Experimental rejection of observer-independence in the quantum world
arxiv.org
arxiv.org
Also not directly related video if you like your mind being blown: https://www.youtube.com/watch?v=8ORLN_KwAgs
Especially given how our understanding of the world changed during the last 200 years, I would say we are still very much clueless. I find that exciting.
Btw, no idea why did researchers get "freedom of choice" involved.
Personally, I don't see any reason it isn't a figment of the mind. Nobody is debating whether love exists. Our ability to model counterfactuals in our mind doesn't have any broader philosophical implications, so I don't see why free will would, when that may as well be all that it is.
Perhaps such a system could predict our actions and intervene when some poor outcome is predicted and direct us toward more positive outcomes?
This statement itself depends on the logic which are you supposing might be deprecated.
"There's a divinity that shapes our ends/ Rough hew them how we will."
Hamlet, to Horatio, in Hamlet Act 5 Scene 2.
The question therefore is: is God’s omniscience equivalent to predestination?
Some think yes (no free will. some Protestants and Muslims). Some think not (therefore we are free. Catholics, ppl who reject omniscience.)
Personally I like Tolstoy’s take on the whole thing in WnP, because it betrays how deep his understanding of physics is (he even scratches at chaos theory).
We are the needle. The needle responds to the environment of the record but can't see outside it's little perspective.
God on other hand sees the record from above all at once.
We are a consciousness needle scratching a path through 4D space-time.
And death is when the needle has scratched the record so much that it has become worn flat, producing just silence. The record is still being played but the effect is all the same.. Until a new record is placed on the player. Do you identify yourself with the record or the needle? ;-)
What’s the definition of “free will”?
Think of God as an observer at the top of the Grand Canyon, looking down into the winding path of the river below. Humans are afloat on the river down in the canyon, unable to see what lies ahead. They can choose how to react to obstacles, for example, yet do no see them coming. God, however, sees our entire path on this metaphorical river laid out for us since birth; each of our individual choices are "perceived"[0] by him simultaneously.
[0]: Note, however, that many theologians would protest anthropomorphizing God in this way. Maimonides, as an example, says that we can speak of God as exhibiting mercy but not being merciful. He simply acts in a way that is analogous to human mercy.
Furthermore, in this metaphor God doesn't just see the river, He created the river, and all the people on it. He laid out its twists and obstacles, with a perfect understanding of how it would affect the humans afloat on it. Moreover, at the moment of creation, He had an unbounded freedom to create the river in any other way He saw fit. If I get smashed on a rock half way down, it's because He chose to create the river in such a way that I would get smashed on it. If He didn't know this, or didn't have a choice, then He is either not all knowing, or not all powerful.
This, for me is the fundamental problem with any theist philosophy that holds that God is both omniscient and the prime mover of creation, but still suggests concepts like free will and morality have any meaning. Not only are "our" decisions predestined, they're a direct result of another being's free will, and the concept of us then being judged on them by that same being is farcical. It's like a puppeteer putting one their puppets on trial.
We have free will to the extent that we can seperate internal and external forces, and to the degree that internal forces play the causal role in any given event. Of course, in reality, this is just an abstraction, and there's no true seperation between internal and external just as there's no real seperation between bodies of water, but the categories are useful, so we use them.
It's actually not. The assumption that determinism precludes moral responsibility has turned out to be false.
When did this happen? I missed that memo.
If you want a review of the failures of every challenge meant to demonstrate how moral responsibility and determinism are incompatible, and that Compatibilism empirically matches our use of moral language and so is what people actually mean by free will, that will take more time.
Compatibilism is a kind of comfortable opinion, because it simply avoids thorny questions regarding logic, and morality. E.g. C-sts belive that a criminal is guilty, when he allegedly acts in accordance with his will, because, as you noted, that's what people mean when they say "free will". However, it still may sound unconvincing when one believes that the criminal's mind contents was predetermined aeons before he was born.
Standards of evidence for HN are lower. Calling something false needs a preponderance of evidence, not a formal proof. Centuries of challenges attempting to show an inconsistency have all failed.
> However, it still may sound unconvincing when one believes that the criminal's mind contents was predetermined aeons before he was born.
Anything can be made to sound unconvincing when it's completely divorced from its context. The point in context, is that bad behaviour felt justified in the moment, and holding a learning agent responsible is the moral feedback needed to correct that error in judgement.
That's exactly where the trouble is. In case of strict logical determinism we correct nothing because everything is predestined. However, choosing limited definitions, and restricting ourselves in comfortable contexts we can certainly avoid discomfort of the thought. Not that I have better idea, but I see clearly why non-compatibilists say "word juggling" towards compatibilists.
