddhism, and before the era of quantum mechanics in physics.
This will make clear that both Madhyamaka and quantum physics
represent achievements in a long process of emancipation with
respect to the naive beliefs of ordinary life. Section 4 will develop
the central parallel between quantum entanglement (or non-
separability) and dependent co-arising. In section 5, a deliberately
deflationary, non-speculative, reading of this parallel will be
presented. Finally, in section 6, a suggestion about the common
root of the Buddhist and quantum critiques of metaphysics will be
made.

3- Historical Development of the Theories of Knowledge in

Buddhism and in Classical Physics

The intellectual history of both physics and Buddhism went

through remarkably similar steps that can be enumerated as
follows :
1. Transition from the paradigm of substance to the paradigm of
causality ;
2. Replacement of productive causality by lawlike successions ;
3. Transition from causal and lawlike successions to co-
emergence ;
4. Criticism of ontological construal of co-emergence, and claim
that co-emergence itself is relative to the cognitive act that
posits it (“co-emergence co-emerges”) ;
5. Silent return to the practices of life, or agnostic return to the
practices of experimental science.
In Western science, the two first steps were taken by classical
physics, whereas the three following steps had to wait for the birth
of quantum physics. This is the reason why this section will be

33
HH the Dalai-Lama, The Universe in a Single Atom, op. cit. p. 67 : “Each of these (physical)
pictures is excellent in its own right and for the purpose for which it has been designed, but if
we believe any of these models to be constituted by intrinsically real things, we are bound to
be disappointed”.

14
devoted to the two first steps, whereas the following steps will be
discussed in the next sections.
The first step from substance to causality, to begin with, was
taken very early by Buddhism. The primary ambition of the
concept of pratîtyasamutpâda was existential ; it was to describe
(across the twelve links of dependent origination) the inexorable
chain of consequences from acts to retribution, or from “ignorance”
(avidya) to aging and death. Yet, it also had an important
epistemological content, since : (i) it involved a strong attack
against the belief that substances and selves are permanent beings,
and (ii) it posited the general rule according to which every
(impermanent) phenomenon arises as an effect of causes and
conditions. At this stage of Buddhist thought, the use of the word
“cause” evoked the concrete and familiar idea of an impelling
force. Indeed, “cause” is the English translation of the Sanskrit
“Hetu”, deriving from the verbal stem “Hi”, which essentially
means “to throw ahead”, “to hurl”, “to set in motion”. So, one
could still figure out that an effect is somehow “produced” by its
causes and conditions, in the sense that causes give an impulsion
that triggers the effect under certain conditions. This elementary
conception of causation is precisely that which came under strong
pressure with Madhyamaka. But even in this elementary form, it
performed a very important epistemological task : it showed that
the apparent regularities we witness in ordinary life can be
explained without assuming that there are enduring and
intrinsically existent “things” or substances. To explain these
regularities, it is enough to assume a strict causal connection
between successive instantaneous events. In other terms, causes
replace substances.
In Western science, a similar turn (though less complete) was
taken during the XVIIth century, with the physics of Galileo and
Descartes. But let me first sketch what was the conception of
science in earlier days, from Aristotle to the middle ages.
According to Aristotle, the aim of science is to show that given
predicates inherently pertain to certain “subjects”, or substances.
This strong connection between each predicate and the
corresponding substance is established through reasoning. In other
terms, it is established through a chain of propositions that starts
from a set of axioms (namely from a set of accepted propositions of
the same type as the one which has to be derived) and stops with
the proposition ascribing the predicate to the relevant substance.

15
Reasonings usually involve two levels of generality. At the upper
level of generality, one establishes that a certain “essential”
predicate is inherent to a certain species. At the lower level, this
predicate is ascribed to an individual belonging to that species.
During the middle ages, Aristotle’s emphasis on substance as the
foundational category of Being, and on inherence of essential
predicates to substances, was taken for granted and many
consequences were (often abusively) drawn from it. One
consequence was the pervasive use of the concept of “substantial
form” : a basic, essential, characteristic of substance that persists
unaltered through change.
René Descartes (the French philosopher, mathematician and
physicist of the first half of the XVIIth century) eagerly criticised
the concept of “substantial form”. He insisted that “Substantial
forms cannot offer us a strong reason for any natural process, since
even their supporters confess that they are occult and that they do
not understand them”34. In other terms, substantial forms (namely
inherent, essential properties of substances) are fake explanations
of phenomena. Instead of them, Descartes advocated mechanical
explanations in which only spatial bulk and velocity are involved.
A typical mechanical explanation then relied on the description of
motion and collisions of extended bodies. Accordingly, Descartes
developed at length the description of how “(…) the bodies that
collide change the state of motion of one another”35. From then on,
phenomena were explained by calculating the sudden impulses that
bodies receive from other bodies that bump into them. Causes, in
the most straightforward sense of thrust or impulsion, were
endowed with the explanatory power which was formerly ascribed
to substances. Substances were still there at this initial stage of
classical physics, but they had no explanatory role by themselves ;
only their relations of mutual contact had one.
This tendency towards more emphasis on causes and less on
substances was amplified by substitution of laws for productive
causes.
In Buddhism, this substitution was rather obvious from the
beginning. It was only amplified to its ultimate consequences by
Madhyamaka, against the residual reifications of Abhidharma. A

