Wittgenstein introduces family resemblances to help us understand how some concepts actually work, how they function in language. Take the classic example of a game. What is a game? How do we decide if this is or isn't a game? Why is this a game but that not? And so on. In short, how do we define ``game''? Solitaire is a game, so is basketball, chess, bingo, poker, pick-up-sticks, Parcheesi, .... If these are all games, we want to say that they MUST have something in common, something in virtue of which they are games. But what would that be?
66. ...--For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don't think, but look!
What emerges is the picture of overlapping sets of features which come together to form an interlinked array.
By the time we get to Game 4, none of the properties in Game 1 are even there--they share no properties, yet they are both games. When confronted with some new activity--Activity 5--we decide whether it's a game not by asking if it has this or that property, but by seeing if it might fit into such an array. We don't find out whether it's a game by looking at the definition of ``game,'' but by seeing where it fits into this ``family'' of activities we call ``games.''
What, then, about ``Moses''? The motivating idea in Wittgenstein is that on any specific occasion of use, the speaker and hearer attach some sense to the name ``Moses,'' connect it with some set of features or properties, but that this sense can vary from speaker/hearer to speaker/hearer and from time to time for any given speaker or hearer.
We're talking about Moses and I say, ``Well, Moses really caused a stir when he drove the money changers from the Temple.'' Puzzled as to whether I meant Moses or Jesus, you ask me--''Who's this Moses you're talking about? What else did he do?'' If, on the one hand, I say something like this: ``Oh, he led the Israelites out of Egypt, was found by Pharaoh's daughter floating in a reed basket in the Nile when he was a baby, was raised at court, parted the Red Sea, never actually returned to Israel himself, and refused to sacrifice his son when God asked him to do so.'' Even if not everything I say is true of Moses, you will take me to be talking about MOSES, and may well go on to tell me that it wasn't Moses who threw the money changers out of the Temple. On the other hand, if I say: ``You know, that guy who's the only son of God, the one who died so that we might be saved. Fed the masses with the loaves and fishes, and parted the Red Sea so they could all get across to see Lazarus rise from the dead,'' you're most likely to say: ``Oh, you mean `Jesus'.'' (Once again you may correct some of my mistakes, but you'll have gotten hold of the person I am actually talking about.)
Let's say it's the first. One of my ``props'' for identifying Moses was the money changer story, losing it doesn't affect my ability to talk about and identify Moses any more than does learning that in addition to chess, checkers, Parcheesi and Monopoly, basketball and poker are games. Before I used the presence of boards as a mark of games; now I see that this doesn't always work.
Taken in this way, the cluster theory is not about the meaning of a name--in the way that ``family resemblance'' explains the ``meaning'' of ``game.'' Instead, it has become a theory--and this is Wittgenstein's point--about the shared nature of language, something common to a language/linguistic community. How can we talk if we don't agree on everything about the words we use? Wittgenstein's answer is: Like this .... We rely on arrays rather than necessary and sufficient conditions, analogies, resemblances, practices and examples.
Searle takes this a step farther: to every name, we/language/the community associate some ``cluster'' of properties gathered from among our understanding and understandings; some of these are true of the referent, some of which aren't--after all, some of us have some true beliefs at some time or another, and some of us have false beliefs. (This, by the way, accounts for Strawson's remark that ``democracy rules.'' If we are all competent speakers, there is no reason to toss out features relied upon by each of us as potential reference fixing features in the sense Donnellan uses in connection with the referential use of names. This doesn't entail that in terms of finding the referent all of them are equally useful, let alone all true of whatever the referent turns out to be.) No one of these properties--we'll ignore essentialism for the moment--must be true of the referent, but ``enough'' of them must. If we talked about someone named ``Finkelstein'' and we all thought we knew this, that and the other about him, but it turned out that no one ever had any of these features, we might well say that we (the community!) were mistaken, there never was a Finkelstein, it's a community myth or whatever.
If we now look at a statement of the form ``N is '' we see that whether this is trivial or informative depends upon the context in which it is used. ``Wittgenstein is a philosopher'': trivial here; significant at your grandmother's bridge club. It's not the ``definition'' of ``N'' which results in triviality, but which of the features in the cluster the audience associates to the name.
Just as Kripke decides to index a priori to specific knowers, it seems that triviality too is indexed (or tied) to specific knowers/listeners/speakers. If I associate ``teacher of Alexander'' to Aristotle as the feature in terms of which I identify him, if someone tells me that Aristotle taught Alexander and I'm feeling nasty, I may well say ``Tell me something I don't know,'' signaling that this is trivial and adds nothing to my knowledge base.