QM allows non-local signaling
Martin Schlottmann
> It's official. Quantum Mechanics allows classical information to be
> sent non-locally. See the article quant-ph/9904075 by Srikanth on
> http://xxx.lanl.gov for details of a thought experiment showing this.
No need to worry, everthing's cool.
Srikanth suffers from the misconception that
photons doing a double slit while being entangled
with another system such that some observation
on the second system might reveal information
about which slit is passed will produce the usual
double slit interference pattern when that
second observation is not made. This is clearly
not the prediction of QM.
--
Martin Schlottmann
Eugeny Kornienko
> See the article quant-ph/9904075 by Srikanth on http://xxx.lanl.gov
> If the position is measured in the remote beam, however, the photons
> will be too localized to go through both slits, and the interference
> pattern will disappear.
I think photon localizes by x in the point of measuring of x, not
everywhere. For the interference will disappear it is needed the
photon's wave function to collapse by x in between the slit and
the screen (or second slit), not only in the first slit.
If even the remote collapse by x occurs in the slit then the "fringy"
pattern on Bob's screen will be provided by indeterminacy in momentum
p. So I think the experiment cannot show non-local collapse of wave
function.
john baez
Matt McIrvin <mmci...@world.std.com> wrote:
>So has Srikanth demonstrated an observable that is an exception to this
>rule? I don't think so, but here I should probably defer to people more
>willing to dive into the details of Young interferometers...
It's probably best *not* dive into the details of Young interferometers,
but simply to say: Srikanth shows that if you modify the usual rules of
quantum mechanics, you can get non-local signaling. It's then an
experimental question to see whether Srikanth's modification of quantum
mechanics is correct. If he hasn't done any experiments to support this
claim, it doesn't seem very likely.
(Greg Weeks)
: It's official. Quantum Mechanics allows classical information to be
: sent non-locally.
I doubt it.
: In short, Srikanth sends one beam of a positionally entangled source
: through a double slit to get an interference pattern. If the position
: is measured in the remote beam, however, the photons will be too
: localized to go through both slits, and the interference pattern will
: disappear.
As stated, this is mistaken. The positional entanglement alone is enough
to eliminate the interference pattern.
Greg
Vesselin Gueorguiev
>
> In article <7g9sd5$bgk$1...@nntp5.atl.mindspring.net>, "John N White"
> <jn...@mindspring.com> wrote:
>
> >It's official. Quantum Mechanics allows classical information to be
> >http://xxx.lanl.gov for details of a thought experiment showing this.
>
> appealing puzzle and it is very short).
I was going to download the article some day but you message made
me do it today.
>
> Srikanth makes an implicit claim that is, in its way, even more radical
> than instantaneous signaling.
[...]
I haven't read the article yet, but Dejanews search on R.Srikanth e-mail
address sr...@iiap.ernet.in shows that he has been posting in
list.sci.lang.constru, so he may be willing to say something in
sci.physics.research about this topic.
If the article is similar to what John N White was discussing I would
like to here more opinions since I have my doubts about the outcome
of the experiment:
The most resent experiment on EPR-paradox and Bell's inequality,
shows that if we look at any one side, we only see random polarized
photons, but when we look at the data for both sides we find strong
correlation as expected.
If we put *polarization filter* at one of the paths of the entangled
photons we will filter only one polarization this increases the probability
for the other photon to be in a certain polarized state but this does not
mean that we will have only one polarization at the double slit where
we have two polarizers in the slits.
My guess is that the efficiency of the *polarization filter* will be
correlated with the intensity of the interference pattern. However,
I could be wrong, and that is why I have set my UnCover alert for
any papers on this topic waiting to see when some one
will do the experiment.
torqu...@my-dejanews.com
Martin Schlottmann <martin_sc...@math.ualberta.ca> wrote:
> Srikanth suffers from the misconception that
> photons doing a double ...
Can't someone not hurry up and produce a non-existence theorem showing, once
and for all, that you can't signal faster than light using QM and a standard
kit of physics parts including lights, slits, polarising filters and
orientation randomisers :-) How come someone hasn't managed to produce such a
result so far? Or have they? And if they have why aren't more people aware of
it so we can do away with all these disappointments about FTL signalling!
--
Torque
http://travel.to/tanelorn
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
ca31...@bestweb.net
mmci...@world.std.com (Matt McIrvin) wrote:
> In article <7g9sd5$bgk$1...@nntp5.atl.mindspring.net>, "John N White"
> <jn...@mindspring.com> wrote:
>
> >It's official. Quantum Mechanics allows classical information to be
> >sent non-locally. See the article quant-ph/9904075 by Srikanth on
> >http://xxx.lanl.gov for details of a thought experiment showing this.
>
This paper raises some fun questions, but there seem to be
a few overly squared corners though:
1) The ideal source is supposedly producing only entangled pairs
2) At a minimum, some number n pairs are necessary to distinguish
the diffraction pattern from the interference pattern
3a) The source produces the n pairs in parallel (simultaneously);
in which case their positions are deterministic
3b) The source produces the n pairs sequentially (coherently);
in which case their momenta are deterministic
3c) The source produces the n pairs as a mixture of serially and
parallely produced pairs but the proportions of these
determine the proportion of the n photons which Alice can
make either an absolute position or momentum measurement
upon. This necessesitates increasing n to maintain the
differentiation of the patterns ?
It would be interesting to see what and how Srikanth would compute
the expected bit-rate of the information transmission.
---
http://www.bestweb.net/~ca314159/
Vesselin Gueorguiev
[...]
> No need to worry, everthing's cool.
>
> Srikanth suffers from the misconception that
> photons doing a double slit while being entangled
> with another system such that some observation
> on the second system might reveal information
> about which slit is passed will produce the usual
> double slit interference pattern when that
> second observation is not made. This is clearly
> not the prediction of QM.
Could you say this differently? I am having problem
to understand it.
Thanks.
john baez
>Can't someone not hurry up and produce a non-existence theorem showing, once
>and for all, that you can't signal faster than light using QM and a standard
>kit of physics parts including lights, slits, polarising filters and
>orientation randomisers :-) How come someone hasn't managed to produce such a
>result so far? Or have they?
They have, they have! For crying out loud, it's built right into the
axioms of quantum field theory! The only way you can communicate faster
than the speed of light is by violating these axioms, and no theorist's
"thought experiment" is gonna demonstrate a violation of those axioms -
the only way to prove them wrong is to do an ACTUAL experiment in which
you ACTUALLY demonstrate that quantum field theory is wrong.
>And if they have why aren't more people aware of
>it so we can do away with all these disappointments about FTL signalling!
Because most of the people who spend their time trying to do
impossible things are not the ones who spend their time learning
the proofs that these things are impossible. Look at the circle-
squarers and angle-trisectors! You might think that before
attempting to square the circle with a ruler and compass, anyone
would want to become an expert in mathematics and learn the proof
that it's impossible (and maybe find a flaw in the proof if there
were one - which there's not). But that would be naive. There are
certain people who, when they decide they're gonna do something,
don't *want* to learn why it's impossible. I think one reason is
that certain people *don't believe in the idea of impossibility*
and reject the whole idea of a proof that something is impossible.
(Note: I'm not saying faster-than-light communication is impossible,
only that it's impossible if quantum field theory is correct.)
Dirk Bruere
>
> >And if they have why aren't more people aware of
> >it so we can do away with all these disappointments about FTL signalling!
> Because most of the people who spend their time trying to do
> impossible things are not the ones who spend their time learning
> the proofs that these things are impossible. Look at the circle-
Quite so. If everyone knew what was impossible and did not make any
foolish attempt to overcome it science and technology would grind to a
halt.
Everyone knew the earth was flat, that heavier than air manned flight
was impossible, that space travel was 'utter bilge' etc.
It's not theory that determines what is and is not possible but Nature.
And the only way that Nature is going to tell us anything is through
experiment. That's what seperates us from the ancient Greeks.
Dirk
Matt McIrvin
>They have, they have! For crying out loud, it's built right into the
>axioms of quantum field theory! The only way you can communicate faster
>than the speed of light is by violating these axioms, and no theorist's
>"thought experiment" is gonna demonstrate a violation of those axioms -
>the only way to prove them wrong is to do an ACTUAL experiment in which
>you ACTUALLY demonstrate that quantum field theory is wrong.
Well, there's another possibility, I suppose-- you could prove that the
axioms are internally inconsistent in some glaring way. If you really
could derive faster-than-light communication from QFT, it wouldn't prove
that you really could do it, but it *would* deal a major blow to quantum
field theory, and that would be intrinisically interesting.
Certainly nobody has proven beyond a doubt that the axioms are internally
consistent, since, after all, nobody has proven beyond a doubt that
arithmetic is internally consistent; and we have much greater reason to
believe that QFT is at best incomplete.
