Does a controversy still exist ?

1. Nov 14, 2005

McQueen

Three hundred years after the controversy had started , the question of whether light is wave like or particle like in nature is still raging. Modern theories of light tend more to the particulate view of light , in spite of the wave like properties associated with light and the generally accepted view of the wave-particle duality of light , wherein light possesses both wave like and particle like properties but can never possess both properties simultaneously. One instance of the general dissatisfaction with the theory of wave-particle duality is that explanations of how light undergoes reflection , refraction , and transmittance through substances is today explained almost entirely in terms of the particle nature of light , while even ten years ago , explanations for the manner in which light underwent , reflection , refraction and transmittance were almost wholly wave based. The fact is that the overwhelming evidence tends towards the view that light , in the form of photons , does interact with matter ( electrons) in a very definable particulate manner. In order to support this view , the conjecture has been put forward that light might be composed of particles but that the particles themselves travel like a wave , this is analogous to the way in which water , which is made up of molecules , assumes a wave like form. The draw back with this point of view is that a wave never interacts with matter in the manner of a particle , while light does. This leads to the saying that light travels like a wave but arrives at its destination as a particle. Thus the debate still rages. Today the widely prevalent view is that Reflection is due to the rapid absorption and re-emission of photons , while refraction is also thought to be due to the result of the slowing down of light as it travels through a medium due to its absorption and emission as it travels through the medium. This raises the extremely interesting question of why , if light can travel through a medium such as glass by being rapidly absorbed and emitted by the electrons in the atoms of the glass , cannot it be transmitted through a metal in a similar manner. The rapid absorption and emission of photons through a glass pane implies that this kind of interaction is due to the conduction band properties of the glass. This being so , why cannot light travel through a metal , a metal has wide open conduction bands , it should theoretically be possible to replicate in a metal the phenomenon which is known to exist in glass , namely the transmittance of light . Why doesn’t it happen ?

2. Nov 14, 2005

tbone

For one thing, light can be reflected off of a metal pain. Secondly, metal is made up of more mass. With this light is is obsorbed but you also have to remember that, when the photons hit the tightly packed electrons, they lose some of their energy. Well with the thickness and tightly packed electrons in a metal you could imagine how many atoms they would hit before they made it through.

3. Nov 14, 2005

Hurkyl

Staff Emeritus
FYI, I stopped seriously reading your article after this introduction. The question has been answered for quite a while now: light is neither a (classical) particle nor a (classical) wave. Light is some quantum mechanical thing to which the classical notions of particles and waves are good approximations under various circumstances.

4. Nov 14, 2005

tbone

You know I think that is the best explaination of light I've heard yet. I'll have to remember that one

5. Nov 15, 2005

Careful

Very amusing :rofl: No, no, light is just a good old fashioned EM wave, no fuzzy QM stuff involved.

6. Nov 15, 2005

vanesch

Staff Emeritus
Aren't you taking a bit your dreams for reality here ? I know that your programme is to show ONE DAY that SOME classical field theory might EVENTUALLY reproduce observed quantum effects, but for sure it will not be good old Maxwell with no additional stuff, right ? Try to explain anti-correlations such as the famous paper by Thorn et al (Am. J. Phys. 72) sept 2004 with *pure classical optics*.

So I'd say that *at least for the moment* the best description of light we have is the quantum-mechanical one and then Hurkyl's statement is very accurate.

7. Nov 15, 2005

marlon

McQueen,

I assure you, there is no problem with QM, what so ever. I think Hurkyl gave you a nice explanation concerning your question. I would like to add that al these "measurement problems" are all just coming from people who are interpreting the result and formalism of QM in the WRONG way.

QM works, Einstein was wrong, "point final"

regards
marlon

8. Nov 15, 2005

Careful

Sorry, don't have immediate acces to library. Can you explain me what the measurement setup is and what the results are?

Cheers,

Careful

9. Nov 15, 2005

Careful

Really, and on what basis do you claim that ?! :rofl: :rofl:

10. Nov 15, 2005

vanesch

Staff Emeritus
It is also available freely here:

http://marcus.whitman.edu/~beckmk/papers/Thorn_g2_ajp.pdf [Broken]

cheers,
Patrick.

