The standard explanation for light "knowing" the angle of diffraction is that actually light just propagates in every direction and then constructive interference is stronger for paths near the shortest path because its length is more consistent when the path is perturbed (meaning the phases of the perturbed paths tend to agree more so they add up instead of cancelling). I don't think you even need quantum mechanics for this; it occurs in classical wave optics.

You can see Feynman explaining mirrors this way in recorded lectures [1]. There's also a recent Veritaseum video explaining why the shortest paths dominate [2].

1: https://youtu.be/SsMYBWpsQu0?si=o1eAEvESwjroTke3&t=2251

2: https://www.youtube.com/watch?v=Q10_srZ-pbs

▲

justonceokay a day ago | parent | next [-]

Thanks for sharing this. I was vaguely aware of this explanation – I believe I watched a Richard Feynman video where he describes this.

The funny thing is that while the standard quantum explanation for this behavior is different than the situation I described in my post, it’s somehow even less intuitive than believing that the light “knew“ where to go before it left it’s source. Instead we are supposed to assume that it checked all possible directions and instantaneously made a (“correct”) decision.

I am fully aware that my sentences above are full of anthropomorphisms and are hopelessly classical. But in my classically oriented mind, my first way of describing the situation and you’re more correct way both involve a violation of what I would consider “causality”.

▲

jimbosis 4 days ago | parent | prev | next [-]

There's another Veritasium video that has a neat experiment ostensibly showing light (both lamp and laser sources) "taking all paths" (or words to that effect--I don't really know what I'm seeing or what I'm talking about!) It starts around 25 minutes in.

https://www.youtube.com/watch?v=qJZ1Ez28C-A

EDIT: Whoops. The YouTube video linked by naasking in this comment: https://news.ycombinator.com/item?id=44771713 touches on the Veritasum video I linked to and goes to some length to explain that it is NOT proving the light is taking "all possible paths." He also brings up and links to a video on the "Looking Glass Universe" channel in which the hostess recreates the Vertiasium experiment and gives a differing interpretation. (Some commenters there have objections to the experimental setup. Oh boy, I may be down a rabbit hole here.)

▲

tsimionescu 5 days ago | parent | prev | next [-]

That only works with quantum mechanics - it's a consequence of the "path integral" idea of QM. In classical optics this wouldn't work, because you'd be able to detect light on the other paths if it really did take all paths.

▲

Strilanc 5 days ago | parent | next [-]

I think you're confusing the distinction between classical ray optics and classical wave optics with the distinction between classical wave optics and quantum mechanics. Quantum mechanics and classical wave optics agree on the explanation for diffraction as a path interference effect. In classical optics, the reason you don't see light coming from angles away from the shortest path is because of destructive interference between the other paths.

For example, note that the Huygens principle predates quantum mechanics by over 200 years [1]. As another example, diffraction gratings (which manifestly require interference between different paths) were being made in the mid 1800s [2] but in physics documentaries you never hear of people being confused about how to explain their behavior. Because they are explained by classical wave optics. Also see this lecture which talks about diffraction in the context of ray optics [3].

Where wave optics disagrees from quantum mechanics is in the dim-light limit, when you start resolving individual photons.

[1]: https://en.wikipedia.org/wiki/Huygens%E2%80%93Fresnel_princi...

[2]: https://en.wikipedia.org/wiki/Diffraction_grating

[3]: https://www.youtube.com/watch?v=5tKPLfZ9JVQ&list=PLB1A0BF14E...

▲

pdonis 5 days ago | parent | prev [-]

Classical optics is just the limiting case of quantum optics when the path length is much longer than the wavelength. In such a case quantum optics predicts basically zero probability to detect light on any path other than the classical path--which is classical optics. So classical optics doesn't say anything that's actually contradictory to quantum optics. It's just a special case.

▲

Strilanc 5 days ago | parent [-]

It's classical ray optics that fails in the path-not-longer-than-wavelength regime. Classical wave optics works in that regime. Where classical techniques fail is at low brightness (because you start resolving individual photons).

	▲	pdonis 5 days ago \| parent [-]
		Yes, but in classical wave optics you're no longer talking about "paths" as they appear in path integrals. Classical wave optics is basically quantum wave optics without the discreteness of detections, i.e., interpreting the wave as a straightforward EM field intensity instead of as a probability amplitude for detecting a photon.

▲

naasking 5 days ago | parent | prev | next [-]

I don't think this "standard explanation" is as standard as it is sometimes portrayed:

https://youtu.be/XcY3ZtgYis0?si=9TyD5-7B00WTLzOH

▲

bookofjoe 5 days ago | parent | prev [-]

>I think I can safely say that nobody understands quantum mechanics.

— Richard Feynman

https://youtu.be/w3ZRLllWgHI?si=bX77FX6BLVnfCuq_