It’s more complicated than that, even in the Copenhagen interpretation: link
The quantum bomb detector experiment has been performed and (big surprise) works as expected, demonstrating that no “conscious” observer (the bomb in the original thought experiment) is required in order to perform a measurement.
This seems like a pretty fringe interpretation, correct? This sounds like the Von Nuemann-Wigner interpretation of the Copenhagen interpretation, and if that’s correct, and the stats I looked at are correct, we’re looking at roughly a 2.5% acceptance rate.
But at the risk of sounding like a certain ex-President, what does it mean to ask “what does it mean”? Can a theory really even have any “inherent meaning”, beyond what predictions it can make? Explanations are slippery things, and humans have a habit of giving “simplifying explanations” that are actually far more complicated than the thing they’re “simplifying”.
So far as I can tell, what people usually mean by “explaining” something is describing it in terms of things that are already familiar to the audience. But by this standard, quantum mechanics cannot be explained at all, because it’s inherently unfamiliar to humans.
I don’t see how that experiment says anything about the observer one way or the other. Can you explain?
It clearly doesn’t prove or disprove any interpretation of quantum mechanics.
No, the Von Nuemann-Wigner interpretation is not a sub-interpretation of Copenhagen. It stands on its own.
And we were talking about the Copenhagen interpretation itself, which is the most mainstream and accepted interpretation.
A live bomb “observes” the photon in that arm of the apparatus, destroying the interference which keeps a photon from entering one of the detectors (Detector D in the linked article) as when there is no bomb. (What is supposed to be interesting about this particular setup, compared to simply blocking the path, is that this can happen even without setting off the bomb.)
Of course not, but I was responding to the post asking whether a non-conscious observer could collapse a wave function (which itself is a particular interpretation of what is going on in these experiments). A simpler way to do it would be to block one of the slits in a double-slit experiment and watch the interference pattern disappear, or have a robot tell you the pattern disappears. But there is no end to the game of possible interpretations.
To try to clarify the situation:
The Copenhagen interpretation does NOT a require a conscious observer. It requires an ‘observer’, but that may be mechanical. That is not an issue of debate in the Copenhagen interpretation.
The problem with the Copenhagen interpretation is that it requires an ‘external’ non-quantum system for observation. It assumes that there is a quantum system which is ‘observed’ by an external, macroscopic ‘classical’ system.
So where do you draw the line between quantum and classical? Where is the ‘cut’, as Pauli wrote to Max Born in the quotation above. At what point does quantum behaviour collapse to classical behaviour, and why? How can there be an ‘external’ system to observe at all? All this is left vague and unexplained in Copenhagen.
I see the ‘actually’ reply has already been made, but just to be clear: this isn’t right. In ordinary interactions, if a system in a superposed state (such as a dead/alive cat) interacts with another system, no collapse occurs; instead, both systems become entangled, as an entangled state really just is a superposition of states of a compound system whose parts may be spatially distant. So the usual interaction of, say, and electron in a superposition of ‘spin up’ and ‘spin down’ states (along some particular axis) with a measuring apparatus would be something like (with the indices e and m labeling the electron’s state and that of the measuring apparatus respectively)
(spin up)[sub]e[/sub] + (spin down)[sub]e[/sub][sub]m[/sub] –> (spin up)[sub]e[/sub](detects up)[sub]m[/sub] + (spin down)(detects down)[sub]m[/sub]
If ordinary interaction were all there is, then we’d be stuck with this final state (after the arrow), in which the measuring apparatus is in a superposed state of detecting the electron as having spin down and spin up. But then, [del]a miracle occurs[/del] what’s variously called ‘process I’ (von Neumann), the collapse, or state-reduction occurs, and one of the components of the state (spin up)[sub]e[/sub](detects up)[sub]m[/sub] + (spin down)(detects down)[sub]m[/sub] just drops away, leaving (for instance) (spin down)(detects down)[sub]m[/sub], i. e. a state in which the electron’s spin is definitely down, and has been detected as such.
How this reduction occurs, and even whether it occurs at all, is the central question of the interpretation of quantum mechanics. This process doesn’t happen in every interaction, but only during measurements—at least, on most interpretations; indeed, if it happened in every interaction, quantum mechanics would lead to a vastly different phenomenology, as the ‘uncollapsed’ state can be distinguished from the collapsed one by a suitable interference experiment, and we indeed observe coherence to persist across particle interactions.