EDIT: FYI, this is not an EPR style experiment. It would be very simple to explain the experiment with bullets, for instance. But with *classical optics* I think it is impossible (unless you modify about all we know about optical devices such as beam splitters in classical optics).
Although this is not demonstrated in the paper, similar setups can show *interference* after recombination of the split beams, so the argument that the beamsplitter sends little packets "left" and then "right" randomly would not do.

Last edited by a moderator: May 2, 2017
11. Nov 15, 2005

marlon

err, realy easy, how about the fact that we have transistors, semiconductors, diodes, ...

How about the fact there is not a single experiment that contradicts with QM ?

marlon

12. Nov 15, 2005

ahrkron

Staff Emeritus
QM is extremely well established now. For sure, there is no controversy among physicists about that.

Actually, the title of the paper quoted by vanesh is
"Observing the quantum behavior of light in an undergraduate laboratory".

Here's a bit of the abstract, in which I color-emphasize some parts:

"While the classical, wavelike behavior of light (interference and diffraction) has been easily observed in undergraduate laboratories for many years, explicit observation of the quantum nature of light (i.e., photons) is much more difficult. For example, while well-known phenomena such as the photoelectric effect and Compton scattering strongly suggest the existence of photons, they are not definitive proof of their existence. Here we present an experiment, suitable for an undergraduate laboratory, that unequivocally demonstrates the quantum nature of light."

i.e., the article is not about a high end, controversial, multimillion dollar experiment, but about how to confirm in school a well established, well known result: that QM is a better description of nature than classical physics.

13. Nov 15, 2005

vanesch

Staff Emeritus
That said, QM faces serious problems too on a more foundational level of which the measurement problem and the incompatibility with GR are the two principal ones. Another difficulty is of course the mathematical inconsistency of QFT - no matter how well it works to crank out numbers that compare to scattering experiments. But that doesn't do away the tremendeous experimental success it has seen in vastly different areas.
As such, I cannot say anything about how 'fundamentally true' QM is, but at least how successful it is as a current description of the workings of nature.

Last edited: Nov 15, 2005
14. Nov 15, 2005

Careful

How, how, I was not aware of this result and I shall study it in detail; I thought people would come up with compton scattering again But I doubt it will be that unambiguous ... as the authors claim it is.

Will come back to this, thanks for the reference anyway Vanesch

15. Nov 15, 2005

marlon

Sorry, but on this one i disagree. Concerning the incompatibility with GR, i do not see how that is an issue for QM ? I mean we do not say this about newtonian physics and QM, right ? QM is not built to explain the GR-phenomena, so why is this incompatibility an issue then ? This is just a matter of physical regimes. I know, in the past, i have stated this before but i really feel that we need to look at it like that ? We should not "create" problems based upon interpretations of the underlying mathematical formalism.

Same goes for this mysterious measurement "problem".

If it ain't broken, do not fix it.

ok, you may find this question to be very stupid, but...what mathematical inconsistency ?

Even if there is one, how is this correlated to QM ?

regards
marlon

16. Nov 15, 2005

ahrkron

Staff Emeritus
Agreed. However, one has to be careful (which you have been) when discussing these, since people sometimes get the wrong impression that QM is not well tested.

What are you referring to in here? renormalization?
I've often heard good theorists say that this is now understood, in a tone of "we now have it solved"...

17. Nov 15, 2005

vanesch

Staff Emeritus
Renormalization is not the problem as such. The trouble starts with Haag's theorem which invalidates in fact the canonical approach to QFT, and which states that the "interaction picture" must always be equivalent with a free field theory if the creation and annihilation operators are to be what we think they are.
Now, you can leave the canonical approach for what it's worth, and switch to the Feynman path integral. But here the trouble is the measure. Nobody has ever been able to define a measure on the space of paths - as far as I understand, there are reasons to think that this is impossible. As such the path integral is an undefined quantity.
Next, you can STILL do a step backward, and consider QFT to be defined as the set of Feynman diagrams. Apart from difficulties of convergence (even after renormalization: in QED, it is now I think established that at best the perturbative series are only asymptotically meaningful, which means that they will start diverging again after a certain order - and as such that the "true" value is never reached) this would put aside a lot of non-perturbative results which clearly play a role.
As far as I know, there is no known axiomatic structure of QFT - this in sharp contrast to non-relativistic QM which was axiomatized by von Neumann.