34
R. Descartes, Œuvres Philosophiques IV, Hachette, 1835, p. 30
35
R. Descartes, Les principes de la philosophie II, in : Œuvres Philosophiques III, Garnier,
1973, p. 195

16
well-known statement attributed to the Buddha himself, and quoted
in many Sûtras, is the following :

“When this is present, that is present

From the arising of this, that arises
When this is absent, that is absent
On the cessation of this, that ceases”36

In the former sentences, there is no question of production,

generation or alteration ; only of association and correlation. This
restriction to the immanence of (correlated) phenomena, as
opposed to the transcendence of productive powers, has often been
compared to Hume’s empiricist theory of causality. Indeed,
according to Hume, the notions of “power” or “productive cause”
are “obscure” ; experience, he writes in his Treatise of Human
Nature, teach us the usual association of phenomena but not their
generative causes (if any).
Now, Hume was expressing nothing else than the lesson he had
gathered from his knowledge of Newtonian physics. True, at first
sight, Newton’s theory of gravitation represented a metaphysical
regression when compared to Descartes, since it appeared to use
the notion of an “occult quality” called the “attractive power” of
massive bodies. No wonder that many French Cartesian
philosophers of the late seventeenth century and early eighteenth
century eagerly criticized Newton on this ground. But in fact,
Newton carefully avoided to rely on speculative claims ; even
though he had developed a very elaborate metaphysics of space and
material bodies during his early career, he was eager to make his
mathematical physics virtually independent of it. According to him,
the really important thing in mechanics is the set of mathematical
laws that connect and unify (astronomical and earthly) phenomena.
Throughout his Philosophiae Naturalis Principia Mathematica, he
insisted that when he used words like “attraction” or “impulse”, he
did not wish to assert something about “the species of forces and
their physical qualities, but their quantities and mathematical
proportions”37. As for the mathematical laws themselves, they were
considered by Newton as little more than an expression of the

36
Samyutta Nikâya, II, p. 10, quoted by: Encyclopaedia of Buddhism, IV, 1, p. 2, Sri Lanka
Government Printing, 1979
37
I. Newton, The Principia (Mathematical Principles of Natural Philosophy), University of
California Press, 1999, Book I, section 11, Scholium, p. 588

17
observed regularities of successive phenomena. In his fourth “rule
for the study of natural philosophy” Newton thus pointed out that
“In experimental philosophy, propositions gathered from
phenomena by induction should be considered either exactly or
very nearly true (…)”38. At the end of the eighteenth century (a few
decades after Hume), Kant corrected this claim by pointing out
that, actually, Newton’s mathematical laws derive as much from
the constructive power of our understanding as from pure
experience. But this remark does not need to be developed any
further in this context.
Our central concern is only to notice the shift of the concept of
cause in physics, from “productive power” to mere regular
succession. Indeed, this shift has crucial consequences, such as
(here again) additional weakening of the substantialist view of the
world. Moritz Schlick, a philosopher of the Vienna Circle at the
beginning of the twentieth century, expressed these consequences
as clearly as possible. According to him, the concept of substance
(which expresses permanence of certain feature of phenomena) is
superfluous as soon as universal mathematical laws of regular
succession are formulated : “Each time one evokes substance in
physics, one asserts nothing else than the existence of a specific
regularity”39.

4- Quantum Non-Separability and Pratîtyasamutpâda

We have just seen how classical physics progressively

underrated the concept of substance (which only subsisted in the
mathematical and non-individual form of the principles of
conservation of mass, energy, charge etc.), and simultaneously
abstracted a concept of regularity law from the ordinary concept of
causality. This closely parallels the Buddhist critique of the
concepts of substance and productive causality.
However, nothing prevented a classical physicist from speaking
(in ordinary language) as if the concept of individual substance
were still in use, and nothing prevented him/her from providing
partial explanations as if the concept of productive causality had
not been replaced with law-like regular successions. By contrast,
quantum physics brought to radical completion the critique of the

38
I. Newton, The Principia (Mathematical Principles of Natural Philosophy), p. 796
39
M. Schlick, Philosophical papers II (1925-1936), Reidel, 1979, p. 26

