## Does God Play Dice?

Interview by Richard Marshall.

‘*The core of quantum mechanics is unsatisfactory in many aspects. First of all, although it can calculate and predict the probabilities of measurement results, it does not tell us how these results are generated. In fact, if assuming the wave function of an electron is a complete description of the electron, and the time evolution of the wave function is always governed by the linear Schrödinger equation, then the theory cannot account for the appearance of a measurement result as a matter of fact. This is the notorious measurement problem of quantum mechanics.*‘

‘*Bohr did believe that atoms are real, but he would not attribute intrinsic and measurement-independent state properties to atomic objects. He did not regard the wave function of an electron as a description of something physically real either.*‘

‘*Feynman is a well-known example. He once said to his students in the 1960s, “Do not keep saying to yourself, if you can possibly avoid it, But how can it be like that? because you will get down the drain, into a blind alley from which nobody has yet escaped. Nobody knows how it can be like that.” This was also a deep puzzle for me when I was an undergraduate. We live in a classical environment after all.*‘

‘*If each wavepacket of the electron is massive and charged, then why do they have no gravitational and electromagnetic interactions with each other? There is only one answer. It is that these mass and charge distributions are effective, formed by the motion of a particle with the mass and charge of the electron which moves throughout the two separated regions very quickly. Since a particle being at an instant has no interactions with itself being at another instant, the two separated wavepackets will have no interactions with each other.*‘

Shan Gao is Professor of Philosophy at the Research Center for Philosophy of Science and Technology, Shanxi University, and Visiting Professor at the University of Chinese Academy of Sciences. He is the founder and managing editor of the International Journal of Quantum Foundations, and is the author of several books including the recent monograph *The Meaning of the Wave Function: In Search of the Ontology of Quantum Mechanics* (CUP, 2017). He is the editor of *Protective Measurement and Quantum Reality: Towards a New Understanding of Quantum Mechanics* (CUP, 2015) and co-editor of *Quantum Nonlocality and Reality: 50 Years of Bell’s Theorem* (CUP, 2016). His research focuses on the philosophy of physics, especially the foundations of quantum mechanics. He also has interests in the philosophy of mind and the philosophy of science. Here he discusses quantum mechanics, what makes it unsatisfactory in many respects, problems of the wave function, continuous vs non-continuous motion, his ontic view of the wave function, protective measurements, the Schrödinger equation, why he thinks God does play dice with the universe, whether causation is an illusion, quantum mechanics and relativism, why he thinks Everett’s many worlds interpretation is incompatible with his own approach, consciousness and quantum mechanics, and why Newton was right when he said reality was particles.

**3:AM:** What made you become a philosopher?

**Shan Gao :** Well, this is also the question my daughter asked me when she began to choose her future career. I think, in essence, it is the beauty and mystery of the universe. They aroused in my heart something like the “cosmic religious feeling” (to quote Einstein), and have been motivating me to unveil the mystery by “thinking on it continually” (to quote Newton).

Every child is born with a curiosity about the world. During my childhood, it had been a wonder for me that the twinkling stars strewed in the night sky don’t fall to the Earth. I had a strong desire to know the whys and wherefores. When I was an undergraduate, I began to read some Chinese popular science books and English academic books about the philosophy of quantum mechanics, although I did not understand what they said so much. I was then entranced by the deep mysteries of the atomic world. I was especially stunned by the fact that the commonsense planetary picture of atoms turns out to be utterly false; the electron in an atom cannot rotate round the atomic nucleus as the Earth rotates round the sun, or else it would soon radiate its energy and fall into the nucleus, and as a result, my body composed of atoms would collapse in a blink.