However, there's the question of how reasonable it is that the trouble
with QFT will include something that gross. Srikanth's derivation is an
even more extreme case; if it were right it would imply gross internal
inconsistency even in nonrelativistic QM, the kind of thing you use to
analyze the hydrogen atom on an undergraduate level. It's always possible
that there's something that blatant that everybody has overlooked, but
given the sheer number of calculations done in the field, it's unlikely.
To knock down the colossus you need a lot of ammunition.
--
Matt McIrvin http://world.std.com/~mmcirvin/
Toby Bartels
>John Baez <ba...@math.ucr.edu> wrote:
>>Most of the people who spend their time trying to do
>>impossible things are not the ones who spend their time learning
>>the proofs that these things are impossible.
>Everyone knew the earth was flat, that heavier than air manned flight
>was impossible, that space travel was 'utter bilge' etc.
Never in the history of science was there a single moment
when everyone knew these things
or anyone proved them from a widely accepted theory.
-- Toby
to...@ugcs.caltech.edu
john baez
>> >And if they have why aren't more people aware of
>> >it so we can do away with all these disappointments about FTL signalling!
>> Because most of the people who spend their time trying to do
>> impossible things are not the ones who spend their time learning
>> the proofs that these things are impossible.
>Quite so. If everyone knew what was impossible and did not make any
>foolish attempt to overcome it science and technology would grind to a
>halt.
No: if everyone knew what was impossible, people would avoid wasting
their time on impossible things and science and technology would move
ahead a little faster.
But I think you misunderstood my post. I explicitly said we *don't*
know that faster-than-light communication is impossible. We *can't*
know for sure that it's impossible, because this is a question for
EXPERIMENT to decide, and we can never be sure we've tried all possible
tricks to communicate faster than light. I merely said that faster-
than-light communication is impossible *given the usual axioms of
quantum field theory* - this is a MATHEMATICAL fact. If people recognized
this, they would stop wasting their time attempting to dream up ways
to communicate faster than light within the framework of quantum field
theory. Instead, they would use their time more productively, e.g. by
doing experimental physics, or by thinking about general relativity,
which lies outside the framework of ordinary quantum field theory.
Consider the Alcubierre warp drive, for example! At present this is
the most promising approach to faster-than-light transportation.
If Alcubierre hadn't turned his attention to general relativity,
he would have never invented this.
bj flanagan
> However, there's the question of how reasonable it is that the trouble
> with QFT will include something that gross. ... It's always possible
> that there's something that blatant that everybody has overlooked, but
> given the sheer number of calculations done in the field, it's unlikely.
> To knock down the colossus you need a lot of ammunition.
bj: Indeed. One can confront intelligent and well-educated persons with the
evidence of their senses, with lucid mathematical certainties, with working
mechanical models ... and they will persist in believing what everyone agrees
is so. And there is at times a more fundamental difficulty than simply
changing one's beliefs: There is the famous experiment wherein college
students were shown a deck of cards in which the colors were reversed--i.e.,
black hearts and diamonds, red clubs and spades. A surprising number of the
students needed quite a bit of time before they could say what was wrong with
the picture, and some of them had to really sweat it out, growing quite
agitated, certain that something wasn't right but not being able to say what
it was. And they had the cards before them all the while.
--
Quanta & Consciousness
http://www.freeyellow.com/members/sentek/index.html
Wolfram Schmied
On Sat, 8 May 1999 15:28:34 GMT, ba...@galaxy.ucr.edu (john baez)
wrote:
>In article <7gqg8l$d90$1...@nnrp1.deja.com>, <torqu...@my-dejanews.com> wrote:
>>Can't someone not hurry up and produce a non-existence theorem showing, once
>>and for all, that you can't signal faster than light using QM and a standard
>>kit of physics parts including lights, slits, polarising filters and
>>orientation randomisers :-)
>They have, they have! For crying out loud, it's built right into the
>axioms of quantum field theory!
Do mean it's an axiom, or are you saying it can be easily proved? I
have absolutely no knowledge of QM, so I'd appreciate a few sources
WRT to this problem. Please remember, what's "easy" for you, might not
be for someone else, like me for instance.
Thanks
Wolfram "neophyte" 8-)#
Michael Weiss
:
:Quite so. If everyone knew what was impossible and did not make any
:foolish attempt to overcome it science and technology would grind to a
:halt.
:Everyone knew the earth was flat, that heavier than air manned flight
:was impossible, that space travel was 'utter bilge' etc.
Showing, I guess, that the folks who try to do impossible things spend
as little time learning history as they spend learning impossibility
proofs, preferring to repeat old canards and urban legends.
The ancient Greeks knew that the Earth was round. Heck, even Aristotle
listed several arguments for this, all based on *observation*, not "pure
thought". Scholars knew this throughout the Middle Ages. Read Dante,
for example.
As for heavier than air flight, or space travel, these have never
contradicted any of the standard physical theories at *any* time, even
as these theories have shifted and changed over the centuries. (Check
out Kepler's "Somnium" for some speculation about space travel.) Of
course, there have always been skeptics about *practicality*, from an
engineering standpoint.
And the skeptics may have had a point. From Da Vinci to the Wright
brothers is what, 400 years?
You want an accurate analogy, don't compare a theoretical impossibility
(like FTL signalling) with an engineering brick-wall. For example, some
folk today argue that human interstellar travel will never occur--- not
because it is theoretically impossible (it isn't), but because the
engineering obstacles are too great.
Never? Well, never is a long time, for something that is not
*theoretically* impossible.
Aaron Bergman
>But I think you misunderstood my post. I explicitly said we *don't*
>know that faster-than-light communication is impossible. We *can't*
>know for sure that it's impossible, because this is a question for
>EXPERIMENT to decide, and we can never be sure we've tried all possible
>tricks to communicate faster than light. I merely said that faster-
>than-light communication is impossible *given the usual axioms of
>quantum field theory* - this is a MATHEMATICAL fact.
I have a question about this. I'm fairly sure it's something simple, but I
haven't gotten a straight answer. It is an axiom that operators commute at
spacelike separation in QFT. However, there are still correlations at
spacelike separation. What exactly does the correlation mean?
Aaron
--
Aaron Bergman
<http://www.princeton.edu/~abergman/>
Dirk Bruere
> >>Most of the people who spend their time trying to do
> >>impossible things are not the ones who spend their time learning
> >>the proofs that these things are impossible.
> >was impossible, that space travel was 'utter bilge' etc.
> when everyone knew these things
> or anyone proved them from a widely accepted theory.
A bit like non-local signalling.
Nobody needed to prove the Earth was flat. Just the opposite.
Even so, it does not seem to stop people stating what is and is not
possible based on the 'obviously absurd'.
Dirk
[Moderator's note: Educated Westerners have known for a very
long time that the earth is round. For example, the reason
King Ferdinand's panel of experts refused to back Columbus'
attempt to sail to India was not because they thought the
earth was flat, but because knowing its radius as computed
by Eratosthenes around 240 BC, they correctly believed India
was very far. For details see:
http://www-spof.gsfc.nasa.gov/stargaze/Scolumb.htm
-jb]
bu...@pac2.berkeley.edu
>Everyone knew the earth was flat,
I know of know evidence that, at any point in history, "everyone knew"
this. Certainly, from well before the time of Aristotle onwards,
educated folks in Europe understood that there was overwhelming
evidence that the Earth was a sphere.
The evidence includes the straightforward observation that ships drop
below the horizon as they sail away; Eratosthenes's famous estimate of
the Earth's radius by noting the way the angular position of the Sun
above the horizon varies with latitude; the fact that the Earth's
shadow on the Moon during a lunar eclipse is always circular (even if
the Earth were a disc the shadow would in general appear elliptical);
and probably lots of other things.
(American students often used to be taught that Columbus was the only
one who thought the Earth was round. I've always assumed that this
was just part of our national mythology.)
By the way, just to inject a bit of physics content (or at least
history-of-physics content) to this post, let me ask a question I've
wondered about for a while. Does anybody know how the scale of the
solar system was first established? My understanding is that there
were pretty good estimates not long after Copernicus. Once you've got
the right (heliocentric) model, relative distances are easy to get
from geometrical considerations, but I wonder how the overall scale
was established.
Can you use parallax from two widely separated points on Earth to get,
say, the distance to Mars? I'm not sure if you can do it without
accurate timekeeping to make sure both observations are made
simultaneously. Maybe you can if the observers are at the same
longitude?
>that heavier than air manned flight
>was impossible, that space travel was 'utter bilge' etc.
I wonder whether either of these was really universally believed
either. The first one certainly wasn't believed by, say, da Vinci.
