What if we could measure superpositions?
Dirk Bruere
"Ralph Hartley" <har...@aic.nrl.navy.mil> wrote in message
news:ar2ul6$rcl$1...@ra.nrl.navy.mil...
> Nick Maclaren wrote:
>
> > I interpreted your question to be roughly "What if we could measure
> > one aspect of the state without affecting the state?"
>
> It is possible to interpret some of the questions I have seen here to mean
> that.
>
> The answer is that it doesn't make any sense.
What makes more sense is the question: "What if we could force the collapse
of a superposition to an arbitrary predefined value?"
Dirk
Nick Maclaren
Yes and no. It is not true that the question doesn't make any
sense; it does, even though a valid answer is "doing that is
inherently incompatible with quantum mechanics as currently
hypothesised."
It is possible to hypothesise a model where measuring aspects of
a state could be done without affecting the state, but it wouldn't
be quentum mechanics as we know and love it. And my remarks on
the consequences of adopting such a variant model stand.
Your question is much more compatible with q uantum mechanics,
I agree.
Regards,
Nick Maclaren,
University of Cambridge Computing Service,
New Museums Site, Pembroke Street, Cambridge CB2 3QH, England.
Email: nm...@cam.ac.uk
Tel.: +44 1223 334761 Fax: +44 1223 334679
Ralph Hartley
> Ralph Hartley wrote:
>>Nick Maclaren wrote:
>>
>>>I interpreted your question to be roughly "What if we could measure
>>>one aspect of the state without affecting the state?"
>>
>
> What makes more sense is the question: "What if we could force the collapse
> of a superposition to an arbitrary predefined value?"
That is at least a sensible question.
I will attempt to give two answers: what it would mean for physics, and
what it would mean for potential technology.
My first answer is that if you could do that, you could prove that either
quantum mechanics or special relativity, or more likely both, were wrong.
This would be about on a par with discovering in 1622 that the world was
flat after all; you would have to explain how the world was circumnavigated
a century before.
When people say that the whole structure of modern physics would fail, they
mean the *whole* structure!
As for practical applications, that would depend on how the physics
actually worked out, but it would be likely to inculde:
Faster Than Light comunication: Prepare a pair of distant particles in an
entangled state, like the ones used for quantum cryptography. "Force" one
to a chosen value, and observe the other. The effect would be simultaineous
regardless of distance.
Communication Backwards in Time: Simutanaiety is only meaningfull with
respect to a reference frame. Send an instantainious signal in one
reference frame, and have it instantainiously sent back to the source in
another, it would be recieved before the original message was sent. This
requires that at least some aspects of relativity still hold, but
relativity is very well tested, so that some remnant of it would have to
remain.
Perpetual Motion: Use Backwards in Time communication to control the door
in a Maxwell's Demon divice.
Paradoxes: Your grandson uses Backwards in Time Communication to take out a
contract on your life.
Don't get too excited about these applications, since there is overwelming
evidence that both Quantum Mechanics and Special Relativity are correct.
I think you would have to agree that breaking quantum cryptography would be
a pretty minor side effect, wouldn't you?
Ralph Hartley
Dirk Bruere
Are these conclusions also true if one could simply bias the collapse of a
superposition towards an arbitrary predefined value, rather than completely
determine it?
Dirk
Ralph Hartley
> "Ralph Hartley" <har...@aic.nrl.navy.mil> wrote:
>
>> Dirk Bruere wrote:
>>> What makes more sense is the question: "What if we could force the
>>> collapse of a superposition to an arbitrary predefined value?"
>>
>> That is at least a sensible question.
>> My first answer is that if you could do that, you could prove that
>> either quantum mechanics or special relativity, or more likely both,
>> were wrong. This would be about on a par with discovering in 1622 that
>> the world was flat after all; you would have to explain how the world
>> was
>> Don't get too excited about these applications, since there is
>> overwelming evidence that both Quantum Mechanics and Special
>> Relativity are correct.
>>
>> I think you would have to agree that breaking quantum cryptography
>> would be a pretty minor side effect, wouldn't you?
>
> Are these conclusions also true if one could simply bias the collapse of
> a superposition towards an arbitrary predefined value, rather than
> completely determine it?
Yes. That would just add noise, and we know how to deal with noise.