That said, QFT as practiced DOES have a huge number of empirical successes on its record. But as far as I understand, it does not make mathematical sense. It is just a bag of phenomenological techniques which, when applied with care and fingerspitsengefuhl, cranks out good numbers which compare to experiment.

18. Nov 15, 2005

vanesch

Staff Emeritus
If you take QM to be a phenomenological theory, I agree with all this of course. However, if you consider QM to be a *fundamental* theory (or better, if you take the founding principle of QM, namely the superposition principle, to be a fundamental principle), then you ARE in trouble. And there are people (like Careful) who come from a quantum gravity background who have seen the problems that arise when you are combining both the principle of superposition and the principle of general covariance, and who prefer to stick to the latter. That said, they shouldn't close their eyes to those parts of QM which bother them if they are supported by experiment ; however QM proponents shouldn't be blind either to the difficulties.

19. Nov 15, 2005

ahrkron

Staff Emeritus
I don't understand the problem here. Why is that equivalence a problem?

That's quite scary. Can you expand on that? or maybe give a reference? If things are expected to diverge again after some order, when can we trust any numbers obtained from it?

Which, in light of what you mentioned, is quite puzzling.

[edit: fixed a quote]

Last edited: Nov 15, 2005
20. Nov 15, 2005

marlon

How is that ?

Besides, one can always chose to do a duality transform in order to go from high coupling constant to low coupling constant, like in the case of QCD. Either way, perturbation theory still holds.

marlon

21. Nov 15, 2005

vanesch

Staff Emeritus
the raw stuff:

http://en.wikipedia.org/wiki/Haag's_theorem

and a more poetic version:
http://www.cgoakley.demon.co.uk/qft/renorm.html [Broken]

Last edited by a moderator: May 2, 2017
22. Nov 15, 2005

vanesch

Staff Emeritus
Just a random search on an article related to it which is freely available:

Last edited by a moderator: May 2, 2017
23. Nov 15, 2005

vanesch

Staff Emeritus
Isn't that only possible in supersymmetric models ? And even then, you can transform high coupling constants in low ones, but for medium values you're screwed, no ? Because if it were so easy, I'd guess that hadron masses would easily be obtained and that one wouldn't have to go on the lattice!

Now, I have to say that I'm not up to level in all this, so you can easily hit me around the ears with lots of technical stuff I'm not aware of. But I don't think that that changes the content of my statement that there is no axiomatic basis for QFT as of today, and that it is a lot of phenomenology which works very well and where one invents more and more useful techniques, but not a crystal clear theory.

24. Nov 15, 2005

George Jones

Staff Emeritus
Although this was demonstrated by Dyson [1] (but not proved) more than 50 years ago, it is not well known.

From [2], page 451: "Thus, QED may have a zero radius of convergence in $$\alpha$$ space."

From [3], page 259: "The belief is that the perturbation series is an asymptotic series for real e at e = 0." Despite its title [3] is NOT a book about rigorous mathematics - it is a book that covers much the same topics as Peskin and Schroeder and at about the same level, and, in my opinion, is one of the best grad-level expositions of quantum field theory.

[1] QED and the Men Who Made It, S. Schweber, 9.17 Divergence of the Perturbation Series

[2] Quantum Field Theory, M. Kaku, 13.5 Does Quantum Field Theory Really Exist

[3] Quantum Field Theory for Mathematicians, R.Ticciati, Remark 9.4.12

Regards,
George

25. Nov 15, 2005

ahrkron

Staff Emeritus
Thanks for the references!!

50 years ago! Boy, do I feel ignorant! : S