I think that's a rather 'meta' example. A better one would be that if someone can reply to your tweets or mention your username, they can make it appear that y

I think that's a rather 'meta' example. A better one would be that if someone can reply to your tweets or mention your username, they can make it appear that you're (rudely) refusing to engage them in legitimate conversation. Or they can retweet you, followed by a series of their own tweets ridiculing you. I'm sure there are more creative ideas than this. One might argue that any harm done here is outweighed by the fact that the target of the harassment won't be able to see any of it, and I find that argument reasonably persuasive. It seems that Twitter's decision has been taken in order to make blocking a 'silent' or 'invisible' action, such that the person being blocked cannot know that they're blocked. Their calculation seems to be that harm is reduced this way, because it introduces an asymmetry of information in favour of the person doing the blocking - they know who they have blocked, but the person they've blocked does not. To call that 'capitulation' is not an accurate reading of the situation, in my view.