A predestined correction is still a correction though. Unless you assume "correction" to have some bizarre and probably incoherent non-deterministic meaning. In context, "correction" means "learning to choosing correctly as opposed to incorrectly", which is precisely what happens in this scenario.
> However, choosing limited definitions, and restricting ourselves in comfortable contexts we can certainly avoid discomfort of the thought.
I dispute this characterization. It's not "limited definitions" and "comfortable contexts" (presumably pejoratively), it's "appropriate definitions" and "relevant contexts".
> but I see clearly why non-compatibilists say "word juggling" towards compatibilists.
I think you'll find that the incompatibilists are just as guilty of this. They've spent centuries arguing for a definition of free will that no one actually holds. Philosophy is sometimes about drawing fine but relevant distinctions, and that's exactly what Compatibilism does: it carves out a fine line behind which we can make sense of holding intelligent learning agents responsible for their choices.
This statement is incorrect, and actually prove the very thing you intended to deny. It's not like adherents of C. held a plebiscite, isn't it? (And if they would, I dare to say they would be in danger of huge disappointment.) They made a choice to call an action free when it's inescapable consequence of a butterfly flapping wings millenia ago on another continent, but not free if it's e.g. equally inescapable consequence of a another human doing something recently, and closely (even if this forcing person is equally inescapably forced by a butterfly wings before). It is not, however, some sort of obvious conclusion, accepted by everybody. It's something C. philosophers prefer to call a common sense, because... well because this way we can make sense of things we do. So it feels much better.
As a side-note, your explanation of 'correction' makes perfect sense if you took the C. pill already, but doesn't make any useful proof of C. One have to accept that the only available variant of future (if we are in a purely determinist universe) still presents a free choice between correct/incorrect. Which is a circular reference to basic presumption of C.
Anyway, thank you for taking time to write the long response!
Addiction and marketing both prove beyond a shadow of a doubt that we don't have "perfect" free will, and this is both (1) true regardless of what you think about determinism and (2) more relevant than any consequences of whether the universe is deterministic or not.
The pragmatically relevant aspect of "free will" is to what extent our actions can be predicted and/or manipulated by others today and in the future. This necessarily involves e.g. measurement and computational limitations that others have, and it's something that occurs on a scale rather than as a binary.
If you have no choice, than that property must have been predetermined to have a specific conspiratorial outcome, essentially since the beginning of the universe. Just to make sure you and your friend make compatible measurements.
Without that assumption, every outcome can be explained as due to a highly unlikely conspiratorial setup.
The "free will" debated in philosophy is not the same as the "free will" physicists talk about. The former is about moral responsibility, the latter is about being able to set experimental parameters independently of the system you're measuring.
From what I understand, this basic idea of 4D geometry also provides some of the more elegant explanations for the effects of relativity when approaching the speed of light, because time dilation effects reduce down to (relatively) simple 3D trigonometry in a 4D graph.
Of course, that raises predestination questions, because you're treating all of time and space as a single "object".
Pardon my basic understanding, but doesn't it meant that it's both simply because it hasn't been measured - and therefor we don't know? Meaning it's either one or the other; at a very fast rate, it's both.
Pardon any incorrectly used terminology above, but I believe my explanation is conceptually sound.
A photon can have one of an infinity of polarizations. These states are describable as a mixture of any two orthogonal polarizations (a "basis"), such as "horizontal" and "vertical". "Left circular" and "right circular" are another choice.
The funny thing happens when you measure "how much" a photon is polarized into any given orientation. It appears as being polarized either in that orientation, or in the polarization complementary to what you measured. E.g., horizontal or vertical. The outcome is determined with a probability based on the actual polarization of the photon prior to your measurement. Then – the funny thing – the photon takes on that polarization. Repeated measurement using the same basis gives the same results. Measure again with a different basis, and the game starts over again.
You can see this in the macro world with a laptop screen and two pairs of polarizing sunglasses. Using one pair of sunglasses, you should be able to block the light from the screen entirely – you are holding them perpendicular to the polarization of the screen. Place the second pair in between the first and the screen, and rotate it. At certain orientations, you will be able to see the screen! The second pair (nearest the screen) is diagonally polarized, and will "convert" the light from the screen into that polarization, before it reaches the first pair (furthest from the screen), which is orthogonal to the screen only.
(Note: Things change completely once a second photon enters the picture. It can be entangled with the first, and their states are no longer independently describable. Measurement of one affects the pair.)
