Bilal asked:

I have a question about Gareth Evans' example of the name 'Turnip'. What two major views are
illustrated in this example and which one seems to be the stronger position?

============

I'll come to 'Turnip' in a minute.

I was lucky to attend Gareth Evans' seminars on Reference in the Summer of 1977 while I was a
graduate student at University College Oxford. The seminars were widely thought of at the time to be
as important as anything that was going on in the philosophy of language — on this side of the
Atlantic. John McDowell, Evans' close collaborator at 'Univ', was my graduate studies supervisor.

I spoke up a lot during those seminars. At the last meeting I approached Evans to ask if he would be
interested in looking at some of my work. He said he would try to fit me in, but the meeting never
materialized.

Some time later — it could have been that summer, I'm not exactly sure of the time scale — I heard a
horrifying story of an incident in Mexico, where Evans survived a bungled kidnap attempt by bandits
on his friend Hugo Margan, the son of the US Ambassador. Both men were shot in the leg. Margan
bled to death. Evans nearly died, enduring a desperate taxi ride from one hospital to another, being
refused admission because the doctors would not treat gunshot victims.

By a cruel twist of fate, Evans died in Oxford not long after the Mexico incident, from lung cancer.

So much for personal reminiscences.

The two theories of how proper names refer are known as the 'cluster theory' and the 'causal theory'.
You might wonder why on earth we need a theory of how names refer to objects. In fact, the two
theories represent sharply divided views on the metaphysical question of how our thoughts relate to
reality, as I will explain.

A baby is born, and given the name 'Turnip' by its loving parents. (This is my version of Evans' story
about Turnip, with a few added details.) Turnip grows up to be quite a character, and many stories
are told about his escapades. As the centuries pass, however, some stories are mistakenly
associated with the name 'Turnip' when in fact they were about someone else entirely. Let's say,
Turnip fought at the Battle of the Boyne. That's true. But Turnip was not, as many falsely believe, the
anonymous author of the bawdy novel, The King's Mare. The true author was Swede, not Turnip.

In recent times, the errors have become so compounded that most of the beliefs associated with the
name 'Turnip' are in fact true of Swede. But what exactly does that mean? When someone says,
'Turnip wrote The King's Mare', to whom are they referring? Is that statement a false statement about
Turnip, or is it a true statement about Swede?

According to the cluster theory, what determines the reference of a proper name is the cluster of
descriptions we associate with it. The object to which the name refers is the object about which a
sufficient majority of the descriptions are true. In that case, 'Turnip' just means'the author of The
King's Mare'.

According to the causal theory, on the other hand, what determines the reference of a proper name is
the initial act of 'baptism' when the object was first given the name, and the causal chain of speakers
who each pick up the name from the previous speaker in the chain, with the intention of referring to
the object to which the name was originally given. In that case, 'Turnip' still refers to the individual
who was originally given the name 'Turnip' even if allthe things we have subsequently come to
believe about 'Turnip' are in fact true of Swede and not Turnip.

The causal theory was first proposed by Saul Kripke in his paper 'Naming and Necessity' (1972).
Gareth Evans wrote a paper, 'The Causal Theory of Names' giving his version of the causal theory.
Up until Kripke, the cluster theory was widely, or possibly universally believed to give the correct
account of the semantics of proper names.

The metaphysical significance of the causal theory is that it rejects an idea which we find quite
plausible when we first think about the reference of a name: that whatever we are talking about
depends upon our knowledge of whom or what we meanto refer to. The causal theory says that's
wrong. The true significance of our thoughts depends not on the way the world looks to us, from
inside our minds, but on an externalview which includes information which we do not possess, or
which at least is not accessible to conscious reflection.

The metaphysical point about the external view is an important one. But it is not sufficient to vindicate
the causal theory. In fact, it seems quite clear to me that both theories are false. In manycases there
is no correct answer to the question, 'Is this a true belief about X or a false belief about Y?' The
answer is simply indeterminate. Don't bother asking the speaker, because they can't tell you, and
no-one else can tell you either. In short, there can't be a philosophical 'theory of proper names' of the
kind that proponents of the causal and cluster theories supposed. Human linguistic intentions are
messy and complex, and refuse to align themselves to any precise theory.

In his Oxford seminars, Evans was quite critical of his original formulation of the causal theory,
pursuing some very interesting lines of inquiry later described in his book, Reference(John McDowell
Ed. OUP), published posthumously. Evans combined the 'external' idea with much tougher line on the
question under what conditions a person reallyunderstands a name. The (sometimes tenuous)
causal 'chain of communication', Evans thought, was merely a reflection of the way we use language
without always grasping the meaning of what we are saying — what Hilary Putnam calls the 'division
of linguistic labour'.

Geoffrey Klempner