It's true that some people who should have known better (most
famously, the NY Times's editorial page) claimed that rocketry was
impossible in vacuum, because there was nothing to push against, but I
doubt very much that "everyone believed" this, at least not recently.
-Ted
Barry Adams
wrote:
>(Note: I'm not saying faster-than-light communication is impossible,
>only that it's impossible if quantum field theory is correct.)
OK, this bring a flood of question to my mind starting with,
How different would modified QFT that allowed faster than light
communication look. I.E. say the axiom with says space-like
seperated observerables commute, didn't hold would a theory
still be constructable or sensible, what axioms could be used
instead? (This question might to do general to ask, but physists
do tend to look for simple modifications of existing theories if for
no other reason than to confirm the existing theory).
Could an above style of theory still fit in with known experiments?
Do the axioms also prevent backward in time signalling? Even one
with only a probablity (but one better than guessing) of being
correct?
Do post QFT theories like loop quantum gravity, spin networks,
and string theory, also have the same axiom, or does the axiom
emmerge from the theory in certain cases?
How is space-like seperation even defined in quantum gravity
theories?
Barry Adams
Alejandro Rivero
Try backwards from the paper. Notice that he needs to ask B to measure both
position and momentum (he says "Energy" there) in order to assert the result.
Then work out why it needs to do so. As an aside, it seems that a lot of
paper on this things come from Indian researchers. I guess they are into
something, but they do not concrete it...
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
Jim Carr
}
} >And if they have why aren't more people aware of
} >it so we can do away with all these disappointments about FTL signalling!
}
} impossible things are not the ones who spend their time learning
In article <3734DF4E...@kbnet.co.uk>
art...@kbnet.co.uk writes:
>
>Quite so. If everyone knew what was impossible and did not make any
>foolish attempt to overcome it science and technology would grind to a
>halt.
Forgetting about the earth-is-flat bit, there are quite a few
experiments that have been realized in the last few years that
I thought would always be gedanken experiments. What Aspect did
was impressive, but the precision of the last round of Bell's
Theorem tests is astounding. It is one thing to manage to count
neutrinos from the sun, it is quite another to image the sun with
them or detect the neutrinos from a supernova.
One person thought about looking for odd effects in Kaon decay,
but gave it up and became a theorist. Now we have all sorts of
data on CP and T violation and are building a B factory to do more.
Physics sometimes moves ahead because a clever experimentalist
tries something that should be impossible, but does it anyway.
--
James A. Carr <j...@scri.fsu.edu> | Commercial e-mail is _NOT_
http://www.scri.fsu.edu/~jac/ | desired to this or any address
Supercomputer Computations Res. Inst. | that resolves to my account
Florida State, Tallahassee FL 32306 | for any reason at any time.
John Baez
Wolfram Schmied <wsch...@mail.blinx.de> wrote:
>On Sat, 8 May 1999 15:28:34 GMT, ba...@galaxy.ucr.edu (john baez)
>>For crying out loud, it's built right into the
>>axioms of quantum field theory!
>Do mean it's an axiom, or are you saying it can be easily proved?
There are different axiom systems for quantum field theory. In
some of them, the impossibility of faster-than-light communication
is an axiom. For example, in the Haag-Kastler axioms, there's an
axiom which says (roughly) that if every light ray going through
a region R of spacetime hits the region S, then anything you can
measure in R can be computed from stuff you can measure in S.
This is a way of saying that no influence travels faster than light.
(Actually, by translating it from math into English, I have made
their axiom vaguer than it originally was, and left it open to
certain misinterpretations. But that's life. The original axiom
says "if R lies in the causal shadow of S, the C*-algebra of
observables for R is contained in C*-algebra of observables for
S".)
>I have absolutely no knowledge of QM, so I'd appreciate a few sources
>WRT to this problem. Please remember, what's "easy" for you, might not
>be for someone else, like me for instance.
I don't know any "easy" references on axiomatic quantum field theory, but
my favorite references are "Introduction to Axiomatic Quantum Field Theory"
by Bogoliubov, Logunov, and Todorov, and "Local Quantum Physics" by Haag.
I'm afraid both make extensive demands on ones knowledge of math and physics.
ba...@math.ucr.edu
Aaron Bergman <aber...@Princeton.EDU> wrote:
>I have a question about this. I'm fairly sure it's something simple, but I
>haven't gotten a straight answer. It is an axiom that operators commute at
>spacelike separation in QFT.
Yes - but by the way, that's not the axiom I was referring to when I
said the axioms of QFT forbid faster-than-light communication. It's
easy to see that the axiom you're mentioning does *not* forbid
faster-than-light communication. For example, you can often interpret a
classical field theory as a quantum field theory in which all operators
commute. Then the axiom you're referring to holds automatically, even
when your classical field theory allows propagation of information faster
than light.
>However, there are still correlations at
>spacelike separation. What exactly does the correlation mean?
I'm not sure what you want to know - it's just a correlation... you're
not mixing up correlation and causal influence, I hope! If I wander
around the world and see which languages different people speak
in different places, I'll observe correlations at spacelike
separated locations. For example, right NOW - pick your favorite
spacelike surface and call it "now" - lots of people all throughout
Germany are speaking German. But this doesn't imply that they
are using faster-than-light telepathy or something to tell each
other "Okay, folks, let's all speak German now!" The decision to
use German spread much slower than light; it just happens to have
spread quite widely by now.
Similarly, the fact that there are correlations between field
operators at spacelike separated points should not seem spooky
or surprising: after all, if points A and B are spacelike separated,
there is always a point C which has both of them in the future
lightcone, which allows us to explain correlation between A and
B as the result of a common influence from C. If you want, you
can probably say this is the "reason" why these correlations exist.
Maybe this is the kind of answer you want? One might be able to
do some computations in free field theory that "explain" the spacelike
correlations this way.
Slightly more spooky is the fact that even in a free field theory,
the vacuum state violates Bell's inequality: the correlations between
fields at spacelike separated points can't be explained *classically*
without resorting to the hypothesis of faster-than-light causal
influence. But of course this is okay, because we're talking about
*quantum* field theory. It's perfectly common for quantum theory
to allow correlations that classically could only be explained by
causal influence.
Laurence Yaffe
>In article <7h59bl$qvk$1...@pravda.ucr.edu>, ba...@galaxy.ucr.edu (john baez) wrote:
>>But I think you misunderstood my post. I explicitly said we *don't*
>>know that faster-than-light communication is impossible. We *can't*
>>know for sure that it's impossible, because this is a question for
>>EXPERIMENT to decide, and we can never be sure we've tried all possible
>>tricks to communicate faster than light. I merely said that faster-
>>than-light communication is impossible *given the usual axioms of
>>quantum field theory* - this is a MATHEMATICAL fact.
>I have a question about this. I'm fairly sure it's something simple, but I
>haven't gotten a straight answer. It is an axiom that operators commute at
>spacelike separation in QFT. However, there are still correlations at
>spacelike separation. What exactly does the correlation mean?
Those space-like correlations are telling you about correlations
in the ground state wave function(al). For example, in a free
field theory, the seemingly trivial statement that there are
no particles in the vacuum means that Fourier modes of the field
*are* correlated -- over arbitrarily long distances.
------------------------------------------------------------------------
Laurence G. Yaffe ya...@phys.washington.edu
Department of Physics
University of Washington 1-206-543-3902 (fax: 1-206-543-9523)
Dirk Bruere
>
> Nobody needed to prove the Earth was flat. Just the opposite.
> long time that the earth is round. For example, the reason
> King Ferdinand's panel of experts refused to back Columbus'
> attempt to sail to India was not because they thought the
> earth was flat, but because knowing its radius as computed
> by Eratosthenes around 240 BC, they correctly believed India
> was very far. For details see:
>
> http://www-spof.gsfc.nasa.gov/stargaze/Scolumb.htm
On a philosophical note, I do not 'know' that the Earth is round, I
infer it at best from the evidence presented by people I don't know
doing experiments I cannot (or will not) do myself. Faith in science...
Probably only astronauts come close to 'knowing' rather than
'believing'.
Dirk
Jos Bergervoet
> wondered about for a while. Does anybody know how the scale of the
> solar system was first established? My understanding is that there
> were pretty good estimates not long after Copernicus. [...snip...]
Much earlier!
The heliocentric idea was launched by Aristarchos (of Samos, 310-230 bc.)
The ancient Greeks also tried to compute the distances. Erathostenes
determined the Earths circumference within 0.2 percent correct.
Ptolemaius obtained the distance to the moon within a few percent
(actually the ratio to the earth's diameter.) Posidonius came within
a factor of 2 of the correct distance to the sun. Most estimates of
this last quantity by other Greeks were 10 to 100 times too low.