18
concepts of individual substance and causality ; so much so that it
became utterly artificial to use them, even for the sake of verbal
communication. This achievement of the critique here again
closely parallels similar steps in Buddhist thought, as we will now
see.
One crucial point of the quantum paradigm is the strict
correlativity of attributes to one another and of entities to one
another, so that the terms of a certain type of relation – let’s call
them its relata – cannot even be ascribed an existence
independently of the relation that connects them (step 3 of our list
in the previous section). But before we document this striking
feature of quantum physics, it is worth stating its Buddhist
analogue straightaway.
According to several early Buddhist texts, the conditioning
which unfolds through the twelve links of dependent origination is
not necessarily one-way ; it is likely to be mutual : “The
conditioning relation can be reciprocal” 40. One argument in favor
of this reciprocity is the cyclic character of the chain of dependent
origination. But the best way to get a flavour of what is at stake in
mutual conditioning is to read the beautiful metaphor which
illustrates it : “(…) It is just as if there stood two sheaves of reeds
leaning one against the other (…) If I were to pull towards me one
of those sheaves of reeds, the other would fall ; if I were to pull
towards me the other, the former would fall”41. Here, the property
“leaning against” of one sheave of reeds has no intrinsic reality, but
only a relational reality : it co-arises with the corresponding
property of the other sheave of reeds. Later on, Nâgârjuna and his
successors of the Madhyamaka-Prasangika school made this
identification of pratîtyasamutpâda with mutual co-arising even
more systematic. They insisted on the equivalence of dependent
origination and Śūnyatā (or lack of own-being), which was already
stated in the prajñapâramitâ-sûtra. They also indicated indirectly,
through the etymology of several quasi-synonyms of
pratîtyasamutpâda, that dependent origination is tantamount to a
mode of being (or rather to an intermediate status between being
and not-being), not to a mere causal connection between pre-
existent beings. One example is the word idampratyayamâtra,

40
Katthâvatthu, XV, 2, quoted by J. Macy, Mutual Causality in Buddhism and General
System Theory, SUNY Press, 1991, p. 57
41
Samyutta-Nikâya, II, 114, quoted by J. Macy, Mutual Causality in Buddhism and General
System Theory, op. cit. p. 55

19
which means literally “the measure (or extent) of going-with-this”.
This word is used by Candrakîrti in the following sentence :
“things exist to the extent of their going-with-this”42. In other
terms, things have no other form of being than their going-with
other things, or their being-related to other things. Other examples
are the terms paratantra (literally “woven-of-the-other” as opposed
to svatantra, “woven of itself”), and paraparasiddha (literally
“established by one another”). Both words are used by Nâgârjuna
to characterize dependent origination, and ultimately Śūnyatā43.
One can understand provisionally these words as saying that
interdependent “beings” are replaced by interbeing ; or that the
mode of being of interdependent “beings” is nothing beyond
interdependence itself.
Co-arising or co-relativity of properties and entities is also one
of the most striking characteristics of quantum mechanics. It was
taken as a central theme around 1935, in papers of Schrödinger44
and Einstein, Podolsky and Rosen45. Schrödinger’s characterization
of what he called the “entanglement” of the states of the micro-
physical systems runs thus : “When two systems, of which we
know the states by their respective (state vectors), enter into
temporary interaction due to known forces between them, and
when after a time of mutual influence the systems separate again,
then they can no longer be described in the same way as before,
viz. by endowing each of them with a (state vector) of its own. I
would not call that one but rather the characteristic trait of quantum
mechanics, the one that enforces its entire departure from classical
lines of thought”46. In other terms, after an interaction has taken
place between them, the two systems can no longer be ascribed two
separate states (they are non-separable) ; their quantum states are
“entangled” insofar as one can only define a single global state by
combining them, and ascribe this global state to the whole.

42
Candrakîrti, Madhyamakâvatâra, Trad. L. de la Vallée-Poussin, Le Muséon, 1911, p. 277-
278, quoted by L. Viévard, Vacuité et compassion dans le bouddhisme madhyamaka, De
Boccard, 2002, p. 44
43
Nâgârjuna, Acintyastava (42-43), in C. Lindtner, Master of Wisdom, op. cit. p. 27
44
E. Schrödinger, “Discussion of probability relations between separated systems”, Proc.
Cambridge Phil. Soc., 31, 555-563, 1935
45
A. Einstein, B. Podolsky and N. Rosen, “Can quantum-mechanical description of physical
reality be considered complete?” Physical Review, 47, 777-780, 1935. About the relation
between the EPR argument and dependent origination, see HH the Dalai-Lama, The Universe
in a Single Atom, op. cit. p. 64.
46
E. Schrödinger, “Discussion of probability relations between separated systems”, loc. cit.;
see a comment in: H.J. Treder and H.H. von Borzeszkowski, “Interference and interaction in
Schrödinger's wave mechanics”, Foundations of Physics, 18, 77-93, 1988

