§79

Is Wittgenstein presenting a 'cluster-concept' theory of names here? It doesn't seem to me to follow from this passage that Wittgenstein need even commit himself to the view that names have a meaning. Isn't the question of whether names have a meaning independent of the considerations here? I might well claim that no names, including the name 'Moses', mean anything at all but still indulge in a discussion about which descriptions would have to be true in order for me to be happy to say that Moses existed.
I don't think that every claim made about Moses is true. I don't think that anyone has been handed stone tablets from God, for example. But that doesn't mean I'm unwilling to contemplate the possibility that Moses existed. It may well be that Moses existed but that many of the things said about Moses are untrue.
I can refer to someone successfully even without saying anything true about them (I might get my embassies/nations mixed up and say "Julian Assange is that guy who is living in the Bolivian embassy" - and somebody might still know who it is I'm talking about). Is that of relevance here?
Wittgenstein suggests that in the case of Moses I might be willing to say that Moses exists if many of the things said of him are true - without even committing myself to which of the things are true. If one thing turns out not to be true I might 'lean on' another description (or set of descriptions).
Wittgenstein himself does connect up the considerations about which definite descriptions would have to be true with the meaning of the name ("I use the name 'N' without a fixed meaning").
Wittgenstein also suggests that a claim might move from being empirical/reporting a contingent relationship to being a definition (which seems right to me).