If I fall into a black hole, can I affect the event horizon after I pass through it? I'm holding two very massive, dense objects, one in each hand at arms length. I fall through the event horizon of a very large black hole. The tidal forces at the event horizon are not large, so I survive.
Now inside the event horizon, I bring the masses that I hold together. This should affect the shape of the event horizon. Using this method, I will communicate information from inside the event horizon, to an observer outside the event horizon, who is closely monitoring the geometry of the event horizon.
This seems natural, and also seems to contradict the idea that information cannot be communicated outwards across an event horizon. Is the description right, and information can be communicated outwards from within an event horizon? Or is the description wrong? In which case, what have I missed?
Thanks for taking a look.
 A: The description is wrong. After you fall past the event horizon no part of the horizon is in your future light cone. You can no more affect the horizon than you can affect yesterday, and for the same reason.
No matter how you wiggle those masses outside of the horizon you cannot send a signal to yesterday, therefore no matter how you wiggle those masses inside the horizon you cannot send a signal to the horizon. It is easiest to see this using Kruskal–Szekeres coordinates where light cones form standard 45 degree lines.
EDIT: from some of the comments below there is a bit of confusion about the relationship between the horizon and the singularity. The singularity is not in the past of any portion of the horizon, so the singularity does not cause the horizon. Also, for technical reasons the singularity is not part of the spacetime manifold.
The standard Schwarzschild black hole is a vacuum spacetime, meaning it does not contain mass anywhere. A more realistic spacetime is the Oppenheimer Snyder spacetime which models the gravitational collapse of a spherically symmetric cloud of dust. This is not a vacuum solution and has mass, but note that in this spacetime the event horizon forms prior to the formation of the singularity. So again it is incorrect to think that the horizon is caused by the singularity or that gravitational waves emitted inside the horizon could reach the horizon.
A: From the point of view of a distant observer, you never reach the event horizon. Gravitational time dilation becomes extreme. You slow down as you approach the event horizon, and essentially become frozen in time. Even after an infinite time in the distant observer's coordinates, you still haven't reached the event horizon.
Thus the observer never sees effects of what you do after passing through the event horizon.
