I’m not sure what to make of this, or if I even completely understand it. The claim is that different observers can have different objective realities. The claim is about things at the quantum level and not at our day to day macro level. That seems contradictory to me, but quantum physics defies common sense to begin with.
Maybe some of the physicists on here can explain it or explain why it’s incorrect. Maybe it’s just typical bad science reporting.
We are sitting across from each other. On the table between us is a box containing the cat and the vial of poison that breaks open at an indeterminate time. The cat is in the superposition of being both alive and dead at the same time until the box is opened and the cat’s condition is observed. At least, it is for you. From my side of the table, the box has a glass panel that lets me observed the condition of the cat. Thus we have different objective realities.
lt is almost exactly like that. Rashoman for leptons. Rashoton.
Different observers do have different realities. Or rather, different experiences of reality, since we construct our realities internally from analyzing our experiences.
in a large swath of areas, it doesn’t matter much. You know what parallax is? Where the position of a stellar object varies depending on the observer’s position? Well, for most of “reality”, the “parallax” between one human observer and another doesn’t amount to much, once they’ve really got all their observations in and fully analyzed.
That’s how we’re able to confirm each other’s experiences of reality and verify our observations. It’s not a case of “gee, reality is whatever the observer conceives it to be, everybody’s got a different reality”. At the same time, it’s technically true in all cases, and in some areas of life it’s quite true in a real sense, that our experience of the same thing is sufficiently different that to get anything approaching an “objective” understanding requires taking the other person’s experience into account.
It’s solid science, but this sort of thing should really be understood as clarifying the consequences of our commitments rather than yielding definite insights into ‘the nature of reality’, or however you may want to phrase that. Bell’s theorem is the original example of such an insight: the conjunction of certain assumptions (usually glossed as ‘locality’ and ‘realism’) implies certain consequences (usually given in the form of an inequality) that are incompatible with quantum mechanics. Hence, given that quantum mechanics is an empirically well-confirmed theory, we can’t any longer believe in the sum total of these assumptions.
What the theory doesn’t tells us is which idea needs to be thrown out. And that’s where the question of the interpretation of quantum mechanics comes in. Bohmian mechanics throws out locality: events can instantaneously influence one another across arbitrary distances. Copenhagen quantum mechanics, to the extent it’s a unified framework, gets rid of a single, observer-independent, unique reality—that is, denies that measured values have meaning beyond the explicit context of measurements (i. e. that you get a value of ‘spin up’ for a measurement on an electron because the spin of the electron just happened to be up, independently of your measurement). Many-Worlds quantum mechanics gets around things in an entirely different way—by denying that experiments have unique outcomes, it keeps a kind of realism—albeit a many-valued one—along with locality.
So experiments don’t finally decide the question of what nature is ‘really like’, at the bottom. And neither does this one—it’s acknowledged in the article: if one allows for superluminal signaling, as is present in the Bohmian framework, we can continue to believe in a single, observer-independent reality.
Well, it’s not quite like that. You’d have to be completely isolated from davidm, in a ‘box’ of your own; and davidm’s description would then be that of an entangled superposition between you having different experiences of whether or not the cat’s alive, and the cat’s actual state. In particular, davidm could no longer assign a simple superposed state to the cat on its own, but only to the joint system of the cat, you, and everything in the box.
So from his point of view, you actually fail to have a definite experience of the cat. That’s the crux of the ‘Wigner’s Friend’-thought experiment that’s at the heart of this experiment. The question is, essentially, whether there is some way—perhaps in a theory more general to quantum mechanics—to have an assignment of definite ‘facts’ to both you in the box, and davidm outside, leaving the disagreement between the two of you merely an issue of description.
An added wrinkle here is that you can tell davidm that you have observed a definite state for the cat, as long as you carefully avoid telling him which one. That means, the facts of the world from his point of view will include both that he describes the box you’re in as being in an entangled superposition, and that you’ve made some definite observation—which suggests that there is some definite fact of the matter of what you’ve observed that the quantum description of the system just doesn’t capture.