20
Schrödinger summarized this situation by pointing out that “the
best possible knowledge of a whole does not necessarily include
the best knowledge of all its parts, even though they may be
entirely separated”47. Accordingly, there has been a widespread
temptation to express quantum entanglement in terms of a doctrine
of holism (inanalysability of a whole into its parts).
But a more immediate and less speculative empirical
consequence of quantum entanglement is the strict correlation of
certain observable properties measured on entangled sub-systems.
This correlation is so strong that it can only be interpreted in terms
of co-relativity of these properties. Let me give an example. A pair
of spin-1/2 particles is prepared in a global state corresponding to
an overall z-component of spin equal to 0. The individual states of
these two particles are then entangled. In this case, whenever one
finds that the z-component of spin of one particle is +1/2, a
measurement of the z-component of spin of the other particle is
bound to yield the value -1/2. Conversely, whenever one finds that
the z-component of spin of one particle is -1/2, a measurement of
the z-component of spin of the other particle is bound to yield the
value +1/2. The crucial point is that, before any measurement has
been performed, there is no sense in ascribing a property (or a
value of the z-component of spin) to each particle taken in
isolation. This means that the property “z-component of spin” has
no existence of its own in each particle, but only relative to the
corresponding property in the other particle. These properties of
microscopic particles clearly behave as the property “leaning
towards” of sheaves of reeds, used to illustrate the reciprocity of
pratîtyasamutpâda.
David Mermin built a complete interpretation of quantum
mechanics on the basis of this remark. According to him,
“Correlations have physical reality; that which they correlate does
not”48. In other terms, in quantum physics, there exist correlations
between properties, whereas the correlated properties have no
existence of their own. However, the claim that “correlations exist”
in the world out there, or, as Mermin writes, that “correlations are
fundamental, irreducible and objective”49, is itself very dubious.

47
E. Schrödinger, “Discussion of probability relations between separated systems”, loc. cit.
48
N.D. Mermin, “What is Quantum Mechanics Trying to Tell Us ?”, American Journal of
Physics, 66, 753-767, 1998
49
N.D. Mermin, “The Ithaca Interpretation of Quantum Mechanics”, Pramana, 51, 549-565,
1998

21
This will be shown below, when we come to the fourth step of our
parallel between dependent origination and quantum entanglement.
But before we criticize Mermin’s relational realist claim, we
must amplify the scope of quantum entanglement by showing that
not only properties but also entities are made relational by it. Let
us consider again a pair of spin-1/2 particles with entangled spin-
states. In addition, let us suppose that these particles are
indiscernible, namely that all the properties they have apart from
spin are identical. If the two particles are, say, electrons, they have
exactly the same mass and charge. Moreover, both can be in a
spatial state for which the distribution of probability of their
position in space is exactly the same. One can then distinguish
them neither by their masses and charges nor by their pre-
measurement position ; and the list of criteria of distinction stops
there. In virtue of Leibniz’s Principle of Identity of Indiscernibles,
one is tempted to declare that since there is no way to distinguish
these electrons, they in fact constitute one and the same object. Yet,
physicists still say that they are two electrons, not one object. How
can we justify their way of speaking ? By invoking Quine’s
principle of “weak discernibility”50 : two objects are weakly
discernible if they stand in an irreflexive relation, namely a relation
that each object cannot have with itself.
A well-known illustration is provided by imagining a universe
made of only two metallic spheres. The spheres have all their
properties in common, which, in view of Leibniz’s Principle of
Identity of Indiscernibles, threatens our initial claim that there are
two of them. However, the spheres also stand to each other in the
relation “being at a distance of 10 meters from one another”. This
relation is clearly irreflexive because no sphere can be at a distance
of 10 meters from itself. It is then in virtue of such an irreflexive
relation that the spheres can and should be said to be two, not one.
Another, more relevant, illustration is provided by our two
indiscernible but entangled electrons. They can be distinguished by
no intrinsic property, not even by the putative property “having a
given value of the z-component of the spin”, since, as we have
seen, no such property can be ascribed to one electron taken in
isolation. However, they can be said to be two entities, not one,
because they stand to each other in the relation “having an opposite

50
Quine, W.V.O., “Grades of Discriminability”, Journal of Philosophy, 73, 113–16, 1976 ;
Saunders, S., “Physics and Leibniz’s principles”, In Symmetries in Physics: New Reflections,
ed. K. Brading and E. Castellani, Cambridge University Press, 2003