However, my conclusions above are sort of like speculation about what the
world would be like if we found out it was flat. Would China be east of
America or west? Where would the south pole fit in? What exactly is the
space station doing every 90 minutes? What's with those time zones?
People think Quantum Mechanics is counter-intuitive and hard to understand,
but that's *nothing* compared to what would have to take it's place if it
was wrong on such a basic level.
Ralph Hartley
Dirk Bruere
______
Something I posted a while back both here and on spr, to little in the way
of critical response.
I would appreciate your opinion.
It flows from a 'naive realism' concerning quantum measurement and embodies
a number of dodgy 'ifs'. The latter lead to a simple experimental test.
It begins with the Delayed Choice expt and a Cramer style interpretation of
a 'backward in time signal' upon measurement that sets a 'real' unique path
for the photon retrospecively. Given that view we have a temporal loop whose
parameters can be changed according to where and when the measurement is
made.
The question that has nagged me is: Is that loop local limited to the
photon/apparatus or does it essentially 're-run' the universe over that
duration?
To provide a test requires another dubious assumption. Namely that, if we
'rerun' the universe over again the results of a quantum measurement will
not necessarily be the same. That they will still exhibit a random quality
ie that re-runs will not be identical on a small enough scale.
Anyway, to the test.
An interferometer with (say) a 4uS transit time (eg monomode fibre in Sagnac
configuration). We also have a random number generator based upon (say) a
radioactive decay providing a single quantum event clocking a 0 or 1 from a
high speed counter.
Over a period of time the count should be fairly predictable statistically
and evenly distributed between 0,1.
However, what we do is to trigger a measurement on the interferometer within
4uS of getting an output from the random number generator of a 1.
If the 're-run' is affecting the universe as a whole the result will be that
the output of the RNG is skewed towards 0.
Any comments and statements of impossibilities are welcome (as long as they
are accompanied by an explanation).
Dirk
Minti
superposition is measured and dd not collapse but rather just gave use
the intermediate value. and that we could just collapse them to any
arbitury value at any instant which can at present be done using Walsh
Hadamard transformations. What would be it good for?
Or are superpostions good and only good.
--
Minti
Nick Maclaren
In article <e87fc4b0.02110...@posting.google.com>,
mintiSP...@yahoo.com (Minti) writes:
|> What if superpositions did not collapse and that if the system in
|> superposition is measured and dd not collapse but rather just gave use
|> the intermediate value. and that we could just collapse them to any
|> arbitury value at any instant which can at present be done using Walsh
|> Hadamard transformations. What would be it good for?
Escaping from the constraints of physics. If you could do that,
the uncertainty principle would not hold, for a start. I am not
enough of a quantum theorist to know what would happen beyond
that, but I should be surprised if you couldn't get a serious
inconsistency in reality.
|> Or are superpostions good and only good.
Well, without the superposition restriction, quantum communication
would not be secure.
Ralph Hartley
> What if superpositions did not collapse and that if the system in
> superposition is measured and dd not collapse but rather just gave use
> the intermediate value.
*Which* intermediate value? The whole *point* of doing a measurement is to
collapse a superposition by giving a definite value.
> and that we could just collapse them to any
> arbitury value at any instant
If you could force a superposition to collapse into a chosen value, that
would let you do lots of stuff verging on magic, e.g. faster than light
communication, communicating backwards in time etc.
The whole structure of quantum mechanics prevents this. It's like perpetual
motion, I don't have to see the schematics to know it doesn't work.
If you meant something else, you will need to be a little more clear.
> which can at present be done using Walsh
> Hadamard transformations.
I don't even see the connection here.
> Or are superpostions good and only good.
Well if we *only* ever wanted superpositions, we would never do
measurements, would we.
Ralph Hartley
Dirk Bruere
> Minti wrote:
> > What if superpositions did not collapse and that if the system in
> > superposition is measured and dd not collapse but rather just gave use
> > the intermediate value.
>
> *Which* intermediate value? The whole *point* of doing a measurement is to
> collapse a superposition by giving a definite value.
>
> > and that we could just collapse them to any
> > arbitury value at any instant
>
> If you could force a superposition to collapse into a chosen value, that
> would let you do lots of stuff verging on magic, e.g. faster than light
> communication, communicating backwards in time etc.