So then, we should expect that there are two facts that we expect to coexist: the entangled superposition of the box (which davidm could verify with an appropriate experiment), and the definite outcome you observed. The theorem by Brukner now shows that this assumption is inconsistent with a couple of others that might seem reasonable, namely, that quantum mechanics is universally applicable (i. e. that not something like a spontaneous collapse of the wave function occurs once you cross some threshold of complexity), that both you and davidm are free to choose which measurement to perform (i. e. that you can do so randomly, and this isn’t ‘rigged’ so to only allow certain measurements in cases where the outcome will be a certain value), and the already mentioned notion of locality (i. e. that distant events don’t instantaneously influence one another).
But wait, how does the locality creep in? We’ve so far only talked about a single box, at a single location. The reason for that is that to derive the inconsistency, Brukner essentially appeals to Bell’s theorem: one needs to set up two copies of the box, entangled in the right way; then, one finds that the outcomes of measurements on each (checking if the cat’s alive versus checking if the box is in superposition) obey statistics that are inconsistent with these facts being locally predetermined.
In other words, given the assumptions, davidm knows (can verify by measurement) that the box is in an entangled superposition, that you have observed a definite outcome, but that you can neither have observed the cat to be dead, nor that it is alive, as assuming that either of these must be true would be in conflict with the violation of the Bell inequality.
But, again, this only tells us that the conjunction of assumptions above is inconsistent with quantum theory; it doesn’t tell us, as such, that there is no observer-independent reality.
In these cases, there is still just one single consistent set of underlying facts, though. If one is adamant on not rebuking the other assumptions going into the theorem discussed above, then it’s just not the case that there is such a set.
I have an issue with the wording, because there would be a lot of people who would be tempted to overreach the definition of “objective reality”, which does not stretch to things like laws or most macroscopic phenomena. That way lies saying “that’s, like, just your opinion, man” to everything.
Like me, which believes in the possibility subjective reality which interrelates to each other. Or perhaps 4 different interlinked realities which people live in, forming a cast system in society that is basically invisible to most people. Each one will experience different things, but in a way that they don’t seem to be in conflict with each other as our minds are not programed to notice the differences. But it is possible to learn to see them or aspects of them (or be thus gifted to see them). In short life is harder for some than others - that we know, this is one way that can explain it.
I always understood that when a probability wave collapsed it collapsed to a definite objective reality (or at least a definite reality for everyone in a given branch if we’re talking many worlds).
If I’m understanding the article correctly then a wave can collapse differently for different observers; that they may each experience the particle taking a different path.
If that’s true, then how can the collective action of particles lead to an objective reality at the macro level?
Am i asking one of those questions that we still don’t have an answer for?
Wave-function collapse is just one of the stories we can tell about quantum mechanics to get our everyday experience out. As you note, many worlds (which features no collapse) is another. In many worlds, the ‘split’ into different worlds is only complete once decoherence of the different possibilities has occurred; as long as (in my example above) eschereal is still in the box, and the box is isolated from the outside, there hasn’t yet been any decoherence from your point of view; only once you interact—once eschereal tells you whether the cat is dead or alive, for instance—does decoherence occur, and the whole situation splits into two worlds.
From the point of view of the observer in the box, of course, the universe ‘splits’ as soon as they check on the cat.
In Copenhagen QM, however, you’re right that measurement is supposed to ‘collapse’ the wave function. But then, what happens if we let an observer (the one in the box) interact with a quantum system, and describe that from the point of view of another observer outside? If we’re allowed to extend the quantum description to an observer—and as they’re presumably made of the same kind of stuff as everything else in the world, it’s hard to see why we shouldn’t be—then we can’t say that the wave function collapses as soon as observer I makes a measurement, as to observer II, the whole thing ought to be describable as a coherent process, and in principle, they ought to be able to perform a measurement on observer I (and anything they’ve interacted with, i. e. the whole box) to verify that yes, they’re in a superposed state.
Eugene Wigner, who came up with the idea, initially thought that surely, this was an absurd state, and wanted to install an absolute ‘buck stops here’-type of deal with observers: he claimed that yes, there was something special about observers—their conscious experience—that sets them apart from all of the rest of the matter we usually describe using the quantum formalism, and hence, in a setup such as the above, the wave function would actually collapse upon observer I’s observation (although he later disavowed this notion). Notably, though, this means that this sort of interpretation leads to in principle observable experimental differences to ‘vanilla’ QM.
