Superdeterminism may have solved the quantum measurement problem
A local, realist, reductionist interpretation with no extra dimensions, universes, or solipsism.
A local, realist, reductionist interpretation with no extra dimensions, universes, or solipsism.
Quantum entanglement is weird, but it is hard to explain why. It seems quite mundane why two particles that were once in contact might be correlated.
Yet, Einstein called entanglement “spooky action at a distance.” And many physicists insist that information is transmitted instantaneously between particles. This information tells one particle both what experiment was done and the result of the measurement at the other.
Measurement correlation hardly seems unusual since if you have a bag with a black ball and a red ball and you reach in, pull out a black ball, and your friend reaches in, you already know they must pull out a red ball. Likewise, if you fire the black ball towards your friend Alice in one direction at near the speed of light and the red one in the other direction towards your friend Bob at near the speed of light, and Alice sees a black ball speeding towards her; Bob must surely see a red ball.
It is easy to show the difference between colored balls and particles using mathematics. The case of the balls is called a classical mixture, while the case of the particle spins is an example of a “pure” quantum state. And the difference is, in some respects, like the difference between having two balls of two colors versus one ball that is simultaneously two colors, yet, when you look at it, it is only one of the two. When particles are entangled, it is like having two balls each with many simultaneous colors. Yet, when you look, each is only one color, and they are always “opposite” colors. On the face of it, it seems as if there is no distinction between this quantum state and a classical mixture, but there is.
Experimentally, how you differentiate between them has a lot to do, not with mathematics, which is just a description, but the counter-intuitive observations that you make when you take measurements. You find that the measurement you are making has as much to do with your measurement as the nature of the ball (or particle). In other words, measured and measurer(s) are intertwined.
To understand entanglement, you have to understand how quantum states differ from classical. Let us look at spin as an example.
Spin is thought to be similar to a particle spinning like a top. A charged particle, like an electron, has a spin that gives it a magnetic field. The electron thus has a north pole and a south pole. If you pass the electron through a magnetic field that is uniform, the north and south poles are equally pulled, and the electron passes straight through. But if you pass it through a non-uniform magnetic field that changes strength in the up and down direction, it will be deflected either up or down.
If your magnetic field strength increases in the upward direction and the electron is oriented with north pole up, then the electron will be deflected up. If the electron is oriented the opposite way, it will be deflected down.
If the electron’s axis is perpendicular to the magnetic field, it should pass straight through, while, if it is at an angle with respect to the field, it should be deflected less.
Suppose we set up this experiment and fire electrons (or silver ions as in the image below; it doesn’t matter as long as they are charged particles) in between our magnets and into a detector on the other side.
If electrons were like spinning tops with magnetic fields oriented any which way, we would expect to see a pattern of dots on our detector showing a more or less uniform distribution depending on how aligned each electron’s spin is with our field. But, instead we see only two distinct groupings of points, an up blob and a down blob.
You might think that the magnet surely is forcing the electrons (or ions) to align with itself and that is why there are only two groupings. This is how compass needles work. The Earth orients them to magnetic north. But that theory would only explain one grouping, the top one. How could it explain the other? Reverse compass needles?
Besides that problem, if the electrons were truly spinning like gyroscopes, their spinning would stabilize their orientation with respect to the magnetic field. A gyroscope maintains its orientation, and this is what we see in experiments with classical charged gyroscopes in magnetic fields.
Maybe then the cathode is only emitting “up” and “down” spin electrons for some reason. Yet, if we turn the magnets, we see that the pattern turns with it. So, the emitter would have to somehow “know” the orientation of the magnets before the electrons reach it.
Experiments like these made many physicists abandon any sense of an underlying reality in nature. Somehow the measuring apparatus is determining the outcome of the experiment. For some, this means that the wavefunction for the spin of the particles is “collapsing” when it hits the detector and necessarily how that collapse occurs is determined by the orientation of the magnets.
For others, it means that, as in Einstein’s relativity, the state of the measurer is as important as what is being measured. There is no collapse. Rather there is a spectrum of potential outcomes depending on who is doing the measuring and how.
The former idea has wrongly been considered the “official” interpretation of quantum physics — the so-called Copenhagen interpretation. But it is more of a strawman invention of its detractors, and only a few extremists actually believed in it at first. Of those who actually work in the field, you would be hard pressed to find a true adherent of Copenhagen.
The latter certainly makes more sense than the former since we already understand Einstein’s relativity this way. Your state of motion determines how you measure time and space. As soon as you look at things in four dimensions instead of three, everything once again becomes concrete and reality is clear. Things do have an absolute underlying reality. It is just in too many dimensions for us to see all at once.
Likewise, in quantum physics, perhaps reality is just in too many dimensions for us to see, and measurer and measured instead fall along a spectrum within yet another dimension, a fifth dimension. Perhaps if we looked at things in five dimensions everything would, yet again, have an underlying reality. Some physicists, such as Mikio Namiki and yours truly, hope for such a solution, despite its disturbing implications on the persistence of history.
