Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Do Gravitons exist?

  1. Dec 5, 2005 #1
    First of all, I’m not well versed in Calculus or other tools required to fully grasp the latest understandings of QM or Relativity, so please take my question with a grain of salt.

    My question is simple. Do Gravitons exist?

    My intuition impels me to think of Graviton and Gravity in terms of Particle/Wave duality (as in Photon/Light). I suppose such visualization is in line with how other particle and wave concepts that are well applied in experiments, but then I hit a wall.

    Consider a Black Hole. We know that not even light can escape the Black Hole from within the Even Horizon boundary (save the Hawking’s Radiation, which is really a QM effect at the fringe of Event Horizon). Why, then, is Gravity (Graviton) allowed to “escape” or “emanate” from Black Holes? While I realize that it is precisely Gravity which makes Black Hole what it is, but what’s so special about Gravity/Graviton, a seemingly close cousin of Light/Photon, that makes it the only remnant of Black Holes?

    I had an earlier post asking about the nature of Space-Time fabric and whether or not the Matter, Gravity and the Speed of Light are but manifestations of the underlying composition and characteristics of such Space-Time fabric’s interaction with Force/Energy (https://www.physicsforums.com/showthread.php?t=102640). Without providing any equations or proofs (I lack the tools), isn’t it possible that Black Hole is a region of Space-Time that counters the isotropic inflation/expansion (due to great Mass) that is the very fabric of Space-Time itself?

    I must be misguided somewhere, but would appreciate some pointers on this and my other post (which is lacking reponses/guidances).

  2. jcsd
  3. Dec 6, 2005 #2


    User Avatar

    Staff: Mentor

    My answer is simple. Nobody knows yet whether gravitons exist, as far as I know anyway.
  4. Dec 6, 2005 #3
    Well, actually you should have asked : do gravitational fields exist and can they be quantisized ? In QFT, the fundamental building blocks are NOT elementary particles but the associated quantum fields of which the fluctuations correspond to particles.

    We do not know if such gravitational fields can be quantisized (they DO exist ofcourse) because we do not have the "observational power" to detect manifestations of this quantization. The required distance scales are far too small for our current technology. I say, you wait 10 to 15 years and we will know a whole lot more:wink:

  5. Dec 6, 2005 #4


    User Avatar
    Science Advisor
    Homework Helper

    They definitely do exist, it's just that we can't see them, not yet...

  6. Dec 6, 2005 #5
    By theory they MUST exist or the theory which predicted other particle properties and force carrier particles is wrong!

    CERN are building (and have nearly finished building) the new particle accelerator that will hopefully show that gravitons exist in 'the flesh' not merely on paper.
  7. Dec 6, 2005 #6
    Sorry, but this is entirely incorrect. gravitons arise when one applies the notions of QFT onto gravity fields. But if gravitons were not to exist, this most certainly does NOT prove that the other interactions are not correctly described by QFT. How can you even say this, when you consider all the successes of QED,QCD, etc etc ???

    Err, i don't think so. Do you know what energies are required to observe gravitons. Let us just be happy if we would detect the Higgs particle, ok ?:wink:

  8. Dec 6, 2005 #7
    I was not saying that by a 'proof' gravitons do not exist, that the whole of QFT falls to the floor going aaaaaaah. Rather I was saying that the graviton almost certainly exists as the theory that predicts its existance has so much experimental support and theoretical support.

    Yet one must remember the quantum world is weird and thus something peculiar may happen with regard to the graviton.

    Gravity waves are being looked for at LIGO - by wave-particle duality is not finding one the same as finding the other!
  9. Dec 6, 2005 #8


    User Avatar
    Science Advisor

    I don't know any theory which predicts the graviton that has any experimental support. (E.g. string theory is far from experimentally verified). Although, I will agree it is very, very likely the graviton exists, since quantum field theory has been so succesfull in so many other areas.