Let's not forget teleportation of macroscopic objects (tunnelling).
Dirk
Minti
> > What if superpositions did not collapse and that if the system in
> > superposition is measured and dd not collapse but rather just gave use
> > the intermediate value.
>
> *Which* intermediate value? The whole *point* of doing a measurement is to
> collapse a superposition by giving a definite value.
I mean that lets say we have I had
1/sqrt(2) [ |0> + |1> ] ----------> *
Now when we measure it it collapses to either |0> or |1>
But Lets say it did not collapse and that when we meausre it would
give the wave function [*]
I know its like performing the double slit experiment and trying to
say that the photon has been detected at both the slits.
But just in case some artificial particle could help use solve this
puzzle.
However I do think that we would have to tune the particle to the best
of our needs.
>
> > and that we could just collapse them to any
> > arbitury value at any instant
>
> If you could force a superposition to collapse into a chosen value, that
> would let you do lots of stuff verging on magic, e.g. faster than light
> communication, communicating backwards in time etc.
>
> The whole structure of quantum mechanics prevents this. It's like perpetual
> motion, I don't have to see the schematics to know it doesn't work.
>
> If you meant something else, you will need to be a little more clear.
>
> > which can at present be done using Walsh
> > Hadamard transformations.
>
> I don't even see the connection here.
In the Gorvers article in DDJ, he used walsh hadamard transormations
in the overall process of sniffing out the right state.
>
> > Or are superpostions good and only good.
>
> Well if we *only* ever wanted superpositions, we would never do
> measurements, would we.
--
Minti
mania...@msn.com
Dr Chaos
>
> In article <e87fc4b0.02110...@posting.google.com>,
> mintiSP...@yahoo.com (Minti) writes:
>|> What if superpositions did not collapse and that if the system in
>|> superposition is measured and dd not collapse but rather just gave use
>|> the intermediate value. and that we could just collapse them to any
>|> arbitury value at any instant which can at present be done using Walsh
>|> Hadamard transformations. What would be it good for?
>
> Escaping from the constraints of physics. If you could do that,
> the uncertainty principle would not hold, for a start. I am not
> enough of a quantum theorist to know what would happen beyond
> that, but I should be surprised if you couldn't get a serious
> inconsistency in reality.
The uncertainty principle is because operators don't commute, not
directly necessarily because of "superpostion".
Then again what is a pure state in one basis is a mixed state in
others (and your choice of basis is somewhat arbitrary), so whether or
not something is in a "superposition" is not a unique physical observable.
Barry Walsh
can't be measured?
Douglas Eagleson
Ralph Hartley wrote:
Why would using a knowledge of some correlated state to
cause your collapse, cause a faster than light magic collapse?
The prior existence of the correlation causes only the apparent
faster than light communication of the selected states.
Indicating a misapplication of the concept of superposition
to the Copenhagen Interpretation, which only has
Copenhagen wave functions, defining your version
of a conception of superposition. Lousy theory.
A bad assumption of the concept of superposition.
Look at an atomic weapon, under compression.
A lot of correlated states. What happens to the wavefunction?
A collapse in a superposition of conditions. And this conception of
conditions is introduced to warn of the method of theory that
you used to define superposition. A condition is a what?
A concept assumed equal to correlated collapse. A word change
unwarranted.
I call just this threads concept, the real theory of superposition.
Also called correlated conditions, here.
Douglas Eagleson
Gaithersburg, MD USA
Minti
> Doesn't decoherence take care of all this and ensure that superpositions
> can't be measured?
Yes indeed... But I am saying is that if the decohrence Did not
occour.
I mean to ask that do we have enough knowledge in physics to prove
that Dechorence WILL occour in microscopic world.
--
Minti
mania...@msn.com
Edward Green
...
> Then again what is a pure state in one basis is a mixed state in
> others (and your choice of basis is somewhat arbitrary), so whether or
> not something is in a "superposition" is not a unique physical observable.
Right. And yet some people speak of "superposition" as if it were a
weird quantum atribute -- the classic example being the cat
superimposed in life and death. Since live/dead are not the
eigenvalues of any observable, AFAIK, we don't have any idea what we
are talking about. Vector sums are not ghostly superpositions of the
constituents, but perfectly good vectors in their own right.
greysky
"Minti" <mintiSP...@yahoo.com> wrote in message
news:e87fc4b0.02110...@posting.google.com...
