Reference

1. Millian and description theories of proper names

The description theory of proper names stands in sharp contrast to the age-old and attractive theory that there is no more to a name’s meaning than its role of designating something. Thus, John Stuart Mill claimed that ‘proper names are not connotative: they denote the individuals who are called by them; but they do not indicate or imply any attributes as belonging to those individuals’ ([1843] 1867: 20) (see Mill, J.S. §2). One problem for this theory is that it makes answering our central question seem hard: if a name does not ‘imply any attributes’ of its bearer, as the description theory claims it does, what determines which object is the bearer?

It was not primarily concern with this question, however, that led most philosophers to abandon the Millian theory and adopt the description theory. They did this largely as a result of the criticisms and counter-proposals of Frege (1893) and Russell (1911).

The most famous criticism of the Millian theory concerns identity statements. The ancient Greeks observed what they took to be a star rising in the evening and called it ‘Hesperus’, and what they took to be another star rising in the morning and called it ‘Phosphorus’. In fact these ‘two stars’ were the planet Venus. Consider the following two statements:

Sentence (1) is true. A comparison of (1) and (2) seems to indicate that they differ sharply in meaning. Various reasons have been adduced in favour of this view. Some philosophers have argued that whereas (1) is synthetic, (2) is analytic; others that whereas (1) is known empirically, (2) is known a priori; still others that whereas (1) is contingent, (2) is necessary. However, probably the most influential reason for thinking that the statements differ in meaning has been Frege’s claim that they differ in ‘cognitive value’: (1) is highly informative, revealing an important astronomical discovery, whereas (2) is uninformative, a trivial piece of logical knowledge (see Frege, G. §3). In any case, it was generally agreed that the two statements do differ in meaning. If so, the only way to explain this seemed to be to attribute different meanings to ‘Hesperus’ and ‘Phosphorus’. Yet, according to the Millian theory, they must have the same meaning as they both have the role of designating Venus. So the Millian theory must be wrong: a name’s role of designating its bearer does not exhaust its meaning.

Another important criticism concerns existence statements. ‘Vulcan does not exist’ is true. Because it is true, ‘Vulcan’ does not designate anything. So, according to the Millian theory, ‘Vulcan’ should be meaningless. So, ‘Vulcan does not exist’ should be partly meaningless. Yet, it is perfectly meaningful. Indeed, if it were not, it could not be true. So, once again, the Millian theory must be wrong (see Fiction, semantics of).

The description theory provides neat solutions to what are problems for the Millian theory. According to the description theory, a name in effect abbreviates the definite description that competent speakers associate with the name: thus ‘Aristotle’ might abbreviate ‘the last great philosopher of antiquity’. So it is easy to see why (1) and (2) differ in meaning: ‘Hesperus’ might be associated with the description ‘the star that rises in the evening’, and ‘Phosphorus’ with ‘the star that rises in the morning’. And ‘Vulcan does not exist’ is fully meaningful because ‘Vulcan’ abbreviates the meaningful description ‘the planet in orbit between Mercury and the Sun’ (see Proper names §§1, 5).

More importantly for our purposes, the description theory provides an answer to our central question. A name designates a certain object because that object is denoted by the definite description associated with the name; ‘Aristotle’ designates Aristotle because ‘the last great philosopher of antiquity’ denotes him. Of course, this answer raises another question (see §6): in virtue of what does the description denote that object? Still, progress has clearly been made because we had that problem anyway.

Some obvious problems with this ‘classical’ description theory – including the problem that speakers differ in the descriptions that they associate with a name – led some philosophers, notably John Searle (1958) and Peter Strawson (1959), to modify the theory. A name is not tied tightly to one description but loosely to many. It can designate its bearer despite the failure of some in its ‘cluster’ of associated descriptions to denote that object: it designates whatever object most of the descriptions in the cluster denote.