22
value of the z-component of their spin”. This relation is irreflexive,
once again, because no electron can have an opposite value of the
z-component of its own spin. As a consequence, not only the spin-
component but also the identity of electrons with entangled
quantum states is relational rather than intrinsic. Even their
existence qua distinct entities is relational rather than intrinsic! In
the same way as the mode of being of interdependent “beings” is
nothing beyond interdependence, here the mode of being of
entangled indiscernible particles is nothing beyond the irreflexive
relation that arises from their entanglement.
Now, some people can still think that the co-arising, co-
emergence, or strict co-relativity which characterizes both
dependent origination and quantum entanglement, is underpinned
by causal influences in which the effect comes later than the cause
(even though the delay between cause and effect might be
imperceptibly small). Is it so ? Interestingly, the answer to this
question seems to be entirely negative in the two domains we are
comparing.
To begin with, Nâgârjuna eagerly criticized the common sense
features of the concept of causality, and especially the idea that a
cause exists before its effect. Here is one of his sharpest remarks
about this point : “A cause has an effect when there is an effect, but
when there is no (effect) the cause amounts to no cause”51. Several
commentaries, based on the well-known example of the seed and
the sprout (the seed which has disappeared when the sprout
appears, and the sprout which does not exist when the seed
exists)52, have developed the meaning of this concentrated
sentence. In short, there is no point calling something a cause
before the effect in dependence of which it arises has manifested
itself. But there is no cause either after the effect has manifested
itself, since, if it has finished its work, or if it has disappeared in
the process, there is no point calling it a cause at that moment. The
basis of the argument is that no present experience can hold
together the cause and the subsequent effect (except indirectly, by
means of memory and interpretation, or by means of mental
projection). In addition to this, the very idea of “producing”
something, of having an effect “produced” by a cause, looks
incorrect or artificial when what arises is empty of own-being :

51
Nâgârjuna, Sûnyatasaptati (6), in: C. Lindtner, Master of Wisdom, op. cit. p. 97
52
Candrakîrti, Madhyamakâvatâra, 6, 17, in : C.N. Huntington & Geshé Namgyal Wangchen,
The Emptiness of Emptiness, University of Hawai Press, 1989, p. 159

23
“When neither existents nor non-existents, nor existent non-
existents are established, how could one propose a ‘productive
cause’”53 ? Co-relativity and interbeing are not only distinct from
causation developing in time and production of being ; they make
them pointless.
In quantum physics, the conclusion turns out to be very similar.
A group of physicists of the university of Geneva (Switzerland)
recently wondered whether the correlations that manifest between
micro-systems with entangled states, can be ascribed to faster-than-
light causal influences54. They pointed out that a necessary
condition for such causal influences to make sense would be that
causes occur earlier than their effects. Accordingly, they arranged
their device in order to test the time-ordering of events by relying
on relativistic effects. And they found that the outcome of the
experiment is incompatible with the very idea of causal influences.
There are correlations, but nothing can be said to cause them. This
result shook the initial conviction of the authors of the experiment,
and prompted them, here again, to adopt a holistic metaphysics :
according to them, the experiment proves that the world is an
inseparable whole. But, as we will now see, other (non-
metaphysical) approaches of this perfectly reciprocal and non-
causal kind of co-relativity are available.

5- Quantum Non-Separability, Pratîtyasamutpâda and

Śūnyatā : a deflationary perspective

A well-documented temptation at this stage is to claim that one

has elucidated the nature of the world. In a Buddhist perspective,
the temptation would amount to declare that the ultimate nature of
reality is emptiness, that the world is a process involving a
meshwork of dependently arisen phenomena. In physics, the
temptation consists in saying that the world is an inseparable whole
existing below the level of space and time, or that reality is pure
structure, or that the relations are real while the relata are not. In
other terms, in both cases one tries to elaborate a new metaphysics

53
Nâgârjuna, Mûlamadhyamakakarikâ, I, 7, in J. Garfield, The Fundamental Wisdom of the
Middle Way, Oxford University Press, 1995
54
A. Suarez, « Quantum mechanics versus multisimultaneity in experiments with acousto-
optic choice-devices », Physics Letters, A269, 293-302, 2000; A. Stefanov, H. Zbinden, N.
Gisin, and A. Suarez, « Quantum entanglement with acousto-optic modulators: 2-photon
beatings and Bell experiments with moving beamsplitters », Physical Review, A67, 042115,
2003

