There are two misunderstandings that I would like to help clarify:
- We do observe objects in our universe (like the M87 black hole), but in reality these objects are not fully formed black holes (the event horizon never forms in a finite time as seen from Earth). Why are they black? Because of something called infinite redshift. Photons that come from the object towards us, are redshifted to infinity so that we cannot detect them. Hence, the object looks black.
The light from that matter is ludicrously redshifted. The redshift doubles in a time comparable to the light-crossing time of the hole, which for the M87 hole is around 1 day, so after a year it's roughly 1 googol. Therefore, unless the hole ate a star quite recently, you can't actually see whatever the ray hits, and it'll appear perfectly black (plus some orange for the glowing semitransparent matter that the ray also passed through).
If an event horizon never forms for an outside observer, then what do (or don't) we see in the middle area on this real image of an actual black hole?
- When an object approaches the black hole, the reason we do not see it anymore is again, infinite redshift. Now what is very important to understand, is that when an object approaches this black hole (which is not fully formed as seen from Earth), it does not need to cross the horizon (or more precisely, we do not need to see it cross) to add to the black hole's mass. The gravitational field of the black hole extends outside the (not even fully formed) horizon. When the object (as seen from our view) asymptotically approaches the horizon, it already adds to the strength of the gravitational field of the black hole. By the no-hair theorem, and the definition (there are a few, but in this case they all agree on this) of the mass of the black hole, the black hole's mass stems from the gravitational field. The gravitational field's strength itself defines the mass of the black hole, now including the object that seems to asymptotically approach the horizon (then disappears because of redshift).
This isn't some accounting trick; it means we will never see an event horizon form. At this point someone will usually pop up and say that means black holes don't really exist. In a sense that's true in our coordinate system, but all that means is that our coordinate system does not provide a complete description of the universe.
How can anything ever fall into a black hole as seen from an outside observer?
So the answer to your question is that yes, we do observe from Earth black holes, and we can observe them growing as they accumulate more matter. But this is not a contradiction because we observe black holes that are not fully formed (in our view), and the objects that seem to asymptotically approach them add to the gravitational field of the black hole thus increasing the (from our view not fully formed) event horizon.