Space-time curvature VS light diffraction

Thanks to YouTube suggestions I saw some videos.

And I noticed some discrepancy. Since light propagates as waves, the experiment with the hidden star can be explained by light diffraction. I've tried to google that issue with no luck. So the question is:
Is the effect of the hidden star can be explained by light diffraction?
And what about Space-time curvature then? is it a fake?

Side note:
I'm neither a mathematician nor physicist. So I do not make statements. But from what I intuitively feel, what Einstein did was the ingenious math trick. He replaced an unknown force, gravity, by some mathematical expression, so that a result of calculation stays correct. And by introducing the concept of space-time curvature he didn't explain the phenomena of the gravity itself. He just swept the issue under the rug. Even if a mass causes space to bend, what force causes the mass to bend space? So we came to where we started. Although this concept is incredible, it's mind-boggling as well. Since human imagination is based on the vision and the vision is three dimensional, hardly anyone can clearly operate this four-dimensional model. So I presume it could be a workaround Einstein devised to avoid gravity :)

Here's a reason why diffraction can't explain what we see which doesn't rely on complicated arguments about the nature of diffraction, followed by some notes on what diffraction effects would look like.

Deflection of light by the Sun

First of all let's take a famous example of something General Relativity predicts, which is that light will be deflected by the Sun. A famous early experiment was done in 1919 by Eddington and this experiment has been repeated several times since.

But this experiment was done when the Sun was eclipsed by the Moon. So, if we suppose the effect is due to diffraction it is diffraction by the Moon and not by the Sun. But there are many times when the Moon is new (ie dark) but not directly in front of the Sun: it should be perfectly easy to observe this supposed diffraction effect then, as well. But it's not: light is not significantly deflected by the Moon: there may be diffraction effects but they look completely different. So it can't be a diffraction effect.

More generally, observationally, if you want to know the correct amount of deflection it turns out you need to know the mass of the object that is doing the deflecting not just how big it is and what shape it is, which tells us that it's a gravitational effect, and not diffraction.

What diffraction effects would be like

One of the comments mentioned this document which describes what diffraction effects from the Moon might look like. There are two interesting things about this:

Firstly, diffraction effects are wavelength-dependent. This means that different wavelengths will be diffracted by different amounts at the edge of the shadow. So if you are observing a star, for instance, you will see its light smeared out based on wavelength: you'll see a spectrum, in fact. Well, again, when observing gravitational deflection effects we don't see this: all the light follows the same set of null geodesics and the star is not smeared out. So again, we can tell that this is not a diffraction effect: it's something different.

Secondly diffraction effects provide very characteristic bands of light and shadow (see the bands computed in the document above): this is not what is observed with gravitational effects at all, and provides another way of distinguishing them.

Thirdly diffraction effects depend on the edge being rather smooth. I have not thought this out hard enough but I suspect that in the case of diffraction by the Moon the edge of the Moon needs to be smooth to well under $14\,\mathrm{m}$ which is the characteristic length computed in the document mentioned above. Well, it's probably not that smooth (there are hills, cratars &c), so any diffraction effects would be much reduced by this. Again, that's not the case for gravitational light deflection, and this would provide yet another way of distinguishing things.

I'll also add that General Relativity makes a fairly large number of predictions, many of which have been tested, including the direct detection of gravitational waves. Any theory which replaces it has to agree with all of the predictions of GR which have been tested.

• The moon does diffract light. See Moon occultations or read this: spiff.rit.edu/richmond/occult/bessel/bessel.html – Bill Alsept Aug 22 '18 at 15:03
• As I understand, every object diffracts light by definition of the light. I.e. if light propagates as a wave, everything that stays on the way of the wave causes its diffraction. From what you've said the GR is correct. But that doesn't prove the impossibility of diffraction. – Markus Marvell Aug 22 '18 at 15:44
• @BillAlsept: thank you. I've added some notes based on that document, which makes it really clear that the two effects are nothing like each other: no wavelength-dependence, no bands &c. – tfb Aug 22 '18 at 16:36
• @tfb How can you be positive the two effects are not alike. Relativity gives no description or physical derivation of what causes mass to bend space. Photon Diffraction is not that clear either. – Bill Alsept Aug 22 '18 at 17:59
• @BillAlsept: There is one mathematical model -- GR -- which predicts what we see, and one -- diffraction -- which does not. So I rule out diffraction and start looking for other tests of GR. If there is some alternative theory of diffraction which predicts all of the tests GR has passed so far, feel free to publish it. – tfb Aug 23 '18 at 7:40