Ultimately, however, all of the stories will agree; whenever observer I is out of the box, so to speak, and tells observer II what they observed, their stories will both agree. So the theory never predicts any observable inconsistencies.
The issue is also not a different collapse for the observers, but rather, that one observer describes a system (the cat) as having undergone ‘collapse’, if that’s how you want to talk about it, while the other describes the system of ‘observer I + cat + box’ as being in a superposed, not-yet-collapsed state (that a system is in such a state can be verified by an experiment that’s in principle a generalization of the famous double slit experiment, where the interference pattern means that the particle must have been ‘through both slits at once’). What’s interesting is then that the outside observer I can’t say that the one in the box has made some definite observation, and I just doesn’t know which—that assumption makes it impossible to explain the violation of the associated Bell inequality.
As I said, ultimately, all the stories of everything that interacts with one another agree; so when you sit down for a cup of tea, you’ll all agree on what happened.
That depends on what sort of answer you mean. We’ve a perfectly well functioning formalism capable of making predictions about the outcome of experiments. But if you want to know what’s really going on below, the issue isn’t that we don’t have an answer, but that we have too many—each of the different interpretations of quantum mechanics will have its own.
Small quibble here, though. As I read the article, it indicated that consciousness is not a requirement for an observer. An instrument that registers and records readings of a given system can be regarded to be an observer, and they even seemed inclined toward extending the definition of “observer” much farther. A thing which is involved with other things might he described as an observer, since that thing’s behavior is regulated by the information available to it. A neutrino might even be seen to be an observer, albeit not a very good one, since neutrinos have the habit of ignoring just about everything almost all of the time.
Yes, very few people today think it is, largely because the dominant philosophy of mind is (some flavor of) physicalism—if the mind itself is physical, there’s not really any room for it to be an exception of quantum mechanics. However, Wigner’s original proposal was indeed that consciousness provides the special sauce to distinguish proper observation from ordinary interaction.
Sure, but the problem is that in doing this, you lose a clear dividing line between what’s an observer, what counts as a measurement, and what’s merely interaction—because ultimately, each interaction leading to a state change can be taken as ‘observing’ some property of the thing interacted with. But then, it’s just not clear when one ought to apply the usual quantum dynamics, or have the wave function ‘collapse’.
Am I correct that the experiment yielded exactly the result predicted? There was no new “surprise” that “objective reality doesn’t exist.” The excitement is just that an experimentalist was finally able to rig a set-up to confirm what quantum theorists already confidently predicted.
The experiment used for Bell’s theorem involved two entangled photons. The GHZ experiment uses four entangled photons to get a crisper result with the paradoxicality more blatant. This latest experiment relies on six photons to make the paradoxicality intrinsic to Bell’s theorem even weirder still. Ho-hum.
(Well, that’s this layman’s vague understanding. Am I all wet?)
As Huw Price discusses, quantum mechanics (with the above “paradoxes”) is simply the kind of theory we ought to have expected when time symmetry is taken to its conclusion. (But I wish I knew a similar argument to understand two-slit experiments.)
Absolutely. I don’t think anybody had any expectation the experiment would show anything other than what it did.
(Tiny detail: GHZ states exist for different numbers of qubits, the smallest being a three-qubit state.)
Well, the novelty isn’t really in the number of photons used (much larger systems have been successfully entangled—uh, well, ‘large’ in the relative sense that really means still quite mind-bogglingly small), but in the observables that were measured, which are analogous to those in the Wigner experiment. Thus, one can interpret these observables as corresponding to something like ‘seeing the cat dead or alive’ and ‘Wigner’s friend being in a superposed state’, roughly.
Now, strictly speaking, this means that tests of this form are a subset of Bell inequality tests. But by connecting with observables that we can give some macroscopic meaning to, rather than appealing to strange quantum notions like ‘spin’, one might consider this to throw the issue into sharper relief.
Or not. I mean, in the end, many of these things boil down to experimenters looking for something cool they can apply their shiny new gadgets to (not to diminish the great work that goes into building and applying said gadgets).