Reading: Searle 'Proper Names' (Chap 21 in Martinich The Philosophy of Language)
Kripke 'Naming and Necessity' (Chap 22 in Martinich The Philosophy of Language)
Russell and Kripke on names in Grayling Philosophy
The clearest exposition of a cluster theory about proper names is found in Searle (although he himself does not term it as such), so I will first discuss his approach to the problem of how we define proper names. Then I will look at Kripke's criticisms of such an approach and assess his claim that proper names are 'rigid designators'. I will argue that, whilst Kripke raises a number of important objections to theories which make use of descriptions in their analysis of proper names, it is not clear that these objections are strong enough to force us to abandon any such analysis. As Kripke himself acknowledges, he does not wish to spell out an alternative set of necessary and sufficient conditions by which to determine to reference of a proper name, but rather to give an alternative way of looking at the problem. He says that,
"It's in virtue of our connection with other speakers in the community, going back to the referent himself, that we refer to a certain man." (P288) However, it would seem that Kripke's theory needs to be clarified a bit further in order to make it more persuasive.
Searle, in his paper entitled 'Proper Names', discusses Frege's notion that proper names have sense and reference. He argues that it seems as though a name must have sense as well as reference in order to correlate it to the object it refers to. We use a name to refer by picking out an object and explaining that this word is the name of that object. But Searle argues that we can only identify the object by description. It can be argued against this that the description of an object may change, but the reference of the name remains the same. Searle responds by saying that this is not a convincing argument because, if all our knowledge about Aristotle turned out to be true of no one, under the theory that names have reference and no sense, 'Aristotle' does not denote an object, and nothing more. However, it seems intuitive to claim that if someone says that Aristotle does not exist, the do not just mean that the name 'Aristotle' does not refer to anyone. Therefore it looks as though proper names have sense necessarily but only have reference contingently (ie. they only have a reference if exactly one object satisfies their sense). Searle argues that, if this view is correct, we must be able to determine the sense of every name. So what is sense? If it is simply a description of the object then the meaning of the name would change with every change in the object, and the name would also have different meanings for different people. Searle tries to determine what the necessary and sufficient conditions are for applying a particular name to a particular object. He arrives at the conclusion that the necessary and sufficient conditions of 'This is Aristotle' are not identity with any object names 'Aristotle', but rather identity with Aristotle.
Searle distinguishes proper names from definite descriptions and demonstratives in the following way: Although all these refer to particular objects, demonstratives are contextually dependent (that is to say that there are contextual conditions surrounding their utterance), whereas proper names are not. Names also differ from definite descriptions in as much as the former specify some of the characteristics of the objects to which they refer, whereas the latter do not. So it remains to be asked what the relation is between the proper name and the object it refers to. The crux of Searle's argument is that using a proper name to refer to an object presupposes that the object has certain characteristics. He states that:
"If we could agree in advance what characteristics constituted the identity of Aristotle, our rules for using the name would be precise" (P276)
However, if this was the case then we would only be able to refer to an object by describing it, which is precisely what proper names help us to avoid doing. He argues that:
"The looseness of the criteria for proper names is a necessary condition for isolating the referring function from the describing function of language" (P276)
So the answer to the question of whether or not a proper name has a sense rests upon what exactly we mean by sense. They do have a sense if this means that proper names are logically connected with characteristics of the objects they refer to, but they lack a sense if this means that proper names are used to describe the characteristics of objects.
Kripke argues that cluster theories about proper names, such as Searle's are misguided because they do not correctly take into account the problem of necessity. He claims that all theories about proper names which at some point make use of the idea of descriptions are flawed and he gives two objections to them. The first is that users may fail to associate a name with any description and the second is his Gödel/Schmidt example, but I will discuss these in detail a little later on. Kripke instead argues that proper names are rigid designators, because it is a contingent fact that, say, Aristotle did any of the things attributed to him, that is to say we can imagine a possible world in which it was not that case that Aristotle taught Alexandra the Great and so on. I will firstly examine Kripke's criticisms of cluster theories in more detail and then elucidate his theory that names are rigid designators.
We are faced with a problem if we claim that a name has one definite description to pick out the referent, as this would lead to the undesirable conclusion that certain statements are necessary truths, when it is obvious that they are not (For example if the name 'Aristotle' was characterised by the definite description 'The philosopher who taught Alexander the Great', the 'Aristotle taught Alexander the Great' becomes a necessary truth). A response to this that many have made is to put forward some form of cluster theory, which holds that we really associate a 'family of descriptions' with a name. There are two ways of regarding this theory, the first being that the cluster of descriptions gives the meaning of the name; and the second being that the cluster gives the reference of the name. Kripke argues that no contingent facts can be the sense of a name and it is simply false that 'Moses' exists means 'The man who did X, Y, and Z exists'. Searle would argue that it is a necessary fact that Aristotle has the sum of all the properties commonly attributed to him. But Kripke states that this is not true and it is just a contingent fact that Aristotle did any of the things attributed to him.
Kripke introduces the concept of necessity in order to explain what he means when he states that proper names are rigid designators. We have to ask, if it is true could it have been otherwise? If it could not then it is a necessary fact, but if it could have been otherwise then it is a contingent fact. A rigid designator is something that designates the same object in every possible world. (This is not to say that it must exist in every possible world, but if it does then it can be termed 'strongly rigid') For example, Kripke claims that it is not the case that Nixon might not have been Nixon (although he might not have been called Nixon). I find this point a bit confusing as I do not see how an object in a possible world could be Nixon if he has none of the same descriptions and is not even called Nixon.
In the second lecture, Kripke posits a number of theses which would make up a cluster theory of names, and then he proceeds to show that they are flawed. A cluster theory would hold that one or some of the properties are believed by A to pick out some individual uniquely. Kripke gives a counter example to this claim by saying that people can use a name without associating it with any description. He gives the example of someone who uses the name 'Feynman' only knowing that he is 'a physicist or something'. According to Kripke, he is still using the name 'Feynman' as a name for Feynman. But it can be objected that he does have a description in mind, because he is referring to the man called Feynman. So I do not think that this criticism is very strong.
Another point that Kripke objects to is the claim that, 'If most of the j's are satisfied by one unique object, y, then y is the referent of 'X'. He uses the counter-example of Gödel and Schmidt and refutes the claim that when we talk about Gödel we are actually always referring to Schmidt because it was he who in fact discovered the incompleteness of maths. He tries to show that this is not the case by showing that when people talk about Columbus as the man who discovered America, it does not matter if it transpires that a Norseman actually discovered America, we would still be referring to Columbus and not to the Norseman. This is quite a persuasive objection, but I would argue that a distinction must be made between what is meant and what is actually said (like with Donellan's objections to Russell's theory of definite descriptions).
To conclude I would argue that the objections to cluster theories that Kripke makes may give a cluster theorist cause to refine their account, but are not strong enough as they stand to provide a conclusive dismissal of all such theories altogether in favour of Kripke's picture of causal chains which provide a reference.
© Anne Witton 1996. No part of this article may be copied without my permission.
