Hacker News Reader

Comment by nimish | Hacker News Reader

Not really. The "roiling sea of particles" is a metaphor for perturbation series, which very specifically does not work for the strong force at low energies.

Take anything involving virtual particles as just that, virtual. They're an aid for computation and cannot be observed directly. They aren't necessary either; lattice gauge theory is always applicable if not practical.

The mass(-energy) being from the strong interactions is still true. And the residual bit of the strong force between protons and neutrons works with the virtual particle/perturbation theory approach pretty well, using pions.

View on HN

I'm glad you brought this up, because I've been reading some stuff lately that makes it seem like virtual particles actually have visible effects, like this one:

https://www.forbes.com/sites/startswithabang/2019/07/12/yes-...

So is it that these articles are wrong, or that I'm reading them wrong, or that the idea that virtual particles are just for calculations is outdated?

We discovered the mesons by photographing their tracks on plates exposed to the sky. They're real, just extremely short-lived.

What a bad piece of journalism, intent on confusing instead of explaining.

Virtual particles were invented because they have measurable effects. Physicists don't go around inventing invisible things for no reason. What they are not is "particles". The particle facade is only there because it fits the math.

(The article seems to be describing an experiment that measured energy-time uncertainty.)

Incomplete analogy that's probably a better starting point: if the tone of a bell is a particle, then the other movements of the bell that don't resonate are virtual particles. Now, imagine we called the thermal motions of that bell “virtual tones”, and you have an idea why physicists always sigh and emphasize that virtual. particles. are. not. particles.

If virtual particles "fit the math" the same as real particles, what makes them not real? What do they not fit?

What makes particles travel long distances as lumps is that they are on-shell. Virtual particles are not on shell.

https://en.wikipedia.org/wiki/On_shell_and_off_shell

You don't _need_ them, and IIRC _can't_ (in general) use them if you deal with non-perturbative effects, unless you do stuff like re-summing infinite series of diagrams and that really makes interpreting the ultimately observable effect in terms of virtual particles interacting difficult.

An example that comes up in semiconductor is the 'hole' virtual particle. It is the absence of an electron. It is not a real and independent phenomenon. But you can treat it like a particle just fine

Electron holes are quasi particles, not virtual particles.

Ah hell, guess I have to start treating the stuff I learned in undergrad into the unreliable narrator category. Thanks for the correction!

Virtual particles are not particles.

Virtual particles are all the disturbances in a field that don't behave like particles do.

It's really a terrible name to have entered the lay vocabulary: “virtual memory is something that behaves largely like real memory” is _exactly_ wrong, it's more like “virtual memory is all the circuitry that doesn't perform any memory function, but is still made of silicon”.

The more correct idea is that particles aren't fundamental. There are fields, and a disturbance in one field will cause a disturbance in the fields that the first couples with.

Most chains of disturbances die out really fast (“virtual particles”), but some combinations will resonate for a significant amount of time before they die out (“unstable particles”) and others don't die out at all (“stable particles”).

And if the lifetime of a disturbance is fairly short compared to the frequency of the disturbance itself, it becomes hard to even make a solid distinction between those types, but again, that's a flaw caused by imposing a categorization scheme (“particles”) based on something that isn't fundamental.

My impression is that the notion of "virtual particles" is a bit of an outdated concept. There isn't really as much of a physical distinction between "real" and "virtual" particles as there is sometimes made out to be. All particles are excitations of some underlying field. Generally speaking, these excitations can have some resonance that allows them to persist for long durations. This resonance occurs when the particle is "on the mass shell" in the jargon --- that is, it has a rest mass that is equal (or at least extremely close) to the observed mass of the particle. Excitations that are "off the mass shell" decay exponentially. But these other excitations do have real observable effects. The Casimir effect is the most famous, but they're also responsible for the Lamb shift and Hawking radiation which have also been observed.

Hawking radiation has been observed? When did that happen?

It's been observed in sonic black holes, which are mathematical analogs of gravitational black holes in fluids (though to be fair the experiment has been disputed, so the evidence is not absolutely unambiguous):

https://www.nature.com/articles/nphys3863.epdf

Thanks for the clarification.

I don't have the background to be confident about this, but aren't the predicates on which Hawking Radiation is based on part of the equivalency framework between sonic and "real" black holes?

If so, then while the observation of Hawking radiation in the model is certainly interesting, calling it an observation of Hawking radiation with regards to real black holes sounds like a stretch.

I agree that observations from analog black holes don't rise to the level of definitive proof of the existence of Hawking radiation around gravitational black holes. But what has been observed demonstrates that a system whose behavior is described by the equations that we believe to be correct will exhibit Hawking radiation.

No, it's a pretty well-defined concept when you stick to its technical definition as an aid in interpreting terms of a perturbation series as Feynman diagrams.

> these other excitations do have real observable effects

Yeah, that's the major thing: virtual particles explain observable effects in a sort-of intuitive way.

But you could (to my knowledge) get the exact same results without involving any virtual particles, via lattice gauge theory. Since you get the same observable results without them, virtual particles, IMO, shouldn't be considered fundamental to any effect, even if they make the explanation a lot easier.

Anything involving complicated interactions with relativity like Hawking-Unruh stuff has an even bigger issue since the notion of a particle/vacuum is observer dependent.

All post-classical effects in quantum field theory arise from various "corrections" that come from what you could define as virtual particles, whether it's from the perturbative treatment or a numerical lattice gauge calculation. In the latter, you need to (formally) sum over all possible field configurations, the majority of which will contain half-witted weird off-shell "particles" and the entire spectrum between stuff you'd never see as a particle and the classical resonances.

It is more clearly visualized in a perturbative expansion, for sure, but it's a bit disingenous I think to argue that there are no virtual particles in a lattice calculation.

Yeah, the 'Standard model of particle physics' actually states that particles don't exist and there are only fields around.

Reply on news.ycombinator.com