On Monday 25 October 2004 18:01, Charles Brown wrote: > I've been mugging up on this too :>) and a physics professor told me: > > "When I discuss quantum physics, I never refer to an observer, but to the > objective recording of an interaction by an instrument.
this goes some way towards healing the quantum-classical split. there are still problems remaining. the reason is this: an "objective recording of an interaction by an instrument", meaning a classical observation, can in principle be analyzed quantum mechanically. so for example, when a photon hits a recording film and it creates a white spot. we can think of that classically as recording the photon's position in some kind of photon propagation experiment. but nothin prevents us from analyzing this film recording event. in that case, there are silver chloride atoms, and an interaction with a photon, and some probabilities for atoms making transitions to excited states, and so on and so forth. there is also the notion of quantum decoherence (charles: we'll talk more about it offlist). this also renders more classical the results of quantum "paradoxes". however, there is still disagreement in the physics community about whether even this solves the problem completely, meaning "in principle". many agree it solves a lot of the problem "in practice". in brief, quantum mechanics does not specify a dividing line between quantum and classical behavior. it is a two-layered approach to physics. there is an almost classical part, like a schrodinger equation, describing the time evolution of a "state", and there is the part which connects "states" with "observations" in the macro (classical) world. classical mechanics does not have this split in principle. its the part that takes a quantum state and renders unto us what we can observe that is the source of the oddities. in classical mechanics, i solve the equations of motion for the earth-moon-sun system in terms of their positions and momentum, and the solution is rendered in terms of those position/momentum variables. those numbers are directly connected, in principle, with our observations, and so we know there WILL BE a total lunar eclipse this coming wednesday evening. > Maybe this is the "unravelling" of the inadequate way the we have "always" > been forced to think of change of position, displacement, as a series of > positions at rest. i think there is something to this, certainly in terms of getting off of the point-particle picture as the total description of material reality. however, QM also contains other kinds of "can't do this" prescriptions, called "incompatible observables", like position and momentum. so, for example, QM also says you cannot measure precisely an electron's intrinsic spin in two mutually orthogonal directions, say east and north. there is no classical analogue to this intrinsic spin, and so can't be analyzed further in terms of some things like a point's position and momentum. daniel davies wrote: > > One important thing to remember (I'm mugging up on this this weekend for an > argument with Steven Landsburg) is that quantum probability is very > different from classical probability. When a particle is superposed in two > states, that isn't at all the same thing as saying that it's in one state > with probability x and in the other state with probability y. the difference is this: the superposition is in the "quantum amplitude", or state, the time evolution of which can be almost classical. however connecting this amplitude to observations requires taking the "complex magnitude" of this amplitude, and so you can get interference patterns and all kinds of goodies. this is what i refered to above as the two-tiered approach. going from amplitude to probabilty of observation is the screwy part. an historical note: Heisenberg came first came to a notion of a quantum state and a quantum kinematical description of micro particles that made sense of radiation lines from excited atoms (among other things). in examining this new description, he and Bohr noticed oddities, and the uncertainty principle was born as a way of making sense of the new description. sometime later, it was asked what is the meaning of this quantum amplitude, and Born answered that the complex modulus was a probability thing. so historically this was taken in two somewhat distinct seperate steps. in very crude terms, Heisenberg came to the idea that something like position was better described as a matrix of numbers rather than a lone number, and Born figured out how to go from this matrix and state description to the measurement and observation of properties. there is a famous -- within the physics community -- quotation from Murray Gell-mann (quarks) about Bohr and his Copenhagen interpretation: "Niels Bohr brainwashed a whole generation of theorists into thinking that the job (interpreting quantum theory) was done 50 years ago." more recently Gerard t'Hooft (Nobel Prize for proof of renormalizability of electro-weak theory) has argued for a new look at the principles and interpretation. for example: http://arxiv.org/pdf/quant-ph/0212095 (for non-physics people, see beginning and end sections). here is a slide show of one of his talks http://online.kitp.ucsb.edu/online/kitp25/thooft/oh/01.html and links to the accompanying talk: http://online.kitp.ucsb.edu/online/kitp25/thooft/ les schaffer
