Sure, but there is no 'pruning of the tree' in case a dead end is reached, so the citations stay allowing the quack to pretend they have more credibility than they do. In fact, the whole idea of these citations is to build credibility where there is none.
Right. The purpose of citation metrics is to measure the intellectual influence of a person's ideas. If an academic's works are never being cited by anyone else, the correct conclusion would be that their ideas have little influence -- regardless of how many self-citing papers they've published.