Actually, we really have no evidence that dechorence actually has occured on
a quantum by quantum basis. QM is a theory of final states involving qroup
order - it has nothing to say concerning the discreet paths taken by
individual quanta to arive at those final states. In some ways it is easier
to assume no dechorence has occured in a system - that all paths are valid.
Our measurements just localize and target a particular path in no particular
order. This throws the concept of time stability out because if you could
reverse time and perform the same exact measurement again, you would arrive
at a different outcome. Most people don't like thinking along these lines
because if you can show time is not a stabalizing attractor in a measured
system of variables, all the other nasty phenomena other posters to this
thread have talked about will also happen. If, while running an N slit
experiment for example, you stop asking which slit did the electron really
go through, and instead realize it can go through all without splitting up,
and see the interference pattern that results as a measure of the
probability of the same electron interfering with itself as it travels down
N paths, quantum phenomena begin to make sense in a very profane way. This
is why the noise level goes up as you increase the number of open slits...
the bifurcation gets so complex it becomes random very quickly. But the
results get very interesting as well. Shoot your electron at, say, a N=5
slit matrix. If the electron has a 20% liklihood of hitting any slit, odds
are for any five tries, there will be 5 discreet pathways and 5 potential
detection events. We usually don't see all these paths, not because we can't
or because of the nature of the equipment being used, but because the noise
level of the experiment interfers with the observation most of the time, but
not always. Because this interference is random, it will not always cancel
perfectly and then you will get a multiple detection event. Decrease the
number of slits to reduce the noise level of the experiment and you also
reduce the number of paths that can be detected. Increase the slits and
unfortunately the nose level also increases and obfuscatres your efforts. A
free traveling electron corresponds to an electron traveling down an
infinite number of paths and also corresponds to the maximum level of
interference possible. If N equals the number of paths possible, though it
is not possible to state the probabilistic level of stochasitc interference,
the results of this interference can be given as (N -(N-1)) {1} in general.
The reciprocal gives you a limit on the 'quiet' amplitude of your
experiment, where observing multiple detection events can occure.
So which path do you choose as being the right one? Does not observing
multiple paths for the same electron necessairly exclude them? Just what is
dechorence, anyway? It's a known though seldom accounted for fact that
thinking while running your experiment will interfere with the results more
than if you postponed your thinking until after you have the results in
hand.
Greysky
www.allocations.cc
Nick Maclaren
I interpreted your question to be roughly "What if we could measure
one aspect of the state without affecting the state?"
And my response was that the whole edifice would fall apart. I don't
know enough about quantum theory to be quite certain that this would
break causality and the speed of light, but I am pretty sure that it
would. It would CERTAINLY eliminate quantum cryptography.
ca314159
>
> In article <e87fc4b0.02110...@posting.google.com>,
> Minti <mintiSP...@yahoo.com> wrote:
> >"Barry Walsh" <bwa...@DIEYOUSPAMMERFUCKindigo.ie> wrote in message news:<2daz9.6$5r1....@newsfep2-win.server.ntli.net>...
> >> Doesn't decoherence take care of all this and ensure that superpositions
> >> can't be measured?
> >
> >Yes indeed... But I am saying is that if the decohrence Did not
> >occour.
> >I mean to ask that do we have enough knowledge in physics to prove
> >that Dechorence WILL occour in microscopic world.
>
> I interpreted your question to be roughly "What if we could measure
> one aspect of the state without affecting the state?"
>
It's something to shoot for though, like Pinnochio wanting to be a real boy
with a short nose.
> And my response was that the whole edifice would fall apart. I don't
> know enough about quantum theory to be quite certain that this would
> break causality and the speed of light, but I am pretty sure that it
> would. It would CERTAINLY eliminate quantum cryptography.
Perhaps quantum zeno/non-demolition/interrogation/non-interaction is
BB's backdoor to quantum crypto:
http://www.physics.uiuc.edu/People/Faculty/profiles/Kwiat/Interaction-Free-Measurements.htm
Remember the "clipper chip" ?
http://www.cpsr.org/program/clipper/clipper.html
Reminds me of Sagan vs. Velikovsky.