Aristarchos' own estimate (19 times the distance to the moon) was
about 20 times too low.
[see: Bertrand Russell, "History of western philosophy and .... " ]
[Moderator's note: I'm just reading John North's HISTORY OF ASTRONOMY
AND COSMOLOGY, which has some nice bits on this topic. It's also very
well-written and provides a fairly extensive account of its subject. -P.H.]
-- Greetings,
-- Jos
Nathan Urban
> The ancient Greeks also tried to compute the distances. Erathostenes
> determined the Earths circumference within 0.2 percent correct.
Is that known? I seem to recall that there was some debate over how long
"stadia" (his unit of measurement) really were. If so, it's probably
more than a 0.2% uncertainty in the units alone.. and without knowing
how big the errors on his distance measurements were likely to have been,
that incredible accuracy of that figure might be overemphasized; it could
be accident that the errors cancelled out right. (I don't disagree that
he got a really good figure, though.)
> Ptolemaius obtained the distance to the moon within a few percent
> (actually the ratio to the earth's diameter.) Posidonius came within
> a factor of 2 of the correct distance to the sun. Most estimates of
> this last quantity by other Greeks were 10 to 100 times too low.
> Aristarchos' own estimate (19 times the distance to the moon) was
> about 20 times too low.
Why did the others go wrong on these distances? I'm just wondering,
because it's conceivable that Ptolemaius and Posidonius got good answers
on a fluke and not because of superior methodology.
John Baez
Matt McIrvin <mmci...@world.std.com> wrote:
>In article <7gtk85$jud$1...@pravda.ucr.edu>, ba...@galaxy.ucr.edu (john baez)
>>For crying out loud, it's built right into the
>>axioms of quantum field theory! The only way you can communicate faster
>>than the speed of light is by violating these axioms, and no theorist's
>>"thought experiment" is gonna demonstrate a violation of those axioms -
>>the only way to prove them wrong is to do an ACTUAL experiment in which
>>you ACTUALLY demonstrate that quantum field theory is wrong.
>Well, there's another possibility, I suppose-- you could prove that the
>axioms are internally inconsistent in some glaring way.
I don't think you can do that without proving large hunks of
mathematics are inconsistent, because these axioms have models.
I.e., there are quantum field theories (free theories) that
have been proved to satisfy the axioms. Thus if you could derive
an inconsistency from the axioms of quantum field theory, you could
use this to derive an inconsistency in whatever mathematical
infrastructure (ZF set theory or whatever) you happen to be using.
On the other hand it might be true that there are no *interacting*
theories in 4 dimensions that satisfy the axioms of quantum field
theory. This is an open question.
>Certainly nobody has proven beyond a doubt that the axioms are internally
>consistent, since, after all, nobody has proven beyond a doubt that
>arithmetic is internally consistent; and we have much greater reason to
>believe that QFT is at best incomplete.
The axioms of QFT are *supposed* to be incomplete because they're
supposed to admit many different models. Incompleteness is fine.
Inconsistency is not.
Anyway, if someone proves the axioms of arithmetic or set theory
are inconsistent, faster-than-light signalling is gonna look like
pretty small potatoes. Read your Egan! :-)
Aaron Bergman
>In article <abergman-110...@abergman.student.princeton.edu>,
>Aaron Bergman <aber...@Princeton.EDU> wrote:
>
>>I have a question about this. I'm fairly sure it's something simple, but I
>>haven't gotten a straight answer. It is an axiom that operators commute at
>>spacelike separation in QFT.
>
>Yes - but by the way, that's not the axiom I was referring to when I
>said the axioms of QFT forbid faster-than-light communication.
Really? The discussion on pgs 27-9 of P&S basically uses that axiom. What
is the statement of the axiom that you're referring to?
> It's
>easy to see that the axiom you're mentioning does *not* forbid
>faster-than-light communication. For example, you can often interpret a
>classical field theory as a quantum field theory in which all operators
>commute. Then the axiom you're referring to holds automatically, even
>when your classical field theory allows propagation of information faster
>than light.
Possibly the source of my confusion.
>
>>However, there are still correlations at
>>spacelike separation. What exactly does the correlation mean?
>
>I'm not sure what you want to know - it's just a correlation... you're
>not mixing up correlation and causal influence, I hope!
Hmmmm. I think that the language of a particle propogating from one point
to another might be what's confusing me. A correlation function means that
there is a probability that a particle can propogate from point A to point
B. Of course that doesn't mean you can do anything with that particle,
then, does it. I guess I'm not entirely sure what the point of the
commutation axiom is then....
I need to think about this some more.
Dirk Bruere
>
> >Everyone knew the earth was flat,
> I know of know evidence that, at any point in history, "everyone knew"
> this. Certainly, from well before the time of Aristotle onwards,
'Generalisation for effect' is my excuse.
However, to say that some people 'knew' the Earth was round is also not
true. They may have believed it, or deduced it, but all the evidence was
indirect and 'counterintuitive' insofar as one can see a flat world out
there (barring hills etc). As for ships dropping below the horizon,
you'd need pretty good eyesight, a big ship and a very calm sea until
the invention of the telescope.
As for other 'impossibilities' such as orbital velocity spacecraft one
has to admit that some people questioned the axioms that belied the
'utter bilge' mentality, in the same way some people are questioning the
impossibility of FTL signalling.
Ultimately it comes down to experiment, not theory.
Dirk
(Greg Weeks)
: Hmmmm. I think that the language of a particle propogating from one point
: to another might be what's confusing me. A correlation function means that
: there is a probability that a particle can propogate from point A to point
: B.
For massive particles, this is indeed true *as an approximation*. But it
is not strictly true for a couple of reasons. First, even for a free field
A, there is no entirely satisfactory definition of "position". Second,
even if you are content to take seriously the Newton-Wigner definition of
position, it is not the case that the wave-function of A(x)|0> is localized
at x. It is approximately localized at x, with an exponential tail.
Greg
W.Ta...@math.canterbury.ac.nz
|> On a philosophical note, I do not 'know' that the Earth is round,
You can do.
|> Probably only astronauts come close to 'knowing'
Strictly speaking, they don't either. They only *see* the earth
as a great big circle. So it *could* be a flat disk! Though they
also can see this from (supposedly) two different places.
You can do the same. Stand on a very tall hill (better, a mountain)
near the seaside, and look at the horizon. You can SEE quite clearly
that it's curved. So just do this from two widely differing places
(to eliminate being on a flat disk) and voila!
----------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
----------------------------------------------------------------------
You are what you remember.
----------------------------------------------------------------------
P.S. It's best just at about sunset, on the dark side of the sky; or
at night on a moonlit night - then you see the earth against
the starry background... "hanging" there in space!
Jos Bergervoet
> In article <...>, Jos Bergervoet wrote:
>> The ancient Greeks also tried to compute the distances. Erathostenes
>> determined the Earths circumference within 0.2 percent correct.
> Is that known? I seem to recall that there was some debate over how long
> "stadia" (his unit of measurement) really were. If so, it's probably
> more than a 0.2% uncertainty in the units alone..
You have to use that one stadion was 157.5 m. The mid-day height of
the Sun differed by 7+1/7 degree between Alexandrie and Syene, acc. to
Erathostenes, and the distance between those cities was 5000 stadia.
The accuracies of all these three numbers enter the result. Especially
the 5000 stadia distance seems a bit like a 'rounded' value.
But on the other hand, it's possible that they defined their distance
unit after doing an elaborate measurement of the Alexandrie-Syene
distance and then calling that 5000 stadia, much like c=299798542 in
our current definition. Otherwise, their '5000' doesn't look more than
only 10 percent (one digit) accurate. Anyone who knows this?
-- Jos
bu...@pac2.berkeley.edu
Dirk Bruere <art...@kbnet.co.uk> wrote:
>there (barring hills etc). As for ships dropping below the horizon,
>you'd need pretty good eyesight, a big ship and a very calm sea until
>the invention of the telescope.
Well, theorist that I am, I got the experiment backwards. You want to
put the observer on the ship and have him look at a fixed, tall
landmark on shore. As the ship goes out to sea, the landmark appears
to sink relative to the local horizon. If the observer is on land
looking out at a receding ship, things are trickier for the reasons
you mention (mostly that ships, especially ancient ones, weren't very
tall). Strabo, who appears to have been the first to write down this
argument for a spherical Earth, got it right even if I didn't.
(Strabo lived in the first century B.C. as well as the first century
A.D., by the way. To paraphrase Mel Brooks, he lived long enough to
find out what the "C" stood for.)
>Ultimately it comes down to experiment, not theory.
Depends on what "it" is, I guess. If we're talking about the shape of
the Earth, then you're right: that question is to be settled by
experiment. There are plenty of other questions out there, including
the one that started this whole discussion [1], that are by their
nature purely mathematical and can (indeed, must) be settled via "pure
theory." The important thing is to keep your eye on which sort of
question you're dealing with.