24
by reifying the very tool used for the criticism of old metaphysics.
But in both cases also, this attempt fails.
The Madhyamika school, to begin with, is very careful to
criticize in advance any reification of emptiness or dependent
origination, which can be so tempting as a replacement of the
reification of entities. According to Nâgârjuna, “If there is no
essence (svabhâva), there is no other-essence (parabhâva)”55.
According to Stcherbatsky’s translation56, which has pedagogical
virtues, this means that if self-existence is lacking, relational
existence is also lacking. In other terms, the idea of relational
existence is relative to its dialectical opposite, namely to the
(conceptual) possibility of absolute existence. Interdependance
(Pratîtyasamutpâda) is no absolute either, but only one term in a
dual system of oppositions (here, the opposition with
independence). Interdependance itself is dependently arisen,
emptiness itself is empty (Śūnyatā is Śūnya).
In addition to that, any system of oppositions underpinning
dependent origination is in turn relative to a certain conceptual
scheme, or to a certain pattern of discriminative criteria. A standard
example, discussed by Candrakîrti57, is the couple of cart and
wheel. Cart is relative to wheels because without wheels it would
not be a cart but only (say) a wooden box ; wheels are relative to
cart, because without a cart they would not be wheels but only
(say) wooden circles or wooden hoops. But in addition to that, both
carts and wheels are relative to their corresponding concepts and to
the conceptual framework which incorporates them. For, if it were
not for their concept, and the standard use of them which is made
according to this concept, they would just “have” thusness
(tathāta).
To recapitulate, it is clear that, in virtue of its universal
applicability, Śūnyatā is no self-existent “thing” to be substituted to
ordinary things : “If there were to be something non-empty, there
would then be something called empty. However, there is nothing
that is non-empty. How could there be something empty ?”58.

55
Nâgârjuna, Mûlamadhyamakakarikâ, I, 3, in J. Garfield, The Fundamental Wisdom of the
Middle Way, op. cit.
56
T. Stcherbatsky, The Conception of Buddhist Nirvâna, Academy of Science of the USSR,
1927
57
Candrakîrti, Prasannapâda, quoted by: C.N. Huntington & Geshé Namgyal Wangchen, The
Emptiness of Emptiness, University of Hawai Press, 1989, p. 50
58
Nâgârjuna, Mûlamadhyamakakarikâ, XIII, 7. Translated by D. Kalupahana, Motilal
Banarsidass, 1996

25
 A similar series of criticisms of relational being has recently
developed in quantum physics. The major target of the criticism is
Mermin’s claim that “correlations have physical reality” or that
“correlations are fundamental, irreducible and objective”. Mermin
explicitly considers that correlation are local element of reality,
namely that they are not influenced by distant correlations. In the
same way, properties of particles were taken by Einstein as local
elements of reality, not influenced by distant correlations. But we
know that the very assumption that particles are endowed with
properties construed as local elements of reality is inacceptable in
virtue of the following reasoning :

a. Such properties are connected by Bell’s inequalities

b. Bell’s inequalities are experimentally violated
c. Therefore, local realism is falsified

A. Cabello59 then wondered whether Mermin’s relational local

elements of reality can really replace Einstein’s local properties. He
wondered, in other terms, whether local realism could still be
defended provided the predicate of “local reality” is ascribed to
relations rather than monadic properties. But the outcome of his
investigation was definitely negative. Cabello found that no
quantum correlation can be isolated from the overall network of
quantum correlations that constitute a physical system, irrespective
of the spatial distance between the various parts of the system.
Relational local realism thus fares no better that property local
realism. Later on, as we have seen earlier, Smerlak and Rovelli60
gave strong arguments in favor of interpreting this failure of
relational local realism not as a failure of locality, but as a failure
of realism. According to them, no feature of physical systems, not
even their correlation encoded in quantum entanglement, has
intrinsic reality. Any such quantum feature is relative to a certain
sequence of experimental procedures worked out by specific
observers. In other terms, even the relations (between quantum
subsystems) are relative (to experimental conditions). Just as, in
Madhyamaka, emptiness of own-being is itself empty of own-being,
in quantum physics relational being is itself relative.

59
A. Cabello, « Quantum correlations are not local elements of reality », loc. cit. », Physical
Review, A59, 113-115, 1999; A. Cabello, « Quantum correlations are not contained in the
initial state», Physical Review, A 60, 877-880, 1999
60
M. Smerlak & C. Rovelli, « Relational EPR », Foundations of Physics, 37, 427-445, 2007