A common misconception about quantum superposition is that the particle "has" some state, and we only use superposition to model that we "dont know". This is called the hidden-variable theory (https://en.wikipedia.org/wiki/Hidden-variable_theory). It's been proven that this hidden variable theory can't be true because of https://en.wikipedia.org/wiki/Bell%27s_theorem.
This is completely non-intuitive. See this video for a laymen's explanation: https://www.youtube.com/watch?v=z-s3q9wlLag (you may also need to watch the prev. video about the EPR paradox https://www.youtube.com/watch?v=5HJK5tQIT4A).
Edit: actually, this is a much better video https://www.youtube.com/watch?v=ZuvK-od647c - less technical and more visual.
It is easier to understand if you look at photons position. Try to calculate the interference between any two waves, and compare to the odds of finding a photon somewhere it they were dispersed proportional to the waves amplitudes at the start.
Let’s modify the thought experiment: Wigner’s friend measures the photon’s polarization and determines that it is horizontal. He then calls Wigner on the phone and tells him the result. Does this affect the result that Wigner sees when he verifies superposition? What if the friend lies? The post you linked seems to specifically call out that the friend cannot reveal the result of his measurement, lest the experiment get spoiled.
This all seems to point to some philosophical/theological Mumbai jumbo with regard to consciousness being a necessary ingredient here, which doesn’t hold water. Yet with every thought experiment like this that I have seen, this kind of clause is included. Can anyone explain it like I am 25?
Somewhat relatedly[0], I can't find an explanation of the Oberth Effect that makes any damn sense. Every so often I try again and leave disappointed. People are happy to provide them. There are lots of them. They attribute it to all sorts of things. Not one of them I've read makes sense.
[0] i.e. is physics, but is not quantum physics. So far as I know, anyway.
It's simpler than people make it. When a spacecraft is moving at a high velocity, the amount of kinetic energy embodied by its fuel mass is a significant fraction of the chemical energy already present. And kinetic + chemical is greater than chemical alone. That's all there is to it.
I'll take a crack at it:
Kinetic energy plus gravitational energy is conserved when you aren't actively accelerating[0]. Rocket fuel has kinetic and/or gravitational energy in addition to it's chemical energy. If you burn the fuel while going very fast[1], you can extract it's kinetic energy by bringing the exhaust to(wards) a stop. When going slowly[2], you end up spending energy to accelerate the exhaust away from you, and you also can't really do anything to get at its gravitational energy[3].
0: Ie, thrusting your engines or crashing; things that involve g-forces; not freefall.
1: That is, when most of its gravitational+kinetic energy is in kinetic form.
2: With energy mostly in gravitational form.
3: Gravitational energy is a function of where mass is at that exact moment; if you're high away from a gravity well, you have lots of energy tied up as gravitational, and no quick way to turn it into a more useful form.
But straight "quantum physics" covers lots of other, more solid, areas. Including understanding how lots of things you can actually manufacture work, like semiconductors and superconductors.
It's definitely worth the 10 minute watch if you want it explained to you.
On the road to Damascus and enlightenment, stunned by a bystander's use the term "Mumbai jumbo", Paul fell off his horse and into the stones beneath a set of railroad tracks. Upon reaching consciousness again he saw an old man who extended a hand, smiling, nodding and, speaking the phrase "mumbo jumbo" re-assuredly and repeatedly, helped Paul to his feet.
Exactly. The universe would be rather boring if we learned everything about it during my lifetime.
https://www.scottaaronson.com/blog/?p=3975
Here's Scott Aaronson arguing that the theoretical work underpinning this experiment is flawed.
It reminds me how sometimes converse is confused with contrapostive.
Reflecting on this a second time, I wonder whether it is possible to interpret this as the universe not caring until the information interacting actually matters. For example, if someone tells me that they have measured a value, they could very well be lying, and until I say "prove it, what value did you measure?" the underlying implementation of the universe is in a superposition where they both did and didn't measure the value (where the friend is telling the truth vs lying).
If they generated a random bit and lied depending on that value, then it would be.
This paper about the experiment is trivially true, there were no surprises regarding to what QM predicts.
Here is a direct quote from Scott: >My comments about the Frauchiger-Renner paper never doubted for a second the result predicted by QM. So an “experimental demonstration” of their setup—by which one really means an experimental demonstration of Hardy’s experiment, since the Frauchiger-Renner one would require superposed conscious observers (!)—has zero effect on any of those comments.
And here is a prescient direct quote from Craig Gidney: >I wonder how long it will be until someone 1) executes the circuit from Hardy’s paradox (which has very few qubits and very few gates, so you could run it on a NISQ machine), 2) confirms that of course the results match the predictions of quantum mechanics, 3) frames the result in terms of Frauchiger et al’s variant, and 4) journalists telephone-game it into something like “quantum researchers prove decisions aren’t real”.