Another possibility is that measurer and measured affect one another because that is simply how the universe is set up. The universe, from the beginning, arranged for the particles to be measured in just that way and for the outcomes to be just so. Nothing is an accident.
If this is sounding like the Calvinist version of quantum physics philosophy, it is. It is called superdeterminism.
Just as double predestination, in which God predetermined who would be saved and who would be damned, causes heartburn for many theologians, superdeterminism is bound to cause a lot of anxiety and anger amongst physicists. This is because is violates a dearly held assumption in all science called the Statistical Independence Assumption. This is the idea that measurements and that which is measured are uncorrelated.
Correlations between experiments and measurement outcomes! Preposterous!
As Sabine Hossenfelder and Tim Palmer wrote in a recent article:
The major reason this path has remained largely unexplored is that under quite general assumptions … any theory which solves the measurement problem in a form consistent with the principles of relativity, makes it impossible to prepare a state independently of the detector that will later measure it. If one is not willing to accept this dependence then — by virtue of Bell’s theorem — one necessarily has to conclude that a local, deterministic completion of quantum mechanics is impossible.
Bell’s theorem is simply the mathematical statement of what I just talked about. The measurement being done and the measurement outcome are correlated with one another. When two particles are entangled and so far apart that light could not reach from one to the other in time to affect how that particle is measured, this violates relativity. This means that if Alice and Bob are light years apart measuring two entangled electrons. How Alice chooses to measure electron A will affect what Bob measures at electron B.
That means that either (1) information is really being transmitted from Alice to Bob instantaneously, (2) the measurements Alice and Bob were going to make and the states of electrons A and B were all predetermined from the Big Bang, or (3) the future can somehow affect the past.
This reminds me of certain time travel stories where the time traveler goes back in time to change something in the past only to find out that they were instrumental in it or had no power to change it because all was predetermined.
Free will seems to be on the chopping block here. Yet, from a philosophical perspective, free will is impossible to define in the colloquial sense anyway. What is free will but having the power to do otherwise given a choice? But my own desires determine what I do. I have no power to change those; for in order to change a desire, I would have to desire something else. This is the conundrum that free will faces when presented as a simple conditional: wanting means having the choice to do what one wants, but one does not have the choice about what one wants. This definition has problems even without superdeterminism.
Alternatively, you can define free will not to include desire at all. Instead, it simply focuses on power. If I have power to do otherwise in any given situation, all things being equal, then I have free will. It doesn’t matter what I want. I could flip a coin and do different things and by that consideration I have free will of a sort. I am not condemned to a certain outcome and the future is not only uncertain but depends on my actions. In this conception, free will depends more on what type of person I am than on choices and desires of the moment.
In a superdeterministic universe, this definition of free will is difficult to support because the future is completely determined. Every action has no alternative outcome. One could propose that there are hypothetical worlds in which one did otherwise that do not exist. But this does not work well because it is very clear that, at any point of decision, you cannot do otherwise in this universe.
This is hardly unique to superdeterminism. Any kind of close scrutiny of either physical and biological processes makes free will a problem, particularly from the perspective of moral accountability. If I have the power to do otherwise because of some random flip of a coin, I have no more accountability than if that outcome were predetermined. Maybe I am simply a good or bad person — as Calvin would have — and whoever or whatever structured the universe as it is is the one responsible. If I were damned to eternal suffering, I think I would take little comfort in that.
Free will isn’t an argument against superdeterminism, and some definitions of free will may be possible even in the context of determinism (called “compatibilist” because they are compatible with determinism).
One potential way out for superdeterminism is to imagine that you are free in the sense that your actions determine your life as a whole. The order of those actions does not matter. In other words, your future choices determine your past experience just as much as your past choices determine your future experience.
In this sense, rather than assuming your life was determined at the Big Bang you could assume that your choices determine (some of) the conditions at the Big Bang. This approach implies that past reality is conditioned on future will rather than the other way around.
Although there is no information transfer from future to past, so you can’t remember the future, there can be causal effects at the quantum level and relativity is not violated provided cause and effect are within light speed of one another. In that sense, you cannot know the future yet it can cause the present and the past. It can change reality itself, switching the electron spin orientation for example, or changing what reality was before you became aware of it.
This is why a better term for superdeterminism is “Future Input Dependency”. Thus, my actions in the future might, counter-intuitively, be determining my actions now rather than the reverse. Moreover, my future actions might even determine reality itself in the present. Thus, how I set up an experiment years in the future might determine the state of an electron emitted now.
Like many interpretations of quantum physics, this idea that the future determines the past might seem like a cheap cop-out from a real interpretation. But, unlike most other interpretations of quantum physics, superdeterminism provides a complete, local, reductionist, realist interpretation of quantum physics without invoking extra dimensions or universes. It does not play fast-and-loose with the definition of “real” or “local” and it does not abandon reductionism (detectors and other macroscopic objects are subject to the same laws as microscopic ones). Nor is it solipsistic, positing that no non-subjective reality exists like Qbism.
Thus, it has become the invisible elephant in the room when discussing quantum measurement theory. It is the one idea that explains everything without abandoning classical assumptions about physical reality.