Dr Chaos
>
> "Minti" <mintiSP...@yahoo.com> wrote in message
> news:e87fc4b0.02110...@posting.google.com...
>> "Barry Walsh" <bwa...@DIEYOUSPAMMERFUCKindigo.ie> wrote in message
> news:<2daz9.6$5r1....@newsfep2-win.server.ntli.net>...
>> > Doesn't decoherence take care of all this and ensure that superpositions
>> > can't be measured?
>>
>> Yes indeed... But I am saying is that if the decohrence Did not
>> occour.
>> I mean to ask that do we have enough knowledge in physics to prove
>> that Dechorence WILL occour in microscopic world.
>>
>> --
>> Minti
>> mania...@msn.com
>
> Actually, we really have no evidence that dechorence actually has occured on
> a quantum by quantum basis. QM is a theory of final states involving qroup
> order - it has nothing to say concerning the discreet paths taken by
> individual quanta to arive at those final states. In some ways it is easier
> to assume no dechorence has occured in a system - that all paths are valid.
> Our measurements just localize and target a particular path in no particular
> order.
Since I believe that measurement devices also happen to be governed by
the laws of quantum mechanics, and their atoms don't know an
'observation' from a hole in their spin, then something about
decoherence (what measuring appears to do) has to be there inside, and
not outside, the dynamics of quantum mechanical systems.
When asked "what is time?", Einstein once replied: "time is what
clocks measure." It may seem like a glib joke, but it is
fundamentally the best and most truthful answer.
It means that the the physics of the atoms that make up clocks (and
everything else) is described by some dynamics with a d/dt on the left
hand side of some equation. In physics, everything "is" what it
"does'. Time is what time does: it has a particular and unique role
in dynamical laws that govern the behavior of atoms in the
universe. Clocks are a human artifact that lets this be viewed in a
reasonably simple way.
Similarly I say: "decoherence is what the stuff in experiments does."
Dr Chaos
> In article <e87fc4b0.02110...@posting.google.com>,
> Minti <mintiSP...@yahoo.com> wrote:
>>"Barry Walsh" <bwa...@DIEYOUSPAMMERFUCKindigo.ie> wrote in message news:<2daz9.6$5r1....@newsfep2-win.server.ntli.net>...
>>> Doesn't decoherence take care of all this and ensure that superpositions
>>> can't be measured?
>>
>>Yes indeed... But I am saying is that if the decohrence Did not
>>occour.
>>I mean to ask that do we have enough knowledge in physics to prove
>>that Dechorence WILL occour in microscopic world.
>
> I interpreted your question to be roughly "What if we could measure
> one aspect of the state without affecting the state?"
Well, I believe you can measure one aspect of the state if the system
is already in a pure eigenstate of the relevant operator.
But in general, no it isn't possible.
Dr Chaos
> Dr Chaos <mbkennelS...@NOSPAMyahoo.com> wrote in message news:<slrnasot1v.iv2.m...@lyapunov.ucsd.edu>...
> ...
>
>> Then again what is a pure state in one basis is a mixed state in
>> others (and your choice of basis is somewhat arbitrary), so whether or
>> not something is in a "superposition" is not a unique physical observable.
>
> Right. And yet some people speak of "superposition" as if it were a
> weird quantum atribute -- the classic example being the cat
> superimposed in life and death. Since live/dead are not the
> eigenvalues of any observable,
Sure it is---use a stethoscope. :-)
OK in truth, you're right---using a stethoscope involves a long
sequence of zillions of actual quantum mechanical "observations".
> AFAIK, we don't have any idea what we
> are talking about. Vector sums are not ghostly superpositions of the
> constituents, but perfectly good vectors in their own right.
Of course.
Schroedinger's point though---since cats aren't ever half dead and
alive in experiment---is that the laws of physics of macroscopic
bodies must somehow appear to further select some kinds of states.
Such a vector sum---just stated by itself---is not theoretically
disallowed by the laws of orthodox quantum mechanics, but still, it
doesn't happen. In classical mechanics you can always 'imagine' some
state (canonical momenta and positions) of some system (take orbital
mechanics of planets) that by itself does not violate any basic requirement,
but the laws of physics could make it impossible or extraordinarily
unlikely that you could evolve from some given starting state to the
hypothesized ending state.