There are interaction-free measurements, of which the bomb tester presented in this thread is an example. The idea is to set things up such that if the probe interacted with the system to be measured, both would be influenced in a particular way (e. g. destroyed), but that even if no such influence occurs, our knowledge about this informs the state we ascribe to the system. A simple example is a negative-result measurement: you shoot a photon at a beam splitter, where it can be transmitted or reflected with equal probability, and place a detector along one possible path of the photon. If you don’t observe the photon having taken that path, you know it must’ve taken the other, and thus, it can’t be in a superposition of both being transmitted and reflected, and consequently, its wave function collapses without ever having interacted with your detector. So it’s not interaction that influences the system.
That a photon doesn’t have a definite position (or momentum) in some given scenario can be more directly probed with a suitable interference experiment: only if a photon is in a superposed state of going down both arms of an interferometer will we observe interference; since we do observe interference, it’s not merely our ‘clumsiness’ in measurement that precludes us attributing a definite position to the photon.
If the experiment were done, then the box would be described by application of the quantum formalism, and would thus enter into a superposed state of switching on its light and not switching it on; only once the box itself is measured (by observing the light) would it undergo ‘collapse’ to any definite state. So this sort of thing only kicks the can down the road further.
The bomb doesn’t act as an observer, or even a measuring device, in that experiment; all it does is block an arm of the interferometer (or not, as the case may be). This isn’t different from simply closing one of the slits of a double slit setup, and observing the interference pattern to vanish. The closed slit doesn’t observe the photon, it merely absorbs it, and there’s no mystery to how that makes the interference pattern go away.
Many explanations refer to things completely alien to everyday experience. The explanation of Special Relativity refers to four dimensional spacetime, which nobody has any intuitive concept of; yet, it clarifies the actual physical content of the theory. One could just as well interpret it differently, in terms of a Lorentz ether, but few believe that to be anything more than a crutch to intuition, even though it is empirically exactly equivalent—most notably, because it doesn’t carry over well to General Relativity. So of those two possible, empirically equivalent interpretations of SR, only one really applies to physical reality, and opens up the door towards a generalized theory—finding an interpretation of quantum mechanics that does the same would thus be a huge deal.
And there are of course many other good reasons for wondering about the right interpretation. One is that this sort of thing has spurned many new developments within the theory itself, from decoherence theory to, arguably, the entire field of quantum information theory, not to mention more niche applications like the quantum De Finetti theorem and the search for mutually unbiased bases. Another is that you might like to know how well your theory fits with others, whether there’s points of tension with the sort of picture implied by the other elements of your scientific world view—for instance, Bohmian mechanics sits more uneasily with special relativity than other interpretations of quantum mechanics do, due to the existence of superluminal action as a distance.
But really, all that’s pragmatic window dressing. The best reason for seeking interpretations of quantum mechanics is that you’d want to know how the world works, not merely what relative frequencies of events a certain mathematical formalism can be tickled into coughing up. I know that this instinct is all but extinguished in physicists, thanks, in part, to the anti-philosophical shut-up-and-calculate bend of Feynman and the other ‘savages’, but just as you don’t climb a mountain to ascertain its height, or for any other pragmatic undertaking, the number one reason for interpretation, and all scientific ventures, is simple intellectual curiosity. Having to find pragmatic justifications for indulging this curiosity is part of why physics experiences the crisis it now faces, I believe.
Also, while I’m at it, it’s not quite right to say that all interpretations are equivalent mathematically, so there’s no ‘real’ distinction between them. Spontaneous collapse-theories are usually considered interpretation of quantum mechanics, and yet, they yield empirically different predictions for certain (typically macroscopic according to some suitable measure) systems. So I guess one could call them simply competing theories, but then again, they’re actually mathematically equivalent to quantum mechanics of open systems (where you achieve an effective collapse by tracing out certain degrees of freedom), with ordinary quantum mechanics being a kind of ‘limiting case’. Furthermore, there’s been interesting research (carried out by, among others, my former advisor) into the thermodynamics of different interpretations, showing that certain interpretations lead to runaway entropy production.