How does the electron move then? It must exist in the atom. It must move in some way there. But more surprisingly, the textbooks of quantum mechanics provided no picture of the motion of the electron, and in fact, they did not tell us what electrons are at all. Moreover, I knew that the two great founders of the theory, Niels Bohr and Albert Einstein, did not agree on its meaning either. I was determined to search for the answer by myself. Then I started on a lonely journey to “trace” the elusive electron, until now. I simply want to know the answer of a naive question. I simply think on it continually. But the exploration has completely changed my life. It shapes my way through the world and finally makes me become a philosopher, or more accurately, a natural philosopher, who aims at understanding the mysterious universe. A more detailed story about this journey can be found in my little book *God Does Play Dice with the Universe* (2008).

For me, doing philosophy is not only a profession, but also, and more importantly, a way of life. Life is transitory. Everybody is a mere mote in the universe. Yet God, the universe, gives us minds, and thus we can know and understand His thoughts. The most happiness is not beyond this. As the great Chinese sage Confucius taught us in *The Analects*, “Hear the Tao in the morning, and it would be all right to die that evening.”

**3:AM:** Your philosophical preoccupations centre on particular problems of physics. A key component of the philosophical questions arising from contemporary physics is the wave function. We need to understand it in order to understand quantum mechanics because quantum mechanics is a non-relativistic theory about the wave function and its evolution. But the way physicists seem to talk about the wave function is more like they’re talking about Hamlet or a poem – there is no physically real Hamlet in Shakespeare but just interpretations living in an abstract theatrical space. Is quantum mechanics like that – particles and waves are not physical real but abstracts living in Hilbert space contacting our physical world in measurements of some kind?

**SG:** Let me first make clear what quantum mechanics is. I assume most readers of this magazine are not specialists in this field. So, I will still discuss the electron that had puzzled me when I was an undergraduate. Quantum mechanics assigns a wave function or quantum state to an electron, and specifies how the wave function evolves with time under various circumstances, e.g. passing through the two slits in the double slit experiment, using the famous Schrödinger equation. Moreover, as a physical theory, quantum mechanics also specifies how the wave function assigned to an electron relate to the results of measurements on the electron. This is (mainly) given by the so-called Born rule, which roughly says that the result of a measurement is in general random, and the probability of obtaining a particular result is given by the square of the absolute value of the wave function. For example, for a measurement of the position of an electron, whose spatial wave function usually spreads in the whole space, the Born rule says that the probability density that the electron is found in a particular position is given by the square of the absolute value of the wave function in the position.

This is the core of quantum mechanics, which is composed of a mathematical formalism and its connections with experience. As can be seen, it is basically a mathematical recipe for calculating the probabilities of measurement results. In this aspect, it is extremely successful indeed. I think this is the way most physicists talk about the wave function and quantum mechanics. But it is not enough.

**3:AM:** If we can produce a mathematical recipe for calculating the probabilities of measurement results why do you worry that this isn’t enough and that we need is a physical explanation? Is Bohr right when he says we shouldn’t be asking about a quantum world because there isn’t one?

**SG:** The core of quantum mechanics is unsatisfactory in many aspects. First of all, although it can calculate and predict the probabilities of measurement results, it does not tell us how these results are generated. In fact, if assuming the wave function of an electron is a complete description of the electron, and the time evolution of the wave function is always governed by the linear Schrödinger equation, then the theory cannot account for the appearance of a measurement result as a matter of fact. This is the notorious measurement problem of quantum mechanics. I will discuss this in more details later.

Next, quantum mechanics admits the mind-independent existence of a physical world, which contains various physical systems such as electrons, atoms and our familiar macroscopic objects. But the theory does not tell us what the physical state of a system is, and especially, whether and how the wave function assigned to the system relates to its physical state. Exactly what is an electron? This is also a question Richard Feynman once asked. For example, quantum mechanics can accurately predict the results of the double slit experiment with single electrons, namely the double-slit interference pattern. But it does not tell us how the pattern is formed, and in particular, it does not tell us what an electron is, e.g. where it is in space, and how each of them passes through the two slits to form the pattern in the experiment. In other words, the core of quantum mechanics does not provide a sensible explanation of the double slit experiment.