    No, gravity waves are predicted already by GR, which is NOT a quantum theory of gravity. The wave-particle duality steems from quantum theory, so you can't use that argument before we have a quantum theory of gravity.
  10. Dec 6, 2005 #9
    I refer to my earlier point - the multiple theories that predcit numerous other particles, have been experimentally verified but not for the graviton - I was implying that their ability to standing up to testing so far puts the graviton in good stead!
  11. Dec 6, 2005 #10
    I believe that the correct statement is that, while there is no direct evidence that the graviton exists, we would all be a bit surprised if it doesn't.

    As for detecting gravitons at the CERN: The LHC certainly is not being built for doing that - it's looking mainly for the Higgs, and more generally for new particles at the TeV scale that can give us insight into fundamental physics.

    However, there are classes of theories - all involving extra dimensions in some way - in which graviton effects are directly observable at the LHC. It's a long shot, but there is some chance and the experimenters will certainly be searching for those effects.
    Last edited: Dec 6, 2005
  12. Dec 7, 2005 #11
    But QFT does NOT predict the graviton.

    The graviton is predicted by string theory, ie the adapted version of QFT that incorporates the basic fondations of general relativity. This is NOT QFT, however !!!

  13. Dec 7, 2005 #12


    User Avatar
    Science Advisor

    Eh, I don't really get you?
    Of course QFT does not predict the graviton. QFT predicts nothing at all until you apply it to a field. It is the different quantum field theories that predicts stuff. QFT is more of a method which has worked very well in building the standard model, but also in other cases like in solid state physics. QFT is not just QED, QCD,...
    In string theory we also quantize the fields, so why shouldn't it be called a quantum field theory?
    Now string theory predicts the existence of a graviton, but of course we cannot yet say it is a well working theory, since much is left to compute, and there are no experimental verifications of the theory.
  14. Dec 7, 2005 #13
    Hu ? What is 'it' ???

    Because the basic fondations of string theory are totally different in nature than those of QFT. QFT is not just about quantizising fields. In QFT you cannot describe gravity because of two reasons :

    1) Heisenberg Uncertainty principle
    2) superposition of wavefunctions

    This is the main reason why string theory is fundamentaly different in nature when you compare it to QFT (all known QF-theories will respect the above two concepts). String Theory is not just another field theory because it is an attempt to reformulate QFT with the incorporation of the fact that gravity does not "know" the above two notions. It is a conceptually totally different theory than all QFT-theories. That is my point.

  15. Dec 7, 2005 #14


    User Avatar
    Science Advisor

    Quantum field theory is the application of quantum machanics to a dynamical system of fields.

    But in string theory we can start from a classical theory of a string, where the space-time coordinates (seen as fields) give a map of the world-sheet (parametrised by sigma and tau) into the physical space-time. Then we quantize these fields in the usual manner, and hence achieves a 2-dimensional QFT. Why is this not a QFT?
    On the other hand, if you want an analoge to how we usually assign an operator to every physical space-time point, but now to every string instead, you have to, as I have understood it, do string field theory.

    Could you please elaborate more about 1) and 2)?
  16. Dec 7, 2005 #15


    User Avatar
    Gold Member

    The direct answer to your simple question is:

    We don't know if they exist.

    Some strong theories certainly predict them. But prediction is not the same as detection.
  17. Dec 7, 2005 #16


    User Avatar
    Science Advisor

    I dont agree with this. QFT doesn't predict 'any' particle, rather we postulate the existance of certain fields satisfying various assumptions (for instance we want our EM field to satisfy maxwells equations) and then try to write down the simplest thing we can think off under the formalism. Poof out pops electromagnetism when we require a field to be invariant under U(1) gauge symmetry. The only thing we can think off that works is a massless spin 1 force carrier, so there you go, the photon.

    Well we know Einsteins equations are right, and so we look for something that satisfies them, and poof we find the only thing that works under the QFT formalism is a massless spin 2 field. Lo and behold it outputs einsteins equations exactly.

