5 ms·

> It's possible that it might be deterministic, but there is not currently any basis for certainty other than the fact that so many things we've seen turned out

> It's possible that it might be deterministic, but there is not currently any basis for certainty other than the fact that so many things we've seen turned out to be deterministic That's how empirical proof works though - you can't "prove" something conclusively based on evidence, you can only disprove it or continue to add examples of the model matching reality. That's falsifiability. > Currently the most widely concidered deterministic theory is that there's a deterministic multiverse None of these are provable either way, because we can't prove theories that rely on hidden information, like alternative hypothetical universes. The only useful thing we can do is to model observable behaviour, and as the other replier stated, we have a functioning deterministic model of QM. so going by the evidence we currently have, the universe appears deterministic and the statistical bridge between QM and classical mechanics is essentially a hack to wire up two different models. > Einstein famously agreed with your interpretation, but his invoking of god to make his case is quite appropriate. He didn't literally invoke god - he used god, as we often do, as a metaphor for the laws of reality.

Retric 5y ago
You can disprove theories based on hidden information. Locality or determinism pick one.
- simiones 5y ago
  I'm assuming you're referring to Bell's inequality, but that has little to do with determinism. Bell's inequality says "locality or hidden variables, pick one". In fact, "superdeterminism" can actually render Bell's inequality moot. That is, you can have locality + hidden variables + superdeterminism. You can't have locality + hidden variables + [statistical independence between the measurement settings and the state you are trying to measure]. The existence of hidden variables is also sometimes called "realism" or "counterfactual definiteness" - since if there are no hidden variables, it means that the question "what would another measurement have shown" or, equivalently, "what state were the particles in before the measurement" doesn't make sense. Indeed, in the Copenhagen Interpretation, particles don't have any meaningful classical "state" before a measurement. For example, if you chose to measure their position, they will have a position, but can have any spin; if instead you measure their spin, they will have a spin, but no definite position. You can't then ask "but what spin would they have had if I had measured their spin instead of their position" - there is no answer to that question according to CI (or, there is an infinity of answers according to MWI). Note that superdeterminism simply means "the measurement settings on the apparatus depend on the real state of the particles". That is, the particle actually has a property lambda such that if, say, lambda=0, when the particle hits the apparatus and the apparatus is set to measure position, it will show a definite position; whereas is the the particle hits the apparatus when it is set to measure spin it will have a spin. For a different lambda, a different pair of <apparatus settings, apparatus results> will be expected.
  - Retric 5y ago
    Superdeterminism also gets rid of the need for photons etc. If you’re instruments are running a predetermined script they can just as easily be playing back a movie as having an actual detector. Really though it’s just another flavor of non-localism that gets around FTL.
    - tsimionescu 5y ago
      Not necessarily, superdeterminism is a concept, not a specific theory/model. You have specific models that can basically explain anything at all as a "grand conspiracy", but you can also have models that don't. There are no useful, testable "superdeterministic" models yet, to be fair. But they can, in principle, exist. Again, all that this requires is a model where the measurement settings depend on the hidden variables themselves.
      - Retric 5y ago
        That’s the problem, we aren’t talking about a specific theory so you need to toss in all the suddenly possible models under stuff super determinism allows. Excluding the ultimate “grand conspiracy” stuff that only applies to scientific instruments it posits that basically every interaction between any particle is following a predetermined script. You can of course still have photons etc, but like two actors in a play such particles don’t need to be having a conversation. The idea also extends into all sorts of strange and untestable territories such as our universe being a copy of a prior one that operated on different principles. Of course none of that means such an idea is false, just that it’s currently as unproductive as saying god did it.
simiones 5y ago
To be fair, going by pure empiricism, it makes more sense to say that the world is non-deterministic. QM only matches observations iff we assume that the measurement process is truly non-deterministic. The Schrodinger equation can only be used, in practice, to measure probabilities, we don't know of any way to use to predict exact results. However, philosophically speaking, this is unsatisfactory, since the behavior of the macroscopic world must somehow be reducible to the behavior of its constituents. If Schrodinger's equation were non-deterministic itself, we wouldn't have a problem - we could simply say "the world is nondeterministic, as far as we can see". If it at least was non-linear, with a possibility of chaotic effects, we could say "the world is probably deterministic, but it's impossible to measure it accurately enough to predict any definite outcomes" - but this is not the case. So, we are stuck with this weird dichotomy between philosophy&logic on one hand and empirical observations on the other. MWI does give an out, but it has extra assumptions that are not directly justified any more than the Born rule (an observer only observes one world), and, more importantly, while it reproduces the Born rule in concept, it neither can derive it quantitatively. Even worse, MWI can't explain how the classical world, where certain specific quantities are definite for all objects, arises out of the quantum world - the preferred basis problem. For example, a world where we observed some balls having a definite position, but others having a definite spin but not definite position, would be consistent with QM and the Born rule - and this indeed can happen in an experiment with particles. And yet we observe the same observables for all classical objects in our day to day lives.
- beaconstudios 5y ago
  I suppose it depends on whether we're simply stuck at too high a resolution of measurability due to Heisenberg uncertainty - either subatomic behaviour has a legitimate statistical factor, or we simply can't look low enough to see the deterministic factors (yet?) This is straying into territory I can only hypothesise in though - for all I know the latter option may have already been ruled out. Does this dilemma map onto hidden variables vs Copenhagen interpretation? My primary point of reference for these things is philosophical as I have an only-slightly-above-pop level of understanding of QM and am leaning more on second order cybernetics and epistemology to reason about observer-system interaction.

