by Vered Arnon
In his article “Description and Identification” Bernard Harrison attempts to do two things at once. The main focus of his paper is setting forth his own revised version of description theory, which attempts to preserve the status of proper names as rigid designators by combining the Fregeian theory with the Millian theory. He claims that it is possible to have an identifying description that does not enable any sort of unique identification of the referent of the name, but merely provides assurance that a bearer exists. The description need ascribe no properties at all to the individual, aside from the label of the name itself (Harrison, 322). Then second point of his paper is to apply his theory to Kripke’s Puzzle About Belief, claiming to solve the paradox.
He develops a notion of two different kinds of identifying descriptions. The first kind, he calls “actually identifying descriptions” and the second kind he calls “referentially identifying descriptions” (Harrison, 324). Actually identifying descriptions ascribe properties to the individual, rather than to the name, and they also enable the individual that uniquely satisfies the description to be singled out. Referentially identifying descriptions do neither of these things, but merely assert that there exists an unique individual whom the name refers to.
Harrison agrees with Mill that names do not confer any properties on the things they name.
“… what a user knows in knowing a referentially identifying description is merely that the circumstances of occurrence of a certain token of a proper name, taking into account the conventions surrounding such occurrences of names, give empirical warrant for the supposition that the name of which this is a token corresponds to a unique real bearer…. We can therefore agree with Mill that a proper name, although it may be borne by many persons ‘…is not conferred upon them to indicate any qualities, or anything which belongs to them in common; and cannot be said to be affirmed of them in any sense at all,…’” (Harrison, 334).
He claims that identifying descriptions are not senses. Names are referential, he argues, and since they are mere labels that contain no properties other than the label itself, to suggest they have senses seems to him unnatural. Referentially identifying descriptions don’t enable the user to single out the exact bearer of the name, they merely assert that there exists one unique bearer (Harrison, 333).
From these theories that he lays out, Harrison then attempts to defend the claims that referential identification is fully capable, via the use of proper names, of successfully establishing reference to unique specified individuals; and also even in cases where actually identifying descriptions are available, the burden of successful reference falls to referentially identifying descriptions (Harrison, 324). Proper names are rigid designators as a result of referential identification, because “the individual is identified purely through the circumstances of occurrence of its name” (Harrison, 332).
He proceeds to give lengthy examples of how a person can refer to a specific individual merely from the basis of a token of the name, without having any direct aquaintance with the individual bearer. “Causality underpins reference: convention determines it,” he says (Harrison, 336). The way in which a person becomes acquainted with a token of a name then sets up a causal effect by which the reference is determined. By logical convention, proper names pick out one individual. The use of names in their occurrences is determined by numerous socio-linguistic conventions, and facts about the world enable these conventions to operate, thus building the causal structure. But a causal theory detaches the reference of a name from the speaker’s intentions and beliefs, because to use the name, a person need have no direct acquaintance with the bearer of the name, but merely have encountered the conventional use of the name which is instantiated by the token. In order to prevent this detachment, Harrison proposes that a weak version of description theory be combined with the theories of reference he has been talking about (Harrison, 332).
To solve the problem in Kripke’s puzzle about Pierre’s apparently conflicting beliefs, Harrison claims that “disquotation gives us the content, but not the object, of a belief” (Harrison, 338). He claims that Kripke’s use of the disquotational principle fails to distinguish between Pierre’s direct and indirect beliefs. According to Harrison, the thing that the belief is about is the object. The properties attributed by belief are the content. If one doesn’t have direct acquaintance with an object, then one’s beliefs are merely about the name and its reference, supposedly, rather than the bearer of the name. He makes a distinction between conditions for reference to a single individual, and conditions for beliefs about a single individual.
“Because reference is a merely logical relationship it is not nececcary, in order for a speaker to refer to a specific real individal, that he somehow be able – by means of a description, or through the existence of a causally connected series of uses – to reach out and touch the individual in question. Belief, on the other hand is an epistemic relation, and hence belief about an individual does require actual epistemic contact with the individual in question” (Harrison, 337).
Here Harrison becomes confused and contradictory. He has claimed up until now that direct acquaintance is not necessary for successful use of a name to refer to a singular individual. In his first example, he demonstrates that merely from acquaintance with a token of a name, a person can develop beliefs and successfully refer to the singular bearer of the name by using the name to assert those beliefs (Harrison, 327). His examples about Marcellus and Feynman (Harrison, 329 – 332) directly contradict his example about Chimborazo (Harrison, 337). His dichotomy of reference to an individual versus beliefs about an individual is flawed, since beliefs about an individual include reference to that individual, and reference to an individual implies belief. According to his dichotomy, the belief implied by a reference to an individual is no more than a belief that some individual instantiates the properties indicated by the content of the belief. But the object of a reference is in fact the bearer of the name, so there really isn’t the dramatic dichotomy he attempts to present.
