What about scatter from the satellite itself? The laser can't be perfectly absorbed on the receiving end. I haven't done the calculations, but I fully expect the spot size is much larger than the receiver.

That said, I imagine that for any serious bandwidth, you won't get enough SNR to decode anything from the ground.

Remember that the only reason a laser works for this kind of link is that it is very highly collimated. The light scattering from the receiving satellite is diffuse so the intensity will go down very quickly with distance. It might be observable with a suitably large telescope, but I doubt it would be practical.

So you have a reflection off random bits of equipment followed by a trip through most of earths atmosphere? That doesn't sound easy at all - and they could choose wavelengths that won't make the trip well if at all.

I guess if you got an observation platform up closer that would help.

More like farther away. Put a spy satellite in a high orbit like traditional satellite internet and it'll be able to see half of Starlink simultaneously with essentially no distortion.

If they were strongly encrypted there may be no reason for a nation state to observe the intersatellite links, but I wouldn't expect this at least for the near future since Elon's focus will probably be to cut corners to get the absolute best performance possible.

If it is encrypted and you drop just a couple bits due to noise you can't read it.

There are multiple layers of encoding to mitigate this. CCSDS [0] is very widely used for satellite communications, and a Google image search for CCSDS protocols [1] will give you a quick view of the protocol stack (which can vary a lot as different CCSDS protocols are used in different applications).

[0] https://en.wikipedia.org/wiki/Consultative_Committee_for_Spa...

[1] https://www.google.com/search?q=ccsds+protocols&source=lnms&...

Edited to add: I'm not making any claims on what protocols Starlink uses specifically. Just describing the general approach.