It is well known that Niels Bohr, one of the founders of quantum theory, thought that there is no picture of how each electron passes through the two slits, and in this sense, there is no quantum world.

Bohr did believe that atoms are real, but he would not attribute intrinsic and measurement-independent state properties to atomic objects. He did not regard the wave function of an electron as a description of something physically real either. As to the double slit experiment, his argument is something like this. Once if we measure how the electron passes through the two slits, the interference pattern will be destroyed, and thus we cannot find how the electron actually passes through the two slits to form the pattern in principle. As a result, we should not ask such a question without an answer either. But today we know that Bohr’s arguments have loopholes, and he did not succeed in convincing us his views are correct and inevitable.

**3:AM:** Why is ‘wave function’ something that seems to require interpretation and given that it’s supposed to be a fundamental part of the universe how can something that seems abstract have a causal role in our physical world?

**SG:** If we think that an electron has an underlying physical state or ontic state, then it will be a natural question whether and how the wave function of an electron relates to its ontic state. After all, the wave function of an electron relates to the results of measurements on the electron, while these results of measurements should presumably be determined by the ontic state of the electron. Does the wave function of an electron directly represent the ontic state of the electron? Or does it merely represent the state of (incomplete) knowledge about the ontic state? In short, is the wave function ontic or epistemic? This intriguing question has been addressed in my recent book The Meaning of the Wave Function: In Search of the Ontology of Quantum Mechanics. Strictly speaking, it is not the wave function itself, which is only an abstract mathematical object, but the underlying ontology, which may relate to the wave function, that has a causal role in our physical world.

**3:AM:** How can the double slit experiment be explained (where a single electron passes through two slits) if neither the continuous motion of a particle or a wave can do the job? Do physicists think it’s a question without an answer?

**SG:** Yes, the classical picture of motion, such as the continuous motion of a particle or a wave, cannot explain the double slit experiment. This indeed leads some physicists to claim that explaining the double slit experiment is an impossible task. Feynman is a well-known example. He once said to his students in the 1960s, “Do not keep saying to yourself, if you can possibly avoid it, But how can it be like that? because you will get down the drain, into a blind alley from which nobody has yet escaped. Nobody knows how it can be like that.” This was also a deep puzzle for me when I was an undergraduate. We live in a classical environment after all.

On the other hand, this is not so beyond expectation either, since classical mechanics is already replaced by quantum mechanics. The former cannot even explain the observed double-slit interference pattern, while the latter can, with huge success. However, as I have just said, the core of quantum mechanics is not enough to do the job. What we need for a genuine explanation of the double slit experiment is an underlying ontology and the dynamics for the ontology. Fortunately, we already have several realistic approaches to explaining the double slit experiment, such as the de Broglie-Bohm theory or Bohm’s theory, Everett’s theory or the many-worlds theory, and collapse theories etc.

For example, in (one interpretation of) Bohm’s theory, an electron is a particle accompanied by a wave. This is the ontology of the theory. The wave spreads continuously according to the Schrödinger equation, and it passes through both slits. The particle moves continuously and passes through only one slit, and its motion is guided by the wave, which is like a ship guided by a radar wave. When assuming an appropriate initial position distribution of all particles and a particular guiding dynamics, the reaching positions of the particles on the screen in the double slit experiment can be consistent with the double-slit interference pattern. In this way, Bohm’s theory provides a realistic approach to explaining the double slit experiment.

Unfortunately, we don’t yet know which one among these competing approaches is right. In my view, these approaches are still not complete and satisfactory.

**3:AM:** Why do you find these approaches unsatisfactory? I presume you feel that they fail to explain something that needs to be explained?

**SG:** I think these approaches fail to provide a satisfactory interpretation of the wave function. They directly assume an underlying ontology and the dynamics for the ontology, but they do not explain why. First of all, they assume the ontic view of the wave function, while they do not give a proof. It is also possible that Einstein’s view is right, namely that the wave function of an electron does not represent the ontic state of the electron, but merely represents the state of (incomplete) knowledge about the ontic state of the electron. If this is indeed the case, then the above approaches will be wrong.