Finally, different interpretations may fail to yield self-consistent pictures—there’s been a huge recent kerfluffle around a paper by Frauchiger and Renner that alleged, in the original version, that ‘single-world’ quantum mechanics can’t consistently be applied by all observers to themselves; I don’t think that the claim really holds up, but it has stimulated lots of recent research
Yes, I think that’s perhaps the best way to describe the Copenhagen interpretation: quantum mechanics can’t be universally applied, but only makes meaningful predictions against a certain, classical reference point that must necessarily be excluded from quantum mechanical description.
Well, the usual attempt at resolving this is by noting that the ‘cut’ is essentially arbitrary and can be placed at different spots. In this sense, ‘fixing’ the observer is a bit like fixing a frame of reference in special relativity (the analogy has been explicitly drawn by Bohr)—all frames are equivalent, but to extract some empirical content (say, the time that a particular observer reads off their clock) requires a choice of reference.
Yes, because, well, if a conscious observer is necessary, then we have a problem with conscious observers, themselves made of bits of the universe, coming into existence in the first place, with nobody yet present to observe that happening.
There’s a variant of spontaneous collapse theory in which the collapse probability is directly linked to the expectation value of an operator linked to ‘integrated information’ which, according to integrated information theory, is a measure of consciousness. There’s nothing paradoxical about that—at least, not any more than for other collapse theories, where the collapse probability is linked to other quantities, typically indicating macroscopicity—mass, particle number, and the like. But it’s of course highly speculative, and I doubt if it can be developed into a full theory.
I would say that physicists nowadays are just as interested as ever with the question of “how does the Universe work?”, but the difference is that we’ve come to peace with the realization that “according to this set of equations” is a perfectly good answer to that question. Because that’s the only sort of answer we’ve ever actually had, even before quantum mechanics.
Most physicists belong to the ‘shut up and calculate’ school of thought, but by no means all.
A good article by a theoretical physicist at CalTech argues that this attitude is impeding further progress in physics:
And one of the social and cultural consequences of this esoteric dilemma is the tremendous inertial power of the so-called ‘spirituality’ with its beliefs in astrology (where way over half of the youth believe it is a type of science), Wicca and witchcraft. Paganism is on a spectacular rise. With so many scientific breakthroughs year after year, I would expect to see such amazing numbers about Westerners turning to science to find answers to their most fundamental questions, but concepts like superposition and decoherence don’t seem to quench too many people’s thirst for understanding the nature of reality. In my opinion the OP seems to show this scientific conundrum and the quirkiness of the proposed solutions is doomed to lose most laymen for good.
It may be the other way around. It seems many people find quantum “quirkiness” appealing.
In fact, as more people become aware of the credibility/viability of the Many Worlds interpretation which Sean Carroll (the Caltech author of the quote above) is pushing, I’d expect an increase in both legitimate philosophizing and outrageous claims about ‘parallel universes’ and their implications. The trick may be to distinguish between them.
Unless you take on the Many Worlds interpretation, and we collapsed into this universe, as a spawn from the universe where the photon went though the detector. 
With regard to gravity being the odd force out where quantum physics is concerned, is it really a coincidence that gravity is weak enough that it doesn’t immediately force a contradiction with such things as two-slit experiments? Or put another way, in a hypothetical universe in which gravity was much stronger, how much stronger could it be before one could definitely say that an electron went through one slit rather than another by noting its gravitational pull?
As with most things involving quantum gravity, we have no idea.
It is supposed that semiclassical gravity provides an accurate approximation of quantum gravity, within certain limits ,and in semiclassical gravity the graviational field of a quantum system is the same as would be produced classically by the expectation value of the stress-energy tensor. In the two slits experiment, it’s clear to see that the expectation value for the postion of an electron passing throught the slits is actual midway between the two slits (NB the expectation value of a particles needn’t be a place where it is possible to find the particle).
Roger Penrose’s view though is that at around the Planck mass (21 micrograms) range the wavefunction sponateously collapses, but his views on this particualr subject are not seen as mainstream.
Semiclassical gravity is what we use because we don’t have anything better. I don’t think anyone actually expects that the gravitational center of the particles going through a two-slit experiment would be midway between the slits.