Kripke's Intentional Turn in Proper Name Reference — Epoche B2
Chains and Intentions: What Holds Kripke's Picture of Reference Together Saul Kripke's Naming and Necessity (1980 [1] ) is usually summarised as replacing a theory of names based on descriptions with one based on causal history. The summary is right about what was displaced and misleading about what replaced it. The account Kripke offers is not purely causal: it contains, as an explicit condition, an intention on the part of each speaker to use a name with the same reference as the person they got it from. This essay works up to that condition from examples, argues that it is doing real work rather than decorating a causal story, and then argues that it cannot be promoted into the leading role. The bolder conclusion — that intention rather than history is what fixes reference — collapses back into the view Kripke was attacking. The defensible claim is narrower: intention makes the chain a chain of reference, without determining what the chain refers to. A first example: 'Aristotle' Begin with a question that looks trivial. When I use the name 'Aristotle', what makes it the case that I am talking about a particular man who died in 322 BC? The classical answer is that a name works by abbreviating a description. Gottlob Frege [2] , in 'On Sense and Reference' (1892), suggested in a footnote that different speakers might attach different senses to 'Aristotle' — one thinking of the pupil of Plato and teacher of Alexander the Great, another of the man born in Stagira. Bertrand Russell developed the idea into a general thesis [3] , holding in Introduction to Mathematical Philosophy (1919) that ordinary proper names are abbreviations for descriptions, and that a sentence containing one is really a sentence about whatever uniquely satisfies the description. The theory explains how a name can be informative, how two names for one thing can differ in cognitive value, and how a name can fail to refer. It also faces an obvious difficulty: speakers associate different descriptions with the same name, and many associate none that is uniquely identifying. John Searle's 'Proper Names' (1958 [4] ) proposed a repair: a name is associated not with one description but with a cluster, and refers to whatever satisfies enough of it. Three arguments against descriptivism Kripke's lectures press three distinct objections. It is worth separating them, because they have different targets and different strengths. The first is modal. Kripke introduces a term of art: something is a rigid designator if in every possible world it designates the same object. He argues that proper names are rigid and ordinary descriptions are not. Aristotle might have died in infancy and taught nobody; in that circumstance he would still have been Aristotle, but 'the teacher of Alexander the Great' would have picked out someone else or no one. So the name and the description differ in what they pick out under counterfactual suppositions, and a name cannot mean a description. The second is epistemic. Ordinary speakers can refer with names while knowing almost nothing. Someone who can say only that Richard Feynman was a physicist still refers to Feynman, though the description does not single him out; and Kripke's famous case of Gödel is stronger still. Suppose the only thing most people associate with 'Gödel' is 'the man who proved the incompleteness of arithmetic', and suppose the proof was in fact stolen from a man named Schmidt. Descriptivism entails that when they say 'Gödel', these speakers refer to Schmidt. That is the wrong result: they refer to Gödel, and say something false about him. The third is semantic. A description may fix a reference without giving a meaning. Kripke's example is the standard metre bar: one may stipulate that 'one metre' is to name the length of that bar, fixing what the term picks out, without making 'one metre' mean 'the length of that bar'. This lets him concede that descriptions play a role in the life of a name while denying that they constitute its meaning. The picture Kripke offers instead In place of descriptivism Kripke sketches what he is careful to call a picture rather than a theory. A name is introduced by an initial baptism: an object is present or is fixed by a description, and a name is attached. Thereafter the name is passed from speaker to speaker, and later users refer to the original object without having to be able to identify it. The crucial passage states the transmission condition: When the name is 'passed from link to link,' the receiver of the name must, I think, intend when he learns it to use it with the same reference as the man from whom he heard it. The condition is not an afterthought, and Kripke supplies his own illustration of what it excludes. Someone who hears the name 'Napoleon' and decides to give it to his pet aardvark is at the end of a perfectly good causal chain leading back to the emperor, and refers to the aardvark. The causal connection is intact; what is missing is the intention to go on referring to what his source referred to. So a bare causal chain is not sufficient, and Kripke never said it was. Where this tempts an overstatement It is tempting to conclude that intention is doing the fundamental work and the causal chain is merely the conduit. On this reading, the baptism is an act of intentional conferral, each subsequent use renews the intention, and the history is just the record of those acts. The conclusion goes too far, and the reason is instructive. If reference were settled by the speaker's intention, we would have to ask what that intention is about. It cannot be an intention to refer to whatever satisfies a description, since that is descriptivism. It cannot be a bare intention to refer to a particular object, because the problem is precisely to explain how a speaker who cannot identify the object manages to talk about it. The only remaining option is the one Kripke states: an intention to refer to whatever one's source referred to. That intention i