Harrison justifies his claim about the necessity of epistemic contact in order for a person to have a belief about a real individual, because one must be able to single out the individual whom the belief is attached to (Harrison, 337). He gives his example about the mountain Chimborazo, and says unless he can single out that actual mountain, he would be forced into an “absurd position of saying that [he holds] a belief about a specific mountain without knowing which specific mountain it is that this belief is supposed to be about” (Harrison, 337). But this position is hardly absurd, since Harrison’s claim would prevent people from holding many beliefs at all. To offer a counterexample: A freshman at UW-Stevens Point has never seen Rik Warsch. She has overheard the use of the name while sitting at the bus station. The sparse snippets of conversation she has overheard make it clear that the name Rik Warsch refers to an individual with certain attributes. The student could not, under any circumstances, single out the individual who bears the name Rik Warsch. Yet the student believes that Rik Warsch is the president of Lawrence University, she believes that Rik Warsch is married, she believes that Rik Warsch spends a lot of money remodelling his house, and so on. The student has numerous beliefs about the individual, Rik Warsch. What would be absurd would be to say that her beliefs are merely about the reference of the name. While Harrison is correct in saying that in order to refer to an individual, no direct descriptions or epistemic contact is necessary. But once the reference is made, the reference itself becomes a description. Every day, in many different situations, individuals entertain beliefs about objects which they have no means of singling out. They have beliefs about specific objects, and knowing which physical objects these beliefs are about is not necessary, since incomplete knowledge does not compromise belief. The UW student believes of Rik Warsch that he is the president of Lawrence. She doesn’t merely believe that the name Rik Warsch refers to an individual who is the president of Lawrence. Harrison’s claims make the belief process shallow and rigid. There isn’t pronounced asymmetry between the conditions for belief and the conditions for reference, as he would like to claim. If it were so, then the majority of people’s beliefs would consist of beliefs about references of objects, and people wouldn’t be able to entertain beliefs about many actual objects at all. When a person uses a name to refer to an object, they are in that instance also holding a belief about that object.
To go back to Harrison’s mountain example, he claims “I am not in a position to entertain any belief about the actual mass of stone and vegetation called ‘Chimborazo’” (Harrison, 337). He is imposing limitations on himself that are not at all necessary or reasonable. He knows that Chimborazo is a mountain. He knows that it is tall. He knows that it exists somewhere out there. His belief that it is tall is a belief that the mountain itself is tall, not merely a belief that the name refers to some mountain which is tall.
The reference of the name is the bearer of the name, as explained above. The claim that “disquotation gives us the content, but not the object, of a belief” is false, since there is clearly an object of Pierre’s belief about the pretty city. The object is the city which he believes is pretty. And he calls that city Londres. There is no distinction between direct and indirect beliefs that needs to be made. Harrison writes, “It is only when…. we ask whether, in believing [‘Chimborazo is a tall mountain’] is believing something about Chimborazo, or something about the reference of ‘Chimborazo’, that obstacles begin to appear in the way of a smooth passage to Kripke’s paradox” (Harrison, 338). But the obstacles in the way of this smooth passage are false constructs, since there is no reason to bring up the difference between a belief about Chimborazo and a belief about “the reference of Chimborazo”. When using Chimborazo as a proper noun in a sentence to express a belief, the name refers to Chimborazo, and the object of the belief expressed is Chimborazo.
According to Harrison, when Pierre assents to the claim Londres est jolie, he asserts that the word “Londres” names a pretty city. Rather than infering “Pierre believes that London is a pretty city” from the French statement, Harrison would infer “Pierre believes that ‘Londres’ names a pretty city”. When Pierre assents to the claim London is ugly, he asserts that London is not pretty, from which is inferred “Pierre believes that London is not pretty”. Pierre’s beliefs no longer seem contradictory because Harrison claims that one is direct and one is indirect. Thus Harrison has made Pierre’s beliefs consistent, and would appear to have solved Kripke’s puzzle.
While he thinks he has solved the puzzle, his premises are faulty. Kripke uses the disquotational principle to derive contradictory beliefs. Harrison attempts to reapply the disquotational theory, and with the use of his own description theory, derive beliefs that are not contradictory.
1. Londres est jolie.
2. London is not pretty.
3. Pierre believes that “Londres” names a pretty city. (from 1)
4. Pierre believes that London is not pretty. (from 2)
5. #3 and #4 are consistent because in #3 he has an indirect belief about the reference of an object, while in #4 he has a direct belief about an object.
But when Harrison derives #3 from #1, rather than accepting Kripke’s derivation of “Pierre believes that London is pretty”, he is misusing the disquotational principle himself. His main assumption that he rests his whole argument on, that “disquotation gives us the content, but not the object, of a belief” is unsound. According to the disquotational principle, “If S assents to ‘p’, then S believes that p.” Pierre assents to Londres est jolie, therefore Pierre believes that Londres est jolie. Nowhere does the disquotational principle imply that from Pierre’s assent to #1 should it be inferred that Pierre believes of Londres that it names a reference which has the attribute est jolie. Harrison’s own use of the disquotational principle fails, and so his argument against Kripke cannot stand.
Kripke’s puzzle is a puzzle, it’s a paradox, and merely coming up with a new description theory fails to solve it. Disquotation does give us the object of a belief, despite what Harrison says, since the structure of the sentence hasn’t been changed, and the name, which refers to the object, hasn’t been removed. Disquotation would fail to supply the object of the belief only if it failed to supply the word which refers to the object.
Harrison, Bernard. “Description and Identification”, Mind (1982) Vol. XCI, 321 – 338.