Next, as I see it, there are also some puzzling phenomena these approaches or theories fail to provide a reasonable explanation. As an example, consider an electron being in a superposition of two separated wavepackets in space. On the one hand, each wavepacket of the electron has gravitational and electromagnetic interactions with another electron. But, on the other hand, the two wavepackets have no gravitational and electromagnetic interactions with each other. (In this sense, the two wavepackets of an electron are really like ghosts; they can pass through each other without being “noticed”.) This seems to pose a dilemma. If each wavepacket of the electron is massive and charged, then it seems that the two wavepackets must have gravitational and electromagnetic interactions with each other, while if each wavepacket of the electron is not massive and charged, then how can it have gravitational and electromagnetic interactions with another electron?

I will talk about this later.

**3:AM:** So you’ve presented a new argument for the ontic view of the wave function. This is the idea that the wave function really does directly represent reality (as opposed to the epistemic view which you mentioned above which merely represents a state of incomplete knowledge). Your approach involves what you call ‘protective measurements’ doesn’t it? Can you say what these are and how they help resolve issues that weak measurements can’t?

**SG: **Yes, my new argument for the ontic view of the wave function or the reality of the wave function is based on an analysis of protective measurements. As we know, there are two familiar kinds of measurements in quantum mechanics: strong measurements and weak measurements. The former is so strong that they disturb the measured system greatly, such as leading to the apparent collapse of the wave function of the system. The result of such a measurement is in general random, being one of the eigenvalues of the measured observable. The latter is weak and almost does not disturb the measured system. The result of a weak measurement is not only random but also imprecise, being close to the expectation value or average value of the measured observable in the measured state. By these measurements we need an ensemble of identically prepared systems to measure the wave function. The method is usually called quantum tomography.

For many physicists and philosophers of physics, protective measurements, an intriguing method of measurement discovered by Yakir Aharonov, Lev Vaidman and Jeeva Anandan in 1993, are new and unfamiliar.

The key idea of protective measurement is that during a quantum measurement, either strong or weak, the wave function of the measured system is kept unchanged by an appropriate protection procedure.

Then, according to the Schrödinger equation, the result of such a protective measurement is not random but definite, being exactly the expectation value of the measured observable in the measured state.

Since the wave function can be reconstructed from the expectation values of a sufficient number of observables, the wave function of a single quantum system such as an electron can be measured by a series of protective measurements on the system.

If the wave function can only be measured from an ensemble of infinitely many systems, then it is undoubtedly more difficult to see that the ontic view of the wave function is right, namely that the wave function directly represent the ontic state of a single system such as an electron. Indeed, although there are several theorems such as the PBR theorem which proves the reality of the wave function, they must be based on auxiliary assumptions. On the other hand, if the wave function can be measured from only a single system as for protective measurements, then it will be more readily to prove that the wave function of the system is a direct representation of its ontic state, and thus the ontic view of the wave function is right. For a protective measurement, however, there is a subtlety. One still needs to show that the result is not generated by the protection procedure. This is indeed the case. Then, the wave function, which can be measured from a single system as the results of protective measurements, can be determined only by the ontic state of the system.

This may prove the reality of the wave function without resorting to auxiliary assumptions.

It is strange, at least for me, that protective measurement was discovered so late, after almost 70 years of the founding of quantum mechanics. One reason, I think, may be that the conventional measurements can be realized more readily in experiments. In fact, even in classical mechanics, when we measure a property of the measured system, we still need to protect the property unchanged as best as we can during the measurement; otherwise what we have measured will be not the actual property of the system. Protective measurement is just a natural extension of such a classical measurement. What is a surprise for us is that quantum mechanics, despite its huge difference from classical mechanics, does permit the existence of such protective measurements. Here I will tell you a piece of good news, that is: the first protective measurement has been recently realized in experiment for photons by an Italian group from Istituto Nazionale di Ricerca Metrologica (INRIM).

