Reference

4. Historical–causal theories

Kripke and Donnellan followed their criticism of description theories of names with an alternative view. This became known as the ‘causal’ or ‘historical’ theory, although Kripke and Donnellan regarded their view as more of a ‘picture’ than a theory (see Proper names §4).

The basic idea of this theory is that a name designates whatever is causally linked to it in an appropriate way, a way that does not require speakers to associate an identifying description of the bearer with the name. Reference is initially fixed at a dubbing, either by perception or description of the referent. The name is then passed on from person to person in communicative exchanges. People succeed in designating an object with a name because underlying their uses of the name are causal chains stretching back to the dubbing of the object with the name. People borrow their reference from people earlier in the chain but borrowers do not have to remember lenders; it is enough that borrowers are, as a matter of historical fact, appropriately linked to their lenders in communication. So people can designate Cicero despite their ignorance of him, or designate Einstein despite their errors about him.

Similarly, Kripke and Putnam proposed an historical–causal theory of natural-kind words. Reference is initially fixed at a dubbing, either by description or perception of samples of the kind. The reference is then to all those objects, or all that stuff, having an internal structure of the same sort as the samples: for example, in the case of gold having the atomic number 79. People at a dubbing lend their reference to others, who can then lend it to still others. People who are ignorant about the kind can use the word to refer to the kind’s members because underlying the use of the word are causal chains stretching back to a dubbing.

In thus removing the epistemic burden on speakers, historical–causal theories are a radical departure from the Frege–Russell tradition. That tradition assumes that those who understand a name must know about its meaning and reference, so that if its reference is determined in a certain way, they must know that it is. The historical–causal theory must reject the assumption that speakers have this privileged ‘Cartesian’ access to semantic facts: the reference of a name is determined by causal chains that are likely to be beyond the ken of the ordinary speaker. Once again, we must conclude with Putnam that ‘meanings just ain’t in the head’. This very feature of the theory has led many to reject it and to work hard to preserve the description theory (or the more general ‘identification theory’ favoured by Dummett; see §3). From the traditional Cartesian perspective, the causal theory’s failure to impose an epistemic burden rules it out as a candidate to explain reference.

The historical–causal theory nicely captures the rigidity of names: the reference of a name is determined by its actual causal relations, something that cannot change when we consider other possible worlds. Less pleasingly, by rejecting any descriptive element to the meaning of a name, the theory may seem to leave no alternative but to resurrect the Millian view, identifying a name’s meaning with its role of designating its bearer. Many philosophers, influenced by the ‘direct reference’ approach to indexicals (see §5), have taken this route, despite the problems for the Millian view (see §1). To avoid this it seems that we must explain a name’s meaning in terms of the particular sort of causal chain that determines the name’s reference.

The theory faces problems arising from various confusions and mistakes that can play a role in forming the causal network underlying a name. And it must explain how the reference of a name can change even though the historical fact of the dubbing cannot change. In developing the theory to deal with these problems, Michael Devitt (1981) has emphasized that a name is typically ‘grounded’ in its bearer in many perceptual confrontations after the initial dubbing: it is ‘multiply’ grounded in its bearer.

The historical–causal theory is returned to in §8.