User talk:Carl A Looper

Actually, talk pages are where welcomes go. So:

Welcome!

Hello, Carl A Looper, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are some pages that you might find helpful:

I hope you enjoy editing here and being a Wikipedian! Please sign your name on talk pages using four tildes (~~~~); this will automatically produce your name and the date. If you need help, check out Wikipedia:Questions, ask me on my talk page, or place {{helpme}} on your talk page and someone will show up shortly to answer your questions. Again, welcome! linas 16:38, 18 November 2006 (UTC)[reply]

Afshar's experiment[edit]

Dear Looper, thank you for spotting a typo somewhere in my paper. I will not re-pload new version just because of this :-), I am curious however to know how do you feel, when your name is associated with an inconsistent mathematical thesis, which you made publicly available at Wikipedia. Regards, Danko Georgiev MD 08:13, 20 November 2006 (UTC)[reply]

Hi Danko, I have no problem with my name being associated with the anti-thesis - it does not bother me at all. I just read that as another typo :) What is important to me is your real argument. And that deserves to be read by as many people as possible. --Carl A Looper 03:35, 23 November 2006 (UTC)[reply]

Dear Carl, your reply really convinced me that you are scientist at first place, and not another "web hero". I have replied to some derrogatory attack by Mike Price on the action of mirrors, etc., which is completely irrelevant to the central argument of my paper (indeed my exposition follows Roger Penrose in the section where I describe the Mach-Zehnder interferometer). I have been always relying heavily on mathematics when I release papers and now I am glad that a computer scientist [mathematician] has been interested in my disproof of Afshar. I will stop posting on Wikipedia for some time, and hope that in the mean time my paper can be verified/peer-reviewed by some real scientists. I will submit it for rigorous peer-reviewing in some journal, yet I am not sure how the references to web posts, will be accepted by journals. Nevertheless, I hope that the topic itself will provoke some interest amongst physicists. Although I will take a break from Wikipedia, I will reply all e-mails on the e-mail address provided in my preprint. Regards, Danko Georgiev MD 09:01, 24 November 2006 (UTC)[reply]

Thankyou very much Danko. I'm going to have a break from Wikipedia myself. I look forward to seeing your paper in a peer reviewed journal. It is a privalage to have crossed your path. All the best. --Carl A Looper 13:46, 24 November 2006 (UTC)[reply]

Danko[edit]

Dear Carl, I see that you have mentioned Danko's paper in the "Ongoing debate" section. I would like to ask you to remove both that remark and the reference at the end of the article. You may not know much about Danko, but he is a certified crackpot. The main thesis of his "paper" (about which I was initially happy, but highly disappointed after I read it) is denial of which-way information in a coherent beam. This argument has been rejected publicly by physicists like Unruh, Motl and Drezet (even though they disagree with me) as well as myself. It's content certainly constitutes OR, and its appearance in a philosophy of science archive makes it that much more suspect! No physicist worth his/her title would accept his entirely ignorant and erroneous arguments, as he lacks even an undergrad. level of physical understanding. If you wish, I can give you the remarks related to him by other experts. Looking forward to hearing from you soon. Best regards. -- Prof. Afshar 02:06, 1 December 2006

One-to-one mapping[edit]

"I can build a computer model of Afshar's experiment and generate the data we otherwise see done in a physical experiment. We can put both images side by side, and if we have done it correctly they will be identical (within reason). But that doesn't change the fact that they were produced by entirely different means. The images are the same but their method of construction is not. " Looper

Dear Carl, the self-quotation that you presented above, is exactly what I have proved and publsihed against Unruh and Afshar. Yes, the image in coherent setup looks like two separate pinholes, in a fashion that will look the image if you have used differently polarized filter for light at each slit. However *the means of construction* of this two peaked image is different. So the final image looks the same in coherent or incoherent setups, but the means of construction are totally different. And it is the "means of construction" that give the name "which way" or "not". For me the old debate is closed, I have already said and published what I wanted to say and publish, even I did much more than this, in my reply to Unruh for first time I have provided 1-to-1 mapping of the waves between Afshar's setup and Unruh's setup. A lot of people were thinking there is "subtle" difference, which I proved is missing. No differences, complete equivalence :-)) I think we should discuss by e-mail, I don't want to post in Wikipedia on Afshar and Unruh any more. Exception is the entry on Unruh's interferometer where I don't lke my words to be mis-attributed to Unruh :-) Danko Georgiev MD 03:13, 29 June 2007 (UTC)[reply]

