Thread: Feynman QED lectures and Relativity

Hello. I am new to this forum, and certainly no expert in quantum physics. I am here to try to learn new things, if possible.

The thing that really struck me after viewing all four of the Feynman lectures was how blatantly it all seems to violate Special Relativity. The Feynman diagrams seem to utterly disregard whether interacting particles are within each other's light cone. Apparently, an electron can be affected by a photon it hasn't met yet, at least in some reference frames, but has met in others. Is this correct?

Maxwell's Equations are relativity-friendly, being preserved under a Lorentz transformation. I would have hoped that the broader theory of QED would be similarly preserved, but maybe I'm wrong.

In the first two lectures, Feynman attributed wavelength as an attribute of the photon, but in the middle of the third lecture he stated this was not really true. There is only one kind of photon, with only the distribution of amplitudes that can vary. This is approximately the point where I got totally lost. Is there a simple explanation of what he meant?

Any comments anyone can offer would be appreciated.

Originally Posted by geo_man

Hello. I am new to this forum, and certainly no expert in quantum physics. I am here to try to learn new things, if possible.

The thing that really struck me after viewing all four of the Feynman lectures was how blatantly it all seems to violate Special Relativity. The Feynman diagrams seem to utterly disregard whether interacting particles are within each other's light cone. Apparently, an electron can be affected by a photon it hasn't met yet, at least in some reference frames, but has met in others. Is this correct?

Maxwell's Equations are relativity-friendly, being preserved under a Lorentz transformation. I would have hoped that the broader theory of QED would be similarly preserved, but maybe I'm wrong.

In the first two lectures, Feynman attributed wavelength as an attribute of the photon, but in the middle of the third lecture he stated this was not really true. There is only one kind of photon, with only the distribution of amplitudes that can vary. This is approximately the point where I got totally lost. Is there a simple explanation of what he meant?

Any comments anyone can offer would be appreciated.

QED does not violate special relativity.

In fact the whole point of QED is to have a quantum theory that is consistent with special relativity. That is really what the subject is about.

Maxwell's equations are indeed compatible with special relativity, and in fact the constancy of the speed of light in inertial reference frames can be viewed as a consequence of Maxwell's equations. However, Maxwell's equations are determnistic and therefore not compatible with a the stochastic nature of quantum mechanics. In order to have a theory that encompasses both what is known of electrodynamics and of quantum mechanics it is necessary to go to quantum field theories, and QED is the model for quantum field theories.

Thank you for this reply. I would imagine that Maxwell's Equations can therefore be derived for large-scale phenomena from QED principles. Is this correct?

You didn't address the issue of non-locality of the Feynman diagrams. With time on the vertical axis and space horizontal, he showed many particle interactions that were horizontal; i.e. at the the same time but across different positions in space. Classical special relativity, of course, would say that such interactions are impossible, since the influence of one particle would have to travel infinitely fast to be felt at the same time by the other. Such "action at a distance" paradoxes were what Einstein and others raised in their so-called EPR experiments.

Perhaps what you really mean is that QED obeys special relativity only for large-scale macroscopic effects, and not at the individual particle level. Is that correct?

The reason I bring this up is that I just finished reading the provocative book by Huw Price, entitled "Time's Arrow." In this book, Mr. Price offers non-causality of interactions as a more palatible interpretation of quantum physics than non-locality. Instead of allowing interactions to occur simultaneously across space, why not instead allow them to occur at the same place but across time. This interpretation of quantum physics has the following advantages:

1) Special relativity is not violated at the individual particle level and
2) Time symmetry of the Shroedinger equation is preserved, meaning that particles are equally sensitive to events in their future as they are to events in their past.

There are no experiments which can favor non-locality over non-causality or vice-versa, according to Price. They are simply different ways of interpreting the results of experiments. Does anyone care to comment on this deep subject?

Well, let just say that I don't know much of the subject. But is it not possible that non-locality can be explained by a space-like dimension bridge between the particles? Is there a reason why two particles can't be separated in three spacial dimensions, but connected in a fourth or more spatial dimensions? I am a bit uncomfortable with an interpretation of time as anything more than a measurement of relative movement (even if it makes mathematical sense).

Originally Posted by geo_man

Thank you for this reply. I would imagine that Maxwell's Equations can therefore be derived for large-scale phenomena from QED principles. Is this correct?

You didn't address the issue of non-locality of the Feynman diagrams. With time on the vertical axis and space horizontal, he showed many particle interactions that were horizontal; i.e. at the the same time but across different positions in space. Classical special relativity, of course, would say that such interactions are impossible, since the influence of one particle would have to travel infinitely fast to be felt at the same time by the other. Such "action at a distance" paradoxes were what Einstein and others raised in their so-called EPR experiments.

Perhaps what you really mean is that QED obeys special relativity only for large-scale macroscopic effects, and not at the individual particle level. Is that correct?

The reason I bring this up is that I just finished reading the provocative book by Huw Price, entitled "Time's Arrow." In this book, Mr. Price offers non-causality of interactions as a more palatible interpretation of quantum physics than non-locality. Instead of allowing interactions to occur simultaneously across space, why not instead allow them to occur at the same place but across time. This interpretation of quantum physics has the following advantages:

1) Special relativity is not violated at the individual particle level and
2) Time symmetry of the Shroedinger equation is preserved, meaning that particles are equally sensitive to events in their future as they are to events in their past.

There are no experiments which can favor non-locality over non-causality or vice-versa, according to Price. They are simply different ways of interpreting the results of experiments. Does anyone care to comment on this deep subject?

QED is concistent with special relativity, but it is not a classical theory, and special relativity is.

Not everything works classically at quantum scales. For instance the speed of light is not really constant, but on average photons travel at c.

The Schrodinger equation does not apply to quantum field theories. One does not have a relativistic wave equation.

It appears that you are trying to interpolate between elementary quantum mechanics Inon-relativistic) and spedcial relativity. Quantum field theory is a bit more complex than that.

I personally recommend that you learn your physics from physicists and avoid philosophers, like Price.

Regarding Kalster's comment: A photon can be thought of as a high frequency oscillation modulated by a bell-shaped envelope. Because of the envelope, the spectrum of the photon is spread across frequencies, while the photon itself is spread across space. If you design an experiment that measures the position of the photon with great accuracy, you narrow its envelope and spread its spectrum over more frequencies. On the other hand, an experiment which measures its frequency very accurately will broaden the envelope and cause its position to be more uncertain.

In the same manner, at a given instant of time, the position of the photon is always uncertain, dependent on the nature of the experiment. But the time it takes to arrive at a given position in space is also uncertain, given the nature of the experiment. The leading edge of the photon will reach another particle it is about to interact with before the main body of the photon does. The greater the coherence between the photon and the particle, the farther into the future the influence will extend. This is an intuitive explanation of how particles can be viewed as being sensitive to events in their future.

Regarding Dr. Rocket's latest reply: Yes, I guess I was confusing classical quantum mechanics with quantum field theory, since I know only a little of the former and nothing of the latter.

If Shrodinger's equation does not apply apply to QED, then there must be something analagous to replace it. Feynman made it very clear that the principle that the squared magnitude of a field amplitude represents the probability of an event is "universal." Maybe the field doesn't obey a strict wave equation, but it must be wave-like, correct?