Quantum mechanics is said to indicate that the universe is not “locally real”, because a particle is not in a defined state before measurement. But if a particle is not in a defined state, what is it ontologically? Can something still exist “in an undefined state”? Does “measurement” of something’s properties fully determine if that thing is “real” or not? Does that philosophically imply that “existence” is exclusively determined as a manifestation of observed properties?

See Philosophical Issues in Quantum Theory - Ontological Issues for reference.

  • 1
    That's a very questionable interpretation of QM. Just because our measurement might affect how it behaves doesn't mean it's not "real". Although my ability to (accurately) go into the details one of the most complex concepts in physics is limited (and that probably also applies to anyone else here). Possibly the one thing that could serve as evidence for anti-realism would be if the laws of reality suddenly change or stop working (but even then, there may just be laws we don't know about yet), but QM was always there, we just discovered it somewhat recently.
    – NotThatGuy
    Dec 17, 2023 at 15:11
  • 1
    I have enough of an idea what the traditional interpretation of QM is. My point is that you're drawing a questionable link from probabilistic states to anti-realism. It sounds like you might be conflating realism and determinism (in a sense, at least).
    – NotThatGuy
    Dec 17, 2023 at 19:30
  • 3
    Realism is the idea that there is a reality, that stuff exists independently of our own perception and thoughts (YMMV, but it's the gist of all "something-realism" theories). All physics theories suppose realism, as studying reality is exactly the point of physics. No interpretation of QM is at odds with realism. Anti realism is the idea that what makes a statement true is that it has been demonstrated, rather than observation. It doesn't have much to do with QM.
    – armand
    Dec 17, 2023 at 23:07
  • 2
    What definition of realism are you using on philosophy SE, if not the philosophical one? What relationship, then, does this question have with ontology or "anti-realism", which is also a philosophical term? If you are using specific definitions of a term, it's fine but please be upfront about it because precision in terminology is paramount. And if your question is about physics ask to physics SE.
    – armand
    Dec 18, 2023 at 5:19
  • 2
    There is a definite state. It's just that it's is a wave function rather than a point particle. The evolution of that wave function, both over time and as a result of interacting with other wave functions, is fully deterministic. In many cases, the wave function is highly concentrated, and you can treat it as "the particle is here" or "the particle is in one of these three places with uniform probability" and handle those cases separately using classical physics (and just ignore all these super-low-probability regions) and get a decent approximation. But the wave function never goes away.
    – Ray
    Dec 18, 2023 at 14:58

7 Answers 7


From WP on anti-realism:

Anti-realism in its most general sense can be understood as being in contrast to a generic realism, which holds that distinctive objects of a subject-matter exist and have properties independent of one's beliefs and conceptual schemes.1 The ways in which anti-realism rejects these type of claims can vary dramatically. Because this encompasses statements containing abstract ideal objects (i.e. mathematical objects), anti-realism may apply to a wide range of philosophical topics, from material objects to the theoretical entities of science, mathematical statements, mental states, events and processes, the past and the future.

Anti-realism means many things in many contexts, and is subtle across all of those contexts. Anti-realism is the coherent response to incoherent findings that an essentialist and therefore realist position struggles to explain. Since you mention QM, let's start with wave-particle duality.

In the macro world, a beach ball resting on the ocean is a clear example that balls are particles and behave like them, and ocean water manifests waves and can be described with waveform mathematics. Beach balls do not behave like ocean water, and ocean water does not behave like beach balls. Thus, the essence of the beach ball and the ocean wave are fundamentally distinct and generally observed to have two distinct sets of properties.

Photons suffer no such constraints of essentialist logic and observation. Under some experimental situations, photons are like beach balls, and under others, they are like ocean waves. Thus, QM creates a conundrum for the essentialist because the ontological properties are not fixed in the same way we expect them to be from our experience in the macro world. Thus, the incoherence of properties manifested under different conditions needs to be addressed. Anti-realist thinking merely rejects the idea that the ontological properties are essentialist in nature.