-Ted
[1] I'm talking about the question that's still enshrined in the
subject line of this thread: Does quantum mechanics allow
faster-than-light signaling? This thread started as a discussion of a
paper that claimed that it was possible to send such a signal, within
the framework of standard quantum mechanics. John B. claimed that
this was impossible, again within the framework of standard quantum
mechanics. That claim is a purely mathematical one: its truth or
falsity does not depend on any experiment. The question of whether
faster-than-light signaling is possible *in our Universe* of course is
to be settled by experiments, but that's a different question.
John Baez
Aaron Bergman <aber...@Princeton.EDU> wrote:
>What is the statement of the axiom that you're referring to?
I said it in another post on this thread. Roughly, it says that
every observable that can be measured in a given region R of spacetime
is a function of observables in any region S with the property that
every light ray going through R hits some point of S.
There are actually a number of axioms in QFT that deal with "locality
and causality". One is the axiom you mentioned: if R and S are spacelike
separated regions in spacetime, any observable that can be measured in
R commutes with any observable that can be measured in S. Another is
the one above. Yet another is the one saying that the spectrum
of the energy-momentum operator lives in the future lightcone. (In
other words: no spacelike energy-momenta allowed!)
It's a little hard to get ahold of, but "Introduction to Axiomatic
Quantum Field Theory" is a darn good place to start reading about
this stuff. Personally I found it impossible to learn the "chug-and-
plug" quantum field theory (the stuff particle physicists do for a
living: computing scattering amplitudes and all that jazz) without
simultaneously ingesting some of the axiomatic stuff.
Matt McIrvin
aber...@Princeton.EDU (Aaron Bergman) wrote:
>Hmmmm. I think that the language of a particle propogating from one point
>to another might be what's confusing me.
In classical mechanics, particle propagation and causal influence can
happen along the same sets of intervals. In quantum field theory, these
two notions are separated from one another, since particle propagation is
described by *time-ordered products* of operators and causal influence is
described by *commutators* of operators.
When events are separated by spacelike intervals, virtual particles can
still propagate because the time-ordered product of field operators does
not vanish; but the actual time ordering in the time-ordered product stops
mattering, since the commutator vanishes. This is precisely what is
needed to maintain Lorentz invariance. Nothing can depend on whether the
particle went from A to B or its antiparticle went from B to A-- which
stands to reason, since the order of those two events is different in
different rest frames.
--
Matt McIrvin http://world.std.com/~mmcirvin/
Oz
>However, to say that some people 'knew' the Earth was round is also not
>true. They may have believed it, or deduced it, but all the evidence was
>indirect and 'counterintuitive' insofar as one can see a flat world out
>there (barring hills etc).
You clearly haven't visited the seaside recently. Certainly if you are
on a ship you can see lighthouses and cliffs appearing from over the
horizon as you approach and people on clifs see ships before those on
the shore. That's just with the naked eye, with a simple telescope it
would be even more obvious. Looking across the staits of Calais you can
see the opposite cliffs from the clifftop, but not from the shore. The
examples are endless.
>As for ships dropping below the horizon,
>you'd need pretty good eyesight, a big ship and a very calm sea until
>the invention of the telescope.
Not at all. They are, surprisingly, clearly visible against the sky on
the horizon. Indeed you don't even need a big area of water. Try looking
across a perfectly still lake from an inch or so above the water level
to something in the opposite shore (nightime and a shore barbeque is
good here).
[My back of an envelope calculation suggests a 2km lake should produce
about a 70mm, or three inch, 'elevation' in the middle so you would need
to be 6" above the water line to see the opposite shore. Phew!]
--
Oz
Oz
<Word...@Rocketmail.com> writes
>bj: Indeed. One can confront intelligent and well-educated persons with the
>evidence of their senses, with lucid mathematical certainties, with working
>mechanical models ... and they will persist in believing what everyone agrees
>is so.
Hmmm. Regrettably completely true. It's responsible for so much trouble
and strife in the world as well. Indeed IMHO the reason why science is
in general attacked by so many pressure groups is that it can often
*show* that statements of deepheld belief are *wrong*.
Many find this ethically reprehensible.
People in general want what they *believe* to be true, to be true
whether it is or not. People who disagree, or worse show it to be wrong,
must of course be punished, preferably by burning at a convenient post.
I suspect this is the reason for much of the anti-science sentiment you
find around the world and particularly amongst the scientifically
illiterate 'intelligensia' of the western world. Science has one god
called nature, and nature is not ethical: it just is. Most people
believe nature is ethical and cannot get their head round the fact that
ethics is a human thingy and nature is not.
[end/ rant/]
--
Oz
Doug B Sweetser
The Greeks were great with trig, but they were weaker with reality
:-) Sunrise and sunset happen a little "early" and "late" because
the light is refracted by the atmosphere. If you thing light only
travels in absolutely perfectly straight lines, then the distance
based on that incorrect assumtion ends up being low. The light
from the Sun is _very_ parallel.
doug
http://world.com/~sweetser
bj flanagan
> my favorite references are "Introduction to Axiomatic Quantum Field Theory"
> by Bogoliubov, Logunov, and Todorov, and "Local Quantum Physics" by Haag.
> I'm afraid both make extensive demands on ones knowledge of math and physics.
bj: I don't know John's first reference, but Haag's book *does* contain
a fair number of sections which ought to be fairly accessible, as does
Umezawa's *Advanced Field Theory*. Ramond's *Field Theory* is pretty
easy as these things go and of course Feynman's *QED* is a terrific
little introductory text which goes right to the heart of the matter
while requiring very little math and not much physics. For a very nice
historical overview you really ought to look at *Inward Bound* by
Abraham Pais.
--
Quanta & Consciousness
http://www.freeyellow.com/members/sentek/index.html
Vesselin Gueorguiev
[...]
> Those space-like correlations are telling you about correlations
> in the ground state wave function(al). For example, in a free
> field theory, the seemingly trivial statement that there are
> no particles in the vacuum means that Fourier modes of the field
> *are* correlated -- over arbitrarily long distances.
If so, can we assume that the search for a common point in
the far far past will lead us to the Big Bang as the cause for
these correlations?
Does it mean that any QFT with causally interacting fields
*must have* singularity like the Big Bang?
Causality is intimately related to the geometry which is gravity,
and gravity is intimately related to the matter fields, thus all
fields determine what is causally interacting fields. So we came
back to the notion of causally interacting fields, it seems that
this is highly complex construction that must be self-consistent.
It sounds like consistent mean filed theory! Is there, any
realistic numerical calculations using such recursion?
john baez
Vesselin Gueorguiev <vess...@baton.phys.lsu.edu> wrote:
> Does it mean that any QFT with causally interacting fields
>*must have* singularity like the Big Bang?
No. On the contrary, the sort of QFT we're discussing here
assumes spacetime is flat, so there's never any sort of "big
bang" or other singularity in spacetime in this context.
To a zeroth approximation, you can think of a free quantum field
as a stochastically wiggling field where the wiggles spread
out in a manner described by the wave equation. The value of
the field at spacelike separated points will be correlated
because they're both feeling the influence of the same wiggles.
There's no point to trying to trace back these wiggles to an
"first cause" in the Big Bang - they're just random.
This picture is only good to a zeroth approximation because you
really need quantum mechanics to understand what's going on.
Thus I hesitate to even mention this flawed picture. But it
can be handy if you are careful.
Laurence Yaffe
>Laurence Yaffe wrote:
>> Those space-like correlations are telling you about correlations
>> in the ground state wave function(al). For example, in a free
>> field theory, the seemingly trivial statement that there are
>> no particles in the vacuum means that Fourier modes of the field
>> *are* correlated -- over arbitrarily long distances.
> If so, can we assume that the search for a common point in
>the far far past will lead us to the Big Bang as the cause for
>these correlations?
> Does it mean that any QFT with causally interacting fields
>*must have* singularity like the Big Bang?
I don't think you can argue that causality necessitates an initial
singularity. However, in the context of cosmology, you can argue that
the development of these long-range correlations is one of the reasons
to believe that our universe went through a period of accelerated
expansion, known as ``inflation'', sometime before the era of
nucleosysthesis. If so, this means that our observable universe today
has been "blown up" from a tiny region inside of which all points were
in causal contact with each other before the inflationary epoch.
I bet John has some references on inflation readily at hand ...
> Causality is intimately related to the geometry which is gravity,
>and gravity is intimately related to the matter fields, thus all
>fields determine what is causally interacting fields. So we came
>back to the notion of causally interacting fields, it seems that
>this is highly complex construction that must be self-consistent.
>It sounds like consistent mean filed theory! Is there, any
>realistic numerical calculations using such recursion?
Fully realistic? No. But there is a lot of current work on
self-consistent field theory calculations which include gravitational
back-reaction (but not fully quantized gravity). I don't have
references handy, but if you want to search for articles on this topic,
go to http://xxx.lanl.gov/find, select the hep-th, hep-ph and/or gr-qc
archives, and search on people like Mottola, Boyanovsky, or de-Vega.
That will at least get you started.
Kenneth Jaynes
>Does anybody know how the scale of the solar system was first
>points on Earth to get, say, the distance to Mars? I'm not
>sure if you can do it without accurate timekeeping to make
>sure both observations are made simultaneously. Maybe you
>can if the observers are at the same longitude?
Aristarchus (c. 300 BC) had estimated that the Sun is about 19 times
more distant than the Moon, based on a VERY rough estimate of how much
the angle between Sun and Moon differs from 90 degrees when exactly
half the Moon is illuminated as viewed from the Earth. He very
carefully explained the geometry of this calculation (which seems
trivial to us today), but seems to have put very little effort into
making an accurate determination of the actual angle (although, to be
fair, it's not an easy thing to measure). He almost arbitrarily based
his calculation on 87 degrees (i.e., 3 degrees shy of a right angle),
which gave the grossly inaccurate result of 19 Moon distances to the
Sun. I've always suspected that he just intended this as an example
to illustrate the method, hoping someone would make a decent
measurement of the actual angle, but his "3 degrees" became the
traditional value, and continued to be referenced until the 1600's.
(For some reason this reminds me of the opening post in this thread,
which announced that "It's official. QM allows non-local signalling."
Heh heh. "It's official." I like that.)
By the way, the Earth-Moon distance was reckoned to be about 70 Earth
radii by Hipparchus (c. 150 BC) based on the fact that a particular
lunar eclipse was total when seen from the Hellespont, but only 4/5
(at its maximum) when viewed from Alexandria. Of course, Eratosthenes
gave a quite accurate estimate the Earth's radius based on differences
in noon shadows at Alexandria and Cyrene.
Nearly 2000 years after Aristarchus suggested that "someone should
really make a careful measurement of this", someone finally did. In
1630 Vendelinus repeated Aristarchus' observation, and found that the
angle Moon-Earth-Sun couldn't be more than about 1/4 of a degree,
which is just 1/12 of Aristarchus' value. This puts the Sun at least
55 million miles away. That's about as much accuracy as you can
reasonably expect from this crude method.
The first really good estimates were based on observations of
"transits" of Venus across the Sun. Every now and then the inner
planets pass directly between the Earth and the Sun, as Halley was
the first to observe. The planet (Venus or Mercury) appears as a
black dot moving across the disk of the Sun, and Venus can take about
7 hours to go all the way across. Now, depending on your vantage
point, the planet will cut across a slightly different chord as it
traverses the solar disk, and these chords will be of different
lengths. Therefore, you don't need to synchronize your clocks
absolutely at different locations on Earth, you need only measure
(as precisely as you can) how long it takes for the black spot to
cross the face of the Sun from various vantage points. If you can
measure this time interval (at opposite sides of the Earth) to within,
say, 3 seconds, then the parallax can be computed to within about 1%.
This is the main reason Captain Cook was sent to Tahiti, to observe
the transit of Venus in 1769. Other observations of that transit were
made at San Domingo, Hudson Bay, several places in Siberia, etc.
Observations like this reduced the best estimate of Solar parallax
down to about 8.7 seconds of arc, i.e., about 1/7 of a degree, which
brings the Earth-Sun distance up to around 93 million miles.
Vesselin Gueorguiev
[...]
> Slightly more spooky is the fact that even in a free field theory,
> the vacuum state violates Bell's inequality: ....
Would you give the reference where I can look up the details?
john baez
Vesselin Gueorguiev <vess...@baton.phys.lsu.edu> wrote:
>ba...@math.ucr.edu wrote:
>> Slightly more spooky is the fact that even in a free field theory,
>> the vacuum state violates Bell's inequality: ....
>Would you give the reference where I can look up the details?
Stephen J. Summers and Reinhard Werner, The vacuum violates Bell's
inequalities, Phys. Lett. 110A (1985), 257-259.
Ralph Frost
> To a zeroth approximation, you can think of a free quantum field
> as a stochastically wiggling field where the wiggles spread
> out in a manner described by the wave equation. The value of
> the field at spacelike separated points will be correlated
> because they're both feeling the influence of the same wiggles.
> There's no point to trying to trace back these wiggles to an
> "first cause" in the Big Bang - they're just random.
You mean "non-local", don't you?
> This picture is only good to a zeroth approximation because you
> really need quantum mechanics to understand what's going on.
> Thus I hesitate to even mention this flawed picture. But it
> can be handy if you are careful.
WHAT is a "zeroth approximation", something no one in the field has
thought of yet?
Also, in the old days, didn't quantum mechanics "really need" classical
mechanics? Why single out just this handy picture as being flawed,
among the many other handy flawed pictures that are in the gallery?
Thanks.
-ref-
Ralph Frost
> On the other hand it might be true that there are no *interacting*
> theories in 4 dimensions that satisfy the axioms of quantum field
> theory. This is an open question.
Could you expand on what this means, please? Might this be similar in
meaning to a statement which says, "an analog model of some phenomena,
U, is better and more likely to exist than an abstract mathematical
model"?
> The axioms of QFT are *supposed* to be incomplete because they're
> supposed to admit many different models. Incompleteness is fine.
> Inconsistency is not.
Would this inconsistency manifest as perhaps working for one
organizational level but not another, which seemingly appears to be the
same as the first?
> Anyway, if someone proves the axioms of arithmetic or set theory
> are inconsistent, faster-than-light signalling is gonna look like
> pretty small potatoes. Read your Egan! :-)
Doesn't the Godel Theorem say that proof requires some expansion of
terms or else it won't happen?
Might it be more likely that the QFT application crashes because of some
previously unnoticed flaw or fairly straightforward general
misconception in the 'translation' of the math lingo to the physics
lingo?
Thanks for your help.
Best regards,
Ralph Frost
ref...@dcwi.com
States are states.
Lorenz [to_email_see_my_sig] Borsche
wrote:
>Someone uncited by Lorenz Borche wrote:
>>[....] that heavier than air manned flight
>>was impossible, that space travel was 'utter bilge' etc.
>
>I wonder whether either of these was really universally believed
>either. The first one certainly wasn't believed by, say, da Vinci.
Much earlier! Just think of Daedalus.
>It's true that some people who should have known better (most
>famously, the NY Times's editorial page) claimed that rocketry was
>impossible in vacuum, because there was nothing to push against, but I
>doubt very much that "everyone believed" this, at least not recently.
That's the same like the "bumble-bees-cannot-fly"-thaang. All too
often one article in one paper is mistaken for the opinion of the
(scientific) world. Scientific myths are a dozen a penny.
--
Lorenz Borsche http://www.borsche.de
eMail? Name: LBsys Provider: dinx Extension: de
-----------------------------------------------------
All professions are conspiracies against laymen (GBS)
Vesselin G Gueorguiev
> Vesselin Gueorguiev <vess...@baton.phys.lsu.edu> writes:
> > Does it mean that any QFT with causally interacting fields
> >*must have* singularity like the Big Bang?
> I don't think you can argue that causality necessitates an initial
> singularity. However, in the context of cosmology, you can argue that
> the development of these long-range correlations is one of the reasons
> to believe that our universe went through a period of accelerated
> expansion, known as ``inflation'', sometime before the era of
> nucleosysthesis. If so, this means that our observable universe today
> has been "blown up" from a tiny region inside of which all points were
> in causal contact with each other before the inflationary epoch.
This makes room for many other such "tiny regions" that are not
causal contact and may never be!
> I bet John has some references on inflation readily at hand ...
He was wiggling me... I have a book, I bought it as a student long ago,
and I took it with me to USA this year. This is one of the many
I have bought with the wish to read it some day, but they may become
too old fashioned. The book is:
A.D.Linde, "Elementary Particle Physics and Inflationary Cosmology", 1990
Is is good? I think it is, but I have not gotten time to read it all.
> [...] there is a lot of current work on
> self-consistent field theory calculations which include gravitational
> back-reaction (but not fully quantized gravity). I don't have
> references handy, but if you want to search for articles on this topic,
> go to http://xxx.lanl.gov/find, select the hep-th, hep-ph and/or gr-qc
> archives, and search on people like Mottola, Boyanovsky, or de-Vega.
> That will at least get you started.
Thanks for the handle on this. I glanced at the e-print by those
names. There is some interesting stuff.