Dear Carl, for one to prove he has "which way" he must prove one-to-one correspondence between slits and images, unless you do that, you cannot say you have "which way". So the ONLY possible logic is closing one slit, then closing the other one, take the statistical mixture, so equivalent to claim "mixed density matrix". Now the fact that quantum amplitudes evolve in linear way according to the Schrodinger eq. is NOT relevant to the complementarity topic. Complementarity is about observables - and observables in QM are non-linearly dependent on the q-amplitudes, as implied by Born's rule

P=|\psi |^{2}

. From now on P will mean observable, and

\psi

q-amplitude. In order to have "which way" you have to postulate "mixture" of two observables, so you have

P=1/2|\psi _{1}|^{2}+1/2|\psi _{2}|^{2}=1/2P_{1}+1/2P_{2}

. I cannot be both right and wrong. This mixture of observables, is easily obtained and shown schematically on the next figure as mixture of two single slit diffractions. In the mixed setup you have mapping

f:(slit_{1})->(image_{s1})

f:(slit_{2})->(image_{s2})

here f is unary single valued function, and you have two sets domain of the function $X={slit_{1},slit_{2},...}$ , and range $Y={image_{1},image_{2},...}$

Now please note that when you open both slits, the pure state leads to new prediction for the observables, so this is nonlinear effect, it is what complementarity is all about. $P=|\psi _{1}+\psi _{2}|^{2}\not =1/2P_{1}+1/2P_{2}$ . In this case you do not keep the one-to-one correspondence between slits and images. In the pure setup you have multi argument function that makes holo-transform

f:(slit_{1},slit_{2},...)->(image_{s1+s2+...})

For f in QM it is not generally true that $f:(slit_{1},slit_{2},...)->(image_{s1}+image_{s2}+...)$ , or in other words in QM it is not generally true that $f:(slit_{1},slit_{2},...)->f(slit_{1})+f(slit_{1})+...$ .

You said above that my model is right, but you are searching for another one. So please prove in the putative "another model" that there is one-to-one correspondence between slits and images. I am curious to know this putative new proof that will show there is one-to-one correspondence and will avoid the assumption of "mixed state" for the observables. Please note that in case where $f:(slit_{1},slit_{2},...)=f(slit_{1})+f(slit_{1})+...$ without knowing what is the original function, you are already at the edge of making erroneous prediction. It is obvious that these equalities are rare (intersection of both functions) and exactly this happens at the image plane of a lens. Now you cannot claim "which way" because if it is not the case, you will make erroneous predictions for the function values everywhere except for the intersection point! This is what Afshar and others fail to see. As always I am ready for cooperation so that I would accept co-authorship from anyone who wants to suggests way to improve my pre-print [fill in details] so that it is able to pass peer-review process and get published. I already have downloaded extensive literature on complementarity from Physical Reviews. So many names that I will challenge ... :-)

Danko Georgiev MD 07:17, 7 December 2006 (UTC)[reply]

Hi Danko,

I saw for a brief moment your images and comments but then they disappeared. Looked great. But I wish you'd stop seeing in your own work proof that "other" models are therefore wrong. They very well could be but until you see such models you should refrain from assuming they are incorrect. I have a very interesting model I'm developing which I think you'll appreciate. And I think Afshar will as well. And by the way, although I enjoy your paper you are wrong to conclude I used the wrong density matrix. I didn't use any density matrix at all. That could very well be a fault but you don't know that. Until you know what method I did use you don't know anything at all. Stay happy. I fully support what you are doing.

Carl

Dear Carl, I think that Unruh's setup is many times simpler to discuss complementarity, and it is easier to discuss complementarity. This is because in lens action, and in the single slit diffraction you have infinite number of points, and infinite number of rays. Basicly if you understand that the image B, in the case of pinhole 1 open, falls into Airy minimum [hence into destructive interference zone contributed from pinhole 1] you will see that this is just Unruh's setup. The main thing is that in order to have "destructive interference" you will need to have at least two paths from pinhole 1 to image B. Also these paths must be indistinguishable! Once you make a claim that there is interference in the space before the lens, you will make the paths distinguishable analogous to the Unruh's setup. So you will have "hidden" inconsistency. I have exposed the inconsistency by asking at first glance irrrelevant question - "are paths 5 and 6 distinguishable or not in Unruh's setup?". So I arrived at inconsistency. The same will happen in the lens action. Nevertheless, I am curious what is your idea? When you are ready please post it, I am not idiot deluded my own theses. As Afshar has used against me, I have withdrawn my work once from ArXiv (in 2003), and once from PhilSci (in 2004, this second work was indeed on Afshar's setup). Danko Georgiev MD 10:04, 7 December 2006 (UTC)[reply]