    So again in that sense gravitons are more or less guarenteed to exist in some sense. Why? B/c special relativity, quantum mechanics and General relativity are correct and these three necessarily imply a spin 2 particle, at least in some linearized sense (which again has to exist at some level of reality).

    Now string theory on the other hand goes further. It doesn't just postulate the existance of this field to satisfy experiment (like how we got the equation for all other particles and forces) and we don't just add it in by hand, it goes one step further and says: You give me an interaction between particle a and particle b, not only do you get whatever *that* is, you also get a spin 2 particle popping out for free. It has to show up for the theory to be consistent. Thats very aesthetically pleasing, it means that gravity in some sense knows about all the other interactions a priori, and we don't have to stick it in ad hoc.
  18. Dec 7, 2005 #17


    User Avatar
    Staff Emeritus
    Science Advisor

    You are off on a wrong track here. Both gravity, and the electrostatic columb force, can "escape" a black hole (assuming the black hole is charged) - they are on equal footing.

    Note that the "force" of gravity would be due to virtual "gravitons", just as the columb force is due to virtual photons, not real photons.

    I would suggest that you read the sci.physics.faq

    http://math.ucr.edu/home/baez/physics/Relativity/BlackHoles/black_gravity.html" [Broken] for more info on this particular question.

    You have probably taken the virtual particle model a bit too seriously. Virtual particles, as the FAQ notes, are not constrained by the light cone - they also cannot be used to communicate information.

    You are also drifting off into realms of philosophy. Before you can answer whether gravitons are real, you have to be able to answer whether virtual particles in general are "real". Before you can answer that, you have to know what "real" means - are you real? How do you know that you're real? If you answer, for instance "I think, thefore I am", you are left asking if particles think, which doesn't seem to me to be particularly productive, though I gather some philophers like Whitehead think the idea is interesting.

    Possibly you aren't really interested in such philophical questions - in which case if you could come up with more specific questions (like whether or not virtual particles can carry information), we can give more definite answers. Just asking if they are "real" is both very broad, very vague, and unlikely to achieve very much except a general discussion about philosophy.

    If you are interested mainly in the philophical questions, on the other hand, there is a place on the forum for philosphical questions - it is, of course, the philosophy forum.

    My best guidance is that this is an "overly speculative post". I'm sorry that you need more tools to understand mainstream theory, but a lack of understanding isn't really a good reason to go out and invent a new theory whose only merit is that you understand it (or at least think you do).
    Last edited by a moderator: May 2, 2017
  19. Dec 7, 2005 #18


    User Avatar
    Staff Emeritus
    Science Advisor

    I generally don't think of gravity in terms of gravitons, nor do I particularly recommend it. I gather, however, that it is possible to do so in a reasonably well defined way. See for example http://arxiv.org/abs/astro-ph/0006423" [Broken]

    The "simple" approach though, requires knowledge of group theory and Lagrangian field formalisms, something that particle physicists take for granted, but would not be particularly "simple" from the POV of someone, like the original poster, who lacked calculus.

    There is one other difficulty with this theory - it is not renormalizable. This used to be a major obstacle, but I gather that this is no longer seen as a major obstacle, that one can come up with an "effective field theory" even if the underlying theory isn't renormalizable. (The resulting theory won't work everywhere, though, just at "low" energies).

    Google just found an interesting reference which probably explains this better than I can (I'm in the position of learning that renormalizability was no longer particularly important for self-consistency from remarks by other posters)

    "Introduction to the Effective Field Theory Description of Gravity"

    Last edited by a moderator: May 2, 2017
  20. Dec 7, 2005 #19
    That really explains quite a bit in layman's terms (perfect for guys like me). Thanks.