26
 This all-pervasive ontological criticism is based on an equally
strong criticism of the common sense concept of time and
succession, both in Buddhism and in quantum physics. After all,
time, substance and causality are deeply related notions. Substance
can be defined as an enduring being, one which does not change in
time when (its) attributes change. Conversely, Kant61 argued that
we can perceive time determinations only by reference to (and by
contrast with) something permanent that we call substance. Similar
conclusions hold for causality. On the one hand, causality relies on
succession of events considered as causes and effects. On the other
hand, it has been proposed, in view of the theory of Special
Relativity, to base the concept of succession on causal
connections62. So, we can easily understand that any attack on the
ordinary concept of time is bound to have a devastating effect on
the concepts of substance and causality.
In Nâgarjuna’s work, the criticism of the ordinary concept of
time and motion develops in chapters II, VII and XIX of
Mûlamadhyamakakârikâ. There, we find roughly three kinds of
arguments against the intrinsic reality of time :
(i) It is impossible to detach time and motion from their
substrate (if any) : The motion, the mover, and the path cannot be
separated from one another except verbally and for practical
purposes. The mover would not exist without motion, and motion
does not exist without a mover. The motion and the mover cannot
be separated, even though they cannot be said to be identical either.
This kind of argument has been compared with similar ideas of the
Greek Megaric philosophers, according to whom the mover cannot
be separated from ongoing motion63.
(ii) Motion cannot be defined at a single instant or in a single
point. Indeed, one cannot find a point of space where motion
begins; and one cannot even find a point of space where it merely
occurs. Nâgârjuna insists on this problem in several parts of his
treatise64. He also criticizes the very idea of a beginning, by
pointing out that a beginning must itself begin65 and so on ad
infinitum. His arguments are remarkably akin to those of the Greek
philosopher Zeno of Elea.

61
I. Kant, Critique of Pure Reason, B278, Hackett, 1996, p. 291
62
H. Mehlberg, Time, Causality, and the Quantum Theory I, Reidel, 1980
63
Aristotle, Metaphysics, 1046b 29
64
Nâgârjuna, Mûlamadhyamakakarikâ, II, 14-15, in J. Garfield, The Fundamental Wisdom of
the Middle Way, op. cit.
65
Nâgârjuna, Mûlamadhyamakakarikâ, VII, 13, 17-20.

27
 (iii) Temporal projections going beyond the lived present (to past
or future) have no reality. This being granted, the very notion of
motion is threatened : “What has been moved is not moving, what
has not (yet) been moved is not moving”66. Here, a partial Western
equivalent is St Augustine67, according to whom time does not
exist because: the past is no longer, the future is not yet, and the
present has no other definition than being a separation between past
and future.
In view of this thorough critique of time, there are even more
reasons to understand dependent origination as co-arising rather
than as a true process of “production” in time.
A similarly articulated criticism of time, succession, and
standard causality has been offered as a consequence of quantum
physics. The argument was first developed in an often-discussed
paper of D. Page et W. Wootters68, and then amplified in
cosmological context by J. Barbour69. Their idea is that (a)
systematic inventory of simultaneous states can replace the study
of the succession of states, and (b) in quantum mechanics
simultaneity is expressed by the entanglement of states of systems
with the state of a standard called “the clock”. The most striking
result of these authors is that the equation of evolution of states
(e.g. the Schrödinger equation in standard quantum mechanics) can
easily be derived by calculating the expectation value of any
observable for a given clock-state. In other terms, careful
consideration of the quantum equivalent of simultaneity replaces
the concept of evolution in time.
To sum up :
• In quantum physics, time appears as a clumsy statement of a
network of relations of simultaneous co-arising of potential values
of observables (as expressed by the entanglement of states) ;
• In the framework of Rovelli’s relational interpetation of
quantum mechanics70, this simultaneous co-arising of potential
values holds relative to the present project of prediction of an
observer.
66
Nâgârjuna, Mûlamadhyamakakarikâ, II, 1.
67
St Augustine, Confessions, Oxford University Press, 1998
68
D. N. Page and W. K. Wootters, « Evolution without evolution: dynamics described by
stationary observables », Physical Review, D 27, 2885-2892, 1983; also, D. Deutsch, « Three
experimental implications of the Everett interpretation » in: Quantum concepts in time and
space, eds R. Penrose and I.J. Isham, Oxford University Press, Oxford, 1986.
69
J. Barbour, The End of Time, Phoenix Paperbacks, 2000
70
C. Rovelli, “Relational quantum mechanics”, International Journal of Theoretical Physics,
35, 1637-1657, 1996; M. Smerlak & C. Rovelli, “Relational EPR”, op. cit.

28
 Here again, time and succession recede to the background
whereas simulaneity, presence, and co-arising tend to replace them.

At this point, we are ready to tackle a recurring argument which

seems to undermine the whole parallel we have just established,
since it tends to display a major discrepancy between Madhyamaka
deconstructive philosophy of Śūnyatā and the quantum
deconstruction of ontology. This argument is nothing less than
quantum indeterminism, interpreted as meaning that certain
physical events occur with no cause at all. Now, this possibility of
occurring with no cause is explicitly excluded, together with
several other possibilities, by the very first kârikâ of Nâgârjuna’s
treatise. Is there any possibility of reconciling these apparently
conflicting statements ? I believe there is, provided we keep in
mind that the ordinary concept of causality is under pressure not
only in quantum mechanics but also in Madhyamaka71. We have
seen that the central teaching of Madhyamaka is dependent co-
arising, itself dependently co-arisen. Ordinary (productive,
successive) causality is then criticized in view of the implications
of dependent co-arising. Similarly, the central features of quantum
physics are : (a) entanglement, (b) relativity of entangled states to
present projects of prediction of observers, and (c) relativity of the
values of observables to the actual workings of experimental
devices. A very remarkable fact about quantum physics is that the
failure of ordinary causality can be shown to be a consequence of
feature (c). Indeed, it was demonstrated by P. Destouches-Février
that any predictive theory bearing on phenomena defined relative
to possibly incompatible experimental contexts is “essentially
indeterminist.”72 To sum up, in quantum physics, the invalidity of
ordinary productive causality follows from the relativity (or
context-dependence) of phenomena.