Polarized filters[edit]

One more note, image 1 can be obtained with different polarization filters R and L on each of the pibholes. In this case Afshar's wires will diffract 6 % of the photons. Remove the polarization filters and you will have 0 % loss by wires. So my logic is that if there were "which way" then additional "which way" labeling should not change anything. This is analogous of insertion of several identical polarization filters one after another. If say the photon is R polarized, the action of the next R filters cannot make it "more R" polarized. Danko Georgiev MD 10:16, 7 December 2006 (UTC)[reply]

Other Models[edit]

Ok. Here we are then. The images are great. I've got something similar happening. Now the first thing I should say is that I'm not making any claims that are contrary to yours. This is an assumption you make - that another model must be inconsistent with the one you are using. But I just want this to be clear from the beginning. When I say another model I do not mean one that contradicts yours.

But I still have to code it. I work in C++ and I write all my algorithms from scratch. I don't use Mathematica or any other package. I'm bit of a purist - a hangover from my youth when no software existed. In them days you had to roll your own.

So it takes time. I'll be putting it in a paper, like yourself. Because otherwise it can get misrepresented and misuderstood. Which I think is what happened to your own work - ie. before it was formalised in a paper.

Carl

Hi, and thank you[edit]

Hi Carl,

After a long gap of not looking at that Afshar discussion, I checked out the archives, and I saw your comments. I just wanted to thank you for your encouraging words. I really got a lot out of the discussion, however Price was just too much. I was interested in that insight you had about the patterns seen at the detectors too... I will look into it more when I have the time. Seasons greetings! -Dino Dndn1011 19:37, 19 December 2006 (UTC)[reply]

Hi DnDn. The paper I'm writing involves your observation of an "interference pattern" in the data. The observed pattern is the shadow cast by the wires. Since the wires are placed at the minima associated with a prior interference pattern these shadows constitute a partial trace or partial reproduction of that prior pattern. Whether this data is of importance in how the experiment is to be understood I'm not yet certain - only that it is data which - insofar as it does not explicitly play a role - beghs the question: in what way does explicitly emperical data regarding the wires fail to play a role?

On a philosophical note I've always been intrigued by the question:

"Is light a wave or a particle?"

A typical answer trys to position the question as ill posed. It responds by saying:

"No. Light is "both a wave and a particle".

It then trys to qualify this by saying "but not at the same time". This was how Bohr sort of framed it. But the way I read Bohr is in terms of the logic that seems to be informing this way of putting it.

The word "or" is normally read as indicating a question - ie. that one must decide between the two options. But as you know from your experience in computing, it can also be understood as an operator (the xor operator) which takes the two operands (eg. waves, particles) and if they are complementary (which they are) then it generates a result of true. So the correct answer to the question would be:

"Yes. Light is a wave or a particle."

Now following this logic, we can't conclude light is therefore "not a wave and a particle". If you look at the truth table for a "not and" operation you will see that it isn't quite the same as the "xor" operation.

NOT( 0 AND 0) != (0 XOR 0)

NOT( 0 AND 1) == (0 OXR 1)

NOT( 1 AND 1) == (1 XOR 0)

NOT( 1 AND 1) == (1 XOR 1)

Now part of the problem I see in many attempts to reinterpret complementarity is in treating "inconsistency" as the antithesis of math. But inconsistency is perfectly math containable as just the logical negation of consistency. Consistency is typically expressed with an "=" sign. Classical physics, and modern physics is preoccupied with consistency, ie. with the "=" sign. But in boolean algebra (ie. math) the "=" sign is just the equivalent of a NOT XOR operation. Inconsistency is therefore just a NOT(NOT XOR) operation, or just XOR since the two NOTs cancel each other out.

The upshot is that one can keep trying to decide between A or B, (introducing all sorts of qualifications about when and where one or the other obtains - or introduce multiple universes in which each obtains independantly of the other), or one can just treat "A or B" as a mathematical statement/operation in no more need of any further qualification. It is a pure mathematical statement in a single universe.

