Searle: The Cluster Theory

The point of Searle's arguments on proper names is important, and it's not entirely clear that Kripke keeps it firmly in mind when presenting his version of the so-called ``cluster theory.''

Like definite descriptions, names refer to single objects, but they do so ``without presupposing any stage settings or any special contextual conditions surrounding the utterance of the expression...they do not, in general, specify any characteristics at all of the objects to which they refer'' (170).

But, Searle goes on to say that ``their referring uses nonetheless presuppose that the object to which they purport to refer has certain characteristics'' (170-1). We don't specify what particular characteristics these are, if we did then names would lose their usefulness and their ability to allow us to communicate with one another despite not sharing every experience, &c. Names work precisely to allow us to refer to objects without describing them. If names had meanings/senses as precise, fixed sets of characteristics, they would cease to have this function and be nothing more than descriptions. In Searle's words, ``They function not as descriptions, but as pegs on which to hang descriptions'' (172). This means that what may seem to some to amount to imprecision is in fact an essential feature of the role names play in language: ``Thus the looseness of the criteria for proper names is a necessary condition for isolating the referring function from the describing function of language'' (172).

As we'll see, it's not at all clear that this is what Kripke presents as the ``cluster theory of names.'' Kripke says, for example, that names correspond to a cluster of properties (1), that the referent of the name simply is whatever satisfies (enough of) the cluster (3) that any statement to the effect that the referent (assuming one exists) satisfies (enough) of these properties is known a priori by anyone who uses the name (5), and that any statement to this effect is a necessary truth (in the relevant idiolect) (6). NONE of these, with the possible exception of (1), under some suitable interpretation, is entailed by Searle's account, and the interpretation required to make (1) a part of Searle's account turns on what is meant by ``corresponds to.''

For Searle this only means that within the community of competent speakers we can identify a number of features which turn up with some regularity in reports of speakers' beliefs; for Kripke, it means that there is some disjunctive definition attached to the name; these are quite different and often at odds with one another.

Now, Searle does say the following.

...though I am suggesting it is a necessary fact that Aristotle has the logical sum, inclusive disjunction, of properties commonly attributed to him: any individual not having at least some of these properties could not be Aristotle (172).

What does he mean by this? Isn't this what Kripke attributes to him? I think the answer to the first question shows that it is not what Kripke attributes to him.

We use `Aristotle' as a name for, let us say, the greatest philosopher of late antiquity who was a student of Plato. This looks like two features, but in fact it brings many features into play: Aristotle is human, a philosopher, lived before ...& after ..., and so on. Later we learn that while Plato had a ``pupil'' called `Aristotle' who lived during the relevant period, it was a dog, not a man, not a philosopher. This dog does indeed have some of the features in the disjunction ($\phi$1 v $\phi$2 v ...v $\phi$n). So the disjunction is true of him, but he's not Aristotle, we agree to that. Why? Well, because he doesn't have enough of them (though this could be false, he might have almost all of them), or he doesn't have the REALLY important ones, or .... All this seems fine, but what are we talking about here: language or the world? Are we talking about how the world is or about how we (linguistic community) have decided to use this (or any) name? Searle believes it's the latter, and part of what this means is that we don't discover that a dog can't be Aristotle, but that we have decided that we are using the name ``Aristotle'' as the name of a man, not of a dog AND it is this which we signal via the ``cluster'' of properties we associate to that name. Even if Plato's dog satisfies all of the properties save that one, he isn't ARISTOTLE, not because being human is part of Aristotle's essence, but just because we have positioned it as the name of a human via our associated descriptions.

I want to come back to this and talk about essentialism, a priori/a posteriori and necessary/contingent in more detail. But for now, let's continue a bit with Kripke's view of the cluster theory of names. I don't think it's Searle's theory, but it might be someone's.