6- The common root of quantum entanglement and

dependent origination : deep cognitive interdependence!

We can now attempt an explanation of the former analogy

between quantum entanglement and Buddhist Pratîtyasamutpâda.
71
HH the Dalai-Lama, The Universe in a Single Atom, op. cit. p. 68
72
P. Destouches-Février, La structure des théories physiques, Presses Universitaires de
France, 1951, p. 260-280 ; see M. Bitbol, “A Cure for Metaphysical Illusions: Kant, Quantum
Mechanics and Madhyamaka”, in: B.A. Wallace (ed.), Buddhism and Science, Columbia
University Press, 2003

29
Once again, it must be clear that what we have to explain is not the
similarity of two views of the world obtained with two different
modes of access (experimental and experiential, technological and
contemplative, external and internal). What we must understand is
the reason why two radical deconstructions of metaphysical views
turn out to be so similar ; we must understand what motivated the
strong convergence of these two deconstructions, and of these two
criticisms.
One simple and general way to formulate the reason why the
Buddhist and the Quantum deconstructions of metaphysics are so
similar is to notice the similarity of the cognitive situations they
have to tackle. In both cases, in the contemplative inquiry of
Buddhism and in the experimental inquiry of quantum physics, the
knower and the known cannot be entirely separated. In
contemplative practice this is most obvious, because the alleged
“object” of the study (the workings of one’s own mind) is by not
truly different from its alleged “subject” ; so much so that any form
of dualism is doomed to failure from the outset. In quantum
physics, this is equally clear, in view of the unsuperable
contextuality of properties and entities. Another, more concrete,
expression of the same point was offered by David Finkelstein73.
This author suggested that the similarities of Buddhism and
quantum physics might well be due to the hyper-sensitivity of their
respective domains of investigation to cognitive probing. When we
observe a thought, it disappears, and when we observe a micro-
state it is “reduced”, “collapsed” or “projected”. In both cases,
events cannot be dissociated from the (mental or experimental)
circumstances of their occurrence. Since these circumstances are
highly variable and are not entirely under conscious or technical
control, an irreducible component of change is involved :
impermanence in Buddhism, and sudden “collapses” in quantum
physics.
But with this irreducible component of change, the central
human method for identifying properties and entities is at least
partly challenged. Indeed, this central method, first described by
Kant and universally applied in physics, is the quest for invariants ;
it is the quest for features that do not vary from one situation to
another, from one circumstance to another. In view of the
irreducible component of change we have just mentioned, the

73
D.R. Finkelstein, “Emptiness and Relativity”, in : A. Wallace, Buddhism and Science,
Columbia University Press, 2003

30
search for invariants must either be renounced to, or deflected at a
level where changes can be neglected for all practical purposes.
In Buddhism, this strategy of deflecting the search for
(approximative) invariants at a coarse level of analysis is well-
known. Madhyamaka authors accept that we can speak of
permanent material bodies, ascribe properties to these bodies, and
consider one body as the productive cause of an alteration of the
properties of another body, as part of the conventional truth useful
in everyday life. But when one pays exquisite attention to the flux
of appearances, and to the subtle variety of phenomena, it becomes
obvious that this kind of truth is indeed only conventional, and that
it is based on an approximation.

Summary and conclusion :

We have just confirmed that there is indeed a thorough and

detailed structural analogy between quantum contextuality and
Buddhist Śūnyatā (emptiness of own-being) ; and between quantum
entanglement and Buddhist Pratîtyasamutpâda (dependent
origination). But this twofold analogy does not show that there is a
single essence of reality which can be disclosed by our reason
irrespective of whether we rely on an experimental approach or a
contemplative approach. The analogy rather shows that, at a
sufficiently accurate level of analysis and careful attention, the
negative conclusions of experimental physicists and contemplative
Buddhists are bound to be similar : both realize the approximative
status of the reified entities of everyday life, and both must cope
with the high amount of instability and lack of self-sufficient
existence (Śūnyatā) of phenomena. Then, if thoroughly applied, the
critical concept of Śūnyatā can be seen to underpin the anti-
metaphysical stances of both Buddhism and certain trends in the
contemporary philosophy of science.

31