**3:AM:** What do you mean when you say that you ‘analyze the origin of the wave function by deriving the free Schrodinger equation’? Can you try and give us non-specialists a flavour of what you’re doing here?

**SG:** This is an attempt to answer the question of why the physical state of a single system is described by the wave function, e.g. why the physical state of an electron with definite momentum and energy is described by a plane wave which spreads throughout the whole space and has the same amplitude everywhere. This description seems very counter-intuitive indeed. If we can derive the Schrödinger equation for the wave function from certain fundamental principles, we will be able to understand the origin of the wave function more deeply.

In quantum mechanics textbooks, the Schrödinger equation is usually derived by analogy and correspondence with classical physics. There are at least two mysteries in this heuristic derivation. First of all, even if the behavior of microscopic particles is like wave and thus a wave function is needed to describe them, it is unclear why the wave function must assume a complex form. Indeed, when Schrödinger invented his equation in 1926, he was also puzzled by the inevitable appearance of the imaginary unit “i” in the equation. Next, we don’t know why there are the de Broglie relations for momentum and energy and why the nonrelativistic energy-momentum relation is E=p^2/2m.

According to the analysis in my new book, the key to unveiling these mysteries is to analyze spacetime translation invariance and relativistic invariance of laws of motion. It turns out that these symmetry principles require that the linear, nonrelativistic time evolution for an isolated system (e.g. a free electron) must be governed by the free Schrödinger equation. As a result, the physical state of a single system will be naturally described by a wave function appearing in the equation.

**3:AM:** For you then, the wave function is the ‘random discontinuous motion of particles.’ Can you unpack this a little for us? By saying that motion is discontinuous are you saying that the classical world of continuous mechanics is an illusion, a mere shadow of reality, and that, to use a famous metaphor, God really does play dice?

**SG:** Yes, this is the most important idea of my new book. Let me first tell you a real story about how the idea came to my mind more than 20 years ago. During my graduate study at the Institute of Electronics, Chinese Academy of Sciences, the puzzle of how the electron moves in an atom had been plaguing me. Day after day, I gradually doubted the reality of continuous motion. But I still felt in my bones that the electron is a particle and it must move in space in some way. Finally, in an early morning of October 1993, I experienced a sudden enlightenment or revelation. At that moment, I felt that my body permeated the whole universe and I was united with it. I “disappeared”. A clear picture then appeared: a particle is jumping in a random and discontinuous way. It is not inert but active; it moves purely by its own “free will”.

Since then the idea of random discontinuous motion of particles has accompanied with me. I have tried to find more convincing arguments for its reality. In my new book, I give two such arguments, one of which is based on a strict analysis of protective measurements in quantum mechanics. This argument is also related to the above dilemma about an electron being in a superposition of two separated wavepackets, and it is something like this.

Protective measurements can measure the mass and charge distribution of an electron. And the results as predicted by quantum mechanics show that the charge of an electron is distributed in the whole space, and the charge density in each position is proportional to the square of the absolute value of the wave function of the electron in the position. (In fact, when Schrödinger introduced the wave function and founded his wave mechanics in 1926, he already proposed this charge density hypothesis. Now protective measurements confirm the hypothesis.) As a result, the two separated wavepackets of an electron are both massive and charged. Visually speaking, an electron is a mass and charge cloud, and the density of the cloud is given by the square of the absolute value of its wave function. This seems to be a surprising result for most people.

Now the puzzle is: If each wavepacket of the electron is massive and charged, then why do they have no gravitational and electromagnetic interactions with each other? There is only one answer. It is that these mass and charge distributions are effective, formed by the motion of a particle with the mass and charge of the electron which moves throughout the two separated regions very quickly. Since a particle being at an instant has no interactions with itself being at another instant, the two separated wavepackets will have no interactions with each other.