Retric 5y ago

You can disprove theories based on hidden information. Locality or determinism pick one.

simiones 5y ago
I'm assuming you're referring to Bell's inequality, but that has little to do with determinism. Bell's inequality says "locality or hidden variables, pick one". In fact, "superdeterminism" can actually render Bell's inequality moot. That is, you can have locality + hidden variables + superdeterminism. You can't have locality + hidden variables + [statistical independence between the measurement settings and the state you are trying to measure]. The existence of hidden variables is also sometimes called "realism" or "counterfactual definiteness" - since if there are no hidden variables, it means that the question "what would another measurement have shown" or, equivalently, "what state were the particles in before the measurement" doesn't make sense. Indeed, in the Copenhagen Interpretation, particles don't have any meaningful classical "state" before a measurement. For example, if you chose to measure their position, they will have a position, but can have any spin; if instead you measure their spin, they will have a spin, but no definite position. You can't then ask "but what spin would they have had if I had measured their spin instead of their position" - there is no answer to that question according to CI (or, there is an infinity of answers according to MWI). Note that superdeterminism simply means "the measurement settings on the apparatus depend on the real state of the particles". That is, the particle actually has a property lambda such that if, say, lambda=0, when the particle hits the apparatus and the apparatus is set to measure position, it will show a definite position; whereas is the the particle hits the apparatus when it is set to measure spin it will have a spin. For a different lambda, a different pair of <apparatus settings, apparatus results> will be expected.
- Retric 5y ago
  Superdeterminism also gets rid of the need for photons etc. If you’re instruments are running a predetermined script they can just as easily be playing back a movie as having an actual detector. Really though it’s just another flavor of non-localism that gets around FTL.
  - tsimionescu 5y ago
    Not necessarily, superdeterminism is a concept, not a specific theory/model. You have specific models that can basically explain anything at all as a "grand conspiracy", but you can also have models that don't. There are no useful, testable "superdeterministic" models yet, to be fair. But they can, in principle, exist. Again, all that this requires is a model where the measurement settings depend on the hidden variables themselves.
    - Retric 5y ago
      That’s the problem, we aren’t talking about a specific theory so you need to toss in all the suddenly possible models under stuff super determinism allows. Excluding the ultimate “grand conspiracy” stuff that only applies to scientific instruments it posits that basically every interaction between any particle is following a predetermined script. You can of course still have photons etc, but like two actors in a play such particles don’t need to be having a conversation. The idea also extends into all sorts of strange and untestable territories such as our universe being a copy of a prior one that operated on different principles. Of course none of that means such an idea is false, just that it’s currently as unproductive as saying god did it.

simiones 5y ago

To be fair, going by pure empiricism, it makes more sense to say that the world is non-deterministic. QM only matches observations iff we assume that the measurement process is truly non-deterministic. The Schrodinger equation can only be used, in practice, to measure probabilities, we don't know of any way to use to predict exact results. However, philosophically speaking, this is unsatisfactory, since the behavior of the macroscopic world must somehow be reducible to the behavior of its constituents. If Schrodinger's equation were non-deterministic itself, we wouldn't have a problem - we could simply say "the world is nondeterministic, as far as we can see". If it at least was non-linear, with a possibility of chaotic effects, we could say "the world is probably deterministic, but it's impossible to measure it accurately enough to predict any definite outcomes" - but this is not the case. So, we are stuck with this weird dichotomy between philosophy&logic on one hand and empirical observations on the other. MWI does give an out, but it has extra assumptions that are not directly justified any more than the Born rule (an observer only observes one world), and, more importantly, while it reproduces the Born rule in concept, it neither can derive it quantitatively. Even worse, MWI can't explain how the classical world, where certain specific quantities are definite for all objects, arises out of the quantum world - the preferred basis problem. For example, a world where we observed some balls having a definite position, but others having a definite spin but not definite position, would be consistent with QM and the Born rule - and this indeed can happen in an experiment with particles. And yet we observe the same observables for all classical objects in our day to day lives.

beaconstudios 5y ago
I suppose it depends on whether we're simply stuck at too high a resolution of measurability due to Heisenberg uncertainty - either subatomic behaviour has a legitimate statistical factor, or we simply can't look low enough to see the deterministic factors (yet?) This is straying into territory I can only hypothesise in though - for all I know the latter option may have already been ruled out. Does this dilemma map onto hidden variables vs Copenhagen interpretation? My primary point of reference for these things is philosophical as I have an only-slightly-above-pop level of understanding of QM and am leaning more on second order cybernetics and epistemology to reason about observer-system interaction.