Natural Order and Science

[Alex K]

Most likely we will cross path again on this forum. I'll leave it up to you if you want to restart afresh or continue this hostility.

Sure, no problem. I'm not the vindictive type.

Anyway, about your definition of virtual particles
Now that I know that you know this stuff well on a technical level, I can word my earlier question/remark more precisely.
Basically all the QFT lit I know treats the concept of "virtuality" of stable particles as the distance of the four momentum from the mass shell. I don't recall encountering a distinction between "real" and "virtual" particles based solely on whether they interact with an detector (which to me smells like mixing interpretations of QM, especially the arbitrary definition in copenhagen where the classical detector starts, and categories in QFT - something I am not used to), and if you could give me a reference i.e. to a textbook that does that, I would be grateful because it would be a new approach to the concept to me. I should have most of the known ones, but would be grateful if I didn't have to leaf through all of them searching for the right bits.

Don't you agree that even objects that in usual parlance would be called "real photons" - because they are emitted, have pretty much lightlike momentum and are registered at macroscopic distances from the source - can be described as a propagator between e.g. an electron in the source and an electron in the detector? (I think even Weinberg does that when talking about cluster decomposition, but I might confuse it with another book). This propagator interpreted as a Fourier transform to position space will contain momentum space propagators with all manner of off-shell four-momenta. Why do you still get a perfectly onshell photon in the S-Matrix element even if no photon ever registered is *purely and exactly* onshell? Precisely because we do a somewhat "unphysical" approximation by going to asymptotic times, right?

This lack of a sharp distinction and the somewhat artificial nature of the "real" in and out states was what I was trying to get at in order to dispel Harris' idea that there is this fundamental philosophical difference between real and virtual particles where one type really exists and the other is just a figment of theorists' imaginations.

Do you disagree?

p.s.
we are not even talking about loops here, which would open a whole new can of worms if we were to map the related physics to simple binary language like "real" and "virtual". But also here I think that the need to cancel soft real emission photons with loop corrections in Bloch-Nordsieck underlines the lack of a sharp philosophical ontological distinction between real and virtual particles. We are also not talking about unstable particles where the difference between on- and offshell is even muddier.

You should have mentioned the Ward identities, but I'll forgive you.

I have mentioned gauge invariance and Gupta Bleuler conditions, that totes counts as a substitute for Ward identities! I have implicitly even mentioned Slavnov Taylor