This argument suggests that an electron is a particle which undergo random discontinuous motion, and more generally, a quantum system is a system of particles that undergo random discontinuous motion in our three-dimensional space. The wave function of a quantum system then describes the state of the random discontinuous motion of the particles of the system, and in particular, the square of the absolute value of the wave function gives the probability density that these particles appear in every possible group of positions in space. At a deeper level, the wave function may represent the propensity property of the particles that determines their random discontinuous motion (RDM in brief).

If motion is really discontinuous and random as argued in my book, then yes, the classical world of continuous mechanics will be an illusion, a mere shadow of reality. And then, quite contrary to Einstein’s expectations, God does play dice with the universe.

**3:AM:** Do we need to adopt durationless instants or gunk for your theory to work – or something else completely?

**SG:** No. In fact, I have argued that in order that the theory is consistent with experience time must be discrete, which means that we need to adopt duration instants. Moreover, when time is really discrete at the Planck scale as many people working on quantum gravity believe, it turns out that the theory is consistent with existing experiments and our experiences of the macroscopic world.

**3:AM:** Does this view of discontinuous motion mean that motion has no cause? Is all causation in nature an illusion?

**SG:** Yes, according to my theory of RDM of particles, motion will have no cause at the deepest level of reality.

However, this does not mean that all causation in nature is an illusion. The RDM of particles will introduce a new stochastic evolution term into the Schrödinger equation. It is this term that leads to the collapse of the wave function after a measurement, as well as the localization of macroscopic objects (with the help of environment). But the original deterministic evolution term in the Schrödinger equation is still there, and it will generate the apparent continuous motion of macroscopic objects. And causation is still real at this level and in our everyday world in general.

**3:AM:** If the quantum world and the relativistic world don’t fit together and getting them to fit is what some future theory of everything will have to do, does your approach mean that the quantum ontology is enough to explain all our definite experiences or does it still need to be revised in the relativistic domain?

**SG:** It is well known that there are two important conceptual issues concerning the unification of quantum mechanics and special relativity. First, the apparent incompatibility between wavefunction collapse (and the resulting quantum nonlocality) and the principle of relativity has been an unsolved problem since the founding of quantum mechanics. For example, it is still debatable whether a preferred Lorentz frame or an absolute frame is needed to solve the incompatibility problem. Second, although the combination of the linear quantum dynamics and special relativity has been obtained in quantum field theory, it is still a controversial issue what the ontology of the theory really is. Is it fields or particles? Or is it other physical entities?

In my new book, I have argued that in order to solve the incompatibility problem it is special relativity, not quantum mechanics, that should be revised (without contradicting experience, certainly). Concretely speaking, the principle of relativity is not universally valid, and there exists an absolute frame in which the one-way speed of light is isotropic and the collapse of the wave function happens simultaneously in the whole space. Moreover, the absolute frame can also be detected by measuring the (average) collapse time of the wave function according to my energy-conserved collapse model in terms of RDM of particles. This seems to be a minor view now. But John Bell once supported this view. In his 1981 paper “Quantum mechanics for cosmologists”, he said, “It may well be that a relativistic version of the theory, while Lorentz invariant and local at the observational level, may be necessarily non-local and with a preferred frame (or aether) at the fundamental level.”

Does my suggested quantum ontology need to be revised in the relativistic domain? This needs a more careful analysis. In my view, although the picture of RDM of particles is seriously distorted by the Lorentz transformation, the transformation does not change the existent form of particles. And thus the ontology of (relativistic) quantum field theory will be still particles. Here I should note that this picture of particles is independent of whether the state of motion of particles can be localized or not. For example, the fact that there are no Lorentz invariant localized states poses no threat to the existence of such particles.

However, quantum field theory does introduce a pair of new processes for the motion of particles, which are the creation and annihilation of particles. In quantum mechanics, the number of particles is conserved and the existence of a particle is eternal. While in quantum field theory, a particle can be created and annihilated.