For a more technical explanation of locality and realism, consider Dcleve's comment:

[Note] Bell ‘s Inequality and how Bell defined realism. He held locality to be interactions that are constrained by a light speed cone. And realism was that a universe has a state that is independent of the observer. Bell's Inequality, if it holds, prevents physics EITHER from holding by locality, or realism (or both). Independently he showed that QM has to be non-local. Realism or not was still an open question, but it IS defined. – Dcleve

(For more information, see Bell's Theorem and read The Universe Is Not Locally Real, and the Physics Nobel Prize Winners Proved It (SciAm))

Just as the properties of photons are a function of the experimental system in which they are located, so too are empirical observations themselves subject to construction by the observers who make them; we see what we expect to see and not necessarily what is "really" there. Perceptual theory-ladenness, for instance, simply contends that our concepts and theories we ascribe to affect what we our visual system presents us with in the world.

There is, for instance, some experimental support for predictive coding which maintains that our perceptual systems are biased to construct visual experience based on past experience. Thus, as McDowell has argued, our perception is constantly under pressure from our pre-linguistic and conceptual experience. Thus, when two logical positivists struggle to define what is real by observation statements and arrive at incoherent views, an anti-realist merely points out that the act of perception, which is often taken as a foundation in empirical methods, is at its core a function of the biases of the observer. Different people literally have different first-person observations. Colorblindness and tetrachromaticity are two flagrant examples, but everyone has subtle variations in their visual system that often go unnoticed.

Thus anti-realism is the quietist's solution to the endless parade of petty arguments about ontological reality. In the spirit of Kant's Ding an Sich, an anti-realist can simply exploit the full ontological ambiguity that is inherent when a term refers to something by acknowledging that, under most circumstances, properties appear to be external the observer precisely because realist conceptions are constructed uniformly among observers, but that in edge cases, such constructions break down. Anti-realist experiences are not a deep mystery about the universe, but rather simply a property of embodied cognition (SEP) that requires us to construct our representations.

You ask:

A) What is a measurement? From what I am aware, this still has not been well defined

B) If a particle is not in a defined state, what is it exactly ontologically? Does it still exist? If it exists in an undefined state, isn’t that still a “state” of existence?

In the anti-realist view, a measurement is an observation that uses a standard. A good explanation of what measurement can be understood as can be found in Measurement in Science (SEP). What is a particle ontologically? First, it is not a macrosystem our brains have evolved to construct experience of first-hand. We see particles through our elaborate tools, procedures, and theory. Secondly, a particle is merely the digital manifestation of a portion of our time-space, and that stands in stark opposition to the analog manifestation, a wave. Thus, the age old debate over waves or corpuscles simply resolves to "photons are both waves and particles depending on how we construct our observations of them", which is fully consistent with the empirical apperception of the observer.

And lastly, a realist is someone who, after being confronted with the modern findings of science, simply clings to the idea of a single "reality" with "real properties" looking for certainty where none can be shown to exist. Realism is a good position in ordinary life, because it comes directly from our experience of naive realism. But cognitive science, incompleteness theorems, quantum mechanics, and other modern empirical disciplines present us with a consistent picture that there is no quantifier variance in our natural language ontology because, as Kant has argued, there is no ontological certainty. The best explanation of that is simply that our brains and bodies construct our internal representations.

  • 1
    Good answer, but it would be enhanced by an explanation of Bell ‘s Inequality and how Bell defined realism. He held locality to be interactions that are constrained by a light speed cone. And realism was that a universe has a state that is independent of the observer. Bells Inequality, if it holds, prevents physics EITHER from holding by locality, or realism (or both). Independently he showed that QM has to be non-local. Realism or not was still an open question, but it IS defined.
    – Dcleve
    Dec 18, 2023 at 0:48
  • 1
    @Dcleve My pleasure interacting acting with you is one where I must evince humility. I'm adding your comment and a link to SciAm.
    – J D
    Dec 18, 2023 at 16:26

Honestly it seems just as appropriate given your post to ask is realism coherent, given unsettled measurement and particles not being in definite states pre measurement.

The standard theory does not fully define measurement, see John Bell’s criticisms on this. Thus in the standard theory (textbook) one can easily be anti-realist given that measurement is not satisfactorily defined.

Honestly that’s all one needs to be at least coherently anti-realist. Measurement is so poorly defined, if it is at all, that anti-realism is perhaps more warranted than not. John Bell is clear that measurement has to go, and that’s like prototypical anti-realism.

Yes one can coherently be anti-realist about the most successful and accurate physical theory ever. Many philosophers are anti-realist about math for that matter eg William Lane Craig or Jody Azzouni.

  • As best I can tell from 10 seconds of Googling, Azzouni rejects anti-realism and is nominalist when it comes to maths, as in he doesn't consider mathematical concepts to actually exist. To call that "anti-realism" seems to greatly muddy the water by conflating a rejection of mind-independent reality with merely rejecting the existence of one thing or one set of things. If I for example reject the existence of a unicorn in my back yard (or an integer, if you will), it seems absurd to suggest that's comparable to rejecting the existence of all of reality outside of one's mind.
    – NotThatGuy
    Dec 17, 2023 at 16:18
  • @NotThatGuy I don't get the muddying claim tbh. If you want to say abstract objects of mathematics aren't real/don't make math true, you are an anti-realist. You can still take mathematical assertions as true but not due to abstract objects, which is what Azzouni does I believe. A nominalist, non-fictionalist (where mathematical statements are literally false).
    – J Kusin
    Dec 17, 2023 at 16:27
  • Metaphysical anti-realism (MAR), scientific anti-realism (SAR) and mathematical nominalism (MN) all seem to exist within fundamentally different categories, with the MAR being a rejection of what is clearly perceptible, SAR seeing scientific models as ... models, which are estimated representation of what we perceive, rather than being what we're actually perceiving, and MN rejecting the idea that the models themselves are objects that exists, independently of what they may be representing. You shouldn't conflate all of those under the banner of "anti-realism".
    – NotThatGuy
    Dec 17, 2023 at 16:40
  • @NotThatGuy Agreed they are all different. "You shouldn't conflate all of those under the banner of "anti-realism" -- I'm not, I specifically said anti-realist about a physical theory, and anti-realist about math, not anti-realism as a single category. I brought up the math anti-realists not to unite with the scientific, but to draw a few parallels, as there are some. One is Craig saying upon learning about the mature anti-realist mathematical position, it solved 15 years of philosophical issues for him. Quite possibly the same is true for mature, coherent scientific anti-realism for someone
    – J Kusin
    Dec 17, 2023 at 16:57
  • 2
    I think the quip of mathematics only increases the salience since QM is to no small extent woven from mathematical physics. Mathematical intuitionism and the rejection of hidden variables are kin, right? Intuitionism says that quantum waveforms are mathematical constructs to describe empirical observations, and the attack on hidden variables says that there are limits to mathematical knowledge from the constraints of empirical observations.
    – J D
    Dec 17, 2023 at 19:19

Is anti-realism coherent?

Realism is fundamentally the idea that there is a reality and that there are a number of invariants in it. Typically, we see the sun going up over the horizon as an invariant. Scientists went after more fundamental, and really invariant, "invariants". The speed of light seems to be an invariant, but elementary particles are not.

The real problem, however, is that our own logic is invariant, at least for each of us individually. We don't change logic over the course of our lifetime. We also don't seem to have changed it since Aristotle. It seems likely it comes with our DNA, so that as long as we stay humans, and the same humans, we won't see our logic change. The point of this is that it is our logic which requires that we identify invariants in the world. This is our way of understanding how the world works. This is crucial not for any metaphysical reason. It is crucial for our survival. If there are no invariants, then anything can happen, and we won't be able to do anything about that. We need that there be invariants for the sake of our own survival.

The history of science is also the history of a shift from one purported set of invariants to another, beginning with common sense invariants like water, fire, earth and air, and now possibly four fundamental forces.

However, there has been a conceptual shift. Although we normally take our invariants to be really invariants, we are nonetheless able to change to different invariants whenever the previous ones let us down. Scientists behave the same. However, some philosophers think in terms of the final, real, ultimate invariants. Thus, they might want to say that there is a real world if there are final, real, ultimate invariants.

This is fine but when do you know that there are or that there aren't? This isn't over till it's over.

It is our nature to look for invariants and we will keep doing that if we are not somehow prevented from doing it. Dropping our invariants when we realise they are not really invariants is what we have done for the last 2,000 years so we can do it again. We may become somewhat despondent if we don't have an alternative yet, but this doesn't mean that there are no final, real, ultimate invariants.

Another aspect of the problem is that while we can explain quite a few things in terms of other things, we cannot explain reality in terms of anything because there is simply no other thing that reality. So, we cannot expect to ever be able to explain, logically, reality itself. And then, maybe reality is the only final, real, ultimate invariant. Maybe everything else is fundamentally variable.

If this is so, this wouldn't be any problem for us humans. What we really need are not final, real, ultimate invariants, but things which are sufficiently stable so that we can make reliable plans to insure our survival. So, provisional invariants are normally good enough. We have dismissed air and water as potential invariants but we are still relying on them for our survival.

The question of whether realism or anti-realism are coherent or not is moot. It is the typical metaphysical question. Whatever the answer is, it won't make a dent on our prospect of survival. We are not ourselves invariants, and we have a rather short shelf-life. All we need is that nature be sufficiently stable and for now it is.

Yet another aspect to the problem is science itself. By looking for more stable invariants, science may well uncover a situation where the more fundamental, the less stable, which is perhaps what quantum physics says with quantum randomness. However, if quantum randomness is real, then it is the final, real, ultimate invariant. And quantum randomness perhaps only produces relatively short-lived invariants at the macroscopic level, which, again, is fine. It that wasn't we wouldn't even be there to talk about it.

There is no good reason to believe that we should be able to explain everything.


Practically, a measurement in QM is an interaction between the quantum object whose properties you want to measure and a bunch of other objects, arranged by the experimenter in a way that yields a particular result. A lot of quantum 'measurements' are indirect. Take the measurement of spin direction, for example. You fire an electron past the poles of a fancy magnet which causes the path of the electron to veer in one of two possible directions, and you actually figure out which by detecting where the electron ends-up (ie you directly measure its position), from which you can work out whether its spin is up or down.

The fact that it has a spin upwards or downwards is an outcome forced upon the electron by the interaction with the magnet, so the act of measuring actually changes the state of the particle. It would be a bit like looking at a pile of empty bottles at a recycling plant, each of which is pointing at a random angle, and saying you are going to 'measure' the angle by shaking the bottles over a grid through which they can either pass pointing up or pointing down. The orientation each bottle has as a consequence of your 'measurement' is not the orientation it had beforehand, but one you have forced it to adopt.

In QM, a particle does have a state before a measurement, just not a state that's necessarily associated with the observable being measured. To use the QM terminology... after the measurement, the state of the particle is an 'eigenstate' of the observable measured, before the measurement it might not have been, and if it wasn't, then the measurement forced a change of state on the particle. (If you are not happy with the word 'forced' there, you can substitute 'induced', 'facilitated', 'coincided with' etc according to your taste.)

QM is a very poor source of ontological insight. However, particles (whatever they are) do have 'states' (in the QM sense of the word) at all times. The states don't have to be associated with clear-cut values of observable properties, and generally, if you perform an experiment to 'measure' a property, what you are actually doing is subjecting the particle to some kind of influence that changes its state. Moreover, there are combinations of properties that are incompatible, in that having measured one, measuring the other changes the particle's state in a way that invalidates the first measurement. Consider the glass recycling plant- having 'measured' the vertical orientation of the bottles by dropping them through a horizontal grid, they fall flat on a conveyor belt and we 'measure' their horizontal orientation by passing them through dividers, as a result of which they no longer have a vertical orientation.