Even the idea that the future can cause the past (which isn’t necessary but may be a useful interpretation) isn’t particularly radical. Besides the 2nd law of thermodynamics, a statistical and not absolute law, all physical laws are time reversible.
For scientists, Superdeterminism has some chilling implications, however, for the independence of experiments. If my experiment and the outcome of said experiment are always correlated, rather than independent, then science itself may be dependent on the sum total of measurements that we were predetermined to carry out in this universe. That may imply that science is meaningless. It is simply a game that the universe is playing with itself.
Consistency in results, a core value of experimental science, no longer has any implication on our understanding of the universe as a whole. We cannot extrapolate those results to experiments not undertaken because all experiments are correlated, not only with what they measure and the underlying reality of those measured particles, but possibly with each other and future events. Our understanding is just based on the interwoven tapestry of a finite set of experiments coupled with a finite set of outcomes, and that tapestry, rather than implying some larger picture, is just one potential picture amid a spectrum of other, potentially unrelated realities.
This argument is called the “conspiracy argument” because it suggests that the universe has conspired to give us a certain reality. It turns out, however, that we already have this problem in physics. It is called the “fine tuning” problem. Simply stated, the fine tuning problem is the problem that the universe is so very finely tuned to produce matter and life and us that it must have been arranged just so.
Yet fine tuning does not necessarily eliminate science’s explanatory power. After all, it is precisely because fine tuning shows up so strongly in scientific experiments that we are even aware of it. Fine tuning and correlations between experiments and results do not preclude us from imagining alternative outcomes. The capacity to imagine alternatives and to quantify those imaginings as state spaces that govern the total range of possibilities gives science its explanatory power. Indeed, the very fact that quantum physics has a measurement problem indicates that the correlations that may be between future and past may not be a barrier to science.
Another argument against superdeterminism has sometimes been called the Tobacco Company argument. This argument goes that if we allow violations of statistical independence in quantum physics, then we should allow it in all randomized trials, including those that link lung cancer to smoking. A tobacco company could claim that because of superdeterminism, statistically independent trials are impossible. Hence, no correlation can ever be demonstrated.
This oddball argument makes no sense to me. The violation of statistical independence in quantum physics means that the setup of the measurement apparatus is determining the state of the particle (or that both are determined by a past event like the Big Bang). This does not happen in classical systems. That is the whole problem.
Now, it is entirely reasonable to believe that statistical independence is difficult to achieve in classical experiments, especially in randomized trials. And this has, oddly enough, been shown to be true so many times that a large portion of randomized trials come to false conclusions.
One of the ways to determine if a trial has hidden statistical correlations is to repeat it, particularly with a different methodology, different team, and different location. That is, we try to remove hidden variables that may create correlations in the system. Classical physics gives us the tools to remove correlations because we can predict most of the hidden variables. With quantum physics, we cannot remove the hidden variables because we cannot predict what they all are nor do we necessarily have power over them because some may only show up at very high energies like during the Big Bang.
Superdeterminism, on the other hand, implies that there are correlations that are built into the structure of the universe that vanish when quantum effects can be neglected. We can show that these effects vanish. Try shooting random charged gyroscopes through a Stern-Gerlach magnet instead of particles. You will not see the quantum result but a messy spray of dots.
Right now scientists are mostly ignoring superdeterminism and have done so for decades even though it is as old as Bell’s theorem. John Bell, whose theorem “proves” that quantum entanglement involves spooky action at a distance, spent considerable time trying to refute superdeterminism precisely because his theorem depended on statistical independence. Yet, if statistical independence itself is a classical assumption that vanishes at the quantum level, then Bell’s theorem falls apart.
Albert Einstein, if he had lived, would almost certainly have been an adherent of superdeterminism because it established everything he wanted for quantum physics: realism, locality, and reductionism. It posited a universe much like his general relativistic one: a predetermined, four dimensional structure where particles and the experiments measuring them are frozen in time like flies in amber.
Whether superdeterminism can be tested experimentally is another story. According to superdeterminism, unlike in standard quantum mechanics, particle states from moment to moment are correlated because they are deterministic. In other words, determinism means that quantum mechanics actually behaves a lot more like classical mechanics than the present theory of quantum mechanics predicts with its non-deterministic, random behavior.
Superdeterminism suggests that a particle must carry forward its hidden state rather than its being determined randomly at the moment of measurement. This may not have been noticed because nobody is making measurements of particle states fast enough for those states to be observed to be correlated. But if you could make measurements very, very quickly while simultaneously keeping your measurement apparatus from changing states in that time period (such as using an ultra-cold one) then you might see these correlations. That would be a clear violation of quantum theory.
In any event, superdeterminism is one of many ways to restore classical locality and realism to quantum physics. Will future physicists embrace it? Perhaps it has already been determined.
100th article! Thank you to all my followers!
Hossenfelder, Sabine. “Testing super-deterministic hidden variables theories.” Foundations of Physics 41.9 (2011): 1521–1531.