This will lead to a new form of RDM of particles. For example, there will be superpositions of states with different particle numbers. When a particle is in such a superposition, the particle may not exist in the whole continuous time flow, but only exist part the time at certain discontinuous and random instants.

**3:AM:** Isn’t Everett’s many worlds interpretation of quantum mechanics an alternative ontic view to your own? What consequences has your approach got for the many worlds theory of Everettian? Are the two views compatible?

**SG:** No, it is. But Everett’s theory has been plagued by the well-known probability problem. If all measurement results are obtained, then it seems that the probability of obtaining any result is just one; this is obviously inconsistent with the Born rule. Although some proponents of the theory, notably David Wallace, think that this problem can be solved by resorting to the decision theory, many opponents are not convinced. In fact, a more serious problem of Everett’s theory is that it does not give a clear ontology for the microscopic world, as Tim Maudlin has pointed out. The suggested “spacetime state realism” can be regarded as a generation of Schrödinger’s interpretation of the wave function in terms of charge density, and it is still not a complete quantum ontology.

At first sight, my approach may provide an appealing ontology for Everett’s theory. According to the picture of RDM of particles, each Everett branch, such as the alive cat branch or the dead cat branch of the Schrödinger cat, exists part the time at certain discontinuous and random instants, and all these instants constitute the continuous time flow. This means that worlds exist in a time-divided way in a many-worlds theory, and the time division is random and discontinuous in nature. Moreover, the measure of the instant set for each branch is proportional to the square of the absolute value of the amplitude of the branch. It seems that this may help solve the probability problem.

However, a more careful analysis shows that this appealing picture of many worlds cannot be true, since it cannot account for the measurement result and the Born probabilities. In order to explain the randomness of a measurement result and the Born probabilities in my approach, there must exist an additional random dynamics besides the deterministic Schrödinger dynamics, which results from the RDM of particles and results in the appearance of a random measurement result. This will lead to the collapse theories, in which the Schrödinger equation is revised by including a stochastic evolution term that generates the dynamical collapse of the wave function. Today, collapse theories, together with Bohm’s theory and Everett’s theory, are widely considered as three main realistic alternatives to standard quantum mechanics. Thus, as far as I can see now, my approach and the many-worlds theory are not compatible.

**3:AM:** Consciousness is another part of the universe that seems to create problems for modern science alongside relativity and quantum mechanics. Do you think there’s a possible link between consciousness and quantum measurement, the kind of thing that David Chalmers, for example, has recently been discussing, that would make consciousness a fundamental rather than an emergent property of the universe?

**SG:** The idea of consciousness collapsing the wave function has a long history. It can be traced back to John von Neumann in the 1930s. Although I am not a proponent of this idea, I think it is an interesting idea which deserves to be further examined. Recently, David Chalmers and his collaborator, Kelvin Mcqueen, have done a good job in making the idea more rigorous. In my view, there is at least a potential problem with the idea. Since consciousness is always a definite property, it seems that the theory cannot account for the randomness of a measurement result. It is arguable and understandable that the randomness must come from a noise source with a certain probability distribution related to the wave function, such as the flashes in the GRW theory or the RDM of particles in my theory. Then, it is reasonable to assume that it is this random noise, not other definite things such as consciousness, that causes the collapse the wave function.

However, I do believe that there is a deep connection between quantum mechanics and consciousness which needs to be studied more deeply. And I am planning to edit an anthology about this. The connection is two-way. On the one hand, as I have argued in my book, since the measurement problem of quantum mechanics is essentially the determinate-experience problem, the problem must be defined at the mental level and based on an assumption about the form of psychophysical connection, namely an assumption about how the mental state supervenes on the physical state. Moreover, an analysis of whether the form of psychophysical connection required by each solution to measurement problem satisfies the principle of psychophysical supervenience may also help solve the problem.