Please see also the relevant discussion in comments here: https://www.scottaaronson.com/blog/?p=4147
My point in linking the original article was that the media discussion of this (in part driven by the paper's own framing) as disproving objective reality is wrong, something Scott clearly agrees with.
This experiment is certainly interesting, and it realizes the basic structure of the multiple Wigner's Friend scenario that a number of theoretical papers have been written about recently, but I don't think it comes anywhere close to probing the actual issues raised by those theoretical papers. The kinds of real measurements that would have to be realized in order to probe those theoretical issues are many orders of magnitude more complex than those in this experiment.
In QBism, the Born rule is normative in that it tells the observer which observations to bet on, and observations are data points to update the observer’s subjective assessment of outcomes. Taking this perspective, a lot of paradoxes in QM seem to get resolved trivially. It’s controversial for sure but mathematically it’s rigorous.
> But this is in stark contrast to the point of view of the friend, who has indeed measured the photon’s polarization and recorded it. The friend can even call Wigner and say the measurement has been done (provided the outcome is not revealed).
A bit of a stretch saying that protons lose their superposition based on the content of a conversation.
What if you kind of hear me and are pretty sure I said 'positive' but not 100% sure? What if you're 50% sure? What two people are listening and one heard me say 'positive' and one heard 'negative'?
It's hard to imagine a physical process being dependant on something external and near impossible to quantify or measure.
The conversation simply represents the idea that the causal cone of the observed system is now interacting with the particles of the observer, making them entangled and so ‘collapsing’ the wave function or whatever you want to calm it.
Information is physical, although I agree that the conversation isn't the important part, it's just a useful metaphor.
What if what they're proving is really the abscence of a superposition state? I never understood how superposition can be verified independent of assumptions. In this case,how can Wigner measure the photon has veritcal and horizontal polarity at the same time?
Lastly,could superposition be explained by human's limit of sampling reality? For example audio recorded at 10Hz means 10/samples of audio a second,a sort of audio-resolution. If our reality-resolution is low,some things might require ability to measure their state at a better time-space resolution than what humans can currently sample (or build equipment that can measure this)? But with limited time-space resolution it can appear as if the object is in two states at the same time. Sort of like watching a movie at a very slow frame rate and you capture a moment where half the screen is of one frame but the rest is of a different frame since the display is only half finished rendering the raster/rows.
Apologies if I am being absolutley clueless.
So it's not a matter of our being unable to make detailed enough measurements; we can actually observe phenomena that directly indicate superposition happens.
[0] https://en.wikipedia.org/wiki/Double-slit_experiment [1] And so are electrons, particles, humans, etc.
Depends what you mean by "exist". All that needs to be true - and is true - is that QM provides an accurate model of the experiment. For many of us, QM is just probabilistic mechanics; i.e. probability theory applied to mechanics[1], and so long as you maintain the distinction between (epistemic) probability space states and (ontic) configuration space states (something the article under discussion here unfortunately fails to do), no ontic existence of [superposition] states is ever assumed in the first place.
[1] More precisely, a natural and necessary generalisation of the classical, Kolmogorovian probability theory you might be familiar with (necessary because of the noncommuting random variables):
https://news.ycombinator.com/item?id=19315457
See also: http://blogs.discovermagazine.com/cosmicvariance/files/2011/...
> We note that, although Bell’s mathematical machinery [9] is used to show the result, the set of assumptions considered here — and therefore the conclusions that can be drawn from a violation of inequality (2) — are different from those in standard Bell tests.
That statement contains a lot of handwavium and in fact the interpretation of the results depends a lot on, if this experiment may be interpreted as another test on locality. Which I strongly presume, it does.
https://motls.blogspot.com/2019/03/six-photons-claimed-to-pr...
How could they record their high entropy (?) solipsistic observations in an immutable datastore in such as way as to have probably zero knowledge of the other party's observations?
Anyways, that's why I only read the title and the abstract.
Wigner's friend experiment: https://en.wikipedia.org/wiki/Wigner%27s_friend
If there ever was a repeatable state, I'd bet it would be at thr big bang, though i am no physicist.
Certainly not hurting anyone to publish a "repetitive" paper. Worst thing that happens is more people have a chance to hear about it.
Experimental confirmations of theoretical predictions are very valuable.
Quick counterexample: By that logic, do you think results showing agreement with Einstein's GR should have never been released? I'm taking a wild guess that your answer is "no". This case is not any different.
Theoretical research leads to applied science which ends up in products. Remember that GPS is only possible thanks to relativity, and e-commerce only works thanks for number theory.