Schoedinger's half-live, half-dead cat in a box seems to be
incorrectly viewed by many in the general public as an actual
description of what quantum mechanics actually says is going on.
His point was quite the reverse, as a "theoretical" experimental
counterexample. "something" must account for the obvious experimental
differences between small quantum systems and the large ones seen
in everyday experience. What is it? And is it already there in
quantum mechanics?
Dirk Bruere
"Dr Chaos" <mbkennelS...@NOSPAMyahoo.com> wrote in message
>
> Since I believe that measurement devices also happen to be governed by
> the laws of quantum mechanics, and their atoms don't know an
> 'observation' from a hole in their spin, then something about
> decoherence (what measuring appears to do) has to be there inside, and
> not outside, the dynamics of quantum mechanical systems.
>
> When asked "what is time?", Einstein once replied: "time is what
> clocks measure." It may seem like a glib joke, but it is
> fundamentally the best and most truthful answer.
>
> It means that the the physics of the atoms that make up clocks (and
> everything else) is described by some dynamics with a d/dt on the left
> hand side of some equation. In physics, everything "is" what it
Which no doubt leads to some ugly non linearities and hard sums.
Dirk
Mark
>Since I believe that measurement devices also happen to be governed by
>the laws of quantum mechanics, and their atoms don't know an
>'observation' from a hole in their spin, then something about
>decoherence (what measuring appears to do) has to be there inside, and
>not outside, the dynamics of quantum mechanical systems.
"Decoherence" refers to the phenomena whereby macroscopic observers
and systems all find themselves on the same (or compatible) set of
basis vectors in Hilbert space because of their mutual entanglement,
so that the macroscopic world ends up being in one (of the infinite
number of mutually incompatible) boolean logics that reside within
the quantum lattice of subspaces.
Such a circumstance is guaranteed by the existence of a long-range
unshieldable interaction; particularly one whose interaction events
are quantized at the microscopic level and so only happen in fits
and starts there, but which operates "continuously" as seen at the
macroscopic level and so provides a continual mutual entanglement
of systems at that level.
Such a force is already well-known. It's called gravity. The
gravity of a macroscopic system provides you with continuous
observation of that system which makes "Schroedinger Cat" type
situations impossible -- except when they involve two gravitationally
indistinguishable macroscopic states (e.g. states only distinguished for
instance by polarization).
It can't be shielded. There's no such thing as a gravitational
Schroedonger Cat's box. So, the decoherence is nearly automatic.
And no, this has absolutely nothing to do with Penrose's "gravity
collapses the wavefunction". So, please don't get this mangled in
with what's being discussed here. It is simply a basic point:
you have a long-range unshieldable force
which works continuously at the macroscopic level
THEREFORE
you ALREADY have decoherence at the macroscopic level,
regardless of what OTHER reasons one may pose for the phenomenon.
Ralph Hartley
> I interpreted your question to be roughly "What if we could measure
> one aspect of the state without affecting the state?"
It is possible to interpret some of the questions I have seen here to mean
that.
The answer is that it doesn't make any sense.
In Quantum Mechanics a measurement of one aspect of the state affects the
state is by making the state agree with the aspect that was measured.
Let's use an analogy.
Suppose I flip a coin, and don't look at it. Assuming that I am a Baysian,
I might discribe the state of the world as <%50 heads %50 tails>. If I
observe the coin, I either see heads and the state becomes <heads>, or
tails and the state becomes <tails>.
What if I could observe the coin without changing this state? That would
mean observing if the coin is heads or tails without finding out if it is
heads or tails.
Quantum mechanics is a lot like Baysian staistics. Probabilities are
replaced by complex amplitudes, and the state takes into account what you
could know as well as what you actually know (so it isn't as subjective),
but it still makes no sense to make a measurement without discovering the
outcome.
Ralph Hartley
Ralph Hartley
> A collapse in a superposition of conditions. And this conception of
> conditions is introduced to warn of the method of theory that
> you used to define superposition. A condition is a what?
>
> A concept assumed equal to correlated collapse. A word change
> unwarranted.
>
> I call just this threads concept, the real theory of superposition.
> Also called correlated conditions, here.
I think the only answer I can make to that is "Huh?"
Ralph Hartley