On the other hand, we also need quantum mechanics to analyze the actual form of psychophysical connection (and maybe also the working mechanism of our conscious brains). What physical state does the mental state supervene on? My recent analysis suggests that the mental state of an observer supervenes on her wave function, not on a certain branch of the wave function as in the many-words theory or other hidden variables as in Bohm’s theory. This leads us to a deeper question: How does the mental state of an observer supervene on her wave function? Especially, what does it feel like to be in a quantum superposition? I have obtained two initial results. The first one is that the answer to this question may help solve the famous tails problem of collapse theories. The second one is more speculative, which is that the answer may also suggest that the wave function of a system with consciousness may evolve in a deterministic nonlinear way, and thus consciousness is a fundamental rather than an emergent property of the universe. I think David would like this result. These analyses will be the main content of my next book, *The Quantum Observer: Towards a Deeper Understanding of Quantum Mechanics and the Nature of Consciousness*.

**3:AM:** How much of this is empirically testable? Some of it looks more like metaphysics than science?

**SG:** Well, different from other quantum theories such as Bohm’s theory and Everett’s theory, collapse theories are becoming a lively field of research in both philosophy and physics. Collapse theories, including my own energy-conserved collapse model in terms of RDM of particles, are empirically testable. They predict that the wave function of a physical system will collapse under certain specific conditions which are determined by the specific theories, and the collapse can be detected in principle in experiments. For example, my collapse model predicts that when the energy uncertainty of a quantum superposition is large enough, the superposition will shortly collapse to one of the energy eigenstates in the superposition. If the details of my model can be confirmed by future experiments, then the picture of RDM of particles will turn out to be true.

Recently more and more experimental tests have been suggested for collapse theories. No doubt, there will be a process during which metaphysics becomes science, whose progress depends on the developments of technologies. I am currently editing an anthology about collapse theories for Cambridge University Press, whose title is: *Collapse of the Wave Function: Models, Ontology, Origin, and Implications*. I hope the publication of the anthology will arouse more researchers’ interest in these promising and testable quantum theories.

**3:AM:** As a take-home then, could you say what you think reality is actually like according to your theories about physics, and could you speculate about what are likely to be the next big discoveries in physics , such as the impact of dark matter on our understanding of the universe?

**SG:** I think Newton was right when he said reality is particles. For instance, an electron is still a particle, not a wave or something else. But the motion of these particles is not continuous as we experience for macroscopic objects, but discontinuous and random in nature. Moreover, spacetime is very likely discrete at the Planck scale.

I speculate that the next big discovery in quantum physics will be that the collapse of the wave function turns out to be a real physical process. And if I am lucky enough, I will be able to see that the above picture of reality can also be verified by future experiments.

**3:AM:** And finally, are there five books, other than your own, that you can recommend for the readers here at 3:AM that would take them further into your philosophical world?

**SG:** I would like to recommend the following five books (in chronological order):

Albert, D. Z. (1992). *Quantum Mechanics and Experience*. Cambridge, MA: Harvard University Press.

Bell, J. S. (2004). *Speakable and Unspeakable in Quantum Mechanics: Collected Papers on Quantum Philosophy*. 2nd Edition. Cambridge: Cambridge University Press.

Ghirardi, G. C. (2007). *Sneaking a Look at God’s Cards: Unraveling the Mysteries of Quantum Mechanics*. Princeton: Princeton University Press.

Ney, A. and D. Z. Albert (eds.) (2013). *The Wave Function: Essays on the Metaphysics of Quantum Mechanics*. Oxford: Oxford University Press.

Lewis, P. J. (2016). *Quantum Ontology: A Guide to the Metaphysics of Quantum Mechanics*. Oxford: Oxford University Press.

**ABOUT THE INTERVIEWER**

**Richard Marshall** is still biding his time.

Buy his new book here or his first book here to keep him biding!

First published in 3:AM Magazine: Sunday, September 17th, 2017.