John Baez
>john baez wrote:
>> There's no point to trying to trace back these wiggles to an
>> "first cause" in the Big Bang - they're just random.
>You mean "non-local", don't you?
No, I meant what I actually wrote.
>> This picture is only good to a zeroth approximation because you
>> really need quantum mechanics to understand what's going on.
>> Thus I hesitate to even mention this flawed picture. But it
>> can be handy if you are careful.
>WHAT is a "zeroth approximation", something no one in the field has
>thought of yet?
No. If
f(x) = a_0 + a_1 x + a_2 x^2 + ...
then we call a_0 the "zeroth approximation" to f(x), while
a_0 + a_1 x is the "first approximation" to f(x) and so on.
In colloquial physics lingo, the "zeroth approximation" to a
quantity is the stupidest, crudest approximation possible:
the approximation you can come up with before you even fire
up your brain and start thinking. The "first approximation"
is the one that requires a little thought.
>Also, in the old days, didn't quantum mechanics "really need" classical
>mechanics? Why single out just this handy picture as being flawed,
>among the many other handy flawed pictures that are in the gallery?
Because I want to steer the readers away from making a bunch of errors
I can easily imagine them making if they're not careful. I can easily
picture them falling off the cliff at this particular bend in the trail.
Vesselin G Gueorguiev
>
> In article <7hrl5a$bou$1...@info.service.rug.nl>,
> Vesselin Gueorguiev <vess...@baton.phys.lsu.edu> wrote:
> > Does it mean that any QFT with causally interacting fields
> >*must have* singularity like the Big Bang?
>
> assumes spacetime is flat, so there's never any sort of "big
> bang" or other singularity in spacetime in this context.
I agree, but I am more inclined to discuss a space where the
fields determine the geometry.
>
> To a zeroth approximation, you can think of a free quantum field
> as a stochastically wiggling field where the wiggles spread
> out in a manner described by the wave equation. The value of
This is fine for flat space, but in general space it may get
quite a wiggling. What is a wave equation then?
> the field at spacelike separated points will be correlated
> because they're both feeling the influence of the same wiggles.
So this is the common cause! (the same wiggles)
> There's no point to trying to trace back these wiggles to an
> "first cause" in the Big Bang - they're just random.
We are very unhappy if our solutions to the wave equation are
not well behaved at infinity (far away). Thus we have our
wiggling field well modulated and localized. I guess that
often we can find continues transformation that makes them
more and more localized (up to some degree).
>
> This picture is only good to a zeroth approximation because you
> really need quantum mechanics to understand what's going on.
> Thus I hesitate to even mention this flawed picture. But it
> can be handy if you are careful.
If you want me to think of this wiggles as vacuum fluctuations,
that's fine with me.
Vesselin G Gueorguiev
<jn...@mindspring.com> wrote:
>It's official. Quantum Mechanics allows classical information to be
>sent non-locally. See the article quant-ph/9904075 by Srikanth on
>http://xxx.lanl.gov for details of a thought experiment showing this.
I was hoping some one would point out what is wrong, what is right,
and what is more or less subject to dispute in this paper. Since I haven'=
t=20
read any posts, satisfactory to me so far, I decided to state my opinion=20
(hoping to be corrected when wrong).
=20
In the abstract we read:
[=85 sender measuring either position or momentum of particles in a pure=
=20
ensemble of entangled EPR pairs. This is shown to leave their EPR=20
counterparts as localized particles or plane waves. =85]
This means that plane wave or localization property are regarded as=20
characteristic of the entangled photon system. This looks OK for plane
wave as far as we don=92t do any measurement (absorbing a photon), but=20
as soon as we detect one of the photons (localize it) at some point x_a=20
the above statement claims that the second photon is localized too at x_b=
.=20
What if we don't put any measurement instruments at x_b? Clearly then the
second photon will just keep on moving (most likely as plane wave).
In section III the state of the entangled system is assumed to be:
Psi(x_a, x_b) =3D Integral(Exp[(2*pi*i/hbar)(x_a - x_b + x_0)p] dp)
This is Dirac delta function ( hbar*delta[x_a-x_b+x_0]). By requiring =
the
integrity (conservation) of this state, no matter what we do on the photo=
ns,=20
the plane wave or localization property of the entangled photon system ar=
e=20
provided. However, I argued that this is unlikely to be the case.
Then, after equation 7 we read:
[=85 if Alice detects a photon at a given slit of K, then the B-counterpa=
rt=20
would be found localized at the corresponding slit in L. Each photon in B=
=20
does not interfere with itself, but passes through one or the other slit =
in L=85]
If it was "found localized at the corresponding slit in L" how come=20
it " passes through one or the other slit in L" ?
Vesselin G Gueorguiev
> Try backwards from the paper. Notice that he needs to ask B to measure both
> position and momentum (he says "Energy" there) in order to assert the result.
The only place in the paper mentioning "energy" is:
[...Bob's screen is designed to record only photons within
a sufficiently narrow energy band so that the resultant
fringes have measurable visibility...]
I think this signals some confusion of author (or me)... Emission of
entangled photons naturally must have very narrow energy band.
That is because of the Energy - momentum conservation.
> Try backwards from the paper....
I did try backward, forward and randomly, and I will post on that later.
Dirk Bruere
>
> >I wonder whether either of these was really universally believed
> >either. The first one certainly wasn't believed by, say, da Vinci.
> Much earlier! Just think of Daedalus.
Yes, but believed for the *wrong* reason. Birdlike, wingflapping, human
flight is impossible (at least until genetic engineering lends a hand).
Of course, that's the flip side of dis-belief.
However, being right for the wrong reasons can be very amusing as well
as annoying. Remember Velikovsky (?) and his theory that comets have
made a big impact on earth? (sorry about the pun)
Dirk
Lorenz [to_email_see_my_sig] Borsche
>> Much earlier! Just think of Daedalus.
>
>Yes, but believed for the *wrong* reason. Birdlike, wingflapping, human
>flight is impossible (at least until genetic engineering lends a hand).
Halfright. Birdlike could be like an Albatross (that's practically
without wingflapping) and is done today. To rise from the island a
little wingflapping owuld have been necessary, so, indeed, that's
impossible. Or seems at least (who knows - before the man-made
Albatross with a bikelike human powersource had lifted off, no one
believed it would ever be possible. But it showed, that 200 Watts are
enough to lift 100 Kg via airfoils. The rest is a question of
mechanics and training, I suppose)
>However, being right for the wrong reasons can be very amusing as well
>as annoying.
You say.
>Remember Velikovsky (?) and his theory that comets have
>made a big impact on earth? (sorry about the pun)
Arrrgh. But I'm sure there's lots of examples in the realm of physics
as well.
john baez
Vesselin G Gueorguiev <vess...@baton.phys.lsu.edu> wrote:
>john baez wrote:
>> In article <7hrl5a$bou$1...@info.service.rug.nl>,
>> Vesselin Gueorguiev <vess...@baton.phys.lsu.edu> wrote:
>> > Does it mean that any QFT with causally interacting fields
>> >*must have* singularity like the Big Bang?
>> No. On the contrary, the sort of QFT we're discussing here
>> assumes spacetime is flat, so there's never any sort of "big
>> bang" or other singularity in spacetime in this context.
>I agree, but I am more inclined to discuss a space where the
>fields determine the geometry.
Okay, but then you're talking about quantum gravity, and much
less is known about that than ordinary quantum field theory.
If one is trying to understand why the vacuum in QFT gives
correlations between field values at spacelike separated points,
it only complicates life to bring in quantum gravity, because
QFT is well-understood and quantum gravity is not. So I don't
want to talk about quantum gravity in this thread. I talk about
it too much already anyway. :-)
>> the field at spacelike separated points will be correlated
>> because they're both feeling the influence of the same wiggles.
>So this is the common cause! (the same wiggles)
Yes, that's an okay way to think about it. The same random wave
pattern can pass through spacelike separated points, yielding
correlations between field values at spacelike separated points.
That's one way to say why QFT has such correlations.
Frank Wappler
> > > [...] even in a free field theory,
> > > the vacuum state violates Bell's inequality
> Stephen J. Summers and Reinhard Werner, The vacuum violates Bell's
> inequalities, Phys. Lett. 110A (1985), 257-259.
Summers and Werner base their presentation on
"the CHSH-version of Bell's inequalities"
(J.F.Clauser et. al., PRL23, 880, 1969),
whose derivation in turn involves the integral
"Int_{ Gamma }_(
| A( a, lambda ) B( b, lambda ) - A( a, lambda ) B( c, lambda ) |
rho( lambda ) d_lambda )",
where "Gamma is the total lambda space",
"lambda denotes collectively a set of hidden variables",
"a and b [and c] are adjustable apparatus parameters", and
"A( a, lambda ) and B( b, lambda ) [and B( c, lambda )] represent
results [... the selection of] one of two channels labeled +1 and -1".
AFAIU, if b and c are distinct, then in any one experimental trial
only either b _or_ c can be "adjusted" (even if the outcomes/values of
B( b, lambda ) and B( c, lambda ) might conincide for some lambda).
Is the integral shown above defined if among the
"set of hidden variables denoted by lambda" is
the trial number of the experiment under consideration?
Analoguously, Summers and Werner consider "states omega",
which in turn "associate measuring devices with space-time regions";
AFAIU, the pairs of outcomes of a pair of devices, _trial by trial_,
satisfy this definition.
For this particular example, how is the corresponding "CHSH-version
of Bell's inequalities" derived in the first place?
Summers and Werner proceed to find those "inequalities violated"
with another particular choice: "the quasi-free vacuum state omega_0".
Does there exist a measurement procedure to identify this particular
"omega_0" in any one trial (and perhaps to discard a trial
in which "the quasi-free vacuum state" had been found different from
that of other trials);
or does this particular "omega_0" involve an (at least implicit)
dependence on the trial number, too?
Thanks, Frank W ~@) R
Sent via Deja.com http://www.deja.com/
Vesselin G Gueorguiev
> > Stephen J. Summers and Reinhard Werner, The vacuum violates Bell's
> > inequalities, Phys. Lett. 110A (1985), 257-259.
>
> Summers and Werner base their presentation on
> "the CHSH-version of Bell's inequalities"
> (J.F.Clauser et. al., PRL23, 880, 1969),
> whose derivation in turn involves the integral
>
> "Int_{ Gamma }_(
> | A( a, lambda ) B( b, lambda ) - A( a, lambda ) B( c, lambda ) |
> rho( lambda ) d_lambda )",
>
> where "Gamma is the total lambda space",
> "lambda denotes collectively a set of hidden variables",
> "a and b [and c] are adjustable apparatus parameters", and
> "A( a, lambda ) and B( b, lambda ) [and B( c, lambda )] represent
> results [... the selection of] one of two channels labeled +1 and -1".
>
> AFAIU, if b and c are distinct, then in any one experimental trial
> only either b _or_ c can be "adjusted" (even if the outcomes/values of
> B( b, lambda ) and B( c, lambda ) might conincide for some lambda).
Hmm, I didn't pay attention to this the first time I was reading the
article, but now it looks important since B(b;l) and B(c;l) are
time like separate and therefore they may be strongly correlated.
This helps me to view the Bell's inequality in more physical way:
as study how much of the P(a,c) correlation is due to events (a,b)
assuming that 'c' data has been taken after 'b' data or vice versa.
Trying to follow this road leads me soon to a big mess...
> Is the integral shown above defined if among the
> "set of hidden variables denoted by lambda" is
> the trial number of the experiment under consideration?
That is very interesting point. As far as I can tell, it should be
still valid since we have hidden variable space which is not explicitly
defined (the integral may turn into a sum). However, whether the trial
number is a hidden variable will have impact on the analysis of an
experimental outcome. The way, P(a,b) and the related inequality
put in correspondence to a data set, requires at least 7 trails. One
has to get three R_{i} i=0,1,2 and four different a,b' and c,b call
this a full run. Then if N-runs are done they may go us statistics for
each R and can be thought as one bigger run. However, in this case
we have P(a,b) being somewhat N-dependent since the probabilities
w(+,+)=R(a,b)/R(inf,inf) may be N-dependent. One have to be sure that
the rates R are well behaving for each N as well as the whole
set of combined data. I would do fractal analysis of R(a,b;t_k) as
function of the time t_k needed to accrue the k-th data. Since we
assume static equilibrium situation I would expect that the
fractal dimension should tend to zero when t goes to infinity.
(fractal dimension is defined as the parameter s such that
y(x)=c x^s, s may depend on x as well.) i.e. this is a fancy way
to say that R(a,b;t_k) should lose any t-dependence as t goes bigger.
Now we get problem :(
We want long t's in collecting data to get closer to the real
probabilities, but for too big t's we are allowing time like
separations among the set (a,b;t) for example events (a) at t=0 may
become past for events (b) at t=t1. This seems to restrict the trails
to only one trail since we would like to have all {a} events to be
space separated from all {b} events. Difficult experiment, isn't it?
> Analoguously, Summers and Werner consider "states omega",
> which in turn "associate measuring devices with space-time regions";
> AFAIU, the pairs of outcomes of a pair of devices, _trial by trial_,
> satisfy this definition.
> For this particular example, how is the corresponding "CHSH-version
> of Bell's inequalities" derived in the first place?
I am also wondering about this because they seem to have drop the
Abs() function from the original "CHSH-version". My guess is that
they have explained this in some of the papers given in the references.
> Summers and Werner proceed to find those "inequalities violated"
> with another particular choice: "the quasi-free vacuum state omega_0".
>
> Does there exist a measurement procedure to identify this particular
> "omega_0" in any one trial (and perhaps to discard a trial
> in which "the quasi-free vacuum state" had been found different from
> that of other trials);
> or does this particular "omega_0" involve an (at least implicit)
> dependence on the trial number, too?
I can't say anything about this. May be they should write a book :).
Notice that in Summers and Werner analysis there is about 30% of all
possible theories that give violation of the Bell's inequality.
We can classify this theories as QM or QFT. This qualitatively agrees with
the fact that we can quite well handle most problems using classical
mechanics and statistics.
Ralph E. Frost
[large wads of unnecessary quoted text deleted]
> >Also, in the old days, didn't quantum mechanics "really need" classical
> >mechanics? Why single out just this handy picture as being flawed,
> >among the many other handy flawed pictures that are in the gallery?
> Because I want to steer the readers away from making a bunch of errors
> I can easily imagine them making if they're not careful. I can easily
> picture them falling off the cliff at this particular bend in the trail.
Could you just the six most common errors you think need to be avoided,
please?
Also, this idea of it being bad to fall off the cliff has always
bothered me, particularly in one trying to understand physics or the
separate field of endeavor, mathematical physics. It bugs me for two
reasons. First, folks learn from mistakes. Making mistakes is the
HISTORY of science. Second, the fact that people can identify a guard
rail and a cliff, to me indicates there is MORE to be discovered or at
least one other way to explore down into lower reaches.
As the new school year starts up, could you summarize the pitfalls you
referred to in your prior post?
Thank you for all of your help.
Best regards,
Ralph E Frost
john baez
Ralph E. Frost <ref...@dcwi.com> wrote:
>John Baez wrote:
>> [...] I want to steer the readers away from making a bunch of errors
>> I can easily imagine them making if they're not careful. I can easily
>> picture them falling off the cliff at this particular bend in the trail.
>Could you just the six most common errors you think need to be avoided,
>please?
I'm sorry, this thread is so old I can no longer remember what was
being discussed.
>Also, this idea of it being bad to fall off the cliff has always
>bothered me, particularly in one trying to understand physics or the
>separate field of endeavor, mathematical physics. It bugs me for two
>reasons. First, folks learn from mistakes.
Some do, yes.
>Making mistakes is the
>HISTORY of science. Second, the fact that people can identify a guard
>rail and a cliff, to me indicates there is MORE to be discovered or at
>least one other way to explore down into lower reaches.
For every guard rail, there is always someone who will take it as
invitation. That's fine too.
Hendrik van Hees
are some which should be avoided by the next generation of physicists in
order to go beyond the knowledge the predecessors made ;-)).
One mistake is to present Quantum theory as wave particle duality +
classical physics. Of course in the beginning of a quantum theory course
one has to start with some history in order to motivate the highly
abstract mathematics (abstract compared to Newtonian point mechanics of
course for mathematicians it's a rather old story with quite simple
vector space structures) needed to understand the picture quantum theory
gives us about nature.
There is no way to describe this picture by classical pictures also not
by a mixture of classical wave or classical particles like the old
heroes of qt did. For me this is the most important lesson we learnt
from quantum theory up to now and it seems to me that further progress
in the understanding of fundamental physics can only be obtained by even
higher mathematical abstraction. At least super symmetry and super
strings seem to be necessary steps towards physics beyond the standard
model although I highly doubt that one can find "the final theory"
without new experimental facts for which to obtain we need new
accellerators. But SSC is dead. Nevertheless LHC at CERN is built for
sure and one might hope that TESLA (near DESY in Hamburg) will come up.
--
Hendrik van Hees Phone: ++49 6159 71-2755
c/o GSI-Darmstadt SB3 3.162 Fax: ++49 6159 71-2990
Planckstr. 1 mailto:h.va...@gsi.de
D-64291 Darmstadt http://www.gsi.de/~vanhees/index.html