Quantum states are therefore quite unlike classical states, where you can say that a particle has a specific position and a specific momentum. That all said, quantum objects do seem to be there, and they go about their quantum business whether we are observing them or not, so in that sense, at least, they seem real enough.

  1. Concerning quantum mechanics (QM) Bell’s theorem states that a local hidden-variable theory cannot explain certain experiments on the level of microphysics.

    Hidden-variables imply realism in the following sense: The outcome of a measurement is causally determined. It is independent whether the observation is actually achieved or not.

    Locality has been glossed as “The direct causes (and effects) of events are near by, and even the indirect causes (and effects) are no further away than permitted by the velocity of light.” In particular, information can be transmitted at most with the speed of light.

    Bell’s theorem derives a mathematical inequality which separates

    • the results from local hidden-variable theories

    • from results of experiments interpreted according to the formalism of QM.

    In the meantime, experiments have been achieved and confirm the results predicted by QM. The key characteristic of these results is that the observation of one particle predicts the outcome of a measurement at a different second particle. This coincidence relies on the phenomenon of entanglement: The two particles are companion particles, forming the two components of a single system.

  2. What is a measurement?

    A measurement is the interaction of two systems such that information about the first system is indicated by the second system. The first system is considered the system to be measured, the second is considered the measuring apparatus. The mathematical formalism of a measurement according to QM has been formalized for the first time by John von Neumann in the 1930s, see any graduate level textbook of QM.

  3. If a particle is not in a defined state, what is it exactly ontologically?

    A mathematical basic concept in QM is the wave-function which solves the wave-equation of a physical system , e.g., the function psi which solves the Schroedinger equation or the Dirac equation. According to Born the wave function of a particle determines its state in the following sense: The square of the modulus of psi = psi(x,t) is the probability density to find the particle at time “t” at position “x”.

    In a radical manner the Copenhagen interpretation (Bohr, Heisenberg) dismisses to attribute any physical property to the particle between two observations. There is only a probability density for the outcome of a future measurement. The measurement then makes the change from the world of possibilites to reality.

    Physics does not answer the question about the ontological status of a particle. But physics keeps the concept of an elementary particle as a separate entity, which can be successfully described by the laws of QM.

    But already the question whether two particles can be distinguished as two individual objects has lost any meaning when explaining scattering experiments according to quantum field theory.


Ernest Rutherford fired alpha particles at a gold foil target and, by observing the deflections, discovered that the gold atom nucleus was relatively small. That was how that proportional property was measured and became an established fact, or in Kantian terms, the existence of the property was actualised. Before it was known it was unobserved and science is fairly conservative about ascribing existence to things that have never been observed, for example Newton's concept of absolute time, discussed in Nature, 1937, still to this day considered non-existent.


To put it as concisely as possible, I would say that when you hear about 'non-realism' in Quantum Mechanics, it's not to be understood as a statement that NOTHING is real - merely a statement that the things we're used to thinking of as real are not real, or at least not as tangibly solidly real as we intuitively think.

Now to put it less concisely:

The reality underlying quantum mechanics is still in debate - there are so many interpretations, each with a different vision of what's really happening - and some of them keep what I call "naive realism" in tact, and others replace naive realism by giving reality to other constructs instead.

Naive realism is just simply the idea that, say, you throw a ball and look away and the ball continues to exist in a definite place and traveling at a definite velocity at every moment in time, even if no one is looking at it. This is also referred to in physics as "classical mechanics" or "classical physics".

So one way of conceptualizing non-realism in QM (which I wouldn't actually call anti-realism, I think that's opening it up to being confused with the metaphysical philosophical position of anti-realism) is, instead of an Electron existing at a definite place and flying at a definite velocity at every moment in time, even when unmeasured, instead think of it as The Universe is calculating all different trajectories and states that Electron could be in. The Universe is calculating what it would be like if it went this way, and if it went that way, and that way -- this is the "superposition".

The question, at the end of the day, that anti-realism poses isn't "is anything real?", it's "what things are real?" Because even if "the electron is at this place traveling at this velocity" isn't real, there's still something about the electron that's real even between times of being measured.

You must log in to answer this question.