    With all due respect, I think the modern-day science, particularly physics, has taken a turn which makes it nearly impossible for layperson to understand. We all know that reality can be stranger than fiction (QM, Relativity, etc.) and unless you have the necessary tools to work out the equations, these "facts" do resemble philosophy. In this respect, I think the scientists can do a better job of relaying the information to other non-scientists in terms that they can relate to. I think Richard Feynmann did an excellent job as best he could, but even that I'm sure threw off many people. As I've already stated in my original post, I do lack the necessary tools to better grasp these concepts, but I certainly have the interest and try to keep up at "layman's" levels. As for your comment on lack of understanding not being a good reason to invent new theory/ideas, I disagree. Humans are prone to organize and categorize their understanding (however lacking) into some sort of model in their mind. All great discoveries, I'm sure, involved lack of understanding which was overcome with some imagination and an attempt to theorize what was going on. My intention is not to make new discoveries. Rather, these "theorizations" help me to better understand what's going on by my attempts to tie them together (as I'm sure the experts do). Perhaps I should be more prudent on what I post (or how I post them), but these forums are certainly great source of information.

    Thanks for all you replies.
    Last edited by a moderator: May 2, 2017
  21. Dec 7, 2005 #20
    This is not entirely accurate. Just check out the thread on "the difference between QM and QFT", where we have debated this issue recently.

    Applying QM does not equal quantizising. QM cannot be applied onto a system with infinite or non-fixed degrees of freedom because one of the basic requirements of QM is a fixed finite number of particles. This explains the difference in interpretation of "an annihilation operator" in QM and QFT.

    What exactly is that ?

    Ok, "1) and 2)" mean the two notions i summed up in my previous post (eg HUP and superposition).

    String Theory is a little more than that. What i meant with 1) and 2) is the fact that the gravitational interaction doees not know the concepts of superposition and the HUP. But these two notions are the basic ingredients of QFT, otherwise we could have omitted the letter Q in QFT. When you "quantizise" a field, this means that 1) and 2) become valid for these fields. But than again, how would you describe an interaction that does not recognize 1) and 2) in terms of fields that do. It is this manifest contradictio in terminis that determines the very foundation of string theory.

    Besides, the most basic property of QFT are the fields of which the fluctuations correspond to elementary particles. In string theory, this basic property are indeed also fields of which the fluctuations produce strings. One of the clues of string theory is how to link quantum fields and strings. The reason that strings are used comes from the theory that governs gravity : General Relativity...Also keep in mind that in QFT, elementary particles are described in a fixed space time, while in string theory the fluctuations of the fields actually express the fluctuations of space time. I would say there is a fundamental difference here

    Last edited: Dec 7, 2005
  22. Dec 7, 2005 #21


    User Avatar
    Science Advisor

    Well, they aren't my own words. They belong to Peskin and Schroeder.
    In fact QM can be seen as a quantum field theory (see Zee).

    Applying QM to fields means using the same quantization procedure as in non-relativistic quantum mechanics but now generalized to an infinite number of degrees of freedom. E.g. like the cannonical quantization is generalized to fields in Mandl and Shaw, or like the path integral formulation in Zee.

    "The classical theory of a string" is what you usually start with in a string theory course. A classical string can be described by the N-G action, or equivalently by the Polyakov action (see e.g. Green, Schwarz, Witten). What did you start your string theory courses with? (This was not sarcastic, I'm just honnestly curious.)

    How do you mean? How could any classical field know about superposition and HUP before quantization?


    You lost me here.

    I don't agree here. For example I would not call field excitations like "phonons" elementary. Remember that QFT can be used in many different areas, including solid state physics.

    Now I have to ask you again what string theory books you've used in your courses? This is not the impression I have got, or maybe I just don't understand what you mean.

    And so is the excitations of the fields (i.e. the physical spacetime coordinates) in the quantization of a string (here the "fixed spacetime" is parametrized by sigma and tau). Right?

    So you are saying a QFT needs to have fields which are functions of the physical spacetime?

    Regards /EL
  23. Dec 7, 2005 #22
    I have Zee's book and indeed what you state here is correct. However, this does not change anything because you need to interprete this correctly. What the above sentence means is this : QM can be seen as a "limited or smaller" version of QFT. Limited means (and here i go back to the actual descripancy that is also described in the Wikipedia website and Zee's book) that the number of particles are constant, more generally : the number of degrees of freedom is constant !!!