--Carl A Looper 02:50, 20 December 2006 (UTC)[reply]

Bohr's "paradox" is just a paradox if you read xor operations as meta-mathematical (or metaphysical). I suspect Bohr fully understood the "paradox" was no such thing. But he was convinced that the language in which QM was to be understood would be incapable of understanding xor. Since a double XOR operation always restores the original context anyone starting from a classical context always ends up back in a classical context. And so Bohr feels safe in articulating complementarity in terms of a classical context - the xor's cancel out - and so one can escape a "paradox".

But xor is not (or is no longer) a paradox. There is no longer any paradox from which it is necessary to escape. Anyway, that's the way I'm reading it at the moment.

But it's not the end of the story. The underlying logic has a role to play in the evolution of QM functions. It is not just there as some sort of double articulated escape clause for EPR conundrums. Or there as some problem to be solved by semi-mathematical obsessions with consistency.

--Carl A Looper 03:42, 20 December 2006 (UTC)[reply]

Hi Carl, there is indeed an obsession with maintaining consistency which is in my opinion sometimes taken to irrational extremes. Seeking consistency is very useful, however there are many mathematical entities which do not have a consistency. Probably the most famous is the value of 0/0. What can we do with this expression? Is it zero or is it infinite? Is it correct to say that it is both (1 1) or neither (0 0) or that we can think of it only as one or the other (1 0 or 0 1)? Is it all of these? Is it some imaginary value that we could make a symbol for (lets say 0/0=u).

We think we can escape with our new u, but we can't because if we say that x/u=y then we can not say that x/y=u, u unfortunately spells the end of mathematics (and therefore the universe... bye!)

What does all that mean? It means that paradox does not in itself spell disaster. Paradox does not mean the end of physics or the end of QM specifically. It can mean one of three things: either that the theories are flawed, incomplete or the paradox should be accepted as part of the theory. It is the third path that mathematics takes, and as wonderfully useful as mathematics is, this is still a pragmatic approach and this third choice is merely an assumption.

To return the question of whether a photon is a particle or a wave, I think that far too little consideration is given to the option that it is in fact neither. It is something that has the charactistics of both but that does not mean that a photon is both. And here is probably the great flaw in the reasoning. There is no evidence to support the assumption that the underlying mechanisms have anything to do with the behaviors of classical waves and particles. It is perhaps completely erroneous to apply classical concepts of waves and particles to the non-classical system of QM. Is not the entire debate based on a fundementally flawed premise here? Perhaps we should forget particles and waves. They may have nothing at all to do with the mechanism of QM, even though these classical theories do correlate to some extent with observed behaviors (something which is not uncommon.. as an example I seem to recal the formulae for charge and discharge of electronic capacitors are identical in form to the classical formulae for transfer of momentum of particles, yet clearly the underlying mechanism is very different).

About complementarity: If Bohr were alive today I do not think he would be trying to keep a vice like grip on the concept. I think it is also entirely possible that the concept of complementarity was used as a strategic stop-gap. Perhaps the hope was that eventually the puzzle would be solved by working away at both ends. I have no idea, I am guesisng. But it does seem that the origin of the concept came from the original dual slit experiement, where measuring the path of the photon is equated to closing a hole. This is very unsatisfactory, because it has not measured the path of the photon, it has altered the experiment and dictated the path. It is a self fullfilling prophecy on the photon's path. Unruh's experiment as I understand it is just the same. It seems very clear to me that the entire concept of "which way" makes no sense at all when both slits are open. There is no "which way", because the photon is neither a wave or a particle. If I am right of course it also means that "which-way" never has validity, even when one slit is closed. How can it? The photon does not pass through anything, because it is neither a wave nor a particle. How can opening or closing a hole change the fundemental property of light?

As for the fact that the wires disturb the pattern of light at the detectors, the significance of this is that wherease all photons are accounted for, some of these photons must have interfered with themselves. Although the debate about whether this breaks complimentarity can probably be argued forever, that debate is actually pretty useless. Stricly speaking, complimentarity is a principle that can never be broken because of the nature of light itself. Since we can never know the path of any photon (we can't, we only know its destination) it can probably be proved that complementarity can never be broken. But this does not mean there is any use in this. Complimentarity is supposed to allow us to avoid the paradox. The question is, does Afshar's experiment display the paradox inspite of complementarity? Does it show complementarity as no longer able to paper over the cracks in QM? I believe it does, because it demonstrates quite unambiguously, in my opinion, that light is neither a particle nor a wave, but something else entirely. So did the original double slit experiement of course. However complementarity allowed an alternative view (the XOR view) to gain acceptance, by nothing more than a logical slight of hand. However, Afshar's experiment demonstrates that some photons must have appeared to have gone through one or other slit while at the same time having gone through both (to interfere with itself). This was not shown in the original double slit experiement. Now that it is shown, complementarity should be consigned to the past. It is no longer a useful concept. Dndn1011 17:39, 24 December 2006 (UTC)[reply]

Noise, Measurement and the principle of complementarity[edit]

THE NOISE FUNCTION.

An important component of quantum theory can be called the "God plays dice" concept. It is the concept of "pure noise", a concept recognised by, but opposed by Einstein.

Pure noise can be regarded as any "signal" without a structure.

We can call it a "signal" in the same way that we can call zero a "number", and just as we have some cautions with zero (as a number) we have cautions with pure noise (as a signal).

For example, we can not use zero as a divisor without, eg. introducing a second number axis in which "zero" has a value of one, or without throwing a "division by zero" error in computer programming, and so on.

Now pure noise, insofar as it has no structure, has no formula.

This does not prevent us from approximating the idea using a formula. Nevertheless we should be careful to remember this - that an approximation of pure noise, using a formula, does not imply pure noise has this or any other formula. It doesn't. The reason is that we have defined it this way. If it has no structure it has no formula. It is for this reason we call "random" functions (in computer programming) "pseudorandom" functions. Random functions are, a priori, a contradiction in terms.

Although pure noise has no formula, it can still be used in formulas. We can use the aforementioned pseudorandom functions as a proxy, or we can use pure noise where such is available. One limitation of pure noise (rather than pseudo-noise) is that it can only be emperically derived - and even then we can never really be sure if it is pure noise. Nevertheless a postulate of quantum theory - and theory in general - is that pure noise, if it exists, can only exist emperically, since it has no structure. No formula.

THE MEASUREMENT PROBLEM.

The so called "measurement problem", associated with quantum theory, is not actually a problem with quantum theory as such, but with the way in which the quantum theoretical formalism is partitioned from the otherwise emperical noise "function".

The measurement problem presupposes that either the noise concept can be rationalised (it can't), or that quantum theory does not possess this concept (it does), or that the measurement problem is not about quantum theory.

The measurement problem, in relation to quantum theory, is a "false problem". It presupposes a solution where none can exist.

But it's not entirely a false problem. We can redirect the problem, if not to quantum theory, then to simulations of quantum theory, where pseudonoise functions are used. How do we simulate the collapse of the wave function? Do we do it as a function of the wave function in space at an otherwise prenominated moment in time. Do we precompute the wave function across space and time and retrospectively throw the dice? Do we modify the shape of the wave function to compensate for particles that have already collapsed? These are all legitimate questions and constitute the range of problems that might be called the "measurement problem".

These problems are not with the theory, but with simulations of the theory (and simulations of experiments).

A fully described experiment is one in which the setup, the wave function, and the emperical random function have all been described. But, as mentioned, only actual completed experiments can provide closure on such a full description since the random function has no a priori structure.

And so while all experiments can be fully de-scribed they can only be partially pre-scribed.

And this is due entirely to the emperical nature of the noise function.

THE PRINCIPLE OF COMPLEMENTARITY.

The principle of complementarity introduces ideas not otherwise implicit in the quantum theoretical formalism. In particular is the idea that "regularitys" otherwise associated with classical physics (eg. the path of a particle) can be partially deployed in the quantum theoretical domain.

Nowhere does Bohr suggest a retrospective path function is in anyway derived from, or limited by the formalism itself. Our only limit is whether we can actually construct such a function, ie. whether we can actually draw an unambiguous path, from a particle detection, back to an aperture (ie. a single aperture).

Furthermore, it is not the formalism which suggests, implys or otherwise imposes on us that a setup which allows a retrospectively constructed path function is also a setup which prevents us from demonstrating the wave function.

It is the principle of complementarity which articulates this wisdom. Not the formalism.

And it is the principle of complementarity that the Afshar experiment challenges. Not the formalism.

--Carl A Looper 00:10, 12 March 2007 (UTC)[reply]

Nowhere does Bohr suggest a retrospective path function is in anyway derived from, or limited by the formalism itself. Bohr was wrong. Feynman showed that the path integral formalism was isomorphic to the conventional formalism for QM. BPC was not required. --Michael C. Price ^talk 00:21, 12 March 2007 (UTC)[reply]

If one used a time reversed version of Feynman's path integral, Bohr's path function could never be constructed. But Feynman's path integral is not normally time reversed. If Bohr is wrong it's not due to Feynman. The path integral is, as you say, isomorphic with the formalism. The formalism is not retro-spective. It is pro-spective. And so is Feynman's. --Carl A Looper 04:05, 12 March 2007 (UTC)[reply]

Information in fully specified experiments[edit]

The quantum theoretical formalism is the "jewel in the crown" of quantum theory. It embodys the fundamental structure of signals (information) we otherwise produce/discover in physical experiments.

But there is a strange problem in relation to Bohr's principle of complementarity.

One of Bohr's requirements for the correct appreciation of the formalism is that setups be fully specified (and in classical terms).

But in a fully specified experiment no information (other than noise) is actually produced since everything we know about the experiment (or think we know) is already embodied in the specification.

This contrasts with otherwise 'real world' experiments in which aspects of an experiment are not fully specified. And as a result such experiments do produce information ie. other than that which is pre-specified.

And it is in relation to these sorts of semi-specified experiments that the formalism becomes useful. We can use the formalism to separate out what is actually information and what is just the formal structure of that information.

Bohr's principle of complementarity is very different. It concerns experiments regarding the formalism itself.

Now Bohr requires that no aspect of such experiments be left unspecified.

The result of this limitation is that no information (other than noise) is produced by the experiment.

MORE TO COME

--Carl A Looper 06:16, 23 March 2007 (UTC)[reply]

Interesting, where does Bohr require that no aspect of such experiments be left unspecified? Dndn1011 00:18, 24 March 2007 (UTC)[reply]

Here is one reference.

"As a more appropriate way of expression, I advocated the application of the word phenomenon exclusively to refer to the observations obtained under specified circumstances, including an account of the whole experimental arrangement." Neils Bohr.

I'm currently writing an essay on Bohr's entire argument regarding "Complementarity" but sporadic insights are being noted here in my talk pages as they come to hand.

This particular one is of direct concern regarding the difference between complementarity and the formalism. The formalism can be used in either a "pure" form when an experiment is fully specified or in combination with conventional probability functions, when aspects of an experiment are left unspecified. But complementarity is (amongst other things) about the formalism within fully specified experiments. The reason for this is not immediately obvious but it emerges when you see what happens within a semi-specified experiment.

For example, in a fully (and classically) specified twin slit experiment it is difficult (or impossible) to specify a version where the slits will be both open and just one open. However, one can do so by simply leaving the status of the slits unspecified (as one can do in a black box experiment - ie. a Shrodinger cat experiment). But in doing so there is introduced an additional level of ambiguity. One which Bohr wants to avoid. He wants to demonstrate the formalism in it's pure state. At the same time he wants to show how - even when everything is specified and there is still a fundamental ambiguity - it is one which is neverthelsss alleviated - by the precise conditions in which the formalism has been isolated. And that will be the basis for his principle of complementarity.

The formalism, as discussed in length by Heisenberg, emerges as far more neutral and wide ranging than complementarity. Complementarity has a specific set of boundarys. The formalism, for Heisenberg, stays at the phenomenological level. And in the main it does so for Bohr too but Bohr sticks his neck out and projects the two sides of the phenomenological coin (separately) into the ontological space of the experiment, on the assumption that only mutually exclusive experiments could bring them back together again - ie. that this is the physical meaning of the otherwise abstract complementarity aspects of the formalism.

MORE TO COME

The concept of "information", within the domain of physics, is not as well defined as it might be. Certainly there is Shanon's law and other related bits and pieces but otherwise it is poorly represented. However information is a well studied component of "semiotics" - a science (or philosophy) as old as physics. The study of signs (semiotics) is normally associated with the study of language but semiotics has it's roots in the study of any signs - irregardless of who or what produces such signs. So although semiotics might be preoccupied with the signs produced by human beings it is equally applicable to signs produced by physical processes. For example, the information embodied in a thermometer is just as much a signal as the information embodied in the Mona Lisa, in a photograph of Marilyn Monroe, or in the dress sense of Albert Einstein.

In semiotics the two fundamental components of a sign are called the "signifier" and the "signified". A third component is known as the "denotatum" but this component remains an ongoing question mark and is generally left to the side. In most situations the "signified" replaces the role otherwise occupied by the "denotatum".

A signifier is a unit of communication, eg. in binary communication the simplest signifiers are "0" and "1". More complicated signifiers are built from the simplest signifiers. Thus "0100101110" is a more complicated signifier.

The signifed can be understood as the "content" of the communication, ie. what is being signified by a signifier (or set of signifiers)

So, in the context of binary numbers, "011" refers to (signifys) the numerical value of three.

The signified is somewhat complicated by the need to involve more signifiers in signifying what is otherwise signifed by the original signifier!

Some theorists conclude from this that there is no such thing as the signified. There are only signifiers!

However this is not the case. Signifieds tend to stablise very quickly due to the already signified context in which signifiers typically operate.

For example, in quantum theory, the meaning of the wave function (ie. what it predicts) is stablised by the already signifed context in which the wave function operates, ie. the experimental setup.

In the case of a single particle in an otherwise empty universe, the meaning of the particle's wave function is undefined. There is nothing for it to predict. Furthermore, the wave function itself (the signifier) can not be instantiated as there is no signified context.

In the case of a particle in a box, the box plays the role of a signified context, so the wave function which characterises the particle's state can be instantiated ("prepared"). And the meaning of the wave function (what it predicts or signifys) is likewise instantiated by specification of any detectors associated with the box.

Without a context the wave function is analogous to the nature of heiroglyphics following the collapse of Ancient Egypt (but before the discovery of the Rosetta Stone) - it is without a known meaning. But we can still call it a signifier - or potential signifier.

Many theoretical physicists do not actually understand this.

The meaning of the wave function is thought of as something other than what the wave function predicts. For example, Paul Davies is a good example:

"Nobody questions what the theory predicts, only what it means" says Paul Davies.

But what the theory predicts is what the theory means.

MORE TO COME

We could (in a very theoretical sense) also postulate signifieds which have no known signifers. (Donald Rumsfield territory here) But without any signifers such a postulated signified would be incapable of being signified. Or rather, if it was capable of being signified, as we appear to have just done, then it's signifiers (or candidates for such) are incapable of being unknown!

A more concrete example will suffice. In QM, the experimental context could be understood as missing it's signifiers. However, if it is missing anything (as we might postulate) it could be postulated that it's only missing it's quantum theoretical signifiers. But this is incorrect. Quantum theory requires that the epistomological conditions (the setup) be expressed using classical signifiers, ie. the setup does have a set of signifiers and is accordingly signified! Nevertheless, the concept of a signified, without a corresponding set of signifiers is a possibly useful if somewhat unthinkable idea.

Ambiguity in Full Classical Specifications[edit]

Bohr's interpretation of the quantum theoretical formalism requires that experimental setups be established in a classical manner. But also - that the setups be fully and unambiguously specified.

However, it would seem to be the case that classical specifications can not be fully and unambiguously specified.

The Afshar experiment demonstrates (as many other experiments do) that the wave function used to predict detector data is unable to "see" particular locations in the classical specification. It is just such locations that are occupied by wires in the Afshar experiment. These locations are where the wave function cancels out due to self interference, ie. where the wave function has zero amplitude. Such locations are called "nodes".

The nodes are invariant with respect to any squaring of the wave function (ie. construction of a probability wave) or complex conjugation of the wave function, ie. they remain zero.

Since the wave function is the result of a transformation on the classical specification, so to are the nodes. The wave function is effectively a "translation" of the classical specification into quantum theoretical terms.

But anything specified as occupying, or not occupying these locations - must be understood as untranslatable - or lost in translation.

It doesn't matter whether there are obstacles (wires) or just empty space specified at these locations. They will fail to be encoded in terms of the wave function.

What we need to do is to go back and mark these locations, in the classical specification , as fundamentally "ambiguous". Or "unspecifyable". Or "unknown". And to re-read any node producing specifications as "semi-specified". The nodes in the wave function will identify precisely where the classical specification is unspecified - or rather - unspecifyable (quantum theoretically).

There are good reasons for doing this. The way we exploit the wave function in a semi-specified setup differs markedly from the way we "use" it in an ideal setup.

--Carl A Looper 07:24, 11 April 2007 (UTC)[reply]