Meanings as referents/denotations
One way to break the definitional circle would be to stress the role of the referent or denotation as the main component of the meaning of a linguistic expression. Under this theory, metalanguage explanations of a meaning should be seen as names of the referents of the object language term. As we saw in Section 1.2.1, ordinary discourse about language in English often seems to make an implicit identification between an expression’s meaning and its referent:
The ‘meaning’ of ‘bridge’, the speaker of (38) seems to be suggesting, is the actual harbour bridge itself. ‘Bridge’, we might say, means what it refers to; its meaning on any one occasion of use is its referent. Outside of the narrow context of (38), we could say that the meaning of bridge in general is just its denotation – the class of all bridges. This identification of meaning and referent/denotation succeeds in breaking the circle because it identifies meaning with non-linguistic objects in the world: the meaning of ‘bridge’ on a particular instance of use is the real bolts and metal structure. Given this interpretation of the meaning of ‘bridge’, it doesn’t matter that we would eventually run out of new words with which to defi ne it, since we can ultimately analyse its meaning ostensively, i.e. simply by pointing at the actual bridge itself.
As was pointed out in the discussion of the semiotic triangle (1.2.1), the referents of expressions must be taken not as actual objects in the world, but as representations in the world as projected by the speaker. This means that in order to understand reference we already have to invoke the realm of speakers’ individual psychologies, the particular ‘versions’ of the world as projected by their psychology. The postulation of the world of projected representations allows us to avoid an objection which might otherwise count against referential theories of meaning. This is the objection that it is often the case that there simply is no referent for a given expression, as in (39a–c), or that the referent is unknown, as in (39d–f):
A theory which identified meanings with real world referents would have to say that the expressions in (39a–c) simply have no meaning, since the things they refer to never actually existed, or are impossible; and it would have to say that the meaning (referent) of the expressions in (39d–f) was unknown, since although we can be confident that all of the things referred to by the expressions exist, we do not know what they are. But if referents are taken to be representations projected within the realm of people’s psychology rather than real objects in the actual world, this problem disappears. Whether or not there is any object referred to by the words Robin Hood’s private helicopter, we can easily think of situations in which a speaker might simply imagine, pretend or otherwise entertain the possibility that such a helicopter did exist. For the speaker of (39b), then, the referent of Robin Hood’s private helicopter can be taken as the speaker’s representation of the helicopter in their projected world. The reader will easily see that similar explanations can be constructed for the other examples in (39).
The identification between meaning and reference may be successful in breaking the definitional circle, but it leads to a very fragmented picture of the nature of language: on the reference theory of meaning, ‘bridge’ has as many different meanings as it has different referents. This variety clashes with our pretheoretical intuition that the meaning of bridge is actually something much more unitary: although there are many different individual bridges out there in the world, the meaning of the word bridge, or, we might say, the concept of a bridge is a unified, single entity.
The idea that an expression’s meaning is its referent is at least easy to understand for nouns referring to discrete, concrete things. But it is much less clear what the referents of other lexical categories might be. What are the referents of abstract nouns like scandal, generosity or impermanence? Since there is no isolable object in the world to which these nouns apply, the notion of a referent is rather hard to invoke. And what about adjectives like sweet, polished or ineffectual, or verbs like to have, to allow or to go? In the case of ‘grammatical’ words the problem is even greater: what is the denotation of of, or of the? These cases all pose problems for the referential theory of meaning: because the words have no referents/denotations, they are left without any specifiable meaning. Yet it is obviously the case that these words do have meanings, which we can paraphrase metalinguistically and explain to others. We will consider this question further in Chapter 6.
A second problem with the theory of meaning as reference is the fact that a single referent may often be referred to by a variety of different expressions. Thus, the expressions in the two halves of (40a–d) each pick out just a single individual:
In (40a) we have alternative ways of referring to Sweden, in (40b) of umbrellas, in (40c) of the Spanish confectionery king Enric Bernat Fontlladosa, and in (40d) of the word bank. Yet we surely do not want to say that the meanings of these expressions are the same. While the objects referred to by the expressions ‘institution for lending money’ and ‘institution for depositing money’ have the same denotation – banks – they clearly don’t have the same sense. We could imagine a bank which suddenly stopped lending money even though it continued to accept deposits: something like this, indeed, happened during the Argentinian financial crisis of 2002 and the global one of 2008. If meaning simply is reference/denotation, then examples like this should not be possible. The fact that linguistic expressions can be identical in reference but different in meaning leaves us no choice but to conclude that there is more to meaning than reference/ denotation.