    I agree, but the interpretation of concepts like creation and annihilation operators changes. I mean the concept of "creating a particle" is not covered in QM. This is one reason why the beta decay cannot be described in terms of QM. One needs QFT to do the job properly.


    Don't worry, it is your very right to ask this question. Actually, i used a version of Gerardus t' Hooft's course on String Theory because the professor at my home university that was in charge with this subject, has a close professional relationship with t' Hooft. The course can be found on his website (or at least an updated version of it, since i took it about 2 years back:blushing: )

    This is true, but my point was this : how do you describe an interaction that does not know both the HUP and superposition in terms of a theory that does. You see ?

    Photons are force carriers ! Not matter particles. Besides, how do you define "elementary". I would say a particle is elementary when there is no "deeper" structure that you can use to describe an interaction. I mean something like :"when it is no longer possible to split it up".

    Photons are elementary force particles because of this reason. This is an uniform convention that you will find in any textbook that covers the difference between elementary field theories and EFFECTIVE field theories.

    Elementary particles are not the very base of how nature works (according to the Standard Model). It are the fields of which the fluctuations behave like particles (but also like waves:wink: ) that are the very base of it all.

    Ofcourse, you are right. What do you mean by this ?
    Sorry, i do not understand your question. Could you please clarify ?

    No no, just the opposite. QFT works independently from space time. Besides, there is NO coupling between space and time, that is a general relativity thing

  24. Dec 7, 2005 #23
    ...an appropriate expression in a discussion about the nature of gravity. :p
  25. Dec 7, 2005 #24


    User Avatar
    Science Advisor

    More precise, it is a 0+1 dimensional QFT. A QFT need not be 3+1 dimensional.

    And it didn't say anything about classical strings? I find that very strange.

    Not really. Do Maxwells equations know about HUP and sp? And yet we can formulate QED...
    No, I don't think I get your point...

    I totally agree with all this! Nice comments, but it had nothing to do with what I wrote.:wink:
    Note that I wrote phonon, not photon.:tongue2: That's why I mensioned solid state physics.
    (Goddamn, I know what a photon is...:wink: )

    Well actually it was more of a statement I just wanted you to agree on.
    I wanted to say that when quantizing a classical string, one also uses a "fixed spacetime", namely the worldsheet of the string parameterized by two parameters sigma (spacelike) and tau (timelike). In this case the "fixed spacetime" does not coinside with the physical spacetime (the physical spacetime coordinates are now considered as fields). Hence the quantization of a string gives a 1+1 dimensional QFT.

    So then you must agree on my last line?

    Sorry? What do you mean?

  26. Dec 8, 2005 #25
    This is true, but again you make the same mistake: The interactions described by the Maxwell equations govern the dynamics of the force carriers of the EM-interaction. The clue is that photons, electrons,... all can be described in terms of wavefunctions.

    I mean that you are creating a field theory for an interaction of which the particles can be described in terms of QM (hence they know 1) and 2)). In the case of gravity, you do not have that !!!

    Opps, my mistake. A phonon exists because one can quantisize the wavelike motion of lattice atoms in a crystal. One does not need field theory to describe this. Ofcourse QFT knows these "particles" because all results coming out of QM must be respected by QFT. Even, one can describe phonons in terms of field-fluctuations, but why would one do that ?

    The many body physics is described in terms of Hartree Fock or Density Funtional Theory. One does not need QFT to describe many body interactions. Ofcourse one can do this in QFT but than i do not see what your point is ?

    This is a very strange approach that i am not familiar with or at least that i do not understand. Check out the course of 't Hooft on his website to see how i was introduced to string theory. But this should not be an issue to be honest. When i say that space and time are uncoupled in both QM and QFT, than i mean that there is no space-time coordinate. You only have this in General Relativity.

    Besides, i wanna ask you how do you define a 1+1 QFT ?

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook